Mechanical Engineering MSc Martin Edin Grimheden KTH Royal Institute of Technology Overview
Numbers • 5 year program • 140 students/year • 20 specializations
Keywords • International perspective • Integration of research and industry applicability • Gender equality • MakerSpace, hands-on Swedish educational system
6 years of age: one preparatory year
7-15 years of age: compulsory school
16-18 years of age: three-year gymnasium (senior high school/upper secondary school o Wide selection of programs: the natural science program is typical as preparation for academic engineering studies Degrees adapted to the Bologna Process
* 120 credits, ** 60 credits
WWW.KTH.SE Bsc and MSc programs at KTH
• 3 year BSc programs • 2 year MSc programs
• 5 year national programs (=BSc + MSc) industrial recognized professional degree B. Sc. theses B. Sc. degrees B. Sc. prog M. Sc. thesis M. Sc. degree
B. Sc. prog M. Sc. prog
12345 ”2 + 3 ≠ 5, 3 + 3 = 5”
BSc program MSc program (specialization)
• Specific preparatory Choose MSc program courses for MSc program • BSc thesis project • Elective courses A wide variety of degree programmes
35 programmes in Architecture, Master of Science in Engineering and Bachelor of Science in Engineering…
• Architecture • Energy and Environment • Biotechnology • Electro Technology • Chemical Science and • Electronics and Computer Engineering Engineering • Civil Engineering and Urban • Information and Communication Management Technology • Computer Science • Materials Design and Engineering • Construction Engineering • Mechanical Engineering • Design and Product Realisation • Media Technology • Electrical Engineering • Medical Engineering • Engineering Physics • Microelectronics • Engineering and Economics • Vehicle Engineering
KTH ROYAL INSTITUTE OF TECHNOLOGY • 2018-01-17 8 WWW.KTH.SE Mechanical Engineering Degree Program Master’s Programmes available for ME Regular profile, BSc-level
Manu- Mechanics Machine Engineering facturing components Economics Technology II
Solid Mechanics Electrical Engineering
Differential Equations Thermodynamics International profile, BSc-level
Manu- Mechanics Machine facturing components Technology II Language
Solid Mechanics Electrical Engineering
Language Thermodynamics Integrating research and education in engineering design programs ASEE 2016, June 27 Martin Edin Grimheden Summary
• Engineering programs in Europe have been adapted (more or less) to the BSc+MSc system • National professional programs, aimed at ”ready to engineer” has been integrated with academic programs preparing students for research careers
• => Conflicting goals
• => A discussion on the integration of these goals Sweden's leading University of technology
• Sweden's oldest and largest University of technology • More than 11,000 full-time students • More than 1,800 PhD Students • Over 4,800 employees • Five campuses in the Stockholm region • Ranked as the 7th best University of technology in Europe by THE
WWW.KTH.SE Degrees adapted to the Bologna Process
* 120 credits, ** 60 credits
WWW.KTH.SE Bsc and MSc programs at KTH
• 3 year BSc programs • 2 year MSc programs
• 5 year national programs (=BSc + MSc) industrial recognized professional degree
Similar setup in many EU countries, i.e. France, Spain, Netherlands, Belgium, Italy. Engineering degrees in Sweden (and Europe)
BSc degrees Professional (3 years) Academic (3 years)
MSc degrees Professional (2 years) Professional (5 years) Academic (2 years) Academic (5 years) Program compatibility Overlapping program learning goals From 2000 until now
Programs with different learning goals have now been integrated. Programs have been extended in time, new courses have been added.
One main issue: thesis projects – are intended to show proficiency in all program areas.
=> Example: Engineering MSc thesis projects now needs to show expertise in conducting research. MSc thesis projects in MSc programs
• 30 ECTS (half year of studies, full time) • Last semester of studies • Usually performed in industry, for professional MSc degrees, or in academia for academic MSc degrees What is a MSc thesis project?
A thesis or dissertation is a document submitted in support of candidature for an academic degree or professional qualification presenting the author's research and findings.
In some contexts, the word "thesis" or a cognate is used for part of a bachelor's or master's course, while "dissertation" is normally applied to a doctorate.
[Wikipedia] Now, to the main issue…
An academic MSc thesis must show competence in research
A professional MSc thesis must show engineering competence
When the degrees are combined, all MSc theses needs to show both. To show Engineering relevance
Show ability to apply knowledge and skills from previous courses into an industrially relevant project.
Easy to implement and to evaluate and assess Scientific relevance
Students are encouraged to:
Provide clear and relevant research questions that a scientific method can be applied upon.
Focus on one of the following: 1. Build your work on previous research, extend the body of knowledge 2. Use a scientific process (in your engineering work) 3. Evaluate your results according to a scientific method Solution
• Let students choose between doing an academic or professional MSc thesis project. • 99% chooses a professional project
Choose one of the three models 1. Use a research approach to your engineering problem 2. Use a scientific process when solving your engineering problem 3. Use a scientific method when assessing your solution to the engineering problem Example 1
Title: The effects of test process automation on a fault frequency for a spreader system. Example 2
Title: Design of intelligent door lock system. Conclusions
1. The aboslute majority of our students select a professional engineering career 2. The engineering projects, solutions etc improves when integrating a scientific approach/method 3. Many students are capable of managing both engineering and research, but not all Thanks Mechatronics at KTH
Martin Edin Grimheden, Carl During Bukhara, Uzbekistan, 2016-11-02 Mechatronics at KTH - keywords
• 40 persons at Mechatronics
• The microprocessor as a machine element – early 80s • Machine design context • Methods and Products • Automotive, Automation, MedTech
• Engineering education, project and problem based education
• Research themes – Codesign: Mechanics – Control – Embedded Systems • Integrated mechatronics design • Architecting autonomous systems • Model based engineering and integration • Design for Safety
MARTIN TÖRNGREN AND JAN WIKANDER, KTH, LEUVEN WORKSHOP – 2014-08-19 2 Autonomous vehicles
GCDC2011
• Architectural gaps and bottlenecks • Autonomy patterns • Key safety issues – many safety aspects rely on driver takeover • Modularization
MARTIN TÖRNGREN AND JAN WIKANDER, KTH, LEUVEN WORKSHOP – 2014-08-19 3 The Integration Problem
Point-to-point Monocultures Maintenance, management, Integrations lock you in and change costs go up over time don’t scale
Ongoing and unexpected costs drain resources
End-user productivity suffers: Either stuck with the wrong tool, stuck doing manual integration; often stuck doing both Creating new Past choices integrations is restrict present Integrations consume more of the IT budget: unpredictable action and integration failures are the top 2 causes future vision of software project delays*
More limited ability to respond to change Constrained by exhausted IT budget and lower productivity * Commissioned study conducted by Forrester Consulting on behalf of IBM.
4 (Source: http://www.slideshare.net/openservices/introduction-to-oslc-and-linked-data Development trends and challenges in Mechatronics Master/doctoral programs/curricula at KTH Trends in Mechatronics Education@KTH
1996 2016 • ~20 students • ~50 students • all from Sweden • From Sweden, China, • 3 Mechatronics courses India, Turkey, Europe, US,… • 8 Mechatronics courses • Students built products, made PCBs, soldered components • Students build products based on Arduino, BBB,…
• 33% Mechanical HW, • 50% SW, 25% modelling, 33% Electronics, 33% SW 25% HW Mechatronics @KTH 2016+
Students are more interested in Mechatronics The complexity of the subject and products is drastically increasing
Increased need for a stronger scientific base (mathematics, control, analytics)
Increased need for a systems-level perspective; systems engineering, modelling skills Mechatronics @KTH 2020+
• We need adaptation courses • Students come from all over the world, we need to establish a ground level of knowledge in programming, modelling, electronics etc • We need more systems engineering • Developing a new course with industry (SAAB, INCOSE) • We need students to specialize • ”tracks”, students can choose between for example SW, Control, … Structureofeducation Mechatronics track
2018-01-17 9
KTH Royal Institute of Technology • 9 www.kth.se The context
• 13.000 engineering students on BSc or MSc- level • 50 MSc-programs
• Mechanical Engineering . Master’s program in Engineering Design o 75 students/year – 2/3 BSc from KTH – 1/3 recruited internationally o 3 tracks –Mechatronics » 35-40 students – Machine Design – Internal Combustion Engines
10 Engineering Design@KTH is:
So far, three separate ”programs”
• Machine Design • Mechatronics • Internal Combustion Engines
11 Core concepts
• Courses on three levels plus thesis project . Basic courses first semester . Advanced courses second sem. . ”HK”-course third sem.
• ”HK”-course: applied, project-based course in collaboration with industry. BIG course.
• Thesis projects during 4th semester: in industry or research team@KTH
12 The Mechatronics track
13 Educational idea
• Production are outsorced outside Sweden, development tend to follow • Sweden has great opportunities for development of complex, knowledge-intensive products . Requires non-hierarchical structures, high technical competence, creativity, innovation
• We train leaders for the development of advanced products . Who understands the technology, the development and the trends . Who can create new products
14 Or
• We like to build robots.
15 Mechatronics
16 Google: ”define mechatronics”
• Mechatronics is an interdisciplinary area of engineering that combines mechanical and electrical engineering and computer science. A typical mechatronic system picks up signals from the environment, processes them to generate output signals, transforming them for example into forces, motions and actions.
• IEEE: “the synergistic combination of precision mechanical engineering, electronic control and systems thinking in the design of products and manufacturing processes”.
17 SYNERGY
18 19 Adaptive knee-prosthesis
20 KTH Eco Cars
21 What is Mechatronics?
Reliability User-interface ”Intelligent” product and safety Actuators Embedded control system Control Mechanics
Software Sensors Communication
Design
Function, flexibility
Creativity, development, realization 22 Educational idea
Production are outsorced outside Sweden, development tend to follow Sweden has great opportunities for development of complex, knowledge-intensive products • Requires non-hierarchical structures, high technical competence, creativity, innovation
We train leaders for the development of advanced products • Who understands the technology, the development and the trends • Who can create new products T-shaped people
Technology depth, core subjects • Embedded systems (SW, HW) • Control theory • Prototyping • Electronics
Combined with general skills
Management of product development • Project management, organization • Methods Even better
• "the synergistic integration of mechanical engineering with electronics and electrical systems with intellingent computer control in the design and manufacture of industrial products, processes, and operations." • Synergistic integration • Synergy
Noun
syn·er·gy /ˈsinərjē/
The interaction or cooperation of two or more organizations, substances, or other agents to produce a combined effect greater than the...: "the synergy between artist and record company" • Synergistic integration -> improve functionality Summary
• Mechatronics is the concept of working smarter – not harder, and to inexpensively get the most done in as little time as possible. Specialist or generalist?
• The mechatronics engineer is a specialist in synergistic integration. - Specialist in developing mechatronics products More wisdom
• A mechatronics engineer is a specialist in
• To solve a problem, create a product, not based on technology but on functionality - HW or SW, EE or ME Forest Machines
Komatsu Forest Skogforsk Parker Fluid Systemteknik, IFS HIAB LiU
Student projects (HK) Thesis projects (BSc, MSc) PhD students (Erasmus) Many divisions, many subjects
”Purely” mechanical vehicle
Sus/C Brake Steer Wheel Diff Trans Clutch Eng Driver
Susp X X Brake X X
Steer X X Wheel X X X X Diff X X
Trans X X Clutch X X
Eng X X Driver X X X X
X – Mechanical relations
Dealing with viewpoint interrelations by Martin Törngren; SimuTools 2014, March 17, Lisbon 35 Fully programmable vehicle!
Sus/C Brake Steer Wheel Diff Trans Clutch Eng Driver Susp P P X+P P P P P X+P
Brake P P X+P P P P P X+P
Steer P P X+P P P P P X+P
Wheel X X X+P X
Diff P P P X+P X+P P P
Trans P P P P X+P X+P P P
Clutch P P P X+P X+P P
Eng P P P P P P X+P P
Driver P X+P X+P P P X+P P
P – Programmable relations X –Possiblechange
Dealing with viewpoint interrelations by Martin Törngren; SimuTools 2014, March 17, Lisbon 36 Mechatronics and Embedded Control Systems - Research themes
Scania, Volvo, OFFIS Maenad; ICTlabs, Safety Scania, Volvo car, SP FUSE
System modeling and Model Architecting integration
Volvo car, Scania, SAAB, Atlas Copco, SKF, Haldex MBAT, ICTlabs, IDIOM
Multiview Design and Integrated integration, optimization ME/control design DSL’s, ADL’s
MARTIN TÖRNGREN, KTH – MECHATRONICS AT KTH 37 Research themes
Scania, Volvo, SP Safety
Design and Architecting optimization; Model integration Volvo car, Scania, Volvo, SAAB Multiview integration, ABB; Ericsson, Scania, Siemens, Volvo AB DSL’s, ADL’s
MARTIN TÖRNGREN, KTH – MECHATRONICS AT KTH 38 Architecture browser for automotive embedded SW, Espresso project demonstrator (Scania and KTH, Oct. 2013) Martin Törngren, ICES/IVA seminarium, 2:a Dec, 2013: Inbyggda system – möjligheter och utmaningar 39 Structural Actuator Transmission Sensor Controller element models models models models models
Concept Select components Dynamic component and configure properties Algebraic performance models E.g. - System weight - Energy consumption Optimization Multi-domain design - System costs criteria and optimization E.g. Requirements Concept - Torque/velocity profile generation - Path following accuracy - Closed loop dynamics
MARTIN TÖRNGREN, KTH – MECHATRONICS AT KTH 40 The Mechatronics track
41 Mechatronics at KTH
Martin Edin Grimheden Bukhara, Uzbekistan, 2014-06-03 Mechatronics Advanced Course Case stories from 2011 - 2013 The capstone course in context
13.000 engineering students on BSc or MSc-level 50 MSc-programs
Mechanical Engineering • Master’s program in Engineering Design – 75 students/year • 2/3 BSc from KTH • 1/3 recruited internationally – 3 tracks • Mechatronics – 35-40 students • Machine Design • Internal Combustion Engines A typical capstone project
Design brief: Develop a product that actively simulates a lung and measures the output from a ventilator
A large company • Competent and engaged staff Substantial budget • Resources from company Structure of mechatronics capstone course
The Design Brief
3 Deliverables = 3 Sprints • Critical function prototype • 80/20-prototype • Final deliverable
Extensive feedback after each deliverable
Design reviews •KTH faculty • Electrical engineering Preparatory courses • Mechanical engineering • Control theory on BSc level • Programming
Year 0 • Mechatronics basic course • Embedded systems 1 Semester 1 • Robust mechatronics • Research methodology
Semester 2 • Dynamics and motion control Year 1 • Embedded systems 2 • Integrated product development Semester 3
• Mechatronics capstone course
Thesis project Year 2 The capstone course
Design methodology, intro to capstone project Semester 2 4 ECTS
Capstone project 20 ECTS Semester 3 [13 weeks full time] 3 Deliverables
Spring 2011 Fall 2011 Winter 2011
Critical Function Prototype
80/20 Prototype Final Deliverable Projects
InVent / Maquet T-Rex / Micronic-Mydata Kimtech / Kimtech T-Solutions / Prevas Spiros / KTH Eco Cars Fumo / Realisator AB Griffin Stalkers / Scania GreenPiece / SweTree Robyn / KIR+FamiljenPangea Komatsu Scania ReformTech HIAB/Cargotech A typical project
Design Brief (first day) • Design reviews SOTA – State of the art-report (May) • Project definition • Critical function identification
Project re-start, new members (Aug) • Design reviews Mid-term presentations (Oct) • Design reviews Final presentations (Dec) InVent – Design Brief
First deliverable, May
Team defined the tasks/project as:
Measure tidal volumes with 1% accuracy and simulate a passive lung (a lung without capability to breathe). There are two behaviours that we need to model: • Compliance (delta V/delta P) – How the lung dimension changes with pressure difference “elasticity” • Resistance – Simply resistance of pushing air into the lung Further we also need to create a calibration system to make sure we fulfil requirements First deliverable, CFP
Team presented CFP as:
We identified that the CFP is to measure tidal volume accurately. We worked with several concepts in the beginning but choose a concept with a fixed-volume that builds on pressure difference.
Team presented CAD drawings of a proposed concept. Company thought the concept was too complex Project re-start, first design review (beginning of september) Martins description
We need a tank of fixed volume. We breathe into the tank and measure the increased pressure
The temperature increases with the pressure We get moisture We need to measure pressure We need to measure temperature
BUT - what do we measure? Toward the mid-term presentation
The team spent LOTS of time with LabView, temperature sensors, pressure sensors and PCBs. A lot of problems with broken LabView equipment LOT of problem with signal conditioning and filters Mid-term presentation
November, one month to go
Status update, Nov 8th, to prepare a design review
Temperature reading We have equipped the tank with nine temperature sensors to see how the temperature differs within it. Further we have also calibrated all sensors with help from a thermometer by putting them all in the same environment (in the tank) and tuned them to show the same output. Next step is to investigate how the temperature in the tank differ by external influence from: • Surrounding temperature • External heaters • Simulated breaths Internal heaters (air pre heating) Instead of use point heaters to compensate for temperature gradient we will heat the air before it “enters” the tank. We will need a large surface to be able to heat the air and therefore we will use a “fin solution” (in Swedish: fläns lösning). By forcing the air through the fins, we can change temperature of the air before it enters the tank and the gradient effect will be lower. Final presentation
Conclusions
Very good team members Good teamwork Good interaction with the company
Teaching team never really understood the problem Team understood the problem too late
Too much focus on getting a system up and running quickly Too little focus on asking the WHY and HOW-questions The problem was badly defined
Mutual learning experiences – mechatronics capstone projects based on Scrum
Martin Edin Grimheden ASEE 2012-06-11 A typical capstone project
• Design brief: Develop a product that actively simulates a lung and measures the output from a ventilator
• A large company - Competent and engaged staff • Substantial budget - Resources from company Structure of mechatronics capstone course
• The Design Brief
• 3 Deliverables = 3 Sprints - Critical function prototype - 80/20-prototype - Final deliverable
• Extensive feedback after each deliverable
• Design reviews - KTH faculty Agile Methods for product development Scrum Why agile methods?
1. Need to prepare students for industrial SW development
2. Students spend too much time on planning and not enough on doing
3. Can we use scrum to teach organization for product development? Agile methods, Scrum
• Self-organizing teams • Visual planning (team-sourcing)
• Less focus on formal methods - Less planning, more action - Less documentation, more prototypes - Less organization, more adaptive Agile methods for SW development
”The agile Manifesto”
…satisfying customer through early delivery…
...welcome changing requirements…
…business people and developers must work together…
…face-to-face conversation…
…self-organizing teams…
…the team reflects on how to become more effective… Scrum – in 1 picture
Roles: •Scrum team •Product owner •Scrum master
Process steps: •Creating a product backlog •Sprint planning •Daily scrum •Demonstration •Sprint retrospective Roles – The Scrum Team
• The Scrum Team - Performs the actual work - Is self-organized and the members have a joint responsibility for the results.
• The team members decide - how the work is arranged - how assignments are distributed.
• There are no set project roles - Everyone should be able to swap tasks with another member. - Naturally, this does not prevent individual members from being experts in a field. Scrum keywords
…empowering the team… …self-organizing teams… …satisfying the customer… (company and teachers?) …the team reflects on how to become more effective
Can the same be applied on LEARNING? Product owner
Product backlog Scrum master
Sprint Scrum team Delivery backlog
”drop”
Sprint
Sprint planning Sprint retrospective The experiment
The implementation Each Sprint
• Course Kick-off • Team (re)organization T-Team • Define task-list/backlog • Reflections • Grading
• Team Kick-off S-Team • Refine backlog and project plan
• Drop (Deliver prototype and report) Teaching team intro etc Mon, week 1 Sprint intro
Thu, week 1 Team kick-off Sprint planning
Pulse meetings, 15 min every Monday (all students and faculty) Sprint 1
Daily Scrum team meetings, 15 min
Mon, week 8 Delivery Sprint reflection Thu, week 8 Reflection day • 5 teams, 5-10 students per team • One scrum-master per team • The companies defined a product owner
• The product back-log was established and refined between the team and product owner • The students ”locked” a sprint back-log
• No one was allowed to interfer with the team during a sprint The companies Results and discussion 3 case stories
1. Finding the core value 2. Breaking the rules 3. Project management 3 case stories
1. Finding the core value 1. The project scope was too large. The Scrum-thinking encouraged the students to redefine the project so that to the functionality contributing the most was prioritized 2. Breaking the rules 1. The size-constraint was impossible to fulfill, or – rather – was not realistic. By iterative prototyping the size constraint was modified 3. Project management 1. Compared to earlier projects, ”other” students organized the teams… This year, not leadership capabilities was promoted, but rather the technical ability of dividing a problem into sub-problems. Course relevance
4
3
2
1
0 1020304050 Scrum as a tool
45 40 35 30 25 20 15 10 5 0 12345 Scrum for mechatronics PD
50 45 40 35 30 25 20 15 10 5 0 12345 60
50
40
30 Se
20
10
0 1234 90 80 70 60 50 40 Series1 30 20 10 0 1234 Top learning outcomes
• Project A - Important to have internal deadlines and milestones - It’s difficult to create the overall system architecture - Try to avoid measuring temperature • Project B -… - Working in a big group -… • Project C - Project planning is non-trivial - Integration and manufacturing takes time • Project D - Stuff takes more time than expected - Hard to coordinate the different parties - Nothing works the first time you test it Results
• Scrum is advantageous for learning mechatronics product development
• Scrum is not necessarily advantageous for doing mechatronics product development Reflections
• The students received greater insight into project organziation than before • Project organization became something different than before; technical and practical focus Thank you
Matrix Organization
Proj A Proj B Proj C Proj D
Project Mgmt (Calle)
System Architecture (Martin T)
Mechatronics Architecture (Jan)
HW – Mechanics (Björn)
HW – Electronics (Micke)
Course Admin (Vicki) Roles - Product Owner Who is your product owner? • The Product Owner - Drives the project from the business point of view • ensures that the Scrum Team works with the right things - Communicates clear vision of the product - compiles all possible functionalities, changes & problems for the product - prioritizes the possible functionalities, changes & problems to be fixed.
• Result of the Product Owner’s work is a Product Backlog - A to-do list where all the specifications for a product are listed according to how profitable they are deemed to be. - Can be constantly reprioritized - The document is visible to the entire organization so that everyone is aware of what to expect in future releases of the product. Roles – Scrum Master Sounds Familiar?
• The Scrum Master - Coach - coaches the development team, - Fixer - removes any possible impediments, - Gatekeeper – protects scrum team from external disturbances. • • Meets with the team every day in brief meetings - Daily Scrums.
• Ensures the team is disturbed as little as possible - When someone outside the project has an important issue to discuss with the team – take it with the Scrum Master first.
• Adopts a here-and-now perspective to the work. - The focus is always on providing the team with the best possible circumstances for realizing the goals fixed for the Sprint.
• Responsible to hold the Sprint Retrospective with the Scrum team Process – Creating a Product Backlog
• What are product backlog items? - Requirements, features, stories, new functions, change requests, bug fixes, ...
• How big/small can the items be? - Small enough to be included as a whole into a sprint. - Consider breaking down a item into sub-items.
• When is the Product Backlog updated? - At any time - Items can added, removed, changed, re-prioritized. - But when it is time to start a new sprint, the foremost items should be stable. Process – Sprint Planning (Creating a Sprint Backlog)
• The first day(s) of each sprint are set aside to create a Sprint Backlog. - Product Owner explains the goal of the Sprint - The highest prioritized items in the product backlog are broken down into detailed tasks in the sprint backlog. - Time-estimate to determine how many prioritized functions can be include. - Only the team estimates. Product owner only needed to give feedback on the functionalities.
• What are sprint backlog items? - All tasks & activities that need to be done to get the product backlog items completed. - Education, analysis, investigations, design, implementation, testing, interviews, …
• At end of sprint planning, (once the tasks and required time has been determined) - The product owner lets go. - The product owner knows what will be delivered at end of sprint.
• As of now the Scrum Team works under its own responsibility.
• Each Sprint enhances the product’s market value and adds new functions and improvements that can be delivered to the customer. Process – Daily Scrum
• Every day, at the same time, Scrum Team has a brief meeting. - Purpose: Eliminate all speed impediments
• Each participant answers 3 questions: - What have you done since the last meeting? - What will you do between now and the next meeting? - Is there anything preventing you from doing what you have planned? • Big & Small problems • Scrum master needs to deal with any impediments.
• Anyone may attend and listen at the meeting, but only the Scrum Master and the team members may speak Process - Demonstration
• Each Sprint finishes with a demonstration during which a functioning system is run
• Who is present? - Product owner -users - representatives for corporate management - ...
• The product owner accepts the product at end of the sprint Process - Sprint Retrospective Did you do this?
• After the sprint demonstration, the team and scrum master meet in a Sprint Retrospective (evaluation meeting) to discuss - what went well? - what to improve in the next sprint? - How did previous improvement suggestions work?
• focus on improvement in the work process of team & organisation.
• Who is present? -Scrum team - Scrum master - NOT the product owner Scrum – in 1 picture
Scrum Team: … Plan Work Demonstrate Evaluate Plan … Product Owner: (re)prioritize & (re)work product backlog Scrum master: Part of Scrum team Work with Product Owner to plan coming sprint(s) When does Scrum work well? - Good for Mechatronics?
• When you have an aim/vision, but don’t always know the exact details of how to get there.
• When you have a deadline you cannot miss - 3 dependent variables: time, money/resources, features.
• When the product requirements keep changing (customer does not know what they want. Needs keep changing) - Modern software, websites, …
• Need to deliver business value all the time, to keep customer happy
• Sounds like … - A HK project! - But this is a Mechatronics project, and not only Software development. So? Fine! But what about HW design?
• Break down scrum ideas into two categories: - Incremental delivery - Work methods (daily meetings, flat organisation, clear roles of owner & project, prioritizing your features, …)
• Work methods – any problems?
• work incrementally - prototyping is not new to HW design - Can be very expensive, but the alternative is even more expensive. • Quicker stakeholder feedback can outweight the costs of prototyping - Need to plan ahead • Have a ”here-and-now” approach, but with a clear future vision. - keep your options open • A fundamenal requirement for successfull incremental development • that only makes your solution more robust for change - You should not be afraid to throw away earlier solutions • Not lost - You will learn. Solutions can be of use in the future • You will need management support - Example, car body design.
- Yes it is hard, but not impossible. You need to be smart. 2nd International CDIO Conference Linköping University Linköping, Sweden 13 to 14 June 2006
DEVELOPMENT OF A NEW COMPUTATIONAL MATHEMATICS EDUCATION FOR THE MECHANICAL ENGINEERING PROGRAM AT CHALMERS UNIVERSITY OF TECHNOLOGY
M. Enelund1 and S. Larsson2
1. Dept. of Applied Mechanics, Chalmers University of Technology, Göteborg, Sweden 2. Dept. of Mathematical Sciences, Chalmers University of Technology, Göteborg, Sweden
ABSTRACT
A new computationally oriented mathematics education is presented. The education combines traditional symbolic mathematics with computational mathematics and programming in the Matlab environment. Engineering applications are explored in computational exercises that are taught jointly with the courses in mechanics and thermodynamics.
INTRODUCTION
The Mechanical Engineering program at Chalmers University of Technology has taken part in the development of the CDIO model of engineering education since 2000. For example, in the courses in mechanics and strength of materials, a common methodology of mathematical modeling and abstract thinking is emphasized. Important goals are: to be able to set up mathematical models, to formulate the models in the form of equations, to simulate the phenomena by solving the equations on the computer, and to analyze the simulations in order to assess the correctness of models and solutions, as well as to improve the learning of basic phenomena and concepts. The interaction between courses is also emphasized. At the same time, new mathematics courses for the engineering education have been developed at Chalmers and implemented in the Chemical Engineering and Bioengineering programs since 1999. These courses emphasize mathematical modeling, simulation, the use of modern computational tools, and interaction with courses in chemistry and chemical engineering. This is achieved by taking a computational (constructive) approach to the teaching of mathematics. A basic idea of the reformed mathematics courses is a full integration of the computational (numerical) aspects of mathematics (including programming in the Matlab environment), and the analytical (symbolical) aspects. The traditional separation of these aspects, where the analytical mathematics is usually presented in the first year and the computational mathematics in a later course, is not adequate. We believe that the computational aspects of mathematics should be presented from the start. This permits the discussion of, for example, nonlinear algebraic and differential equations, and hence also the introduction of realistic applications from applied subjects such as mechanics and thermodynamics. This creates motivation for studying the analytic mathematics and raises the level of both the mathematics and the applied courses. The students write their own equation solvers in Matlab based on the algorithms presented in the lectures, which focus on related analytic concepts such as convergence, limit, linearization, and derivative.
1 In this paper we present our ongoing work on developing such mathematics courses for the Mechanical Engineering program at Chalmers. If they are accepted, the new courses will be launched in the academic year 2007-08. More specifically, our work involves:
• Developing course materials for computational mathematics to supplement the traditional textbooks that are used; • Developing computer-oriented projects and exercises that will be used simultaneously in the mathematics courses and the courses in mechanics and thermodynamics; • Developing a new introductory course in Matlab-programming.
Mathematics is a fundamental subject in the engineering education and our goal is not a “mathematics for mechanics” toolbox-based course. However, both subjects can benefit from interactions and exchange of examples.
COURSES
We list the mathematics courses in the first year of the Mechanical Engineering program together with the accompanying engineering courses. The year is divided into four periods (quarters of eight weeks).
Period 1. • Programming in Matlab (4.5 ECTS) • Introduction to mathematics (7.5 ECTS) • Introduction to mechanical engineering (7.5 ECTS)
Period 2. • Analysis and linear algebra A (7.5 ECTS) • Introduction to mechanical engineering (7.5 ECTS), continued • Thermodynamics (7.5 ECTS)
Period 3. • Analysis and linear algebra B (7.5 ECTS) • Mechanics and solid mechanics I (7.5 ECTS)
Period 4. • Analysis and linear algebra C (7.5 ECTS) • Mechanics and solid mechanics II (7.5 ECTS)
The third year also contains two mathematics courses: Mathematical statistics and Transforms and differential equations. These are not considered at the moment and are therefore not included in the work reported here.
The following is a short overview of the contents of the courses.
Period 1. Programming in Matlab. Introduction to Matlab. General programming concepts and techniques. Introduction to mathematics. Number systems, elementary functions, derivative, integral, graphs, geometry in space, vector, elimination method. Common project: function gallery.
Period 2. Analysis and linear algebra A. Nonlinear algebraic equations. Derivative. Integral. Ordinary differential equations. Matrix algebra, linear systems of equations, linear independence. Common projects with thermodynamics: steady state heat equation,
2 equilibrium equations. Mechanics. Statics. Principal stresses. Differential equations for bars and axles.
Period 3. Analysis and linear algebra B. Eigenvalue problem for matrices and differential equations. Linearization and stability for systems of differential equations.
Period 4. Analysis and linear algebra C. Analysis in several variables. Introduction to partial differential equations, boundary value problems. Finite element method. Mechanics. Elasticity, plane problems. Beams and plates. Stability. Fracture mechanics.
TEACHING
The teaching is organized as follows.
• Lectures. • Exercises, both analytical and computational. • Computational assignments common to both the mathematics course and the accompanying engineering course.
From the viewpoint of the mechanics and thermodynamics courses the purpose of the common computational assignments is:
• To allow more complex applications compared with the traditional analytical methods or handbook methods. • To illustrate phenomena such as deflection curves, stability, stress concentrations, stress distribution and fracture, make parameter studies. • To give an introduction to working with realistic, complex, engineering problems.
From the mathematical viewpoint the purpose is:
• To give a better understanding of mathematical concepts. • To motivate the study of mathematics.
An overall purpose is to strengthen the ability to apply a modern way of working based on modeling, simulation, analysis; an approach which is generally applicable in all engineering subjects, not just mechanics. Several courses in the second and third years involve computational simulation, e.g., Mechanics, Mechatronics, Machine design, Machine design project, Manufacturing, Finite element methods (elective), Control theory and Fluid dynamics.
MATLAB
Matlab (Matrix Laboratory) is a software for numerical matrix computations. It can be used at many levels, from a simple calculator to a rather advanced interactive programming environment. It contains advanced tools for graphics and for creating graphical user interfaces, as well as ‘toolboxes’ for many problem areas of science and engineering. We have chosen to use Matlab as our programming environment. We do not use any software for symbolic calculation such as Mathematica or Maple. We are aware that this may be viewed as an inappropriate promotion of commercial software to the students, in particular, because the license cost has increased dramatically recently. However, the lack of
3 competitors with the same flexibility, as well as the long tradition among researchers and lecturers, motivates the choice of Matlab as our software. We use Matlab in three ways: (i) for illustrating mathematical concepts using ready- made interactive programs; (ii) for letting students write their own software for implementing the numerical algorithms that are used in the courses; (iii) for presenting results in the form of graphs, plots and simulations. The motivation for (i) and (iii) is obvious. The second one may require some comments. Even though Matlab contains tools for most of the computations that we need to do, we let students write their own programs, from the simple bisection algorithm to the rather advanced finite element method. We believe that there are pedagogical advantages with this. Clearly it gives the students skills in programming and implementation of numerical algorithms. But it also forces understanding of the algorithms and the mathematics behind them; Matlab is more unforgiving against logical errors and sloppy typing than any teacher. Our constructive approach to mathematics based on numerical algorithms gives the computer exercises a natural and strong connection with the other forms of teaching. Finally, we hope that the students will gain self-confidence from using their own software, based on mathematics that they understand, to model advanced engineering systems, instead of running “black box” simulations with ready-made programs.
GENERAL VERSUS SPECIAL EQUATION
The use of numerical algorithms makes it possible for us to discuss equations in their most general forms and to use them in mathematical models in engineering and science. In contrast, a discussion of the solvability of general ordinary differential equations is outside the scope of a traditional first year mathematics course, and numerical methods are often treated in a later course. Since there is little that can be said, or done, about the general case, all emphasis is put on special cases that can be solved symbolically. We try to reverse this order of priority. A general algebraic equation is of the form f(x)=0. A special case is x2+ax+b=0, which is solved symbolically by the formula
x = -a/2 ± (a2/4-b)1/2 .
Note that this is not an “exact solution” because it is no more accurate than the accuracy in computing the square root, which done by solving f(x)=x2-2=0 numerically. However, it provides a formula, which we can manipulate by analytical techniques. A general ordinary differential equation is of the form u'(t) = f(t,u(t)). A special case is u''+ω2u=0, which is solved symbolically by the formula
u(t) = Acos(ω t) + Bsin(ω t).
Again this is not an “exact solution”; it expresses the solution in terms of well known special functions, which can only be computed numerically. The possibility to solve general equations numerically does not diminish the importance of special solutions such as cos(ωt) or exp(at). On the contrary, they help us interpret the numerical results and to understand complex behavior in terms of simple well- known cases. But there is no reason to pursue symbolical computation to its extreme; it is the simple cases that are useful.
A CONSTRUCTIVE APPROACH
The most important feature of the reformed mathematics courses is that we try to take a constructive approach based on numerical algorithms. For example, after we have studied the bisection algorithm, we note that it proves the intermediate value theorem, as explained
4 below. We believe that this constructive/computational approach has the following advantages:
• it makes the mathematics more understandable; • it makes it possible to discuss general equations, not just simplified special cases; • it makes it possible to do applications early in the curriculum; • it makes it possible to reach advanced applications at the end of the curriculum; • it allows open-ended project work.
This approach is based on the textbook [1] and has been implemented since 1999 in the Chemical Engineering and Bioengineering programs at Chalmers University [2]. However, the book [1] has proved to be somewhat too difficult for the students and we plan to use traditional textbooks complemented by lecture notes. We illustrate the approach by describing the bisection algorithm and the intermediate value theorem here. A function f satisfies the Lipschitz condition if
|f(x)-f(y)| ≤ Lf |x-y|.
We use Lipschitz continuity in all proofs even though most results are valid for continuous functions. The advantage of the Lipschitz condition is that it simplifies the difficult epsilon- delta argument; the Lipschitz condition is an explicit relation between epsilon and delta, telling us directly how small |f(x)-f(y)| is, if we know how small |x-y| is. A sequence xn of numbers is called a Cauchy sequence if it satisfies Cauchy's convergence condition:
|xn-xm| → 0 as n,m → ∞ .
We identify the real numbers R with the set of all decimal expansions as follows. On the one hand, writing the terms of a Cauchy sequence xn in decimal form and using Cauchy's condition, we easily see that the decimals are fixed as n is increased. On the other hand, if we have a decimal expansion, then we can form a Cauchy sequence xn by truncating it after n decimals. Note also, that a real number cannot (in general) be specified exactly; but it can be specified it up to any desired accuracy. The use of numerical algorithms to construct new objects is fundamental to our program. The students first encounter such a constructive argument in the classical construction of √2 by solving the algebraic equation f(x)=x2-2=0 by means of the bisection algorithm. Each student is instructed to write a Matlab program implementing the bisection algorithm. The results of two computations with the program are shown in the table. The results indicate that the decimals are fixed as we go down the table and that therefore xn and yn are Cauchy sequences.
xn yn 1.50000000000000 1.50000000000000 0.75000000000000 1.25000000000000 1.12500000000000 1.37500000000000 1.31250000000000 1.43750000000000 1.40625000000000 1.40625000000000 1.45312500000000 1.42187500000000 1.42968750000000 1.41406250000000 1.41796875000000 1.41796875000000 1.41210937500000 1.41601562500000 1.41503906250000 1.41503906250000 1.41357421875000 1.41455078125000 1.41430664062500 1.41430664062500 1.41394042968750 1.41418457031250
5 1.41412353515625 1.41424560546875 1.41421508789063 1.41421508789063 1.41416931152344 1.41419982910156 1.41419219970703 1.41420745849609 1.41420364379883 1.41421127319336 1.41420936584473 1.41421318054199
-n Indeed, the students will realize that, by construction, |xn-xm|≤K 2 if m≥n, so that the algorithm always generates a Cauchy sequence. We therefore have constructed two real numbers (= decimal expansions = Cauchy sequences)
x = 1.4142... and y = 1.41421...
The students are then instructed to use the Lipschitz condition to prove that x and y solve the equation in the sense that the residuals tend to zero, i.e.,
f(xn) → 0 and f(yn) → 0.
In other words, f(x)=f(y)=0. Finally, we note that f is strictly monotone for x>0, so that the equation f(x)=0 cannot have two positive solutions. Therefore, x=y=1.4142... . This new number is so important that we give it a name: √2. Note the following steps in the constructive argument: • an algorithm that generates a Cauchy sequence; • a proof that the limit solves the equation; • the uniqueness of solutions. A solution is constructed in the first two steps. The last step means that all constructions lead to the same result. The bisection algorithm proves the intermediate value theorem: a (Lipschitz) continuous function f : [a,b] → R attains all values between f(a) and f(b). Proof: if y is an intermediate value, then solve f(x)-y=0 by means of the bisection algorithm. Our study of algebraic equations then proceeds to the fixed point algorithm xn+1=g(xn) for equations written in the form x=g(x), which is immediately generalized to systems of equations. Each student writes a Matlab program and proves the fixed point theorem: the system of equations x=g(x) has a unique solution if the Lipschitz constant Lg<1. Proof: show that xn is a Cauchy sequence. A similar development can be made for the Newton-Raphson method and the inverse function theorem.
EXAMPLES OF APPLICATIONS
Here we present two examples of joint computer assignments from mathematics and mechanics. The first one is placed in period 3 while the second one is placed in period 4. The assignments will be tutored by lecturers and assistants from both mathematics and mechanics. The first assignment considers a static analysis of a plane truss by the displacement- based matrix method. The truss is shown in Figure 1. The task is to determine a value of the area parameter A so that the magnitude of the normal stress is less than half the yield stress in the truss. Further, using this information the normal stresses in all members and the joint deformations are to be calculated. The deformed truss should be displayed and a figure showing the relative stress in each member should be produced. The student needs to handle matrices with dimensions up to 20×20 and a Matlab code should be written. Examples of aims are: (1) To provide an in-depth understanding of fundamental principles used for static analysis and corresponding computational procedures. (2) To serve as an introduction to the application of the finite element method in structural mechanics. (3) To train the student in (Matlab) programming, from problem definition to working code. (4) To
6 use graphical tools for presentation of results and for better understanding. (5) To understand the mathematical treatment of large linear systems of equations.
Figure 1: Left: plane truss to be analysed. Right: the line width indicates the relative stress in the members (bars), red colour indicates pressure while blue colour indicates tension.
The second application considers a two dimensional stress analysis of a thin plate with three holes subjected to uniform stress at the vertical boundaries, see Figure 2. Plane stress conditions are assumed. The analysis is carried using the finite element method and the PDE-toolbox in Matlab. The task is to calculate the stress concentration factor Kt which is defined as
Kt = σmax / σnom
By varying the distance b in the figure, the students should be able to decide whether the stress concentrations near the holes are correlated or not. Further, the calculated Kt should be compared with tabulated values from handbooks.
Figure 2: Left: plate with three holes. Right: the color represents the stress distribution in a quarter of the plate. The largest principal stress is presented. The applied stress is σ0=1. Note the higher stress near the holes and the mesh refinement around the holes. Note also the deformed geometry.
To reduce the number of elements, symmetries should be used and only a quarter of the plate needs to be considered. This means that special attention needs to be put on the boundary conditions. There are several aims with this assignment, for example: (1) By visualizing the stress distribution, the students can develop an intuition about stress distributions and how the stress is increased due to abrupt changes in geometry. (2) Motivate the need to study the governing equations of elasticity. (3) It serves as an introduction to the finite element
7 method. (4) It provides an introduction to error estimation and adaptive mesh refinement in the finite element method.
DISCUSSION
From the small scale that this has already been implemented in the mechanics courses, as well as experiences from the Chemical Engineering program [3], it is clear that the students appreciate the ability to work with realistic models and the possibility to gain insight and understanding of the of the behavior of the systems studied. Further, we strongly believe that the proposed education also has the potential to increase the interest for the underlying mathematics. Clearly, the proposed approach strengthens the connection between the applications and mathematics. This is very important for the engineering education, having in mind that mathematics is the fundamental tool for most of the engineering students. This kind of mathematics makes it possible to solve the complete problem: from modeling and solution to simulation of the system and comparison with physical reality. This is, as far as we understand, one of the corner-stones in the CDIO- curriculum. However, our experience also shows that it is important to emphasize symbolic hand calculation and basic programming concepts in the teaching, so that these are not lost in the excitement over the possibility of doing simulations.
REFERENCES
[1] Eriksson, K., Estep, D., and Johnson, C., “Applied Mathematics – Body and Soul”, Springer, 2003.
[2] Website: http://www.math.chalmers.se/cm/education/courses/
[3] Öhrström, L., Svensson, G., Larsson, S., Christie, M., Niklasson, C., “The pedagogical implications of using Matlab in integrated chemistry and mathematics courses”, Int. J. Engrg. Education 21, pp. 683-691, 2005.
8 Australasian Journal of Engineering Education
ISSN: 2205-4952 (Print) 1325-4340 (Online) Journal homepage: http://www.tandfonline.com/loi/teen20
Integration of Education for Sustainable Development in the Mechanical Engineering Curriculum
M Enelund, M Knutson Wedel, U Lundqvist & J Malmqvist
To cite this article: M Enelund, M Knutson Wedel, U Lundqvist & J Malmqvist (2013) Integration of Education for Sustainable Development in the Mechanical Engineering Curriculum, Australasian Journal of Engineering Education, 19:1, 51-62 To link to this article: http://dx.doi.org/10.7158/22054952.2013.11464078
Published online: 16 Nov 2015.
Submit your article to this journal
Article views: 20
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=teen20
Download by: [Kungliga Tekniska Hogskola] Date: 11 September 2017, At: 06:31 51
Integration of education for sustainable development in the mechanical engineering curriculum*
M Enelund†, M Knutson Wedel, U Lundqvist and J Malmqvist Chalmers University of Technology, Gothenburg, Sweden
ABSTRACT: This paper presents and analyses the integration with progression of education for sustainable development in Chalmers University of Technology’s MScEng programme in Mechanical Engineering. The program has an aim and structure that emphasises employability, integration of general engineering skills, authentic engineering experiences with a focus on holistic view of the complete lifecycle of products and systems. The realisation of these aims stress the need of an integrated and adaptable sustainable development education for mechanical engineering. To reach this goal, we applied a combined top-down and bottom-up education development process that started with the formulation of program vision and program level learning outcomes. Faculty meetings and workshop were used to formulate the course learning outcomes and to map the program level outcomes to courses in which the outcomes are satisfi ed followed this. The strategy integrated specifi c sustainability topics in courses where it is appropriate and to have a separate course in sustainable development to ensure that general aspects of sustainable development are included and that a team of faculty takes full responsibility for this. Design-build-test project courses are shown to be suitable arenas for integrating teaching and learning of sustainable development. Results from a student survey on perceptions of the relevance and quality of sustainability education are discussed. Finally, continuing challenges in the area are identifi ed.
KEYWORDS: Education for sustainable development; mechanical engineering; curriculum development; program development; integrated curriculum.
REFERENCE: Enelund, M., Knutson Wedel, M., Lundqvist, U. & Malmqvist, J. 2013, “Integration of education for sustainable development in the mechanical engineering curriculum”, Australasian Journal of Engineering Education, Vol. 19, No. 1, pp. 51-62, http://dx.doi.org/10.7158/D12-018.2013.19.1. Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 1 INTRODUCTION effi cient use of materials, land and other resources. Mechanical engineers need to take active and leading The mission of a mechanical engineer is to create and roles in solving these challenges associated with the operate products and systems that improve safety transformation to a sustainable society. and quality of life for a growing population. This mission should be achieved using a minimum of Stakeholders and students are expecting engineering resources to ensure we do not limit the possibilities programs to prepare the students for the challenges for coming generations to continue to develop their described above and the education must continuously quality of life and safety. The challenges in the next be developed to meet these needs, see, eg. Hannig et decades are huge; new technologies, systems and al (2012). In a report from the National Evaluation solutions for energy supply and transportations are of Engineering Programs, the Swedish National needed, the growing global population requires more Agency for Higher Education (2006) argued “training in engineering and natural sciences is generally * Reviewed and revised version of paper originally presented at suffi cient, but social, economic and environmental the 8th International CDIO Conference, Queensland University applications of engineering are poorly provided for”. of Technology, Brisbane, 1-4 July 2012. † Corresponding author Prof Mikael Enelund can be contacted These needs are thus being formalised into at [email protected]. requirements that are included in national and
© Institution of Engineers Australia, 2013 Australasian Journal of Engineering Education, Vol 19 No 1 52 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist
international requirements on engineering degrees. sustainability. A program-level is essential in order Specifically within the CDIO context, the CDIO to identify appropriate learning events to include syllabus 2.0 clarifi es the position of sustainability in sustainability into, and to develop a progression the syllabus, bringing forward topics such as design between such learning events. This perspective is for sustainability, for sustainable implementation the focus of this paper. and for sustainable operations (Crawley et al, 2011). The aim of this paper is thus to bring forward Moreover, the EUR-ACE standards (ENAEE, 2008) experiences and knowledge on how to develop and require that a 2nd cycle engineering degree graduate integrate a program specifi c sustainable development “demonstrates awareness of the health, safety and education into a mechanical engineering curriculum. legal issues and responsibilities of engineering The context is the CDIO-based MScEng program practice, the impact of engineering solutions in a in mechanical engineering (“the M program”) at societal and environmental context, and commit Chalmers University of Technology. Specifi cally, the to professional ethics, responsibilities and norms objectives are to: of engineering practice”. As yet another example, the Swedish national degree requirements for • provide an example of a mechanical engineering the “Civilingenjör” degree (Master of Science in curriculum where sustainability is systematically Engineering) state since 2007 that (Ministry of contextualised and integrated Education and Research, 2006): • describe the process to develop such a curriculum, To be awarded the Civilingenjör degree the student applying the CDIO approach should be able to demonstrate: • evaluate students’ perceptions of the changed • Ability to design and develop products, education for sustainability processes and systems with consideration of • discuss how to increase the awareness of human prerequisites and needs and the society’s sustainable development among mechanical goals for economically, socially and ecologically engineering students. sustainable development. The remainder of the paper is structured as follows. • Ability to formulate judgements considering We start by describing the current design of the M relevant scientifi c, societal and ethical aspects, program, including its program learning outcomes and demonstrate an awareness of ethical aspects and curriculum, and elaborate on how sustainability on research and development work. is addressed in the program. We then describe • Insight into the possibilities and limitations the development process of reaching that state. of technology, its role in society and the The evaluation section discusses data from recent responsibility of humans for its use, including sustainability courses. A discussion section identifi es social, economic as well as environmental and some continuing challenges in the area and the paper occupational health aspects. is wrapped up with a list of conclusions. The challenge for educational developers is then to design the education in such a way that the 2 EDUCATION FOR SUSTAINABLE requirements are fulfi lled. DEVELOPMENT IN THE M PROGRAM Earlier work on the topic of addressing sustainability 2.1 The M program: Sustainability in CDIO programs starts with Jeswiet et al (2005) program learning outcomes who compiled a list of sustainability topics that should be included in a CDIO program, including
Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 The M program is a fi ve-year program divided into lifecycle assessment, design for environment and two cycles in accordance with the Bologna structure. remanufacturing. Knutson Wedel et al (2008) The fi rst cycle consists of three years of full time studies discussed the implementation process of engineering and corresponds to 180 ECTS and ends with the degree education for sustainable development into CDIO of Bachelor of Science. The second cycle is a two-year programs and present and analyse the relation (120 ECTS) master program. After completing both between the concept of sustainable development cycles the student is awarded the Swedish engineering and the CDIO approach and, in particular, the CDIO degree “Civilingenjör” (MScEng) as well as the degree syllabus. Silja et al (2011) pointed out that several of Master of Science (MSc). CDIO standards are very amendable to sustainability education, including Standard Five (Design-build The M program aims at developing the knowledge, Experiences) and Standard Eight (Active and skills and competence required to participate in Experiential Learning). In addition, several authors and lead the development and design of industrial have presented innovative sustainability project products, processes and systems for a sustainable course models that can be included in a CDIO society (Enelund, 2011). Sustainable development curriculum, for example aiming to develop water of products and systems is thus a vital part of the M sanitation solutions (Ng Huiting et al, 2009) or green program. The program also prepares for positions design competitions (Hussman et al, 2010). Few in other areas of the society where skills in analysis earlier works provide a program-level perspective on and processing of complex open-ended problems are
Australasian Journal of Engineering Education Vol 19 No 1 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist 53
of great importance. During the studies, the student courses in Year 3 are shown in table 4. Courses having shall be able to develop her/his personal skills learning outcomes related to sustainable development attitudes that will contribute to professional integrity corresponding to at least 1 ECTS are marked grey. and to a successful professional life. In the second cycle, the M program students can The M program is described in a CDIO-based choose between 15 different master programs for integrated program description (Enelund, 2011). the degree of “Civilingenjör”, see table 5. Eight of A number of the program level learning outcomes the 15 approved master programs are organised in related to sustainability are seen in table 1. The close connection to the fi rst cycle of the M program. courses are designed to meet these learning outcomes, This means that the program management is and the design of the curriculum in relation to the responsible for content, level, quality, budget and learning outcomes is displayed in the program study environment of both the fi rst and the second design matrix using a ITU scale (I = introduce, T = cycle. All the approved masters programs contain teach and U = utilise), see table 2. integrated sustainability learning experiences. Master programs shaded in light grey in table 5 offer 2.2 The curriculum substantial advanced development of knowledge in sustainability connected to the programs’ domains. Table 3 and 4 show the M program’s curriculum for The M students are also able to choose specifi cally the fi rst cycle. The academic year is divided into four sustainability-focused programs as the specialisation study periods, quarters of eight weeks. The elective of their studies, marked in dark grey in table 5.
Table 1: The program learning outcomes related to sustainable development in the M program.
The Master of Science in Mechanical Engineering graduate shall be able to ... Lead and participate in the development of new products, processes and systems using a holistic approach for the entire process; from stating requirements and formulating the concept, to design, 3 manufacturing, operations and phase-out/shut-down. Following a systematic development process that is adapted for the current situation does this. This requires for instance: • select materials with an understanding of how such choices will affect the manufacturing process, 3.6 product behaviour and environmental impact during the life of the product • compare and evaluate different product solutions with respect to function, environmental impact, 3.7 production and cost • analyse, design and select production systems and machining processes with consideration to 3.8 effi ciency, work motivation, safety and work environment • describe and estimate the economic, societal and environmental consequences of a product or 3.9 system through its lifecycle. Understand and estimate how human behaviour affects the climate on earth as well as its 4 ecosystems. Identify the available energy resources (renewable and non-renewable) and explain how these can 5 be transformed to other energy forms, along with their limitations and environmental impact.
Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 Table 2: Excerpt from the program design matrix. The links between program learning outcomes and the courses are displayed (I = introduce, T = teach and U = utilise).
Learning outcomes/courses (mandatory) 3.6 3.7 3.8 3.9 4 5 Introduction to Mechanical Engineering (project) I I TI Strength of Materials II Materials Science T Machine Elements U Material and Manufacturing Technology T Thermodynamics and Energy Technology T Sustainable Product Development TT TTT Integrated Design and Manufacturing (project) T UU Industrial Production and Organisation T Engineering Economics T T T
Australasian Journal of Engineering Education Vol 19 No 1 54 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist
Table 3: The M program plan for years 1-3.
Quarter Year 12 3 4 Programming in Calculus in a Single Calculus in Several Linear Algebra Matlab Variable Variables 7.5 ECTS 4.5 ECTS 7.5 ECTS 7.5 ECTS Introductory 1 Course in CAD Statics and Strength of Mathematics 4.5 ECTS Strength of Materials Materials 7.5 ECTS 7.5 ECTS 7.5 ECTS Introduction to Mechanical Engineering 7.5 ECTS Thermodynamics and Industrial Production Mechanics: Energy Technology and Organisation Machine Elements Dynamics 7.5 ECTS 6 ECTS 7.5 ECTS 7.5 ECTS Integrated Design and Manufacturing 2 7.5 ECTS Material & Sustainable Product Material Science Manufacturing Industrial Economics Development 7.5 ECTS Technology 4.5 ECTS 4.5 ECTS 7.5 ECTS Mechatronics Automatic Control Bachelor Diploma Project 7.5 ECTS 7.5 ECTS 15 ECTS 3 Fluid Mechanics Elective 1 Elective 2 Mathematical Statistics 7.5 ECTS 7.5 ECTS 7.5 ECTS 7.5 ECTS
Table 4: Elective courses in Year 3 of the M program.
Quarter 2 Quarter 3 Energy Conversion Heat Transfer Finite Element Method Logistics Machine Design Materials and Process Selection Simulation of Production Object Oriented Programming Sound and Vibration Transforms and Differential Equations Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017
Table 5: Master programs approved for the national degree of “Civilingenjör” in Mechanical Engineering.
Master programs belonging to the M program Other Master programs approved by the M program Applied Mechanics Engineering Mathematics Automotive Engineering Learning and Leadership Industrial Ecology Nuclear Engineering Materials Engineering Quality and Operations Management Production Engineering Sound and Vibration Product Development Supply Chain Management Naval Architecture and Oceans Engineering System, Control and Mechatronics Sustainable Energy Systems
Australasian Journal of Engineering Education Vol 19 No 1 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist 55
2.3 Sustainability learning experiences The solution to an industry problem is developed or reconstructed from idea to prototype. Part of Let us now in some more detail discuss the the coursework consists of determining the effect sustainability teaching and learning experiences on the environment the product can have from a included in the bachelor part of the M program. lifecycle perspective through the use of tools from the The basic idea is that sustainable development is Sustainable Product Development course. The basic integrated into courses when applicable. The main idea is that students should integrate sustainability sustainability course is the Sustainable Product considerations to improve their solutions. Also, in Development course during the second year. the Industrial Production and Organisation course, However, sustainability is included already during the students are tasked to analyse, design and choose the first term in the Introduction to Mechanical production and manufacturing systems with special Engineering course. This course includes lectures consideration given to effi ciency, work motivation, on general aspects of sustainability in product safety and working environment. development and in the choice of materials for The sustainability learning experiences are these products. The course also includes a design- summarised and classified in table 2. Table 6 build-test project in which the students consider includes examples of learning outcomes that show their choice of material for the fi nal product and progression between courses in the program. its impact on the environment. During the first year, the Strength of Materials course, discusses the role played by “Strength of Materials” in a 3 THE CHANGE PROCESS technological, economical and socially sustainable society: about 80% of all mechanical breakdowns 3.1 University-wide strategy and approach today are estimated to stem from fatigue which can be related to inadequate calculations of the strengths, Adopting a 10-year perspective of the development insuffi cient knowledge of the phenomena or of other of education for sustainability in the fi rst few years characteristics of the chosen materials. This costs the focus was to put sustainability on the agenda. society approximately 4-5% of the yearly GDP. One There was an increased interest in sustainability of the main purposes of knowledge in the strength from students, industry and faculties as well as from of materials is to be able to create durable, lean and the management of Chalmers. Results from alumni effi cient products. In Mathematics, Mechanics and surveys and interviews with stakeholders stressed Strength of Materials we focus on computations the need for improved and extended education for and simulations and tools for this. These tools are sustainability. Consequently, Chalmers’ management essential to the design of lean products. developed a strategic framework that guides the integration of sustainability knowledge and skills In the second year, the Materials Science course in its programs. The framework identifi es certain discusses the choice of materials as well as options components that should be included in all programs for the collecting of waste products. The Machine but an essential element of the strategy is also to Element course continues the discussion initiated connect sustainability education very closely to in the Strength of Materials and Materials Science applications and decision-making in the student’s courses by teaching students how to design durable study fi eld, for example mechanical, chemical or machines and products for long life, low friction civil engineering. This approach aims to ensure the and low energy consumption as well as effi cient relevance of sustainability education experiences use of material. The following Sustainable Product
Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 to all fi elds and to increase student motivation to Development course begins with general treatment acquire sustainability knowledge and skills. of the environment and sustainable development focusing on global issues. Analytical tools such The strategic framework comprises four building as lifecycle analysis and multi criterion analysis blocks, as illustrated in figure 1: fundamental, are introduced to help determine the effect that integrated, advances and specialised sustainability products and processes have on the environment. knowledge. The fundamental element typically is In addition, strategies and methods are presented implemented in a fi rst or second year course and to help the student gain a view of environmental develops knowledge of some common sustainability and sustainability issues that are necessary for the topics and defi nitions along with some domain- development of future products and processes. In the specifi c sustainability concepts. There is, thus, a parallel Thermodynamics and Energy Technology common core for all Chalmers students but already course the students are taught the boundary there is some adaptation to the fi eld of study. In conditions for our society’s energy consumption and the integrated elements, teaching of sustainability its connection to the climate issue. Limitations and takes place inside a disciplinary course or project. effects on the environment of different energy sources Sustainability learning can then be closely connected are discussed as well as ways of minimising these to learning experiences aiming at mimicking effects. The Integrated Design and Manufacturing authentic analysis or decision-making situations project course runs in parallel to the other courses. in a fi eld, such as materials selection considering
Australasian Journal of Engineering Education Vol 19 No 1 56 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist
Table 6: Examples of course learning outcomes that show progression between courses in the M program. The columns include examples of learning outcomes for each course. The rows show learning outcomes that are connected.
Introduction to Sustainable Product Thermodynamics and Integrated Design Mechanical Engineering Development Energy Technology and Manufacturing Discuss how different environmental values can Describe some basic give different interpretations perspectives for the role of environmental issues and of engineers in society –– how this can have impact on ... in connection to demand, function and need environmental issues of product design in the role of engineers Discuss advantages and problems for the Describe and use general combination of materials, methods as well as strategies product geometry, joining –– for a sustainable product and manufacturing in the development perspective of sustainable development Describe limitations Describe cause effect and environmental – chains for some known effects for the use – environmental problems of different energy technologies and fuels Perform a basic analysis Chart the product of the environmental and life cycle from an – sustainability impact – environmental with the use of life cycle perspective assessment Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017
Fig ure 1: Chalmers strategy for sustainability education.
performance, lifecycle load and cost constraints. At In order to facilitate for programs to implement the Master level, all master programs are required these ideas, a group containing specialists from to include learning experiences that further advance different disciplines at Chalmers was formed to the student’s sustainability knowledge. This can support faculty and program managements in the take place through dedicated courses or integrated integration of sustainability in courses and programs. learning experiences, or both. Chalmers has further The group conducted a series of workshops for designed its total offering of master programs so program management, faculties and student that there for each student can select at least one representatives. These workshops included inspiring relevant sustainability-specialised master program. lectures by specialists and presentations of good For example, among the 15 master programs that practices of course and program development work mechanical engineering students can choose from, at Chalmers. The presentations were followed by two are focused on sustainability: Sustainable Energy discussions in small interdisciplinary groups on Systems and Industrial Ecology. how sustainability could be integrated and taught
Australasian Journal of Engineering Education Vol 19 No 1 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist 57
in courses and programs. This approach successfully that a team of lecturers takes full responsibility for put sustainability on the agenda for course and this. We found that the CDIO approach was benefi cial program developments, created engagement and when designing and integrating the education for involvement and increased the general awareness sustainable development. The existing structure with (see also Holmberg et al (2012)). program description, program learning outcomes and program design matrix was successful and 3.2 Program level experiences from the integration of general skills such as communication and teamwork were used Realising the education for sustainable development as a template. strategy on the program level involved several The program management pinpointed a group of two challenges. First, we need to formulate specific engaged and committed faculty from two different program learning outcomes for the particular departments to develop the fundamental course. engineering discipline, eg. mechanical engineering. The separate course was then developed in close Second, we need to create specifi c courses or/and cooperation with the program management to make integrate in existing course and plan for progression. certain that the students as well as the lecturers on Third, we need to create legitimacy for lecturers as the program understand the relevance of the course well as students to focus on or include sustainability and its relation to the rest of the program. At the same in courses and projects. time the education for sustainable development in the In 2006 an Energy and Environment course (7.5 Introduction to Mechanical Engineering course was ECTS) was launched in Year 3 of the Mechanical increased and focussed on material issues. The energy Engineering program. At the same time the related sustainability issues were then transferred to integration of sustainability in the program’s courses the Thermodynamics and Energy Technology course. was mapped. Evaluations of the program showed Because the M program is CDIO-based there already that general engineering aspects of sustainability existed product development courses, or design-build- and, in particular, energy related issues such as test project courses, centred round the realisation the climate and impact of different energy sources of a product. Those courses are natural arenas for were covered satisfactorily. Evaluations also teaching, training and practising of general skills and showed that sustainability needed to be more suitable arenas for integrating teaching and training distinctly integrated with clear learning outcomes of sustainable development. The fundamental course and planned for progression. The evaluations in sustainable development Sustainable Product pointed out that environmental aspects of the use Development is taught simultaneously and in close of materials and the environmental impact of the cooperation with the second year design-build-test product development process needed to be included project course Integrated Design and Manufacturing. in an extended and more distinct fashion. Further, The idea is that the fundamental course should student course evaluations and student interviews provide tools and methods for sustainable product pointed out the need for education on tools to development that will be used in the design-build- estimate the products’ environmental impact in the test project to improve the students’ solutions and product development projects. products. Moreover, lecturers and former students Based on this, the M program applied a combined of the Integrated Design and Manufacturing course top-down and bottom-up education development fully supported a simultaneously taught course on starting with formulating program vision and sustainable product development and had in fact
Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 program level learning outcomes and followed by earlier asked for such a course. faculty meetings and workshops to formulate the To summarise, in the case of the M program the course learning outcomes and map the program change has been gradual over several years, see level outcomes to courses in which the outcomes are Knutson Wedel et al (2008) and Malmqvist et al satisfi ed. Education for sustainable development has (2010). It appears at this stage that it has been been a standing item on the program advisory board successful based on student surveys and the input meetings and the program level learning outcomes from faculty. According to the model presented by were outlined at those meetings. The learning Knoster (1991) to lead and manage a complex change, outcomes were then presented and scrutinised at it needs to be consensus, skills, incentives, resources, the program faculties meetings. At those meetings and an action plan. In the present case the faculties’ the links between the program learning outcomes workshops were the most important meetings to and the courses were established and the program create consensus. The necessary sustainability skills design matrix was fi lled, see table 2. were established through program management The strategy became to integrate sustainable involving and inviting the Education for Sustainable development in the courses where it is appropriate Development (ESD) group that was active at the and to have a separate, “fundamental” course in university 2006-2009 and by fi nding faculties with sustainable development to ensure that general a special interest in the area. The skills necessary to aspects of sustainable development are included and undertake the pedagogical reform were present since
Australasian Journal of Engineering Education Vol 19 No 1 58 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist
many years of CDIO reform has presented many opportunities to gain experience. As mentioned, incentives were given both by the national degree requirements, the vision of the university and also the CDIO process, which all highlighted the importance of skills and the societal context of engineering. Resources were, as usual, not abundant, but there were some fi nancial support to develop courses and to engage the ESD group. The most Fig ure 2: Results from course evaluation in important, however, was the action plan developed Sustainable Product Development by the M program advisory board (who discussed course – How important is it for an to gain consensus) and the use of the CDIO model MSc in Mechanical engineering to to implement this plan. have competence in sustainability?
4 EVALUATION AND RESULTS
The renewed education for sustainability is being evaluated. Some of the fi rst course evaluations are reviewed below. A survey was included in the course evaluation of the Sustainable Product Development course. The response rate was 54% (138 students). We notice that after this course, over 80% of the Fig ure 3: Results from course evaluation in 2nd year M students regard sustainability to be an Sustainable Product Development important competence for professional mechanical course – Has it facilitated your engineers, see fi gure 2. “I liked the course and its learning that sustainability has content was relevant”, read one free text comment. entered several courses? The percentage of students that regard sustainability to be of no importance has decreased signifi cantly compared to previous years 10-15%. The majority of the M students consider that the program strategy to integrate sustainability is successful and that it facilitates learning and that it provides a better understanding, see fi gure 3. Moreover, the questionnaire reveals that the 2nd year M students consider the M program to provide them with Fig ure 4: Results from course evaluation in competences in sustainability that are relevant for Sustainable Product Development their profession, see fi gure 4: “I have an interest in course – Has the Mechanical the environment and the course was a good fi t in Engineering program provided both private and professional aspects.” you with competences regarding These results from the questionnaire are very sustainability that you believe are promising and encouraging, in particular considering relevant in your career as mechanical that the students are in the middle of their fi ve- engineer?
Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 years education. Clearly, the M students have a genuine interest in sustainability and they regard sustainability to be an important and relevant competence for their careers as mechanical engineers, see also Hannig et al (2012). The students claim that the program provides the relevant competences and that they have received a fairly clear picture of what competences regarding sustainability that mechanical engineers need, see fi gure 5. One may consider the Fig ure 5: Results from course evaluation in fact that about 30% of the students do not have a clear Sustainable Product Development picture of what expertise in sustainable development course – Did you get a clear picture a mechanical engineer needs to be explored. Having of what expertise in sustainable in mind that the that the students are in the middle of development a mechanical engineer their 5-year education and that they will continuously needs? be subjected to sustainability issues during the education, we consider this not to be a problem. with the Sustainable Product Development course) Faculty teaching in the Integrated Design and verify the general picture of M students’ interest in Manufacturing project course (taught in parallel sustainability and claim that the students include
Australasian Journal of Engineering Education Vol 19 No 1 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist 59
aspects of sustainability in their reasoning to a 1. Improved previous knowledge and dynamic higher extent compared to previous years. Since the progression. It is a common among engineering project course is still running and the students have faculty to complain about students’ entry- not fi nalised their products it is too early to discuss level knowledge and skills, in particular with possible effects on the environmental impacts of the respect to mathematics knowledge and hands- developed solutions and products. on mechanical skills. However, with respect to sustainability, the opposite situation seems to be The results of the students in the Sustainable Product at hand. Current students know, due to inclusion Development course were extremely good, 99% of in elementary and high school curricula, and in the students passed the course and the mean grade the general public debate much more about the was 4.3 where grade 5 is the highest. Such results are subject that students did, say 10 years ago. This is unusual for mandatory courses in the M program at refl ected the responses to the course evaluation “I Chalmers. The lecturers were very pleased with the think the course was too easy, I did not have study students’ efforts and the cooperation with lectures at all to pass the course”. University educators of the course taught in parallel. need to understand what the current students However, despite these favourable data, the students bring in terms of previous knowledge and ranked Sustainable Product Development course consciously build their courses on that platform. low. The average overall satisfaction with the course There is also a cascading effect to consider: If new given by the students in the course questionnaire students bring more advanced knowledge to our was low, 2.7 out of 5. The low overall satisfaction basic courses, we should take advantage of that may partly be due to that the course was given for in those courses, but we also need to change our the fi rst time, some administrative information was advanced courses to take advantage of the revised late and students missed old exams etc. But more basic courses. In summary, a dynamic approach important, students argued that the level and the to sustainability learning progression is needed. content of the course were too basic. They maintained Continuous change is a factor and a challenge, but that the course was not challenging enough and in this case with the positive situation of taking that the contributions to their competences and advantage of improved pre-knowledge. skills in sustainability were minor. Moreover, 2. Challenges to the program ethos. Edwardsson the students asked for a more clear focus on the Stiwne & Roxå (2009) discussed what they product development process and the corresponding call “program ethos”, as the idea of what an environmental impacts: educational program stands for and argue that I would have appreciated more on tools to analyse in addition to what is explicitly stated, there are the environmental impact of different materials and many unspoken elements in the ethos, which processes, etc. The Thermodynamics and Energy are carried by students and faculty. If a change Technology course covered this much better. proposal challenges the unspoken ethos, it will be difficult to introduce and sustain. In the At the same time the students are, so far, generally engineering context, part of the unspoken ethos satisfi ed with the integrated elements of sustainability has to do with the diffi culty and workload of in the other courses: engineering education. To be admitted to the I have learnt very little from this course but courses program and completing your studies refl ects such as Thermodynamics and Energy Technology, high ambitions and a sense of achievement when Material Science and Machine Elements have given you are fi nished. Edwardsson Stiwne & Roxå Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 me useful (sustainability) competences and skills. (2009) argued that this is positive when applying for jobs and for self-confidence. However, a The criticism by the students will be taken seriously course that does not comply with this ethos runs and there are obvious possibilities for improvement the risk of devaluing the subject in the students’ of the Sustainable Product Development course and perceptions. Again we quote the students who continue to transfer focus towards estimations of stated “I think the course was too easy, I did not environmental impacts from the products created have study at all to pass the course”. Another and designed during the design-test-build projects. student wrote “An important subject but unclear approach and too little fact-based information 5 DISCUSSION made the course fuzzy”. While these viewpoints can be problematised in terms of being “right” From the course evaluation responses, it is apparent it seems to be a reality, and subjects which are that the integrated approach has been successful in perceived this way continue to be considered some ways, especially concerning the awareness of as something that lies outside of the core of the the relevance of specifi c sustainability knowledge in education and is thus less important. The question the mechanical engineering. However, the results also becomes to what extent sustainability education point to areas that need to be carefully considered. should be aligned with the dominating practices Let us point out two key challenges: of engineering education and to what extent a
Australasian Journal of Engineering Education Vol 19 No 1 60 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist
different approach should be applied. If so, a this ethos, it runs the risk of being criticised by considerable amount of effort needs to be put into students just for being taught in a different style. motivating why the differences are necessary. And Awareness of this risk and an appropriate strategy this task is not the sole responsibility of the faculty to deal with it is essential. responsible for a single sustainability course. Rather, the program manager and the faculty collective need to embrace the cultural change. REFERENCES Crawley, E., Malmqvist, J., Lucas, W. A. & Brodeur, 6 CONCLUSIONS D. 2011, “The CDIO Syllabus v2.0 - An Updated Statement of Goals for Engineering Education”, Sustainability should be addressed in many Proceedings of the 7th International CDIO Conference, courses in a mechanical engineering program. A Copenhagen, Denmark. suitable strategy includes at least one fundamental sustainability course and a systematic approach to Edvardsson Stiwne, E. & Roxå, T. 2009, integrated sustainability in many other courses. A “Programethos och dess betydelse för studenters program-level perspective is important to maintain lärande och personliga utveckling inom tekniska links to overall program learning outcomes and to utbildningar (Programme Ethos and its Signifi cance ensure progression. for Engineering Students’ Learning and Personal nd An open process for developing and implementing Development)”, Proceedings 2 Utvecklingskonferensen sustainability elements was applied in Chalmers för Sveriges ingenjörsutbildningar, Lund, Sweden. M program. The combined top-down and bottom- up approach created arenas at different levels for Enelund, M. 2011, Programme Description for development work which gave the possibilities for M.Sc. in Mechanical Engineering Programme, www. program management to communicate and discuss am.chalmers.se/~mien/programme_descr_mec_ the program level vision and gave ample room for eng_may2011.pdf. individual involvement and engagement. The CDIO framework with an integrated program description, European Network for Accreditation of Engineering the tools for creating an integrated curriculum and Education (ENAEE), 2008, EUR-ACE Framework inherent design-build-test learning experiences was Standards for the Accreditation of Engineering found to facilitate the process substantially. Programmes, www.enaee.eu/eur-ace-system/eur- ace-framework-standards, accessed 4 May 2011. In the reformed curriculum, sustainability elements are pervasive and adapted to context. It is shown that this has increased students’ awareness of the topic Hannig, A., Priem Abelsson, A., Lundquist, U. & and clarifi ed their view of what specifi c sustainability Svanström, M. 2012, “Are we educating engineers competence that is applicable and relevant in their for sustainability? Comparison between obtained fi eld. The integrative approach facilitates learning competences and Swedish industry’s needs”, of sustainability topics by connecting them closely International Journal of Sustainability in Higher to professional engineering tasks, such as decision- Education, Vol. 13, No. 3. making in design projects. Holmberg, J., Lundqvist, U., Svanström, M. & The work described here is the fi rst steps towards Arehag, M. 2012, “The university and transformation an integrated and domain-specifi c sustainability
Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017 towards sustainability: Lessons learned at Chalmers education for mechanical engineering at Chalmers. University of Technology”, International Journal of The strategy described here will be used in a Sustainability in Higher Education, Vol. 13, No. 3. continuous development setting to improve and extend the education. Hussman, P. M., Trandum, C. & Vigild, M. 2010, Many challenges still remain, though, including to: “How to Include Sustainability in Engineering • adapt sustainability goals and contents to Education – The ‘Green Challenge’ at DTU is one the increasing level of pre-knowledge and Way”, Proceedings of the 6th International CDIO understanding that current students bring from Conference, Montréal, Canada. high school. This is in contrast to the common situation where teachers are challenged by Jeswiet, J., Duflou, J., Dewulf, W., Luttropp, C. weaker pre-knowledge, eg. in mathematics and & Hauschild, M. 2005, “A Curriculum for Life with respect to hands-on mechanical skills. Cycle Engineering Design for the Environment”, • consider if and how to bring the sustainability Proceedings of the 1st International CDIO Conference, courses closer to the ethos and pedagogy Kingston, Canada. of engineering programs, which typically is characterised by high ambitions, a high workload Knoster, T. 1991, “Presentation in TASH Conference”, and tough exams. When a course deviates from Washington, DC.
Australasian Journal of Engineering Education Vol 19 No 1 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist 61
Knutson Wedel, M., Malmqvist, J., Arehag, M. & Ng Huiting, C., Sale, D. & Yeo, A. 2009, “CDIO as a Svanström, M. 2008, “Implementing Engineering Force for Good – A Water Sanitation And Hygiene Education for Environmental Sustainability into Community Service Project In Myanmar”, Proceedings CDIO Programs”, Proceedings of the 4th International of the 5th International CDIO Conference, Singapore. CDIO Conference, Gent, Belgium. Silja, K., Hurskainen, L. & Irma, M. 2011, “Curriculum Malmqvist, J., Bankel, J., Enelund, M., Gustafsson, Development for Clean Technology”, Proceedings of G. & Knutson Wedel, M. 2010, “Ten Years of the 7th International CDIO Conference, Copenhagen, CDIO – Experiences from a Long-term Education Denmark. Development Process”, Proceedings of the 6th International CDIO Conference, Montréal, Canada. Swedish National Agency for Higher Education, 2006, “Evaluation of Civil Engineering Programs Ministry of Education and Research, 2006, Higher at Swedish Universities and Institutions of Higher Education Ordinance. SFS 1993:100, with amendments Education”, Högskoleverket, Rapport 2006:31R, up to SFS 2006:1054, Stockholm, Sweden. Stockholm, Sweden. Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017
Australasian Journal of Engineering Education Vol 19 No 1 62 “Integration of education for sustainable ...” – Enelund, Knutson Wedel, Lundqvist & Malmqvist
MIKAEL ENELUND
Mikael Enelund is professor in structural dynamics at Chalmers. He obtained his PhD from Chalmers in 1996 and was appointed professor in 2012. His main research interests are fi nite element methods, modelling of damping and engineering education developments. Enelund is head of the MScEng program in Mechanical Engineering at Chalmers. Under Mikael’s leadership, the Mechanical Engineering program at Chalmers was selected as Centre of Excellence of Higher Education 2008 by the National Swedish Agency for Higher Education and awarded the distinction of “Best Engineering Education 2012” by Teknikföretagen (the Swedish engineering employers’ organisation).
MARIA KNUTSON WEDEL
Maria Knutson Wedel is Vice president of undergraduate and graduate education at Chalmers. She is a professor in Engineering materials at the Department of Materials and Manufacturing Technology at Chalmers and obtained her PhD in Physics from Chalmers in 1996. Her main research activity has been electron microscopy and the correlation between mechanical properties and microstructure. Her work encompasses a range of materials such as ceramics (Si3N4) aluminium, steel and intermetallics (TiAl, FeAl, MoSi2) as well as mechanical alloying. Her engagement in sustainability and engineering has led to cross-disciplinary research on microwave pyrolysis for recycling of Waste Electronic Equipment, as well as development of methods to improve sustainability awareness in research.
ULRIKA LUNDQVIST
Ulrika Lundqvist is a senior lecturer and vice head of department responsible for undergraduate education at the department Energy and Environment at Chalmers. Her research is in the fi eld of Industrial Ecology. She is also engaged in education for sustainable development at Chalmers’ Learning Centre. Downloaded by [Kungliga Tekniska Hogskola] at 06:31 11 September 2017
JOHAN MALMQVIST
Johan Malmqvist is chair professor in product development at Chalmers. He obtained his PhD from Chalmers in 1993 and was appointed chair professor in 2005. His research focuses on development methodologies and IT support for product development. The results include new design methods and IT tools, but also empirical studies of product development practices. The research is conducted in close collaboration with Swedish industry. Current projects investigate methods and tools for development of product-service systems and for product confi guration. Johan is also strongly engaged in the renewal of engineering education. As a dean of education, he is responsible for Chalmers education programs in mechanical, automation, industrial design engineering as well as the naval program.
Australasian Journal of Engineering Education Vol 19 No 1 TEN YEARS OF CDIO – EXPERIENCES FROM A LONG-TERM EDUCATION DEVELOPMENT PROCESS
Johan Malmqvist
Johan Bankel
Mikael Enelund
Göran Gustafsson
Maria Knutson Wedel
Chalmers University of Technology Gothenburg, SWEDEN
ABSTRACT
The paper describes and analyses a long-term education development process – the CDIO- based reform of Chalmers University of Technology’s M.Sc. in Mechanical Engineering programme. The initial goals of the reform programme and the changes that it has lead to are reviewed. The results of various kinds of evaluations – CDIO self-evaluations, external evaluations, student and faculty views and costs – are reviewed. A number of critical success factors for sustainable educational development process are identified.
KEYWORDS
CDIO, education development, long-term, critical success factors, mechanical engineering
INTRODUCTION
Education development is sometimes seen as a “project” where a reform is initiated and implemented during a period of a few years. However, one can also argue that the project phase should be seen as merely the start of a long-term development process, and that a large potential for improvement lies in the ability to sustain an enduring improvement process, resembling industrial continuous improvement approaches mastered by leading manufacturing firms such as Toyota.
In this paper, we examine such a case: In 2000, the M.Sc. Mechanical Engineering programme (the “M programme”) at Chalmers University of Technology teamed up with programmes from The Royal Institute of Technology (KTH), Linköping University and Massachusetts Institute of Technology to form the Wallenberg CDIO project, which later evolved into the CDIO Initiative, which has many more participating universities [1][2]. This was the starting point for an
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 education development process, which has now lasted for ten years, and comprised many changes including the introduction curricular, pedagogic and learning environment innovations. This reform thus provides a rich empirical base for a study of long-term education development.
Sustainable education reform can be viewed as a strategic process, which according to Lissack and Roos [3] can be maintained by continuously updating descriptions of (a) the current business (who are we and what do we do, our identity), (b) the vision (an image of the desirable future) and, (c) the strategic plan (important sub-goals and actions to reach the desirable future). Persistent changes are further strongly coupled to cultural changes. Such changes require according to Bennich-Björkman [4] agents inside the organisations, a programme for reform or at least some guiding principles, a structural opportunity when the organisation is weak, and changes to the explicit and implicit norms that govern the organisation.
We thus view long term education development as a strategic process governed by a continuously updated strategic plan, actions to realise the strategic plan, and with essential elements of cultural change. The literature on such processes is scarce. One notable exception is Edvardsson Stiwne et alia’s [5][6] studies of students from Linköping University’s Applied physics and Electrical engineering (Y) programme. They examined changes in student’s perceptions of their education and future profession in the context of the Y programme’s CDIO reform. However, they did not focus on describing curricular changes in detail nor on how the implementation of educational changes progressed. These aspects are focused in this paper.
Our general aims in this study are to bring forward experiences and knowledge from a long-term education development process. These, in turn, could constitute a base for proposing improvements in education development practice. In particular, the paper aims to:
• Provide a detailed account for a long-term education development process, highlighting aims and goals, changes made in order to reach these, important events, organizational structures, successes, failures, delays etc. • Evaluate the result: in what way is the programme different and better today? Have the goals and intentions of the initial project been realized? Have the goals changed over time, and if so, how and why? How are the achievements measured? How do different stakeholders view the result? • Examine how the M programme’s development has affected reform of other programmes at Chalmers. What are the catalyst and barriers for spreading educational innovations within a university? • Identify critical success factors for achieving a sustained programme development process over a long time period.
The remainder of the paper is structured as follows: We will first account for the research approach applied, followed by a description of the current design of the M programme. In a retrospective section, we revisit the starting point for the development: the M programme of 2000. We then account for the process up to the current date and summarise the future development plans for the programme. In the following section, we evaluate the outcomes of the reform. We start by reviewing the goals that were stated in the beginning of the CDIO project and assess ho well they have been fulfilled. Further, qualitative and quantitative evaluation data from CDIO self-evaluations, Swedish National of Higher Education evaluations, interviews with students and faculty as well as costs are reviewed. We then present a list of critical success factors for sustainable education reforms and wrap up the paper with conclusions.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 RESEARCH APPROACH
As was stated in the introduction, our general aims in this study are to bring forward experiences and knowledge from a long-term education development process. Education development is a complex activity affected by a great number of dynamic factors and interesting phenomena; we have adopted a qualitative systems approach in the research. This approach requires a detailed documentation of the case and a rigorous data collection, in order to identify underlying factors, to minimise bias, and to increase the transparency of the observations made.
Here, the principle of multiple information sources has been adopted. Various documents have been studied, and individual and group interviews have been carried out to map out the progress, outcomes and perceptions of the process. The documents have included development proposals [7][8], the programme description [9][10][11], course syllabi [12], course evaluations, school year evaluations, budgets, external evaluation reports [13][14], CDIO self-evaluations and a number of papers that have been published during the development process. Two group interviews with students and three group interviews with faculty were carried out (6 student and 13 faculty participants in total).
It should be pointed out that the authors have played central roles in the development process, as dean of education, programme directors, education coordinator and as teachers. There is thus a risk that the results are perceived as biased in a positive direction. To some extent, this is mitigated by the inclusion of evaluation reports by external agencies and interviews with faculty and students in the analysis.
THE M PROGRAMME OF TODAY
This section reviews the current version of the programme, including the programme aim and idea, curriculum, learning environments and management processes. More detailed information can be found in the programme description [10] and at Chalmers’ website [12].
The aim of the programme
The M programme aims at developing the knowledge, skills and competence required to participate in and lead the development and design of industrial products, processes and systems for a sustainable society. The programme also prepares for positions in other areas of the society where skills in analysis and processing of complex open-ended problems are of great importance. During the studies, the student shall be able to develop her/his personal qualities and attitudes that will contribute to professional integrity and to a successful professional life.
The programme idea
The vision of the M programme is outlined in the programme idea statement. It states that the vision of the programme is to offer a relevant, stimulating and advanced level engineering education with a holistic view, which emphasizes both engineering fundamentals and practice. The well-being of the students is in focus as well as the students’ attractiveness for prospective employers. Programme characteristics are:
• The “main thread” of the programme is a holistic view of product and system lifecycle development and deployment.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 • The introductory course of the programme provides a framework for the practice of engineering in product and system building and introduces the students to the engineering profession. The students participate in a small team project in developing a product and producing a prototype. A report is to be presented in writing as well as orally. • The base of the programme is the fundamentals of mathematics and mechanical engineering with emphasis on common principles. This is achieved by having joint projects and assignments between mathematics and the basic courses in mechanics and strength of materials. The projects include the full view of problem solving, from selecting a model and setting up equations, describing the model to solving equations and simulating and assess quality of the choice of model and accuracy of the solution. The purpose of working with the full view, joint projects and the sequence of courses is that education and learning of a topic shall not be isolated in a specific course. • Computer based tools for modelling, analyzing and simulation of real designs, products and systems are early introduced and utilized in the programme at an early stage. • Fundamental engineering courses are introduced early in the curriculum to prepare the students for upcoming Design-Build projects where the assignment is to realize realistic and relevant products and systems. At least one project is included in the curriculum each academic year. • Development of the students´ teamwork and communication skills is integrated in the courses with a distinct progression throughout the programme. • Aspects of sustainable development are emphasized, and the focus is on product development and energy supply. • The fundamentals of the programme together with the elective courses in the third year prepare the student for the concluding two years of study at the master’s level in mechanical engineering as well as adjoining areas such as acoustics, industrial economy, mathematics and mechatronics. • Teaching is partly executed in cooperation with industry through guest lecturers or teachers from industry. Student assignments, project management models and laboratory experiments are designed together with industry. • The level of the education in the two years of study at the master’s level shall prepare for doctoral studies (third cycle). • The syllabus of the programme is continuously improved in cooperation with teachers, students, and administrators as well as in the advisory board where representatives from industry take part.
Curriculum
The M programme is a five-year programme divided into two cycles in accordance with the Bologna structure. The first cycle consists of three years of full time studies and corresponds to 180 credits (cr) and ends with the degree of Bachelor of Science. The second cycle is a two years (120 credits) international master programme. After completing both cycles the students is awarded the Swedish degree Civilingenjör as well as the degrees of Bachelor and Master of Science. Teaching at the bachelor level is generally in Swedish, while the teaching language in the master programmes is English in order to cater for incoming international students with Bachelor degrees.
The programme plan for the first three years of the M programme is shown in Table 1 (the year is divided into four study periods, quarters of eight weeks). The Design-Build-Test courses are marked grey and jointly taught projects between courses are indicated by grey ellipses.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Table 1 M programme plan for years 1-3
Year 1, Quarter 1 Quarter 2 Quarter 3 Quarter 4 Programming in Matlab Calculus in a single Linear algebra Calculus in several 4.5 cr variable 7.5 cr 7.5 cr variables 7.5 cr
Introductory course in mathematics CAD 4.5 cr 7.5 cr Mechanics and statics Strength of materials 7.5 cr 7.5 cr Introduction to mechanical engineering 7.5 cr Year 2, Quarter 1 Quarter 2 Quarter 3 Quarter 4 Thermodynamics and energy Industrial production and technology organization Mechanics: Dynamics Machine elements 7.5 cr 6 cr 7.5 cr 7.5 cr Integrated design and manufacturing 7.5 cr Material and Material science manufacturing Sustainable development Industrial Economics 7.5 cr technology 4.5 cr 4.5 cr 7.5 cr Year 3, Quarter 1 Quarter 2 Quarter 3 Quarter 4
Bachelor diploma project Mechatronics 7.5 cr Automatic control 7.5 cr 15 cr
Fluid mechanics Elective 1 Elective 2 Mathematical statistics 7.5 cr 7.5 cr 7.5 cr 7.5 cr
The first cycle (first three years) begins with the course Introduction to mechanical engineering, which serves as the introduction to the programme and to the role as a professional mechanical engineer. The course includes a team project. The assignment is to identify, define and solve an everyday problem and design and build a prototype for it. An example is a pizza cart holder for bikes. Lectures and exercises in teamwork and communication are integrated in the course. The students are then introduced to general aspects of communication, report writing and oral presentations. Moreover, an introduction to sustainable development is given and the link to product development and material selection is discussed.
Matlab is the general programming and simulation tool in the programme. The mathematical education is computationally oriented and focussed on engineering applications. It combines traditional symbolic mathematics with computational mathematics and programming in Matlab. Engineering applications are explored in computational exercises taught jointly with the courses in Mechanics and Strength of materials. Fundamental engineering courses such as Mechanics, Strength of materials, Materials and Machine elements are introduced early in the curriculum to prepare the students for upcoming design-build-test projects. In particular, the Finite Element Method is taught and used in the courses Calculus in several variables and Strength of materials.
In the second year the students’ communication and teamwork skills are strengthened and practiced together with project management in the design-build-test project course Integrated
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 design and manufacturing. The project task is taken from the industry. Relevant analyses are carried out using principles, knowledge and methods learnt in fundamental engineering courses.
The third step in the training of communicative and team work abilities is implemented at the end of the third year when the students complete their Bachelor thesis projects involving team work, report writing and presentations. Moreover, a minor course in theory and science methodology is integrated in the Bachelor diploma project.
The M programme students can choose between 14 different master programmes for the degree of Civilingenjör. Eight of the 14 approved master programmes are organized in close connection to the first cycle of the Mechanical engineering programme. This means that the programme management is responsible for content, level, quality, budget and study environment of both the first and the second cycle. The approved master programs are listed in Table 2. All master programmes include at least one team project. The master programmes admit both domestic and international students holding Bachelor degrees in mechanical engineering or similar.
Table 2 Master programmes approved for the degree of Civilingenjör in Mechanical engineering
Master programmes belonging to the M Other Master programmes approved by the M programme programme Advanced engineering materials Engineering mathematics Solid and fluid mechanics Management and economics of innovation Automotive engineering Nuclear engineering Industrial ecology Quality and operations management Production engineering Sound and vibration Product development Supply chain management Naval architecture System, control and mechatronics Sustainable energy systems
Learning environment
The M programme has its own prototype laboratory and workshop. It consists of fully equipped metal and wood workshops, a mechatronics lab and a paper working area. The lab and workshop are fundamental resources used throughout the M program. The students build physical models or prototypes of their own designs, e.g., from the simple first year projects in the introductory course and the industrial solutions in the second year course Integrated design and manufacturing to a complete racing car in the fourth year Formula student course. All M programme students take a basic course in safety and handling of basic tools since it is required to be allowed to work in the workshop. More advanced workshop courses in, e.g. welding, operating of NC machinery, metal and wood shaping, are offered to the students. After taking these courses the students get licensed to operate the facilities in the workshop. The students are also allowed to use the workshop for private work during after school hours, and there is a student run organization, which is joined by those who are particularly interested in that.
The students together with the programme management have re-built and furnished the study hall and the cafeteria “Bulten”. The study hall is built and furnished to create a stimulating environment for studying and for social activities. It is very much appreciated by the students and frequently used from early in the morning to late in the evening.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Programme management
At Chalmers, a buyer-supplier setup is applied for managing the education. The programme thus “buys” courses from several departments to compose a programme. The departments are suppliers of courses. The head commissions courses from the departments through an agreement with the vice head of the delivering department. In the agreement, content, pedagogy and budget are specified. The most important reason for this “buyer-supplier” organization is to ensure that the multidisciplinary programmes are well composed and unified. The organisational structure further enables the programmes to optimize goals and content to meet the demands of the society rather than departmental considerations. Another reason is to separate the departmental economics from the economy of the programme.
The head of the M programme is responsible for the programme, including budget, overall planning and quality of the programme, as well as the study environment and the safety and health of the students. The goals of the programme are established through a continuous process lead by the head of the programme in collaboration with the advisory board, students and teachers. The students are very active in the running and the development of the program. The views of the students are considered very important. The program management and the students meet regularly.
Quality assurance system
The programme’s quality assurance system follows a plan-do-check-act cycle. In short, the cycle consists of the following actions:
Plan: Establish programme description and course plans, Outline and confirm links between courses, outline cooperation and common projects between courses. Advisory board meetings, teacher meetings, student-program management meetings are the fora for discussing these issues.
Do: Teaching, learning and assessment in courses and projects.
Check: Course evaluations, class evaluations, follow-up of course delivery agreement, alumni survey, CDIO-self evaluation and benchmarking of the program, student results follow-up, study environment health and safety review and study social environment review are used to evaluate the state of the programme.
Act: Revision of the programme description and course plans, update course delivery agreement for next academic year
Each academic year starts with a meeting with the teachers, programme management team and student representatives. The agenda consists of last academic year experiences and feedback from each quality system, e.g. course evaluations, class follow-ups, students’ results self- assessment, benchmarking and follow-ups of the annual agreement with the departments. The programme description is discussed and established. Links between courses and cooperation between courses are outlined and confirmed.
The major body for programme development issues is the advisory board of the programme, which consists of representatives from industry, students, Chalmers teachers and administrators connected to the programme. Meetings with the advisory board are held two to three times per semester. All strategic questions are discussed in detail at the meetings. Further, the
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 programme content and the achievements of the students are discussed and analyzed regularly. A standing item at the meetings is the students’ view on the education and the study environment.
An essential tool in the quality assurance system is the programme description. The programme has a CDIO-centred integrated programme description [9]. The programme description framework can be described as a template for programme development. It captures the programme aim, idea, goals, plan and it is shown in what courses the goals are fulfilled in a programme design matrix. The development of the programme description is the most important tool for the unification of the programme as well as for the design of courses and teaching activities. It generates a common terminology and helps to shift the emphasis on programme development discussion from specific courses towards high-level issues such as programme purpose, goals, idea and the teaching/learning of generic skills. Suggested changes can be put in relation to the goals of the programme.
The programme’s quality assurance system further comprises a set of planning and evaluation tools including the agreements between programmes and department on courses deliveries, which specify course content, pedagogic, assessment, labs, etc. as well as course budgets are specified, course evaluations, alumni surveys, and CDIO self-evaluations.
THE TIME-LINE
Following this account for the current state of the M programme, this section provides a chronological report of the development, starting from the programme as it looked ten years ago and the initial goals for the CDIO adoption. We will then discuss the main events during the process, ending with future development plans. Figure 1 is referred to throughout.
Starting position
The development that is discussed in this paper started in January 2000 at a meeting at the Royal Institute of Technology (KTH) in Stockholm, Sweden. At the meeting, representatives from MIT, Chalmers, KTH and Linköping University outlined an application for funding of an education development project with CDIO as the basic idea [7].
Chalmers’ M programme had then just graduated the first students from an earlier programme revision – M2000 – that aimed to prepare students for the requirements on engineers of 2000. Educational changes introduced in that revision included several project courses, engineering science already in year one and new profiles at the master level. Design was given a more prominent role in the education. However, the end results of the design projects were, with some small-scale experiments as exceptions, limited to paper and computer prototypes. We did not have the means to run design-build-test projects that went all the way from need to manufactured product, certainly not in classes with 150 students.
We were thus initially attracted by the CDIO idea because we thought that design-build-test projects was “the” missing element in our programme, the element that if included would lift the programme from good to great. Over time CDIO has evolved to a more comprehensive concept, but in 2000 we considered design-build-test experiences as it most distinguishing feature.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Main phases & events
The development can roughly be divided into three phases: CDIO planning, CDIO basic development & piloting, CDIO implementation & further development, see Figure 1.
CDIO planning
This first phase lasted from early 2000 to mid 2001. It comprised initial planning and culminated in an external review that was a requirement for the final approval of the major part of the funding.
During the CDIO planning phase, we stated an initial set of goals for the project, concretized the CDIO concept, made stakeholder surveys [15], and benchmarked our programmes against the CDIO syllabus [16]. We also introduced design-build-test experiences on a small scale. These were essential for demonstrating the CDIO concept to our faculty, and to achieve some short- term wins that are important for a successful change process [17].
CDIO basic development & piloting
The phase of the development lasted from mid 2001 to Fall 2004. During the period CDIO learning experiences and workspaces were developed and successively deployed into the programme. Major efforts included the design and construction of the prototyping lab, the design-build-test learning experiences that use it, and planned learning sequences for integrated learning of teamwork and communication. Early results included design-build-test projects in for the first [18] and second year of studies [19].
In 2002, Chalmers decided to adopt a Bologna-inspired 3+2 education structure, with the last two years consisting of Master programmes taught in English. The reform brought on changes also in the Bachelor part of the programme. As a consequence, the transition to a CDIO and 3+2-based programme became tightly coupled. The new education structure was introduced in 2004. Students from this cohort are the first M programme students that followed a “complete” (but not “final”) CDIO programme.
By 2004, the programme had the basic tenets for running a CDIO programme in place: the learning environment, the key courses, the staffing.
CDIO implementation and further development
In the next period (2004-2008), the M programme implemented, evaluated and refined the basic elements of its CDIO concept. Simultaneously, a number of development activities were carried out: a mathematics course emphasizing modelling and simulation using MATLAB [20], a planned learning sequence for sustainability [21], a further developed 2nd year design-build-test project course, refined programme goals [9], a new bachelor thesis project course [22], and education in English on the master level.
During this period, the M programme was twice evaluated by the Swedish National Agency for Higher Education. The first evaluation was the 2005 national evaluation of all of Sweden’s Civilingenjör programmes [13]. The second was an evaluation of appointment as a Centre of Excellent Quality in Higher Education in 2008 [14].
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010
Figure 1. Time-lime for Chalmers M programme evolution.
During this phase, ideas from CDIO were to some extent spread to other Chalmers’ programmes. For example, the adaptation to the Bologna process at our university meant the start of 44 Master programmes and it was agreed upon that the CDIO model provided structured tools applicable for programme development and for writing programme descriptions. The programmes were required to write detailed programme goals, and to use programme design matrices were also created to demonstrate that the programmes addressed their goals [23]. This adoption essentially focused on CDIO as a tool for systematic programme development. Some programmes embraced the content components of CDIO, e.g., design-build-test projects courses, whilst others chose not to.
Recently, Chalmers’ Computer, IT and Electronic engineering programmes started a reform of their programmes and much of the pedagogical concept that they have chosen to introduce is inspired by CDIO, and by the M programme.
In parallel with the completion of the CDIO implementation, the programme realized that it needed to find new challenges and improvement ideas, in order to continue a positive development. It was decided to create a new vision for the programme based on the requirements on engineers graduating in 2020.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Future plans for the M programme – M2020
As a result of the recognition as Centre of Excellence in Higher Education 2008 [14], the M programme was given an opportunity by the Chalmers’ presidency to further develop. To set up the direction for the developments the program management together with the advisory board arranged a two-day workshop Shaping the Future of Mechanical Engineering Education at Chalmers to discuss the requirements on an M engineer graduating in 2020 and to create a road map for the program [24]. The workshop comprised expertise in engineering education from academia, industry, trade and government from Asia, Europe and North America together with teachers and students from Chalmers.
The panel suggested directions for the long-term program content and stated that teaching methods need to be reformed. The programme vision needs to be updated to match the requirements on an engineer graduating in 2020. The role of the teacher will be transferred to more of leading a dialogue with students than a traditional lecture giving traditional lectures. The panel argued the need to make the education more efficient and to use interactive learning and assessment to a larger extent. The curriculum needs to be more flexible in order to accommodate changes. New technologies and materials (e.g., micro/nano, bioengineering, IT) must be included in the curriculum and used in design-build-test projects. The process of integration of sustainability issues in the curriculum needs to be continued and extended. The projects based education needs to be further developed to include more training of creative abilities, innovation and entrepreneurship. The process of integration of sustainability issues in the curriculum needs to be continued and extended. The panel concluded that the engineers of 2020 need to be broadly educated with specialists skills within an narrow field and be able to take leadership in the transfer to a more sustainable society. Further, they need to be well prepared for the global labour market and the global competition.
The work with the long-term development of the program is running as an active project with strong involvement of the advisory board and the students. An international review committee is appointed to monitor the program’s progress and to provide advice and feedback,
In the short term, the program develops according to a continuously improvement philosophy based on CDIO. Intellectual properties rights will be regarded as a generic competence taught in integrated fashion in the project courses. A new CAD course will be launched next semester. The course includes a complete design chain, from sketching, 3D modelling, drawing and creating data as the basis for rapid prototyping to produce a physical model. The material science and manufacturing technology courses are being re-designed. The new courses will have a product focus and thus be more integrated in the program. The sustainable development education is strengthened through a separate course that includes more about materials selection and lifecycle analysis. The course will be strongly connected to the second year Integrated Design and Manufacturing project course. Further a plan for more distinct integration of sustainability issues is under development. The mathematical education is continuously improved and the cooperation between fundamental mechanical engineering courses is strengthened. Next semester will the first year mathematical course use a virtual learning environment for training, exercises and tests. The emphasis is thus made on individual work of the students complemented by teaching support in computer labs.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 EVALUATION
In this section, we summarise data from a number of evaluations that the programme has undergone during the period.
The original goals and their fulfilment
As described above, the programme began its CDIO reform in 2000-2001 with planning and some pilot experiments. The bases for these activities were two: The first was the then-current state of the programme where the M2000 reform had led to the introduction of early engineering experiences and more projects and new profiles. The second basis was the CDIO concept as it was described at the time. In the original project proposal from 2000 [7], the three basic goals of CDIO are identified along with four areas that need to be addressed in order to meet these goals: curriculum, pedagogy, assessment and workshops. In the proposal, some sub-areas to the four main areas are identified, but the only codified element of the CDIO model that existed at the time was the CDIO syllabus [25].
The original goals for the M programme reflected the opportunities in the CDIO project (with its large initial amount of funding) to realize certain ideas of the M2000 reform that were not realizable then due to financial constraints. Focused goals included the introduction of design- build-test learning experiences and the physical infrastructure to support them. Table 3 summarises the CDIO “model” of 2000 and the M programme’s planned efforts to realize the goals, as described in the original proposal [7] and a follow-up proposal [8], and the achievements to date.
From Table 3, and the description of the programme’s evolution since 2000, it can be noted that many of the original goals have been met and are now key features of the programme, including design-build-test experiences, the prototyping lab, and early engineering experiences.
Some original goals have not been met, notably some that related to technology. The IDE studio was envisaged as a state-of-the-art virtual engineering lab, enabling video-conferencing collaboration between geographically distributed student teams. However, the technology was not mature when the studio was opened (2003) and the experiments were not successful. The studio was subsequently closed. We also failed to find teacher who were interested in applying the electronic system for in-class feedback in their courses. The system was therefore not acquired.
Some goals have been realized but took longer time than planned. These include the construction of the prototyping lab and the reform of the mathematics. The first was due to a rent conflict between Chalmers and its landlord. The second was due to that we first failed to find a mathematics teacher who was interested in reforming the mathematics courses.
It is further noticeable that several elements have been added to the CDIO reform (noted as accomplishments for which the mid-column box is empty): These include key features of the M programme today that have been implemented after 2000 (e.g., the systematic programme design, integrated learning of communication and teamwork, sustainability, programmatic assessment) were poorly visible both in the CDIO model anno 2000 and in the M programme’s plan of that date. An important factor is that the knowledge and skills needed to develop these elements have emerged as a result of CDIO project, rather than planned from the start.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Table 3 An early version of CDIO “model”, the initial goals of the M programme, and the main achievements during the process
CDIO model 2000 Initial M programme goals Main accomplishments Curriculum Benchmark goals and curriculum against Benchmarking reported in [15] and [16] - CDIO syllabus CDIO syllabus - Early engineering Develop sequence of design-build Design-build-test learning experience experiences experiences sequence stretching through all year - Disciplinary linkages developed - Design build Develop course in systems engineering Not accomplished experiences - CDIO skills education Early engineering experiences Reformed introductory course Link courses in mathematics and Collaboration between courses in engineering science mathematics, mechanics and strength of materials Systematic approach for programme goal-setting and design developed and applied [23] Own computation oriented mathematical education with focus on applications developed [20] Teaching & learning Introduce mud cards techniques In use in some courses methods Study teaching & learning in large groups Accomplished - Concrete and hands on learning Develop courses for faculty in CDIO skills Courses in group dynamics and project - Problem formulation work model arranged - Active learning Active, problem formulation focused Matlab introduced, accepted and used - Feedback learning of mathematics as general simulation tool - Pedagogic scholarship Planned learning sequence for - Faculty skills integrated learning of communication and teamwork Integrated learning of sustainability Assessment Develop methods for assessing creative Creative aspects are assessed in - Clear and measurable skills design-build-test projects but no disciplinary goals common method has been adopted - CDIO skills assessment A comprehensive programme quality - Creative skills assurance system has been developed assessment and introduced - Programmatic The programme advisory board has assessment taken a very active role in programme development and follow-up Programming courses with final examination on-line in computer lab Workshop Develop physical prototyping lab Achieved, and used in many courses - Browsing laboratories - Sharing of research Develop IDE studio Studio was built but failed to meet laboratories expectations - System/product Utilize electronic system for in-class Not accomplished realization labs feedback
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 The effects of the reform programme and indications of its quality are also evidenced in high- level success indicators such as:
• The industry is contacting the programme for cooperation as well as for hiring students • The highest number of first priority applicants of all mechanical engineering programmes in Sweden • Comparatively low number of student drop outs from the program • Comparatively high rate of completed degrees • Teachers interested in pedagogy, course development and reforming education are looking to the programme and are eager to teach and participate in the development of the program • Recognitions for high quality from alumni, industry, universities all over the world, the engineering union as well as the employers’ associations • The high and recognized quality of the proposed solutions in project courses including the reports and presentations
Internal evaluations
Below, we will discuss results from the internal quantitative evaluations that we have conducted during the period: a set of CDIO self-evaluations and an alumni survey.
CDIO self-evaluation evolution
Throughout the period, the CDIO self-evaluation tool ([1], chapter 9) has been used regularly to monitor, guide and visualize the evolution of the programme.
A CDIO self-evaluation implies a valuation of the programme’s status vs. fulfilling the twelve CDIO standards. A five-level rating scale ranging from 0-4 is used. A “0” implies that there is no implementation of the standard, whilst a “4” implies that there is a complete programme-level plan for the standard, that it is comprehensively implemented and programme and course level, and that the implementation has been evaluated and improved.
Table 4 Progress of M programme CDIO self-evaluations during the period 2000-2010
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 The ratings for the M programme at different points in time are shown in Table 4. Initially, the programme’s fulfilment of the CDIO standards was low. For many of the standards, the rating was “1”, which should be interpreted as that some experiments and pilots were already on- going, but there had not been a systematic attempt to adapt a set of principles for education design. The highest rating was given for standard 5 – Introduction to engineering. Such a course had been developed as part of the M2000 reform programme.
In the first few years, a major focus was the development and implementation of a set of design- build-test learning experiences and the prototyping lab, as is indicated in the higher ratings for standards 4-6 from 2003. In the period 2003-2005 there was an additional focus on integrated learning of communication and teamwork skills. Programme context, goals and quality system were implemented in 2005-2008. During the last period of the chart (2008-2010) assessment of CDIO skills has been focused.
The chart also highlights some areas in which the programme has not been able to reach the requirements for a “4” rating – essentially standards 9 and 10. The standards relate to faculty competence. In the buyer-seller system that Chalmers applies, faculty competence is the responsibility of the departments. The programmes thus have limited powers to influence hiring and promotion.
One of the programme’s major development efforts during recent years is the implementation of a set of computer-based mathematics courses. This effort has been substantial but only influences the “active learning” standard (8). This aspect, and the circumstance that the standards only have five levels and thus the rating tends to plane out even though improvements are still made indicate the need to complement the CDIO standards self- evaluation with other methods if the method is to fulfil its purpose, i.e. to guide and visualize the evolution of a programme.
Alumni survey
If a CDIO-based programme is effective, the results should ultimately be discernable in alumni surveys that students complete some years after graduation. A fundamental problem with using alumni surveys to evaluate education reform is the time lag. The first M programme students who have gone through a “complete” CDIO programme graduated in 2009, and Chalmers will not survey their views on their education until 2012. Below, we will discuss data from Chalmers’ alumni survey from 2009 that was sent to students who graduated in 2006 [26]. In our case, students take on average 5.7 years to complete their degree. This means that many students who graduated in 2006 commenced their studies prior to the CDIO project. However, as CDIO learning experiences have been phased in across all years of the programme, they will have undergone at least a partial CDIO programme. It is therefore likely that they have been affected by the CDIO programme, and it is relevant to examine if any measurable effects can be found in the alumni survey.
Chalmers’ alumni survey comprises about 40 questions related to the background of the respondent, first job, current job, the importance of the respondent’s education for the job, the respondent’s views on his/her education, and a self-evaluation of his/her knowledge and skills in the areas of the education. On an overall level, the alumni survey shows that Chalmers’ graduates are satisfied with their education and that their employability is very good. As strong points of their Chalmers experience, they typically point to specific courses, along with the breadth of the education. As areas in which Chalmers can improve, the survey respondents point out contacts with employers and teacher’s pedagogical competence.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Tables 5 and 6 present some of the alumni survey data where the responses from graduates from the M programme are compared with those from the average from all of Chalmers Civilingenjör degree programmes. Table 5 shows the responses to a question in which the respondents were asked to provide a self-evaluation of their knowledge and skills related to a number of goals for the education. Table 6 shows how satisfied the graduates were with their Chalmers education. A scale ranging from 1-10 is used. For many of the attributes, the M graduates are similar to those of other Chalmers graduates. However, there are some dimensions for which the differences are significant including teamwork, communication, contact with employers, and to some degree design. These are dimensions that are targeted by the CDIO approach.
It remains to see if these differences are robust over time. An uncertain factor is how the overall experience of a CDIO programme can be discerned in the presence of major external variations, e.g., the business climate. However, the results from our alumni survey are principally similar to those from other CDIO programme-level evaluations, for example Linköping University’s longitudinal survey of the CDIO project and their Y programme [6]. The alumni survey further provides a basis for evaluating if the goals of the CDIO reform are relevant by alumni. This seems to be the case: When asked “What can Chalmers improve?”, the most frequent answer is “contacts with industry”. When asked if they missed something in their education, the most common subjects are economics, project management, organization and leadership. At the same time, when asked what could be removed to make place for other subject, the most common answer is “nothing”. This is a clear illustration of the need for dual learning, where subject knowledge and professional skills are co-developed. From Table 5, it seems that the M programme is able to strengthen the learning of professional skills (teamwork, communication, design) without compromising the learning of subject knowledge (mathematics, science and disciplinary knowledge). Again, it must be emphasized that the data is still very preliminary and given by graduates from a partial CDIO programme.
Table 5 Self-evaluation of alumni’s knowledge and skills (Only a subset of the surveyed knowledge/skills is displayed. Discipline-specific knowledge refers to engineering knowledge in the discipline of the programme, e.g. electrical engineering)
Knowledge/skill Mechanical engineering Average for all Chalmers graduates Civilingenjör graduates Mathematics 7.1 6.9 Science 7.1 6.9 Discipline-specific knowledge 7.7 7.4 Design 7.2 6.8 Teamwork 7.7 7.1 Communication 7.5 7.0
Table 6 Alumni survey respondent’s evaluation of their Chalmers education
Attribute Mechanical engineering Average for all Chalmers graduates Civilingenjör graduates Overall satisfaction 7.8 7.6 Employability 8.6 8.3 Contacts with industry 5.4 4.5
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010
Faculty perceptions - Catalysts and barriers for spreading CDIO to other programmes
In order to identify faculty perceptions of CDIO, loosely structured group interviews were performed with three groups of faculty:
• Members of Chalmers’ university-wide pedagogy committee. These persons have not taken part in our reform, but have a strong interest in pedagogy. • Leaders from the group that are currently reforming Chalmers’ Computer, IT and Electronic engineering programmes. This basic ideas of this reform are CDIO-inspired, but the design of the programmes is not finalized at this time. • Faculty members from the mechanical engineering programme who had taken part in the CDIO reform and are active teachers in the programme.
All interviewees were queried about their perceptions of CDIO and asked to point out catalysts and barriers for its adaptation.
A common response from all interviewees was that the perceived focus of CDIO on the professional role of engineers is the most important catalyst for its adaptation. “- CDIO forces us to reflect on what the respective profession looks like”.
Further, the interviewees emphasized the value of the strong structure of CDIO, pointing out that even though they had projects of industrial relevance and a pedagogical aim before, CDIO helped substantially with process and structure issues; e.g. creating an organized set of learning objectives and providing a language for faculty to communicate about the programme and its goals and content. The pedagogy committee members pointed out that the structure of strategy of CDIO facilitates the integration of learning of generic competences in the curriculum.
The programme leaders from the programmes currently under revision appreciated the notion of CDIO as a toolbox of relatively independent parts, from which certain parts could be picked and adapted, whilst others were not chosen.
The pedagogy committee members, however, also pointed out that the strong structure could have negative associations, implying a top-down programme design that would constrain the autonomy of individual faculty members. However, the programme leaders did not see this as such a barrier. These contrasting statements reflect a critical issue in successful programme development – the need to have a strong programme level to be able to integrate non- disciplinary knowledge and skills in the curriculum, but also the need to provide the faculty with a sense of ownership of the programme.
Another barrier for the adaptation of CDIO is constituted by the interpretation of the CDIO concept in the disciplinary context of the programme. The pedagogy committee members, in particular, pointed out that when an idea is labelled, like CDIO is, there is a strong risk for introducing a variety of misconceptions. It can be brought together under the expression “CDIO of what?”. The definition of “the product” of the programme can be difficult to agree on as well as the conceive-design-implement-operate process. There is a barrier in reaching agreement on this fundamental viewpoint of the education: Programmes with a strong emphasis on preparing for a research career may have serious objections towards the forwarded position of “design” that has motivated much of the CDIO reform. The mechanical engineering faculty interviewed had also experienced this ambivalence towards CDIO. For product development and automotive CDIO was more or less tailor-made and did not infer a need to change attitude. However, there
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 were faculty who disliked CDIO from the start, especially those who wanted to emphasize the specialist role and competence of the graduating engineers higher than generalist skills needed to view their subject in a new manner.
The general opinion of the interviewed mechanical engineering faculty was that faculty was as a whole content with the reform. In particular, the reform was appealing for those who were interested in pedagogy. For example, the Standards raised the pedagogical awareness. “- I learned that it was about what the students do, not what I do”. They suggested that students of today act much more engineering oriented.
The mechanical engineering faculty suggested that the main barrier for the CDIO reform was that teachers needed to work in a new manner, differing from their earlier experiences. “- But our faculty have not been working engineers, that is probably a reason why they are hesitant”.
They further identified a number of pedagogic challenges: e.g., regarding how to act in project based courses; to differ between success in learning and success in project task: – “they work so hard in the project that they do not even buy or read the course literature”. They concluded the discussion by agreeing that many barriers were linked to pedagogical improvements known to be difficult already, but the CDIO reform had officially problematized the issues. –“Before you did your written exams and was content with that”.
The results from the faculty interviews can be summarized as follows:
Catalysts for CDIO adaptation:
• An engineering concept with the profession as the core focus for the education • The structured education concept • The strategies for integrating generic competences into the education • The toolbox, CDIO contains many parts that can be used independently
Barriers for CDIO adaptation: • The structured CDIO education concept implies a stronger programme level and a top-down perspective • Interpretation and translation issues. CDIO is a comprehensive model with risks for differing perceptions and misconceptions • Programmes and faculty members that are strongly oriented towards preparing for a research career or an analytical profession may feel that CDIO lacks appreciation of analysis. • CDIO involves significant changes for both teachers and students.
Student experiences and perceptions
A structured group interview with six (four male and two female) third year students at the M programme was performed. The interview was focussed on evidence of CDIO, general and professional skills.
In general the students were very satisfied with the M program. This is also verified by a recent survey about the study environment. About 95 % of the students declared that they were very satisfied or satisfied and less than 5% declared that they were not so satisfied or not satisfied at
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 all. They pointed out that an environment conducive to studies and social activities are of highest importance for a successful programme.
The students related CDIO strongly to the design-build-test project courses and the fact that they make use of their fundamental skills and competences in mathematics and mechanics. The students found the project courses important, relevant and stimulating. They stated that the courses add realism to the education and that they train their creative abilities in a natural manner. Further, the courses connect theory to practice and enforce the students to utilize and practice knowledge and skills from earlier fundamental engineering courses
The students strongly appreciated the strong link and progression between the courses but pointed that if one course fails, the consequences in following courses may be severe. The students also pointed out that the strong focus on project courses and the continuous assessment and grading in the courses make the students work very hard and sometimes to hard. They may put less effort on the course taught in parallel and experience high pressure and stress.
The students expressed that they are well trained in and prepared for work in teams. In multidisciplinary projects with students from other engineering programmes they have noticed that they are much better prepared to work in teams. Consequently, they had noticed that M students take leading roles in such projects. The students expressed strong confidence in their report writing skills and considered it as a natural component in the learning process.
The students considered it natural to use the computer and programming in the basic course of mathematics. The students showed a very positive and constructive approach to mathematics. The students further stated that they had had use for all mathematics taught in later courses and projects. They appreciated the training in solving more open general problems rather than repeating known solutions to very specific problems. They strongly felt that the M programme’s mathematical education prepared them better for studying advanced level M courses as well as for their professional careers as engineers. They saw their mathematical education as a natural part of their CDIO based mechanical engineering education.
The students expressed strong support for the continuous improvements philosophy of the program and the strong and active student involvement.
Finally, the students showed much self-confidence and had high expectations on their upcoming master’s programmes. They strongly believed that the CDIO based M programme have prepared them well for work as a professional engineers as well as for research studies within the field of their specialization. Further, they think that their employability is very strong because of the CDIO-based education.
Costs
A typical issue in the investigation of a CDIO-based education is that of cost. There are concerns that a CDIO-based education is prohibitively expensive, both in terms of initial investment costs and in terms of operating costs. Indeed, a CDIO programme requires certain learning environments and some courses will be more teacher-intensive and thus more expensive. However, earlier research has indicated that there is a wide range in costs for design-build-test project courses, from 1.0 to 1.8 of that of an average course at the institution in question [27]. In our case, we had the benefit of external financial support for a major part of the initial
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 investments. Investments after the first five-year period and operating costs through the whole period have been financed via our regular budget through redistribution of money.
Investment costs
The investments in the Chalmers M programme case can be split into physical infrastructure and programme and course development. The physical infrastructure investments were mainly to (a) a prototyping lab and (b) a “study hall” for individual and group studies. The latter was not technically part of the CDIO project but is included here as it fills a role that otherwise would have been important for the CDIO project to address, namely that of socializing workspaces for students [28]. Table 7 summarises our investment costs.
Operating costs
The operating costs for the prototyping lab and the more teacher-intensive CDIO courses have throughout been covered by the programme’s ordinary budget. These costs can for 2010 be estimated to 4.3 MSEK for the prototyping lab and 1.2 MSEK redistributed money to teacher- intensive courses, totalling 5.3 MSEK per year. These costs need to be related to the overall budget for the programme. The programme has about 1100 full-time students. For each student, the Swedish government pays approximately 91,000 SEK. Chalmers total income for the programme is thus about 100 MSEK. From this amount, about 30 % is allocated to central costs (administration, IT, certain facilities costs etc). The programme thus has about 70 MSEK at its’ disposal to buy courses from departments. In conclusion, the programme finances its CDIO workspaces and activities by re-distributing about 8 % of its total budget, as compared to a programme that would not contain any CDIO elements at all.
Table 7 Investment costs for CDIO reform of Chalmers mechanical engineering programme. All numbers are in MSEK. 1 SEK = 0.139 USD / 0.103 EUR (March 26, 2010).
Time period Physical infrastructure Programme and course Total development Prototyping lab Study hall Period 1 (2000-2005) 9.8 10 7.2 27 Period 2 (2006-2010) 1.5 - 3.1 4.6 Total 21.3 10.3 31.6 Average 2.1 1.03 3.16
External evaluations
In 2005/2006 the Swedish National Agency for Higher Education (HSV) evaluated all Swedish five-year engineering programmes (Civilingenjör programmes) [13]. The evaluation was based on the CDIO principles. The National Agency concluded that CDIO is a favourable model for engineering education and a tool for developing and reforming educational programmes. The evaluation process consisted of a self-assessment and a site visit. The M programme was highly appreciated in the evaluation. In particular, the work with the CDIO based program description with the connection between the programme goals and the learning outcomes of the courses and the integration of general engineering skills were recognized as outstanding and inspiring.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 In 2008 universities and other institutions for higher education in Sweden were invited to apply to the Swedish Agency for Higher Education for recognition as Centre of Excellence in Higher Education. Internally, the presidency at Chalmers pointed out its M programme to apply for the recognition, which resulted in an application from the programme [11].
The applications were evaluated by an international expert panel. Following the evaluation three applications were chosen by the National Agency for detailed assessments and site visits. In conformity with the expert panel’s proposal, the National Agency appointed the M programme as Centre of Excellence in Higher Education 2008 [14]. The National Agency and the expert panel appreciated in particular:
• The Chalmers ‘‘buyer-supplier’’ organization, • The strong and devoted management team and advisory board, • The study environment, • The elaborate quality assurance system, • The strong teaching and the support of the students’ learning processes, • The strong involvement of the students in the running and development of the program, • The integrated curriculum structured around learning outcomes and competences with emphasis on professional skills as well as fundamentals (CDIO) and, • The close links to industry and research.
Finally they conclude that over time, the concepts of integration, a holistic view, and system as well as process thinking have become best practice through a focus on design-build-test projects following the CDIO model and the combination of this with the organisational structure and systematic quality system.
Discussion
This section has brought together the evaluation data that we have gathered over the period. We can argue that the main goals of the reform have been met, but perhaps more importantly, that new goals have emerged during the process. The sources for these new goals are both a more in-depth understanding of the CDIO concept and internally initiated initiatives, such as mathematics reform and sustainability. The emergence of new goals and elements during the process is characteristic for a sustained education reform, characterised by continuous improvement over a long time period. The views from government agencies, faculty and students provide support for a claim that the programmes holds a high quality, has evolved positively over the period, and that it has influenced other programmes at Chalmers. Some barriers to adaptation of CDIO are also identified.
CRITICAL SUCCESS FACTORS FOR SUSTAINABLE EDUCATION REFORM
This section aims to synthesize our experiences and proposes a set of critical success factors for sustained education development. A critical success factor (CSF) is the term for an element that is necessary for an organization or project to achieve its mission [29]. The set is based on the lessons that we have learned over time combined with recommendations from the literature, including Kotter [17], Crawley et al. [1], chapter 8 and Lissack and Roos [3]. The critical success factors are summarised in Figure 1. Let us now discuss these starting from the top.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Clear purpose, goals and strategy. A programme needs an identified purpose in order to be able to elaborate clear goals and to be able to create and communicate strategies for addressing those goals. A CDIO programme has a stated purpose in educating future engineers who can take leading roles in the conception, design, implementation and operation of products, processes and systems. Clear purposes, goals and strategies are also essential for shaping and discussing the identity of the programme.
Ability to continuously set new challenging goals. However, clear goals are not enough to drive a sustained education development process. Initial goals will at some point be met, and the programme will need to find new goals to direct a continued development. This is particularly challenging for a successful programme, where there is no crisis that can be used to motivate a change. Aspirational goals need to be set. This is in line with Lissack and Roos [3], who argue that a good strategic process is cyclic, continuously revising goals and actions. The ability to set new goals also hinges on openness to the outside world as a source for ideas and benchmark.
Strong programme level. Given the goals and strategy, the next step is system-level design. An engineering programme needs to form a coherent whole and address goals for disciplinary knowledge as well as generic skills. A strong programme level is essential for designing in multidisciplinary topics in the programme, and for resolving conflicts between different disciplines. A strong programme level is also essential for monitoring the programme as a whole.
Well-defined and motivated changes. A programme will during a long-term development process undergo many changes, as is illustrated by the M programme. One by one, these changes need to be well-defined, delimited and anchored in the overall strategy.
A purposeful quality assurance system. A quality assurance (QA) system that monitors and guides the development of the programme is essential. The QA system further needs to be designed so that it measures the intended development. For CDIO programmes, the CDIO standards self-evaluation tool is a central part of the QA system. However, it needs to be complemented with other tools in order to obtain a comprehensive view.
Continuity and multi-year perspective. Many of the changes implemented in the M programme have taken three years or more to design, implement and refine. Long-term change processes require a degree of stability with respect to rules, organisation and staffing. The programme management and the faculty involved need to be aware of the time perspective of the change and not to view the effort as a one-off, nor their engagement as short-term. Over time, the faculty involved in the M programme have also come together as an effective cross- departmental team. This facilitates the introduction of new education innovations that require collaboration between courses and a willingness to take responsibility for programme goals that lie outside of the disciplinary aims for a specific course.
Faculty competence development. Many of the educational changes implemented in the M programme rely on faculty competence that was not available in the start of the project. New and changed faculty competence needs range from the professional (ability to run design-build-test projects) to the multidisciplinary (sustainable development) to the pedagogical (active learning methods). In this context, there is a risk that the reform becomes dependent on a few teachers with critical competence. If one of those teachers leaves the university, there may be no one left who can take on the course. A comprehensive plan for faculty professional competence will provide more teachers to take responsibility for, for example, integration of sustainability aspects in the courses, making the programme less vulnerable.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 Clear purpose, goals and Close strategy Ability to contacts with continuously stakeholders set new goals & employers
Strong Management programme support level
Sustained Active student Well-defined participation education changes development
Empowered Purposeful change quality agents system
Continuity & Faculty sense multi-year of ownership perspective Faculty competence development
Figure 1. Critical success factors for sustainable education reform.
Faculty sense of ownership of the programme. We highlighted the importance of a strong programme level above. However, this needs to go hand in hand with endowing the faculty with a strong sense of ownership of the programme. There is little that a programme manager can achieve without support from faculty: Enthusiastic individuals can realize small, specific changes but long-term changes that are implemented across the programme require broad support.
Empowered change agents. The programme manager needs to be empowered to drive the change process. Elements of this empowerment include funding for development efforts, moral support from management, and the mandate to decide on certain changes. On an individual level, some faculty will take the lead and will be more willing to experiment with their courses. They must be supported in several ways, not least financially and given time to develop their teaching.
Active student participation. Active student participation is essential for education reform, not only as users and evaluators of proposed and implemented changes. Students are only sources for educational innovations that faculty might not come up with: In our case, the Study Hall was originally a student proposal for an improved learning space. Students can also require that an innovation that they are exposed to in one course is picked up by other course. Further, the cultural element of education reform is not limited to changing faculty attitudes. The students of
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 a programme also tend to form a culture with a strong ethos with conceptions of what the programme is about that can be very challenging to change [30].
Management support. Education reform requires programme-level changes and cannot be carried out only on a course level [1]. The programme head will need to balance demands from different disciplines. This can cause conflict. The programme head needs support from management to drive changes in these circumstances. Reform will further require some investments. Allocating money to investments in education development is a key responsibility of university management. University management also needs to understand the time perspective of education change.
Close contacts with stakeholders & employers. Finally, sustained education reforms is very much dependent on influences from the outside. Close contacts with employers will challenge as well as support the education, if it is responsive to employer’s needs. Other stakeholders such as accreditation agencies pose other requirements. In the CDIO Initiative, the collaboration with other universities has provided a continuous source of ideas, perspectives and benchmarks.
CONCLUSIONS
This paper has summarised and evaluated Chalmers University of Technology’s M programme’s evolution during a ten-year period when it has developed, implemented and refined a CDIO- based education.
During the period, the programme has introduced a large number of educational innovations. Some of the educational innovations were driven by goals set in the beginning of the period, but others have been driven by insights gained and external factors during the process.
Internal evaluations such as CDIO self-evaluations, alumni surveys and student interviews and external evaluations such as those conducted by the Swedish National Agency for Higher Education verify that the programme has developed positively during the period and holds a high quality.
The CDIO model has provided the programme with a number of strategies and tools that have been essential for this development: a clear vision and strategy, the professional role of engineers as the focus of the education, a toolbox of adaptable learning experiences, and a quality assurance system that has guided the process. Through working with CDIO, the programme has also established a structure and working practice that has facilitated the introduction of other educational innovations, e.g. in the area of sustainable development.
The approaches developed by the M programme have been spread to other Chalmers programmes in two different ways: Some programmes are adopting CDIO as an idea and are using it to shape their programme goals and content. More broadly, CDIO is adopted as a toolkit. CDIO tools are used to document and communicate programme goals, ideas and structure also for non-CDIO programmes. The main driving factors behind these adoptions are the programme-level thinking of CDIO, its focus on the professional role of engineers, and its applicability as a strategy for integrating learning of generic competences in a programme.
However, CDIO is also challenged by certain barriers to its implementation including its perceived top-down focus and difficulties in interpreting the CDIO concept in certain disciplinary
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 fields and for programmes with a strong research orientation. Lacking faculty competence in and experience of practical engineering work is also pointed out as a barrier to CDIO adoption.
Finally, the paper proposes a set of critical success factors for sustainable education development.
Acknowledgements
The Knut and Alice Wallenberg Foundation financed the first phase of this development project (2000-2005). This support is gratefully acknowledged.
REFERENCES
[1] Crawley, E. F., Malmqvist, J., Brodeur, D. R, Östlund, S. Rethinking Engineering Education – The CDIO Approach, Springer-Verlag, New York, 2007.
[2] The CDIO Initiative, CDIO Home Page, www.cdio.org, accessed on April 20, 2010.
[3] Lissack, M., Roos, J., “Be Coherent, Not Visionary”, Long Range Planning, vol 34, pp 53-70, 2001.
[4] Bennich-Björkman, L., Organizing innovative research: The inner life of university departments, Elsevier Science, Oxford, 1997.
[5] Edvardsson Stiwne, E., “The First Year as an Engineering Student – The Experiences of Four Cohorts of Engineering Students in Applied Physics and Electrical Engineering in Linköping University”, Proceedings of the 1st International CDIO Conference, Kingston, Canada, 2005.
[6] Edvardsson Stiwne, E., Jungert, T., ”Engineering Students Experience of the Transition from Studies to Work”, Proceedings of the 3rd International CDIO Conference, MIT, Cambridge, Massachusetts, 2007.
[7] Chalmers University of Technology, Royal Institute of Technology, Linköping University and Massachusetts Institute of Technology., Improved Engineering Education – A Project to Make the Conception-Design-Implementation-Operation – CDIO – the Context of Engineering Education, Funding application (revised version) submitted to the Knut and Alice Wallenberg Foundation, 2000.
[8] Chalmers University of Technology, Royal Institute of Technology, Linköping University and Massachusetts Institute of Technology., Application for Funding for Program Years 2-4 of the Wallenberg CDIO Program for Engineering Education, Funding application (revised version) submitted to the Knut and Alice Wallenberg Foundation, 2001.
[9] Malmqvist, J., Östlund, S., Edström, K., “Using Integrated Programme Descriptions to Support a CDIO Programme Design Process” World Transactions on Engineering and Technology Education, Vol. 5, No. 2, pp. 259-262, 2006.
[10] Enelund, M., Programbeskrivning för civilingenjörsprogrammet i maskinteknik (Programme Description for M.Sc. in Mechanical Engineeeing Programme), Chalmers University of Technology, Gothenburg, Sweden, 2009. In Swedish.
[11] Enelund, M., Bankel, J., The Mechanical Engineering Programme at Chalmers University of Technology – Application for Excellent Quality in Higher Education, Chalmers University of Technology, Gothenburg, Sweden, 2008.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 [12] Chalmers University of Technology., Course syllabi for M.Sc. in Mechanical Engineering Programme Course, http://www.student.chalmers.se/sp/programplan?program_id=570, accessed on April 20, 2010.
[13] Swedish National Agency for Higher Education (HSV)., Evaluation of Civil Engineering Programs at Swedish Universities and Institutions of Higher Education, Högskoleverket, Rapport 2006:31 R, Stockholm, Sweden, 2006.
[14] Swedish National Agency for Higher Education (HSV)., Centres of Excellence in Higher Education 2008, Högskoleverket, Rapport 2008:38 R, Stockholm, Sweden, 2008.
[15] Bankel, J., Berggren, K.-F., Blom, K., Crawley, E., Wiklund, I., Östlund, S., ”The CDIO Syllabus: A Comparative Study of Expected Student Proficiency”, European Journal of Engineering Education, Vol. 28, No. 3, pp 297-315, 2003.
[16] Bankel, J., Berggren, K.-F., Engström, M., Wiklund, I., Crawley, E., Soderholm, D., El Gaidi, K., “Benchmarking Engineering Curricula with the CDIO Syllabus”, International Journal of Engineering Education, Vol. 21, No. 1, pp 121-133, 2005.
[17] Kotter, J. P., Leading Change, Harvard Business School Press, Boston, MA, USA, 1996.
[18] Gustafsson, G., “Experiences from the Transformation of an Engineering Education Introductory Project Course”, Proceedings of NordDesign 2004 – Product Design in Changing Environment, Tampere, Finland, 2004.
[19] Evertsson, M., Bankel, J., Enelund, M., Eriksson, A., Lindstedt, P., Räisänen, C. “Design- Implement Experience from the 2nd Year Capstone Course “Integrated Design and Manufacturing””, Proceedings of the 3rd International CDIO Conference, Massachusetts Institute of Technology, Cambridge, MA, USA, 2007.
[20] Enelund, M., Larsson, S., “A Computational Mathematics Education for Students of Mechanical Engineering”, World Transactions on Engineering and Technology Education, Vol. 5, No. 2, pp 329-332, 2006.
[21] Knutson Wedel, M., Malmqvist, J., Arehag, M., Svanström, M., “Implementing Engineering Education for Environmental Sustainability into CDIO Programs”, Proceedings of the 4th International CDIO Conference, Gent, Belgium, 2008.
[22] Saalman, E., Peterson, L., Malmqvist, J., “Lessons Learned from Developing and Operating a Large-Scale Project Course”, Proceedings of the 5th International CDIO Conference, Singapore, 2009.
[23] Malmqvist, J., Arehag, M., “Experiences from Using Integrated Program Descriptions to Support Program Development”, Proceedings of 3rd International CDIO Conference, Massachusetts Institute of Technology, Cambridge, MA, USA, 2007.
[24] Enelund, M., Shaping the Future of Mechanical Engineering Education at Chalmers, Workshop documentation, https://student.gate.chalmers.se/sv/studier/programinformation/maskinteknik/shaping%20educatio n/pages/default.aspx, accessed on April 29, 2010, 2010.
[25] Crawley, E., The CDIO Syllabus: A Statement of Goals for Undergraduate Engineering Education, MIT CDIO Report #1, Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA, USA, 2001.
[26] Malmqvist, J. Chalmers Alumnienkät 2009 – riktad till arkitekter, civilingenjörer och högskoleingenjörer med examensår 2006 (Chalmers Alumni Survey 2009 – views from 2006 graduates), Chalmers University of Technology, Gothenburg, Sweden, 2009. In Swedish.
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010 [27] Malmqvist, J., Young, P. W., Hallström, S., Kuttenkeuler, J., Svensson, T., “Lessons Learned from Design-build-test-based Project Courses”, Proceedings of Design-2004, Dubrovnik, Croatia, 2004.
[28] Young, P. W., Malmqvist, J., Hallström, S., Kuttenkeuler, J., Cunningham, G., “Design and Development of CDIO Student Workspaces – Lessons Learned”, Proceedings ASEE-05, Portland, Oregon, USA.
[29] Daniel, D. Ronald, "Management Information Crisis," Harvard Business Review, Sept.-Oct., 1961.
[30] Edvardsson Stiwne, E., Roxå, T., “Programethos och dess betydelse för studenters lärande och personliga utveckling inom tekniska utbildningar (Programme Ethos and its Significance for Student Learning and Personal Development in Engineering Education)”, Proceedings 2:a Utvecklingskonferensen för Sveriges ingenjörsutbildningar, Lund Sweden, 2009. In Swedish.
Biographical Information
Johan Malmqvist is a Professor in Product Development and Dean of Education at Chalmers University of Technology, Gothenburg, Sweden. His current research focuses on information management in the product development process (PLM) and on curriculum development methodology.
Johan Bankel is coordinator for the Mechanical Engineering programme at Chalmers University of Technology in Gothenburg, Sweden. He has a M.Sc. in Naval Architecture, a Lic. Eng. in fluid dynamics and a teacher certificate in natural science.
Mikael Enelund is an Associate Professor in Applied Mechanics and head of the MSc programme in Mechanical engineering at Chalmers University of Technology, Gothenburg, Sweden. His current research focuses on modelling of viscoelastic materials and the development of damping devices for railway applications
Göran Gustafsson an Assistant Professor in Mechanical Engineering and vice head of the Department of Product and Production Development at Chalmers University of Technology, Gothenburg, Sweden. His research interest focuses on tools and techniques for lean product development.
Maria Knutson Wedel is Professor in Engineering Materials, Director of the Masters Programme in Advanced Engineering Materials and member of the pedagogical committee at Chalmers University of Technology. Her research is focused on characterization of engineering materials regarding the correlation between microstructure and mechanical properties and cross- disciplinary research on materials response to microwaves in recycling technology.
Corresponding author Professor Johan Malmqvist Department of Product and Production Development Chalmers University of Technology SE-412 96 Gothenburg, SWEDEN +46 31 772 1382 [email protected]
Proceedings of the 6th International CDIO Conference, École Polytechnique, Montréal, June 15-18, 2010
!"#$% &' ()#%"*$%%&+ (, --.../ 0 /-- !
0
1 2
3 ! " # $% &'() *+%(,(-$&./)$('/.
34
!
" # $ % & ' %
! & ( )
* ) % % &
+ % & * )& % &
, - #.+ /%&& % &% / % )))% / & # %& ' % 0 / #% 1' % + 2& Higher Education Research & Development Vol. 27, No. 2, June 2008, 95–106
Doing course evaluation as if learning matters most Kristina Edström*
The Royal Institute of Technology (KTH), Stockholm, Sweden TaylorCHER_A_280577.sgm10.1080/07294360701805234Higher0729-4360Original2008HERDSA272000000JuneKristinaEdströ[email protected] andEducation Article (print)/1469-8366Francis 2008 Research Ltd & (online)This Development paper investigates barriers for using course evaluation as a tool for improving student learning, through the analysis of course evaluation practices at The Royal Institute of Technology (KTH), a technical university in Stockholm. Although there is a policy on development-focused course evaluation at KTH, several stakeholders have expressed dissatisfaction with its poor results. Interviews were conducted with faculty and student representatives to investigate the perceived purpose and focus of evaluation and its current utilization. Results show that evaluation is teaching- and teacher-focused. As course development is not in the foreground, evaluations merely have a fire alarm function. It is argued that course evaluation should be regarded as a component of constructive alignment, together with the intended learning outcomes, learning activities and assessment. Finally, the concept system alignment is proposed, extending constructive alignment to the institutional level. The evaluation task can generally be said to be: 1. to describe what actually happens in that which seems to happen 2. to tell why precisely this happens, and 3. to state the possibilities for something else to happen. (Franke-Wikberg & Lundgren, 1980, p. 148) Keywords: course development; course evaluation; evaluation policy; system alignment
Introduction Current evaluation policy Since 1997, The Royal Institute of Technology (KTH) in Stockholm has a policy stating that a course evaluation, called course analysis, must be undertaken for each course (KTH Handbook 2). A course analysis consists of: – Quantitative data (number of students registered, completion rate). – Students’ views on the course, appropriately documented, for instance through a questionnaire, minutes from a meeting with student representatives, or interviews. – An analysis by the teacher, with a brief comment on the quantitative data and the results of the survey, including proposed measures and deadlines. The course analysis should clearly show a course’s development from one year to the next. The course analysis should be communicated to the vice dean of education and the dean/board of the school. However, there is widespread dissatisfaction with the course evaluation system.
*Email: [email protected]
ISSN 0729-4360 print/ISSN 1469-8366 online © 2008 HERDSA DOI: 10.1080/07294360701805234 http://www.informaworld.com 96 K. Edström
● In 2005, the student union invited management and educational developers to discuss the results of course evaluations. The student representatives wrote: ‘The feeling among the students is that the results of course-level quality processes are small in relation to the effort spent by teachers, students and administrators.’ (Lindbo, 2005) ● The Swedish National Agency for Higher Education criticised KTH when investigating course evaluation practices (Swedish National Agency for Higher Education, 2006). The agency found a lack of feedback of evaluation results to students and identified a need for guidelines to be applied consistently across the university. ● Many teachers complained that the compulsory evaluations constitute a meaningless burden. Some suffer from being exposed to inconsiderate student comments, even to the extent that it can be a workplace health issue.1
The igniting spark for this study The author teaches a staff development course in which faculty develop their own courses using the concept of constructive alignment (Biggs, 2003). It contains a workshop around the motto: ‘Evaluation is often viewed as a test of effectiveness – of materials, teaching methods or whatnot – but this is the least important aspect of it. The most important is to provide intelligence on how to improve these things’ (Bruner, 1966, as cited in Ramsden, 2003, p. 223). Participants bring their own evaluation questionnaires to the workshop. These almost invariably have a focus on teaching, especially the teachers’ performance, often asking the students to rate the teachers. At the workshop, participants formulate new questionnaires intended to inform course development. Drawing on what has been studied in the course previously, the focus is now on investigating what the student does; for example, volume, timing and appropriateness of studies, and indica- tions of approaches to learning.2 Evaluation can then be seen as a tool for course development. The first time this workshop was planned, the author considered it a straightforward appli- cation of ideas and principles, of which the participants already have shown a firm grip. But participants were very surprised and excited over this novel way to use course evaluation. One participant exclaimed: ‘This takes the venom out of course evaluation!’ The episode sparked the author’s interest in how evaluation practices influence conditions for teaching and learning, and the present study is a first step in learning more about views on and utilization of course evaluations.
Investigation Interviews were undertaken with two elected student representatives and six teachers, two of whom are or have been directors of study. Two additional teachers were interviewed but they were not included in this analysis as they had participated in the author’s staff development course. All respondents were active in the same engineering programme. The study was an open and explorative investigation into the views on course evaluation held by teachers and students, and there was no intention to classify individual respondents according to any pre-existing model or theoretical framework. Respondents were asked to bring a course evaluation questionnaire (teachers and students) and their latest course analysis document (teachers). These documents were on the table during the interviews and were read and discussed as part of the interview. The purpose of using their own course evaluations as a starting point for discussions was to focus on respondents’ actual practice (theory-in-use), rather than their general views on things (espoused theory). Inspired by Kvale (1997, p. 122), interviews were conducted as loosely structured conversations using open questions such as: ‘Please take me through the evaluation data and explain it to me’; and ‘How do you think the course could be improved?’ Higher Education Research & Development 97
The underlying research questions were:
● What purpose are course evaluations perceived to serve? ● What focus are course evaluations perceived to have? ● How are course evaluations utilized? ● What are the respondents’ views on teaching and learning?
Each interview lasted approximately 45 minutes and was recorded and transcribed. From the transcripts, recurring themes were identified through a process of iterations between describing the themes and revisiting the transcripts. Finally, quotes were selected to illustrate the themes.
Findings Teacher ratings The object of evaluation is predominantly the teaching and the teacher. When the students are asked to rate the lectures and there is more than one lecturer, this question is always split so that each individual teacher’s lectures can be rated separately. The same is true for problem-solving tutorials and other types of learning activities. This suggests that it is actually the individual teacher who is rated, even when the question concerns the learning activity. In the questionnaires there are no criteria and, often, not even labels for the rating, and there is little reflection among students or teachers on what the rating actually means. When the student representatives were asked to explain the meaning of a low or high rating, they explained that it corresponds to the teacher’s attitude. A feel-good factor is rewarded.
‘Well you must be able to catch people’s interest, be happy and positive and show that you want to be there…it depends mostly on your will to teach.’ KE: Good or bad in what way? ‘Well…No but just generally that is.’ KE: OK, what is it about [a teacher with high ratings] that is good then? ‘I think it is his interest in [the subject]. And he shows that he, he likes it. He is passionate about it. He is good at teaching it out (sic). Yes.’ The teachers have plenty of thoughts about what results in a low or high rating. ‘This will benefit a course which is welcoming, a bit pre-schoolish…’ ‘The problem is, if you have a difficult exam they will be annoyed…’ ‘Many write comments like…“They seem generally irritated, and want to write something mean.”’ ‘If you have a bad reputation it doesn’t matter much what you do, because they enter with the wrong views, so to speak.’
Teaching-focused view The total impression from the interviews is that teachers and students both express a teaching- focused view, where transfer of content in lectures is the central activity. The teacher’s role is to lecture. This is consistent with the practice of rating teachers as lecturers because, according to this view, the key to learning is the individual teachers’ performance in lectures. The teaching-focus is expressed even in the language used. Almost consistently among the teachers, the words talk about are used to mean teach. 98 K. Edström
‘This topic is much more fun to talk about’; ‘They haven’t understood what I talk about’; ‘Is it worth it to talk about it if nobody understood?’; ‘Did I talk in the right way?’; ‘A colleague was criticized that what he was talking about was too elementary’; ‘But don’t assume that they know this, what I have talked about’.
Students’ own work: nothing to do with evaluations When students’ own work is discussed with the teachers, it is obvious that surface approach to learning, low time-on-task and procrastination are widespread problems. These problems are attributed only to the students. ‘Blame-the-student’ thinking is commonly expressed. Many teachers spontaneously suggest a percentage of students who should not have been accepted into the programme. ‘I would start by accepting 20–30 per cent fewer students…’ ‘We have here a motley crowd…’ The teachers show little awareness of how to influence students’ work through course design, and evaluation is not seen as an issue related to student learning. In several interviews the teacher expresses the view that the students’ work is a completely different topic to discuss, unrelated to course evaluation or to course development. Respondents were surprised that the interview was to cover this topic ‘too’. KE: So what you are saying is that the students must start studying earlier in the course and not just for the exam? ‘Yes. Not that this has anything to do with course evaluation, though.’
Course development is not in the foreground The espoused purpose of evaluation is development, at least every single respondent mentions development when asked directly. But, to the teachers, course development is in the foreground only when the course is new. After that, evaluation turns into a completely empty, but still compulsory, routine. ‘I don’t really need to hand out [the questionnaires] because we already know what they will answer. Maybe it’s meaningful the first and second year when a course is new. But after that it’s just going through the motions because I know exactly what they will say.’ One barrier to course development is that teachers think it would always be more expensive or require more work to organize the course in any other way than it is done presently. Given the current resources, they can’t see any alternatives. They sometimes don’t even see the opportunity to think about potential development. ‘Every course must break even. This locks us in. And the mere thought of reorganizing the course scares people off, because it is so expensive.’ KE: You mean just to pause and consider? ‘Exactly.’ When asked how they would like to develop the course, many respondents express that the key to better teaching is for teachers to develop their lecturing technique, but they had almost no concrete ideas on how to improve this aspect. ‘Well…I could try to vary my voice more.’ This suggests that a development-focused evaluation loses meaning when the teachers lack the intention to develop the course. Several possible reasons why course development was not in Higher Education Research & Development 99 the foreground were exposed, including that it is not considered necessary (problems are attrib- uted to the students), and it is not considered possible because of the lack of resources (even time to think) or lack of ideas on how to go about it.
Negative course development There were a few signs that evaluations put pressure on the teacher to develop courses in direc- tions that might be questionable from a learning perspective. The use of detailed lecture handouts attracted positive comments in evaluations but seemed to make literature superfluous (in the narrow sense that students can pass the course without reading the course book). Another exam- ple came from a course using a peer-teaching method, an uncommon learning activity in this programme. As could be expected, a small minority of students expressed a preference to being lectured to properly. This teacher had no problem standing up for this teaching method within the course and with the students but still expressed concerns about what colleagues and the director of study might believe if they saw these negative comments. It obviously requires strength for a teacher to do anything that might attract negative comments.
‘Fire alarm’ function only The student representatives undertake a separate mid-course survey, the result of which is shared with the teacher. Unless the result is exceptionally bad the student representative does nothing more. Towards the end of the course the teacher distributes another questionnaire, creates the course analysis document and sends it to the director of study. Directors of study mainly collect the documents and consider the utilization of evaluations as the responsibility of the individual teachers. Many teachers comment on a perceived lack of feedback. The student representatives, teachers and directors of study all seem to regard evaluations as having mainly a ‘fire alarm’ function. If ratings are extremely low, they may investigate the causes, although no party claims to do that systematically. Their use of evaluations is limited to their function as a fire alarm. ‘So this works more like an emergency…to see if there are any big problems in a course.’ (Student representative) ‘Well, if something is totally disastrous, you will usually find out here.’ (Teacher)
Discussion Purpose of evaluations: audit or development? The two classic purposes of evaluation are audit and development. The same dichotomy also has been referred to as accountability and improvement (Bowden & Marton, 1998), appraisal and developmental purpose (Kember et al., 2002), judgemental and developmental purpose (Hounsell, 2003), or quality assurance and quality enhancement (Biggs, 2003). The relationship between these two purposes is discussed by Bowden and Marton (1999), who state that ‘if improvement is addressed properly, evidence for accountability will be devel- oped automatically. The reverse is not necessarily the case’ (p. 228). A similar idea is expressed through the concept of audit through self-audit (Franke-Wikberg, 1992). The purpose of an evaluation process must be determined at the outset. On a practical level, audit and development need rather different investigations. As Patton (1997) observes ‘The same data seldom serves both purposes well’ (p. 78). But more fundamentally, there is, in this case, a tension between audit and development because teachers are themselves responsible for performing the evaluation of their own work. If the purpose is development, teachers will want to 100 K. Edström investigate aspects that can be improved – such things are also known as problems. If, on the other hand, the purpose is audit, the task is to create a basis for fair judgement. Then it is disadvanta- geous to focus on problems, as this critical stance can be turned against the teacher. If the audit and development purposes are combined in the same evaluation activity, the inherent forces of the audit risk undermining, or even completely overriding, the development purpose. And it doesn’t matter much what the institution’s intended purpose is. What is important is what the individual teachers perceive to be the purpose. It is sufficient for the developmental purpose to be undermined if the teachers think they are being subjected to an audit. Therefore, when a course evaluation system is intended for development and not expected to be an audit, it must be expressed very clearly.
What is the intended purpose of the KTH course analysis policy? The policy on course analysis was the result of an investigation. The working group identified the tension between evaluating the teachers’ performance and evaluating for course development. In their report, they coined the term course analysis. To many teachers and students the concept of course evaluation is associated only with questionnaires to the students at the end of a course. These traditional course evaluations often focus on the perfor- mance of the individual teacher. Experience shows that they are of limited value for long-term improvement of the course. The working group is introducing the concept course analysis, which includes more than the traditional course evaluations, although these can be part of the data. (Proposal for Course Analysis, 1995, p. 2) The intention behind the KTH policy is apparently course development, as indicated by the phrases ‘proposed measures including deadlines’ and ‘clearly show a course’s development from one year to the next’. The procedure to send the course analysis upwards in the organization might suggest that there is also an element of audit (or audit through self-audit). Nonetheless, the matter is complicated by another policy. The much later policy on appoint- ment and promotion (KTH Handbook 4) states that ‘course evaluations should be included’ in teaching portfolios used for evaluating teaching merits for teachers seeking promotion. It is not known what they are expected to show by including course evaluations in the teaching portfolio, as there are no public criteria for the assessment. Although the intentions behind the course anal- ysis policy clearly oppose a ‘focus on the performance of the individual teacher’, in the promotion process, the individual teacher is a necessary and legitimate focus. Is the term course evaluations used on purpose (rather than course analysis), to indicate that here it is – the rating of the teacher’s performance that should be presented? Perhaps ratings are seen as a reasonable proxy for teaching excellence? Then teaching excellence is restricted to stage performance and charisma, on which the ratings probably focus. The main problem is that this new use of evaluation data injects uncertainty as to the purpose of course evaluation – maybe the purpose is appraising teachers rather than developing courses? The system for appointing and promoting faculty is a powerful instrument for shaping the culture of the institution. Its influence should not be underestimated as it carries dramatic consequences for individuals, who can be recognized, selected and rewarded – or rejected. Although the two policies both serve the same long-term aim – to improve the quality of education – they are not aligned, and the course evaluation policy risks being undermined by the promotion policy. It is argued in the present paper that the course evaluation system has been hi- jacked because its data are being used for another purpose. The irony is that the actual implemen- tation of the teaching portfolio therefore can, in fact, counteract its own long-term aim. If the teaching portfolio policy was consistent with the course analysis policy, the teachers should, instead, be asked to include their course analyses in their portfolios, in order to show their ability Higher Education Research & Development 101 to analyze the student learning experience and the quality of student learning outcomes, and to improve these through adequate course development measures.
Teaching or learning focus? What is the object of evaluation? What is being evaluated? It seems reasonable to suppose that the focus of evaluation should be what is perceived as the keys to the quality of the education. So, by looking at evaluation practices, we may catch glimpses of the present theory-in-use for teach- ing and learning. In teaching-focused evaluation, objects that could be evaluated are the teacher and the teaching (process). If evaluation is learning-focused, possible objects to evaluate would be learning outcome or process. These four possible objects of evaluation are discussed in turn.
Teaching (teacher) The investigation showed that evaluation questionnaires always included rating of teachers, and that ratings seemed related to a feel-good factor. But popular teachers cannot be conflated with good teaching, so teachers can do much better than settle for being popular. How does the wide- spread teacher rating influence the teaching and learning climate at the institution? Does pressure help teachers develop? Is there anything one can actually do with a score? As the rating seems related mostly to teachers’ personal charisma, it must be very difficult to know how to improve. Good ratings could maybe serve as general encouragement, but bad ratings will hardly help teach- ers improve their teaching. No one is going to be frightened into becoming a better teacher by the threat of student-ratings. … Lecturers who feel anxious about their teaching – perhaps because they feel under pressure to do it better, perhaps because they know they do it badly – are the least likely to change. (Ramsden, 1992, pp. 232–233) It is argued in the present paper that the stress created by the focus on the teacher as individual is detrimental to the development of courses and programmes, and that this taken-for-granted practice is created by, and subsequently helps uphold, a teaching-focused culture at the university.
Teaching (process) The investigation showed that it is common to focus on the teaching activities as such, and ask students to rate items such as lectures, labs, recitations and course literature. These data are turned into histograms, showing, for instance, that the first lab scores an average 3.35 and the second lab scores 4.15 on a 1–5 scale. But then what? Can such data be used productively to inform improve- ment? (See also Bowden & Marton, 1998, p. 230.) As shown above, a fire alarm function is the main outcome – a pitiful result of the whole evaluation process. In addition to the quantitative rating, evaluations solicit qualitative comments. It is obvious that many teachers will do what they can to eliminate negative comments. But although comments – positive or negative – can be useful for inspiring improvement, following student comments is no fool-proof strategy for improving student learning. The interviews provided the example of courses in which the use of lecture handouts generated positive student comments, creating a situation whereby handouts were virtually replacing the literature. It can be counter- productive to shape teaching according to the wishes of students with a passive or surface approach to learning (Kember & Wong, 2000) or with less mature attitudes toward knowledge (Edström et al., 2005). What about asking the students to suggest how teaching activities could be improved? This will provide teachers with many ideas for development. But suggestions must be interpreted 102 K. Edström carefully, as it is not always appropriate for teaching to conform to student expectations. Often, student suggestions will reveal their views on teaching and learning (valuable information, indeed) and, as a bonus, some of the suggestions may be very useful. There is still a risk, however, when evaluation focuses on teaching in itself, to confirm the views of students who see themselves in a (passive) consumer role.
Learning (outcome) It would be possible to ask students to rate to what extent they think they attained the intended learning outcomes. But the function of determining student learning outcomes is performed by assessment rather than evaluation. Therefore, we must recognize that evaluation is by no means the only input. It may be more relevant to take the quality of learning outcomes as the most impor- tant starting point for course development. As one participant in the author’s staff development course (not a respondent in the study) reflected:
The irony is that course evaluations don’t show the weaknesses [of the course], as students are gener- ally satisfied and positive to the content and lectures. But we see in reports and in exams that the students have difficulties making conclusions from their results or assess how reasonable an answer is. When students come back a year later to do their thesis project, it is not uncommon to discover that they display abysmally poor understanding. Based on the quality of student learning outcomes, we should ask: What can explain the results we see today? What improvement in student learning do we desire? How can the course design help the students do better?
Learning (process) The respondents in this investigation were taken aback when asked about the students’ own work – it was not seen as something the teachers have much influence on. But, as Thomas J. Shuell states (as cited in Biggs (2003)): If students are to learn desired outcomes in a reasonably effective manner, then the teacher’s funda- mental task is to get students to engage in learning activities that are likely to result in their achieving those outcomes…It is helpful to remember that what the student does is actually more important in determining what is learned than what the teacher does. (Title page) With this view, evaluation can contribute to finding out about students’ own work so that learn- ing activities and assessment can be improved to better support the desired learning. Evaluation should investigate the learning process, in a wide sense (Prosser & Trigwell, 1999). This includes time on task, distribution of work over the duration of the course, appropriateness of learning activity, indications of approaches to learning, how students perceive the demands in the course, students’ conceptions of learning, and approaches to learning. Evaluation is then a form of system- atic inquiry. ‘The evaluation task can generally be said to be: 1. to describe what actually happens in that which seems to happen; 2. to tell why precisely this happens; and, 3. to state the possibilities for something else to happen’ (Franke-Wikberg & Lundgren, 1980, p. 148).
The classical utilization problem of evaluation How can we explain the student union’s view that ‘the results of course-level quality processes are small in relation to the effort spent’ (Lindbo, 2005); a view which is certainly consistent with the findings of the present investigation. Utilization, or rather the lack thereof, is an important theme in evaluation literature, and there are various views on how or whether utilization can be improved. Higher Education Research & Development 103
A stimulating perspective is given by Dahler-Larsen (2005), who asks why evaluation, despite its poor track record in terms of results, is such a taken-for-granted thing to do. He quotes James March, who calls evaluation a protected discourse, impossible to question. Drawing on modern organizational theory, Dahler-Larsen argues that we cannot understand evaluation using a logic of consequentiality. It is simply not the results that motivate its existence. But, using a logic of appropriateness, we see that evaluation fits in a normative way; it is expected in our culture. Evaluation is portrayed as a ritual whose main role is to create an appearance of rational- ity and accountability. Dahler-Larsen points out that this is not to say that evaluation does not have effects. Even as a ritual it will influence reality, contributing to what constitutes the organi- zation’s culture, individual’s self-understanding, and so forth. Dahler-Larsen’s picture of evalu- ation certainly rings true when we consider the results of the present investigation. We recognize the meagre results of evaluation. We also recognize the interplay between evaluation practices and, for instance, teachers’ self-understanding or students’ views on the teacher’s role. Kember et al. (2002) report that in a university in which a standard student feedback question- naire was used for audit purposes, feedback data were compared over time and it is shown that no improvement takes place. Some of the possible reasons discussed are faculty’s perception that there is little incentive to improve teaching; that the process focuses on audit, which is detrimental to the developmental role of evaluation; and that the questionnaire lacks flexibility and appropri- ate focus, that is, it is teaching-focused and not applicable to more innovative forms of teaching and learning. Kember et al. (2002) argue for a system in which teachers are encouraged to devise their own ways of evaluating their teaching innovations. It is interesting to note that the course analysis policy at KTH was designed to be exactly such a system, but practices still suggest that the theory-in-use lags behind. It seems that changing the policy alone has not achieved the intended shift in how the system is used. There are studies (Piccinin et al., 1999) that show improvement measured in terms of higher student ratings in a setting where teachers receiving student feedback (ratings) are given individ- ual consultations with a staff developer. At the University of Sydney (Barrie et al., 2005), the quality assurance system has both an assurance and improvement purpose. Starting in 1999, the quality processes at different levels were aligned with each other, and with an explicit student learning perspective. On the level of the individual unit of study, a questionnaire (Unit of Study Evaluation) is devised centrally for the whole university, but it also has space for instructor’s own items. An important feature has been the strategic and systematic use of quality assurance data within the university. Recently, the quality assurance system has started to show signs of improvement in the quality of the student learning experience (still measured as student ratings). These cases suggest that improvement is possible when evaluation is combined with interven- tions, such as support to teachers, or otherwise is used as part of an aligned set of strategies. Then it is not the evaluation system in itself that has produced the positive effects.
The weak connection between evaluation and development The idea that evaluation should be an integral component of course development is hardly new. But how to make evaluation, in practice, support development is far from trivial. ‘Collecting data is not the same as improving or judging teaching’ (Ramsden, 1992, p. 232). Why is the connection between course evaluation and course development so weak at KTH? What are the barriers to using course evaluation as a tool for course development? First of all, there seems to be some confusion as to the purpose of evaluation. The course analysis policy clearly states a development purpose and the teachers claim that their purpose is course development. Still, the questionnaires used look like they were rather designed for audit, and the data have little value to inform development. Why is that? It seems difficult to 104 K. Edström find inspiration elsewhere, as most evaluation questionnaires described in the literature (Rich- ardson, 2005) are quantitative ratings of dimensions of teaching and seem better suited to serve an audit purpose. In fact, the theory-in-use could be that, after all, the purpose of evaluation is audit. The practice of rating teachers and the requirement to include ratings in teaching portfo- lios can be contributing factors in shaping such a theory. This is a barrier for using course evaluation as a tool for course development, and the very least that must be done at KTH is to align the policy on teaching portfolios with the developmental purpose of the course analysis policy. Rather than ratings, teachers should be asked to include their course analyses in the teaching portfolio in order to show their ability to analyze both the student learning experience and the quality of student learning outcomes, and to improve these with adequate course devel- opment measures. However, the main barrier seems related to the underlying views on teaching and learning. For course evaluation to become a tool for course development, there must be an idea of what development is possible and desirable, something like a theory for teaching and learning. What variables teachers and students will regard as possible and acceptable to manipulate (Vedung, 1993, p. 230) will depend on, and especially be limited by, their views on teaching and learning. The potential for development will depend on how phenomena are interpreted, what is seen to cause problems, and what interventions are seen as possible and desirable. The interviews clearly show that there is a limited view on how courses can be developed, which is related to a limited view on teaching and learning. Åkerlind (2007) identifies five qualitatively different approaches to developing as a teacher: (1) Building up one’s content knowledge (improving what to teach). (2) Building up practical experience (improving how to teach). (3) Building up a repertoire of teaching strategies (becoming more skilful as a teacher). (4) Finding out what strategies work for the teacher (becoming more effective as a teacher). (5) Increasing one’s understanding of what works for the students (becoming more effective in facilitating student learning). These categories are hierarchically inclusive. In the first three levels, student feedback is treated as an indicator of success or lack of success in teaching. It is not until level 4 that teach- ers see student satisfaction as important input for active teaching development, and only in level 5 do teachers see actual student learning outcomes as the primary indicator of teaching effec- tiveness. The teaching-focused views revealed in this investigation are mainly consistent with the earlier levels and no respondent shows a focus on student learning. Although teachers describe many problems related to student learning (the quality of both the learning outcomes and process of studying), they simply lack a framework to make useful interpretations of these problems and, consequently, they have almost no course development strategies with which to address them.
Conclusions Blame-the-students thinking and rate-the-teacher practice The practice to rate teachers, and the blame-the-student thinking displayed by teachers, appear as two sides of the same coin. A student may attribute success or failure to good or bad teachers, much as a teacher attributes the results to good or bad students. These thought patterns allow students and teachers, respectively, to focus on the other party’s contribution in the teaching and learning process and shy away from discussing (or even seeing) their own responsibilities. Changing the course evaluation practices means challenging these comfortable positions and, therefore, resistance may be expected. Higher Education Research & Development 105
Evaluation is a component of constructive alignment When we look at a student’s situation within a course, we know that factors like the students’ conceptions of learning, together with how they perceive the situation and context (Prosser & Trigwell, 1999), will influence how they go about their studies. Constructive alignment (Biggs, 2003) helps us design the course so it will bring about appropriate learning activity. The main idea is that the course as a whole should encourage the student to take on their studies appro- priately, adopting a deep approach to learning. This is supported by the purposeful design and relation between the course components, notably intended learning outcomes, learning activities and assessment. The constructive alignment concept represents a system view on courses. A conclusion of this paper is that course evaluation is also a part of that system; it is also a component that should be in constructive alignment. One aspect of constructively aligned course evaluation is that the expectations on students and teachers that are communicated indirectly through evaluation must be aligned with appropriate roles. If we see the teacher’s role as facilita- tor of learning, but then ask the students to rate the teacher as lecturer–entertainer, that evaluation is clearly misaligned. Another aspect of constructively aligned course evaluation is that evalua- tion practices must support the improvement of student learning.
System alignment The present case study showed how two policies related to course evaluation are in conflict and, therefore, on the level of the individual teacher, the purpose of evaluation is made unclear. Is the purpose to improve student learning, as the course analysis policy suggests, or is it to judge teacher performance, as suggested by the teaching portfolio policy? If, instead, the policies were aligned, they could potentially reinforce each other and better further their mutual long-term purpose – to improve student learning. In parallel with the view of the course as a system, the author proposes the idea to consider the university as a system, whose components must be tuned to influencing teachers to take on their teaching appropriately. The concept system alignment is proposed, a parallel to constructive alignment on the system level. The system components of a university are any macro-level structure, such as organization, infrastructure, work processes and policies, espe- cially those that regulate issues where the rubber meets the road, such as hiring, promotion and funding. The system alignment concept may help us analyze the different processes at the university from a student-learning perspective and identify clashes that need to be addressed. The compo- nents that create the conditions for teaching and learning must be in alignment with each other and with the long-term direction in which we wish the university to move.
Acknowledgements The author thanks the teachers and students who were interviewed; all colleagues at KTH who have partic- ipated in the staff development course; course leaders and participants in Strategic Pedagogic Development, supported by the National Agency for Higher Education, especially Åsa Lindberg-Sand (Lund University) and Anna-Karin Magnusson (now with the National Agency for Higher Education). Special thanks to Stefan Hallström and Adam Edström.
Notes 1. Oral communication, KTH Human Resource office. 2. Inspiration is taken from sources such as Gibbs (1999) and the investigative tools of the Formative Assessment in Science Teaching project. 106 K. Edström
References Åkerlind, G. (2007). Constraints on academics’ potential for developing as a teacher. Studies in Higher Education, 32(1), 21–37. Barrie, S., Ginns, P., & Prosser, M. (2005). Early impact and outcomes of an institutionally aligned, student focused learning perspective on teaching quality assurance. Assessment and Evaluation in Higher Education, 30(6), 641–656. Biggs, J. (2003). Teaching for quality learning at university: What the student does. Buckingham, UK: SRHE and Open University Press. Bowden, J., & Marton, F. (1998). The university of learning: Beyond quality and competence in higher education. London: Kogan Page. Dahler-Larsen, P. (2005). Den rituelle reflektion – om evaluering i organisationer. Odense, Denmark: Syddansk Universitetsforlag. Edström, K., Hallström, S., El Gaidi, K., & Kuttenkeuler, J. (2005). Integrated assessment of disciplinary, personal and interpersonal skills – student perceptions of a novel learning experience. Proceedings of the 13th Improving Student Learning conference. Oxford, UK: OCSLD. Formative Assessment in Science Teaching Project (FAST), Investigative tools.Retrieved October 11, 2007, from Open University and Sheffield Hallam University Web site: www.open.ac.uk/fast/. Franke-Wikberg, S. (1992). Utvärderingens mångfald. Projektrapport 1992: 4. Stockholm: UHÄ. Franke-Wikberg, S., & Lundgren, U. P. (1980). Att värdera utbildning Del 1, En introduktion till pedago- gisk utvärdering. Stockholm: Wahlström & Widstrand. Gibbs, G. (1999). Using assessment strategically to change the way students learn. In S. Brown & A. Glasner (Eds.), Assessment matters in higher education (pp. 41–53). Buckingham, UK: SRHE and Open Univer- sity Press. Hounsell, D. (2003). The evaluation of teaching. In H. Fry, S. Ketteridge & S. Marshall (Eds.), A handbook for teaching and learning in higher education: Enhancing academic practice (pp. 200–212). London: Kogan Page. Kember, D., Leung, D. Y. P., & Kwan, K. P. (2002). Does the use of student feedback questionnaires improve the overall quality of teaching? Assessment and Evaluation in Higher Education, 27(5), 411–425. Kember, D., & Wong, A. (2000). Implications for evaluation from a study of students’ perceptions of good and poor teaching. Higher Education, 40(4), 69–97. KTH Handbook 2. Policy for Course Analysis. KTH Handbook, Vol. 2, Tab 14.1. KTH Handbook 4. Policy for Appointments. KTH Handbook, Vol. 4, Tab 11.1, App. 1.2. Kvale, S. (1997). Den kvalitativa forskningsintervjun. Lund, Sweden: Studentlitteratur. (English Translation available: Kvale, S. (1996). Interviews: An introduction to qualitative research interviewing. London: Sage Publications.) Lindbo, D. (2005). Quality development on a course level – Invitation to discussion meeting. KTH Student Union, March 4, 2005. Patton, M. Q. (1997). Utilization-focused evaluation: The new century text. Thousand Oaks, CA: Sage Publications. Piccinin, S., Cristi, C., & McCoy, M. (1999). The impact of individual consultation on student ratings of teaching. International Journal for Academic Development, 4(2), 75–88. Proposal for Course Analysis. (1995). Unpublished report, October 18, 1995. Stokholm, KTH. Prosser, M., & Trigwell, K. (1999). Understanding learning and teaching. The experience in higher educa- tion. Philadelphia, PA: SRHE and Open University Press. Ramsden, P. (2003). Learning to teach in higher education (2nd ed.). London: RoutledgeFalmer. Ramsden, P. (1992). Learning to teach in higher education. London: Routledge. Richardson, J. T. E. (2005). Instruments for obtaining student feedback: A review of the literature. Assess- ment and Evaluation in Higher Education, 30(4), 387–415. Swedish National Agency for Higher Education. (2006). Tillsynsbesöket vid Kungl. Tekniska högskolan 2005, Report 2006: 36 R. Vedung, E. (1998). Utvärdering i politik och förvaltning. Lund, Sweden: Studentlitteratur. This article was downloaded by: [Kungliga Tekniska Hogskola] On: 05 February 2015, At: 03:13 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
European Journal of Engineering Education Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ceee20 PBL and CDIO: complementary models for engineering education development Kristina Edströma & Anette Kolmosab a School of Education and Communication in Engineering Science, KTH Royal Institute of Technology, Stockholm, Sweden b Department of Development and Planning, Aalborg University, Aalborg, Denmark Published online: 20 Mar 2014.
Click for updates
To cite this article: Kristina Edström & Anette Kolmos (2014) PBL and CDIO: complementary models for engineering education development, European Journal of Engineering Education, 39:5, 539-555, DOI: 10.1080/03043797.2014.895703
To link to this article: http://dx.doi.org/10.1080/03043797.2014.895703
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions 2015 Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February European Journal of Engineering Education, 2014 Vol. 39, No. 5, 539–555, http://dx.doi.org/10.1080/03043797.2014.895703
PBL and CDIO: complementary models for engineering education development
Kristina Edströma∗ and Anette Kolmosa,b
aSchool of Education and Communication in Engineering Science, KTH Royal Institute of Technology, 2015 Stockholm, Sweden; bDepartment of Development and Planning, Aalborg University, Aalborg, Denmark
(Received 7 May 2013; accepted 6 February 2014)
This paper compares two models for reforming engineering education, problem/project-based learning (PBL), and conceive–design–implement–operate (CDIO), identifying and explaining similarities and dif- ferences. PBL and CDIO are defined and contrasted in terms of their history, community, definitions, curriculum design, relation to disciplines, engineering projects, and change strategy. The structured com- parison is intended as an introduction for learning about any of these models. It also invites reflection to support the understanding and evolution of PBL and CDIO, and indicates specifically what the commu- nities can learn from each other. It is noted that while the two approaches share many underlying values, they only partially overlap as strategies for educational reform. The conclusions are that practitioners have much to learn from each other’s experiences through a dialogue between the communities, and that PBL and CDIO can play compatible and mutually reinforcing roles, and thus can be fruitfully combined to reform engineering education.
Keywords: CDIO; problem-based learning; project-based learning; PBL; educational development; curriculum development; change strategy
1. Introduction
1.1. Background
There are many strong drivers for curriculum change in higher engineering education, seeking
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February to establish alternatives to traditional programmes consisting mostly of disciplinary theoretical courses (Sheppard et al. 2009). The background is a need to enhance quality and improve the processes and results of education, for instance to increase attractiveness to prospective stu- dents, decrease attrition, to improve preparation for professional practice, and better contribute to sustainable development, innovation, and job creation. In addition to drivers for change within universities, there is also political pressure and pressure from employers (National Academy of Engineering 2004; Royal Academy of Engineering 2007; Litzinger et al. 2011). The higher edu- cation environment in general has been significantly reformed in recent years, spurred on by
∗Corresponding author. Email: [email protected]
© 2014 SEFI 540 K. Edström and A. Kolmos
for instance the Bologna process and the Accreditation Board for Engineering and Technology (ABET), both of which have been influential far beyond their formal scope. Changes include the structure of degrees and a switch to outcomes-based principles for curriculum, accreditation, and evaluation systems. There is a great variety in how institutions go about the change (Graham 2012). Some projects are designed, carried out, and reported purely internally, while others are inspired by more estab- lished approaches, sharing and discussing the problem analyses, methodologies, and results in wider communities. There are several organisations, networks, and communities focused on cur- riculum change in higher engineering education worldwide. The aim of this paper is to analyse and compare two educational development approaches with organised international communities: problem/project-based learning (PBL) and conceive–design–implement–operate (CDIO). Our main questions are: What are the differences and similarities? How are the approaches related?
2015 What can they offer to engineering education development and to each other? We see two reasons for making this comparison. The first is to offer a knowledge base for anyone wishing to learn about PBL or CDIO, in particular educators or institutions considering reform. The second is to invite reflection to support the understanding, critique, and evolution of PBL and CDIO, by its practitioners. As we need to constantly reflect on and develop our methods and their underpinning arguments, the aim is to provide a contrast conducive for better understanding also one’s own familiar practices. In particular, this paper will make an effort to identify what one of the communities can potentially learn from experienced practitioners of the other, and areas with potential interest for future collaborations.
1.2. Methodology
The framework for the analysis was generated through an inductive, participatory, and iterative approach. The authors started with a dialogue to generate a gross list of relevant aspects to compare. To support the aims to invite learning and reflection, the comparison needed to examine the under- lying ideas and how they are evident in practice. As PBL and CDIO represent system approaches to curriculum development, the comparison also emphasises the organisational and the curriculum level view. The resulting list of aspects was the basis for generating descriptions of PBL and CDIO through studies of the literature and documents, and drawing on our own experiences in PBL and CDIO, respectively. The first iterations of the comparison were discussed and gradually refined in three conference workshops with about 70 highly experienced practitioners (Kolmos and Edström 2011a, 2011b; Edström and Kolmos 2012). The workshop discussions indicated that points
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February of comparison were the most salient and productive in generating insights by revealing similarities, differences, surprises, misconceptions, or unreflected assumptions. The resulting framework converged around the following core aspects: the history, community, definition, curriculum design, relation to disciplines, engineering projects, and change strategy. In the following, these aspects are examined and analysed for PBL and CDIO, respectively, then contrasted. Describing CDIO and PBL in ways that enable comparisons is not a straightforward task, due to fundamental differences in their nature. Since there is much variation in PBL interpretation and implementation, there are multiple definitions and perspectives and the full diversity of PBL practice cannot be covered here. When the description had to be narrowed down, the approach practiced in engineering at Aalborg University was emphasised, but with an effort to demon- strate awareness of and sensibility to other traditions. Because of its more cohesive organisation, it was somewhat easier to define CDIO, at least sufficiently for the practical purpose of this paper. To invite further exploration beyond these short descriptions of PBL and CDIO, references are provided. European Journal of Engineering Education 541
2. History
2.1. PBL – histories
The late 1960s and early 1970s were a period of experimentations and expansion in educational systems. When reform universities were established, inventing new educational models, the result was several forms of PBL. The problem-based learning model was implemented especially in health education at McMaster University (founded 1968) and Maastricht (founded 1972). The problem-based and project-organised models were practiced at Roskilde University (founded 1972) andAalborg University (founded 1974), in a wide range of subject areas such as engineering, science, social science, and humanities (Illeris 1976; Neville and Norman 2007; Kolmos and de Graaff 2013). The pedagogy was developed from a critical stance in student movements, and
2015 added to a theory of learning with cognitive, emotional, and social dimensions (Illeris 2007). The PBL universities are well documented in all aspects of curriculum development, learning, and competence development (Schmidt and Moust 2000). Nowadays, PBL is implemented all over the world. In health and law, the McMaster and the Maastricht models are often used, whereas theAalborg model is applied in engineering and science (Graham 2009; Savin-Baden 2003). In the PBL curriculum, projects are the platform for students to achieve competences, and to relate disciplines to each other in analysis and identification of problems as well as the problem-solving process. Process skills such as self-directed learning, project management, collaboration, communication, and collaborative knowledge construction are taught in an integrated way by letting students reflect upon their practice. A fundamental principle is that the students are owners of the learning process and the facilitator guides the students by presenting several ideas, methods, and tools.
2.2. CDIO – starting point
CDIO started at Massachusetts Institute of Technology (MIT) in the late 1990s as a reaction to conventional engineering education, observing that in many institutions engineering science was replacing engineering practice as the dominant culture. Crawley (2001) summed it up: ‘Education of engineers had become disassociated from the practice of engineering.’ Ever fewer faculty members had professional engineering experience and values related to practice were weakened in the education – affecting graduate qualities. Industry feedback stated the need for change (Gordon 1984; Augustine 1994; The Boeing Company 1996) and similar requirements came from new outcomes-based accreditation standards emphasising a wider set of skills (ABET 1996). Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February This sparked an investigation into the question: ‘What is the full set of knowledge, skills, and attitudes that engineering students should possess as they graduate?’The CDIO Syllabus (Crawley 2001; for version 2.0, see Crawley et al. 2011) lists and categorises desired qualities of engineering graduates, based on stakeholder input and validation. The acronym refers to engineering practice: conceiving, designing, implementing,andoperating products, processes, and systems. The early work at MIT struck a chord with Swedish educators and industrialists, and in 1999, the CDIO Initiative was formed by MIT, Chalmers, KTH Royal Institute of Technology, and Linköping University, with four years of funding from the Knut andAlice Wallenberg Foundation. They adopted the aim to educate students who:
• Master a deeper working knowledge of technical fundamentals. • Lead in the creation and operation of new products, processes, and systems. • Understand the importance and strategic impact of research and technological development on society. 542 K. Edström and A. Kolmos
The project partners set out together to develop pilot programmes at each university (Brodeur et al. 2002; Bankel et al. 2003), thereby creating, implementing, and documenting the CDIO approach, a methodology for engineering education reform.
2.3. Comparing the history
PBL and CDIO both advocate broader learning outcomes compared to traditional academic edu- cation, emphasising student development of skills and personal development, the process of becoming a professional. With its longer history, PBL should be recognised as a milestone for student-centred education, also preparing the ground for CDIO. A difference is that PBL emerged across disciplines, while CDIO was developed within engineering. Moreover, while PBL was an alternative pedagogy created in new reform universities, CDIO was designed by established 2015 institutions, for reforming existing programmes. One fundamental difference is that the means/ends logic is almost the opposite. CDIO aims to align the intended learning outcomes with professional practice – and the focus on more appropriate processes for teaching and learning comes as a consequence of that. For PBL, it was the learning process that was aligned with professional practice, in a highly student-centred interpretation consistent with the social movements in the 1960s and 1970s. The CDIO Initiative was established much later and embodies more recent trends such as outcomes-based education, and explicitly uses references to external stakeholder interest to challenge traditions within the institutions. These differences can be attributed to the spirit of the times when respective approach was developed.
3. Communities
3.1. PBL – organised communities
The size of the community practicing PBL cannot be estimated, due to the different lev- els of implementation, ranging from individual instructors applying PBL in a single course and programmes where PBL is applied to some extent, to whole institutions built around the model, with researchers specialising in PBL evidence. There are several international networks for sharing experiences, none of them with a formal membership structure. Among the most established are:
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February • The UNESCO Chair in Problem-Based Learning in Engineering Education (UCPBL) runs the PBL Global Network with research symposia every second year. The UCPBL has declared a strong emphasis on research and is based on philosophy and learning principles across different PBL practices (Maastricht and Aalborg), derived from educational research and practice. • The International PBL Symposium is organised by Republic Polytechnic, Singapore – the hub of an Asian community with an international symposium every second year. • The Pan-American Network for Problem-based Learning – international conferences each second year.
There is a rich literature documenting PBL, including specialised journals such as the Interdisci- plinary Journal of Problem-Based Learning and Journal of PBL in Higher Education. Plenty of the literature reviews indicate success (Dochy et al. 2003; Beddoes, Jesiek, and Borrego 2010). Results show that employers rank PBL education highly, stating that graduates are able to work from day one. Students from a PBL programme achieve a higher level of skills and competences, deeper learning, and increased motivation. Compared to traditional universities, the retention rates European Journal of Engineering Education 543
increase, students get higher grades and higher salary (Kolmos and de Graaff 2013). Some voices also warn about risks such as lack in disciplinary knowledge (Kirschner, Sweller, and Clark 2006).
3.2. The CDIO Initiative
Soon after the four founding institutions started developing a methodology for reforming pro- grammes, other expressed an interest in participation. They were welcomed and presently the CDIO Initiative has grown to a large global organisation consisting of over 100 institutions as ‘CDIO Collaborators’. There is a formal structure where the CDIO Council grants the status as collaborator and controls key documents (CDIO Syllabus and CDIO Standards). From 2013, the council members are elected, replacing the original organisation where the first 10 collaborators had permanent seats. 2015 Knowledge generated through the experience of developing engineering education is shared and disseminated within and outside the CDIO community.The annual international CDIO Conference started in 2005. Also annual and open is the worldwide working meeting. The early collaborators authored a book on the CDIO approach (Crawley et al. 2007, the second edition Crawley et al. 2014) and the CDIO website (cdio.org) contains resources and contact information as a starting point. While some of the conference publications could arguably be categorised as educational research, and some are published in peer-reviewed journals, the majority of contributors are engineering faculty documenting their educational reform work and very few authors have a position as educational researchers. Lately, interest in educational research has increased, and peer-review of conference papers was introduced from 2009.
3.3. Comparing the communities
On the spectrum between a well-defined and centralised organisation, and an inclusive and decen- tralised community, it is safe to say that PBL consists of clusters of communities of practice that are open and inclusive, whereas the CDIO Initiative is an organisation with at least some control over defining documents and collaborator status. CDIO is mainly an education development community where most participants are instructors and leaders of engineering education rather than educational researchers. Because of the longer history and the wider range of subjects, there are researchers focusing on PBL. Some networks are essentially research communities, and more research publications document the effect of PBL
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February than of CDIO.
4. Definitions
4.1. PBL definitions: three learning principles
Since the first pioneering institutions, existing universities have adapted or partially adopted problem-based and/or project-based models (de Graaff and Kolmos 2003; Savin-Baden and Howell Major 2004) leading to a variety of implementations worldwide. PBL is applied in differ- ent cultural settings, subject areas and at different levels in the educational system ranging from schools to universities and continuing education. The scope of implementation ranges from the institutional, to programme and single course level. This diversity leads to a continuous debate on what should count as PBL (Savin-Baden 2003). 544 K. Edström and A. Kolmos 2015
Figure 1. PBL learning principles (Kolmos, de Graaff, and Du 2009).
Local practices will, and should, constantly evolve with regards to content and educational methods. It is therefore short sighted to define PBL based only on practice – the concept should be dynamic and based on both theories and practices. The UCPBL also recognises cultural differences and has developed an understanding of PBL based on diverse practices and learning theories. The learning principles below (Figure 1) are intended to broadly guide practice, and are consistent with the McMaster/Maastricht and Aalborg models (de Graaff and Kolmos 2003). The principles are related to the approach to cognitive, collaborative, and contents aspects (Barrows 1996; Illeris 2007). Problem orientation indicates that learning starts by analysing and defining problems, be they open and ill defined, or well defined. The choice of problems depends on the learning objectives – to learn methodologies will require open problems, and when the
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February aim is to achieve specific methods, more narrow problems will be suitable. de Graaff and Kolmos (2003, 2007) call this the cognitive learning approach. Problems are the starting point for learning processes; they are placed in a context, and based on the learner’s experience. If the course is also project-based, the task involves more complex and situated problem analyses and problem-solving strategies. The interdisciplinary dimension and the theory–practice relation concern the content in the curriculum. Theory is used in analysis of problems and problem-solving methods. Interdisci- plinary learning, and the fact that problems come mostly from practice, may create challenges for organisation of the curriculum. Another key aspect of the content approach is that problems, no matter how they are chosen, have to be exemplary to the overall learning outcomes and serve as the means for understanding at a deep level and to transfer methodologies to similar areas (Kolmos and de Graaff 2013). Finally, the social approach is crucial. In team-based learning, the learning process is a social act where learning takes place through dialogue and communication. The students are not only learning from each other, but they also learn to share knowledge and organise the process of European Journal of Engineering Education 545
collaborative learning and collaborative knowledge construction. The social approach also covers the concept of participant-directed learning, which indicates a collective ownership of the learning process and, especially, the formulation of the problem. Ownership is key for students’motivation. These principles should be regarded essential, so that any claim to run PBL should mean that the practice reflects all three learning principles. For instance, individual projects fall outside this definition – there must be a team aspect. The PBL curriculum is organised and designed differently, in, e.g. Maastricht and Aalborg. Maastricht students analyse cases and organise the learning process by seven steps procedures, whereas in Aalborg, they collaborate in teams and gradually learn project management skills. Also, the assessment systems are different. However, the three learning principles apply to both implementations.
4.2. The CDIO Standards 2015
As the CDIO Initiative grew, a wider diversity of programmes and institutions were needed to be accommodated. At the same time, stakeholders sought clarification about the distinguishing features of CDIO programmes, and there was an apprehension that CDIO could lose meaning if ‘anything goes’. What was chosen (in 2004) as the defining feature was the educational reform process (Crawley et al. 2007). After answering what students should learn (with the CDIO Syl- labus), the next question must be: ‘How can we do better at ensuring that students learn these skills?’ The working definition of CDIO is ‘how can we do better’ captured in the 12 CDIO Stan- dards (Table 1). The value or novelty lies not in any single standard on its own, but in defining a comprehensive and holistic approach, listing available drivers of change, and supporting the alignment of strategies. Seven standards are considered essential (with asterisks), describing a minimal approach for developing a CDIO programme. The CDIO Standards were later equipped with rubrics for programme rating (CDIO 2010), thus indicating dimensions both for bringing about and systematically monitoring the development (Malmqvist et al. 2006; Kontio et al. 2012; Malmqvist 2012; Munkebo Hussman et al. 2012). It is worth noting that the CDIO Syllabus is not a defining feature of CDIO. Each institution must formulate programme goals considering, e.g. stakeholder needs, national and institutional context, level and scope of programmes, and subject area. To accommodate diversity, the CDIO syllabus is offered as an instrument for specifying local programme goals by selecting topics and making appropriate additions in dialogue with stakeholders. As such, it has served as a reference for a multitude of engineering programmes and for diverse contexts and purposes (Bisagni et al. 2010; Edström et al. 2013). Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February 4.3. Comparing the definitions
An obvious difference between CDIO and PBL is the degree to which their essentials can be defined at all. The PBL principles proposed here are evidence-based; they are known to be conducive to learning. However, a multitude of definitions exist for PBL, and there is arguably no forum where consensus could be established. On the other hand, the CDIO Standards express a more formal definition, codified and controlled by the CDIO Initiative, but with much room for variation in collaborating institutions’ practice. Another notable difference is the nature of what these working definitions set out to define. The PBL principles form a broad philosophy of teaching and learning focusing exclusively on the learning process,thatis,how students should learn, and not on what they should learn. Therefore, the principles can be applied on course, programme, or institutional level, in different fields of education, and any level from school to university. Conversely, CDIO takes its starting point in the learning outcomes of higher engineering education, and how learning should be facilitated 546 K. Edström and A. Kolmos
Table 1. The CDIO Standards.
Standard 1 – The context* Adoption of the principle that product, process, and system lifecycle development and deployment – Conceiving, Designing, Implementing, and Operating – are the context for engineering education Standard2–Learningoutcomes* Specific, detailed learning outcomes for personal and interpersonal skills, and product, process, and system building skills, as well as disciplinary knowledge, consistent with programme goals and validated by programme stakeholders Standard 3 – Integrated curriculum* A curriculum designed with mutually supporting disciplinary courses, with an explicit plan to integrate personal and interpersonal skills, and product, process, and system building skills Standard 4 – Introduction to engineering An introductory course that provides the framework for engineering practice in product, process, and system building, and introduces essential personal and interpersonal skills Standard5–Design–implement experiences* A curriculum that includes two or more design–implement experiences, including one at a basic level and one at an
2015 advanced level Standard6–Engineeringworkspaces Engineering workspaces and laboratories that support and encourage hands-on learning of product, process, and system building, disciplinary knowledge, and social learning Standard 7 – Integrated learning experiences* Integrated learning experiences that lead to the acquisition of disciplinary knowledge, as well as personal and interpersonal skills, and product, process, and system building skills Standard 8 – Active learning Teaching and learning based on active experiential learning methods Standard9–Enhancement of faculty competence * Actions that enhance faculty competence in personal and interpersonal skills, and product, process, and system building skills Standard 10 – Enhancement of faculty teaching competence Actions that enhance faculty competence in providing integrated learning experiences, in using active experiential learning methods, and in assessing student learning Standard 11 – Learning assessment* Assessment of student learning in personal and interpersonal skills, and product, process, and system building skills, as well as in disciplinary knowledge Standard 12 – Programme evaluation A system that evaluates programmes against these 12 standards, and provides feedback to students, faculty, and other stakeholders for the purposes of continuous improvement
is mainly a consequence of what students should learn. The CDIO Standards were developed to define the agenda for structured programme development in engineering. The methodology can also inspire educational development in other fields, e.g. in teaching professions (Fors et al. 2007). Comparing the essentials has made it clear how CDIO and PBL overlap. CDIO Standard 5 prescribes a curriculum with at least two design – implement experiences of increasing levels of
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February complexity. These learning activities are problem-based and project-organised, and students learn from authentic engineering practice. Thus, the introduction of a specific type of PBL elements in the curriculum is an essential feature of CDIO.
5. Curriculum design
5.1. PBL – the Aalborg curriculum model
There are many examples of curricula based on PBL in the world. One of the most complete and institution-wide implementations is the Aalborg model, see Figure 2. It is a hybrid model in the sense that students attend courses half their study time. Thus, the disciplines are mainly in taught courses. The relationship between courses and projects can vary depending on the learning objectives. In some semesters, there is a tight coupling between courses and project: the disciplinary knowledge European Journal of Engineering Education 547 2015
Figure 2. The Aalborg curriculum model in 2011.
is applied in the project. In other semesters, the project is more independent, and students use their learning from the courses only as needed for the project.
5.2. CDIO – the integrated curriculum
The CDIO Standards describe the process of designing an integrated curriculum, starting by establishing a vision of the graduates, informed by stakeholder needs, the context, and conditions (Standard 1). In the light of this vision, programme-level learning outcomes are formulated for both engineering skills and disciplinary knowledge, to be validated with stakeholders (Standard 2). After designing the curriculum structure, the programme learning outcomes are mapped with curriculum elements (Standard 3). This is a negotiation process where the intended learning
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February outcomes serve as the ‘currency’for defining the contribution of a course to the programme goals. Each course is thereby assigned an explicit function in the programme, and it is made clear which courses together carry the responsibility for each programme learning objective. Note the solution- independent approach: after the course learning outcomes are negotiated with the programme, the course design can be implemented in many different ways, because it is possible to address the same objectives through different teaching and assessment methods. The principle is the same for reforming an existing programme – after any changes in the curriculum structure, the new course learning outcomes are negotiated. This methodology has been useful for reinforcing specific competences, see Figure 3. Examples include communication skills (Carlsson, Malmström, and Edström 2010), computational mathematics (Enelund, Larsson, and Malmqvist 2011), and sustainable development (Knutson Wedel et al. 2008; Enelund et al. 2012). A tool to document the integrated curriculum in a structured way is the integrated programme description (Malmqvist, Östlund, and Edström 2006; Malmqvist and Arehag 2008), supporting faculty and other stakeholders to share their understanding of the programme design. 548 K. Edström and A. Kolmos 2015
Figure 3. Systematic integration of specific competences.
The final stage is to develop the courses as integrated learning experiences (Standard 7), where students simultaneously develop disciplinary knowledge and professional engineering skills (Crawley et al. 2005, 2007). Since the intended learning outcomes address both disci- plinary knowledge and professional skills, this should be reflected in the learning activities, and assessment system (Biggs and Tang 2011). The pedagogical principle is that integrated learning calls for integrated assessment (Standard 11) (Edström et al. 2005).
5.3. Comparing curriculum design models
CDIO is a concept for the curriculum level, a methodology for outcomes-based programme design making the coupling between programme and course level explicit. AsCDIOisdefined on the programme level, it is not applicable to say that a single course is a CDIO course, as we can say of a PBL course. CDIO is fully outcomes-based, and though active and experiential learning methods are emphasised, there is also a recognition that the same learning outcomes can be reached through different pedagogical methods. Taking its starting point in the learning process, PBL rather demonstrates that different learning outcomes can be reached using the same pedagogical philosophy based on projects and problems. Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February While CDIO essentially is a model for curriculum development, it is not so straightforward to say that PBL implies a curriculum model at all, given the wide diversity in PBL implementations, from course to institution level. Research evidence (Thomas 2000) suggests that PBL works best when it is implemented consistently across the curriculum, when everything from institutional support systems to buildings are aligned with the educational model. It is also possible to argue that it is better for the students to have a few instances of PBL in their education than none at all.
6. Relation to disciplines
6.1. PBL and disciplines
Kolmos, de Graaff, and Du (2009) defined elements of PBL curricula: objectives, types of prob- lems and projects, progression, student learning, academic staff, space and organisation, and European Journal of Engineering Education 549
Table 2. Dimensions of PBL curriculum elements (Kolmos, de Graaff, and Du 2009).
Curriculum Discipline and teacher- Innovative and learner- element controlled approach centred approach
Objectives and knowledge • Traditional disciplinary objectives • PBL and methodological objectives • Disciplinary knowledge • Interdisciplinary knowledge Type of problems and projects • Narrow • Open • Well-defined problems • Ill-defined problems • Disciplined projects • Problem projects • Study projects • Innovation projects • Lectures determine the project • Lectures support the project Progression, size, and duration • No visible progression • Visible and clear progression • Minor part of the curriculum • Major part of course/curriculum Student learning • Acquisition of knowledge • Construction of knowledge Academic staff and facilitation • No training • Training courses • • /
2015 Teacher-controlled supervision Facilitator process guide Space and organisation • Administration for traditional course • Administration supports PBL cur- and lecture-based curriculum riculum • Traditional library structure • Library to support PBL • Lecture rooms • Physical space to facilitate teamwork Assessment and evaluation • Individual assessment • Group assessment • Summative course evaluation • Formative evaluation
assessment. In principle, there are two extremes in interpreting and implementing these elements: a discipline and teacher-controlled approach, and an innovative and learner-centred approach. It is important to emphasise that there is no institution that practices a pure PBL curriculum, but rather a mix of traditionally taught courses and PBL. For instance, Aalborg University uses a hybrid model in the sense that the students attend courses half their study time. Table 2 illustrates the poles, and most PBL practices represent mixed or hybrid modes. The main point is to create awareness in the implementation of PBL – in a whole institution or a single course.
6.2. Discipline-led learning in CDIO
Recognising the need to educate for professional practice has not led the CDIO community to advocate a fully problem/project-based education. For sure, CDIO implies sharp criticism against poorly designed curricula, at worst consisting of disciplinary courses disconnected from each other, and as a whole, loosely coupled to espoused programme goals, professional practice, and student motivation. But if and when they work well, discipline-led courses provide conceptual understanding of systematically organised knowledge – a basis for solving real problems. Indeed, the first aim of CDIO is a deeper working understanding of disciplinary fundamentals. Strategies Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February for improving student learning in discipline-based courses include active learning methods and assessment practices, conducive to conceptual understanding. The fundamental idea of CDIO is the integrated curriculum, where discipline-led and problem/project-led learning are meaningfully combined. For existing programmes, it is often necessary to increase the share of PBL activities. But that is not sufficient; a curriculum is not inte- grated just because it contains both problem/project-led and discipline-led courses. The synergy comes from integrated learning experiences, where students simultaneously acquire disciplinary knowledge and professional engineering skills. Table 3 lists values from these complementary modes of learning, with potential synergies.
6.3. Comparing the relations to disciplines
Both PBL and CDIO represent curriculum models to support students in integrating and apply- ing their disciplinary knowledge, and in developing the skills and working modes relevant for 550 K. Edström and A. Kolmos
Table 3. Contributions of discipline-led and problem/practice-led learning in the integrated curriculum.
Discipline-led learning Problem/practice-led learning • Well-structured knowledge base (content) • Integrating, applying, and synthesising knowledge • Knowing what is known and not • Open-ended problems, ambiguity, trade-offs, contexts, • Understanding evidence/theory, model/reality and conditions • Methods to develop new knowledge, the scientific • Professional skills (work processes) process • CDIO, or ‘create the world that never has been’(von • Interconnecting disciplines Kármán) • Knowledge building of the practice …connecting to problem/practice-led learning: • Deeper working understanding, i.e. conceptual under- …connecting to discipline-led learning: standing and functional knowledge • Drawing on disciplinary knowledge, seeing through • Knowledge with consideration for use the lenses of problems • Embedded development of skills, e.g. communication • Reinforcing disciplinary understanding and collaboration • Creating a motivational context for learning disci- plinary fundamentals 2015
professional practice. Therefore, both emphasise the problem-led component in education as alternatives to traditional purely discipline-stacking programmes. Note that in PBL, the problem- centred approach is the defining element, while it is the broader set of learning outcomes that leads CDIO to advocate better use of both discipline-led and practice-led learning activities.
7. Engineering projects
7.1. PBL project types
Based on the type of problems that students are working on and their relation to disciplinary learning outcomes, Kolmos (1996) defined three types of projects: (1) The assignment project – relatively narrow learning outcomes and little freedom for the students to influence the learning process. Such projects are going on in laboratories at many universities and as they do not meet the learning principles they fall outside the definition of PBL. (2) The discipline project – addressing disciplinary learning outcomes. Students are working on practical problems to apply theoretical knowledge. The discipline is the frame for raising problems and the project is limited to the discipline borders. (3) The problem project – where the problem takes its departure in the contextual and societal
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February dimension. The problem will determine the disciplines that are involved in analysis and solutions – both interdisciplinary and cross-disciplinary. The problem project normally starts with ill-structured problems with a certain level of complexity (Jonassen and Hung 2008). These three types do not cover the full variety of projects and in the literature there are other taxonomies for problem types, for instance making the distinction between practical, empiri- cal, and theoretical problems. Furthermore, there are typologies of projects, e.g. distinguishing analytical projects, where the aim is new knowledge on given problems, from design projects and construction projects aimed at new technological devices (Algreen-Ussing and Fruensgaard 1990).
7.2. CDIO – the design–implement experience
CDIO programmes contain various problem- and project-based learning activities, but the defin- ing element is the Design–Build Experience, where students design and implement products, European Journal of Engineering Education 551
processes, or systems. Projects take different forms in various engineering fields, but the essential aim is to learn through near-authentic engineering tasks, in working modes that resemble profes- sional practice. Standard 5 implies a sequence of design–implement experiences, with progression in several dimensions. Early in the education, smaller teams apply engineering knowledge of lim- ited breadth and depth, while advanced project teams can involve over 10 students working over an academic year, drawing on a range of disciplinary knowledge and engineering skills. Projects concern increasingly complex and open-ended problems and later problems are ill defined and full of tensions, contextual factors, and stakeholder interests, resembling technical tasks new graduates might encounter in working life. From a learning perspective, it is key that students bring their designs and solutions to an operationally testable state. To turn practical experiences into learning, students are continuously guided through reflection and feedback exercises supporting them to evaluate their work and
2015 identify potential improvement of results and processes. Furthermore, assessment and grading should reflect the quality of attained learning outcomes, rather than the product performance in itself (Edström et al. 2005).
7.3. Comparing the engineering projects
When comparing the project components, it is obvious that PBL comprises a broader scope of problems and projects, and that the PBL mode carries a greater part of the learning in the PBL curriculum. CDIO is born out of an engineering design environment and thus design projects and the near-professional engineering projects are important. But while CDIO proposes a curriculum with a sequence of project-based learning activities, it does not mean that the role of disciplinary courses is downplayed. In the projects, students reinforce their disciplinary understanding by applying the knowledge, and the practical experiences are intended to increase their motivation for learning theory. Furthermore, the problems will often prompt them to learn new theory just in time, as needed to create solutions – but in CDIO, the projects are not intended to replace discipline-led courses as the primary site to learn systematic disciplinary knowledge.
8. Change strategies
8.1. PBL change strategies
It is no coincidence that some of the most sustainable implementations of PBL were created when Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February new universities were started around these principles. Organising learning around problems and stressing interdisciplinary learning often makes PBL perceived as challenging the traditions, and through its history it has provoked substantial resistance in institutions. As a consequence, the change management perspective is always present (de Graff and Kolmos 2007). An important legitimising strategy has been to provide research evidence for the positive effects of PBL, as the absence of evidence makes is difficult to defend investment in change.
8.2. CDIO change strategies
CDIO seeks legitimacy (Suchman 1995) as a cultural insider in engineering education institutions. The educational philosophy is dressed in engineering clothes – created by engineering faculty for engineering faculty, speaking the same language. Curriculum development resembles engineering design, with concrete pedagogical strategies adapted to engineering education. It is also part of the insider strategy to call for more appropriate contributions from discipline-led learning 552 K. Edström and A. Kolmos
without fundamentally challenging the role of disciplines. Recognising that deep understanding of disciplinary fundamentals is crucial for engineering practice, the CDIO community proposes discipline-led courses for deeper and more relevant learning outcomes. This is also a pragmatic strategy – bringing about change by taking advantage of the strengths in the existing culture, not by being iconoclastic. Another fundamental strategy for legitimacy is to involve stakeholders outside academia, e.g. professional organisations and employers, and students (Edström 2012; Edström et al. 2003). The wider dialogue can validate the goals with students, employers, and society. When other stakeholders are present, some arguments heard internally are easily exposed as self-serving or sub-optimising. The CDIO community has had to challenge the assumption that CDIO programmes would require unreasonable resources. The main strategy in CDIO is to put existing resources (e.g.
2015 facilities, instructor and student time) to better use, not just adding new practices on top of the old. While a change project can often find extra temporary resources, the new ways of working that it establishes must be sustainable on normal funding. There is proof-of-concept for sus- tainable approaches, making even design–implement experiences cost-neutral in the steady state (Hallström, Kuttenkeuler, and Edström 2007; Edström, Hallström, and Kuttenkeuler 2011).
8.3. Comparing the change strategies
While evidence of effectiveness is seldom demanded from existing practices, the burden of proof seems to rest on those who want to introduce any change. Therefore, in both PBL and CDIO com- munities, strategies for making change legitimate, and thus possible, have been widely discussed. PBL, and to some extent CDIO, will partly challenge some academic traditions and identities. While problem-led learning aims to align with professional practice, discipline-led learning is better aligned with the organisation and structures of most institutions. Problem-led learning will therefore by its nature go against discipline-based organisational principles. We find the strategies for legitimacy somewhat different. CDIO conforms more to the culture of engineering and works (mostly) within the disciplinary structure of institutions and curriculum, while PBL makes appeals to academic culture first and foremost through research evidence of its positive impact.
9. Conclusions
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February The comparison has shown many similarities between PBL and CDIO. The two approaches for reforming engineering education share the main underlying values and goals – the emphasis on development of professional skills through learning processes that are similar to authentic practice. The difference is that PBL emerged from rethinking the process, while CDIO was developed from rethinking the outcomes. It is further shown that the implementation of CDIO and PBL is partly overlapping, as elements of PBL pedagogy are a defining feature of CDIO (design–implement experiences). Our first conclusion is therefore that PBL and CDIO prac- titioners should have much to learn from each other’s experiences through a closer dialogue and exchange. Obvious areas of mutual interest include the pedagogy of problem-, project-, and design-based learning experiences and the lessons learned around organisational change strategies. Another area of mutual interest and collaboration is the emerging field of engineering education research, where the CDIO community can find much inspiration from the evidence produced around PBL. The comparison also showed that PBL and CDIO are of quite different nature. While the PBL philosophy relates to the learning process in problem/project-based parts of the curriculum, CDIO European Journal of Engineering Education 553
contains a methodology to develop the whole curriculum including disciplinary courses. Our sec- ond conclusion is therefore that CDIO and PBL are not mutually exclusive, but complementary. For an institution that plans to create an innovative engineering curriculum, there is no need to make a choice between the two approaches as they can be productively combined. The CDIO approach supports a structured process of setting the high-level learning outcomes and systemat- ically translating them into a curriculum, and any combination of CDIO and PBL pedagogy will support the development of appropriate learning experiences. The approaches should be seen as compatible and mutually reinforcing.
Acknowledgements
The authors would like to express warm thanks to all participants in the workshops where this comparison was discussed,
2015 as well as to Åsa Lindberg-Sand, Lund University, and Lars Geschwind, KTH Royal Institute of Technology, for helpful comments on a previous version of the paper.
References
ABET. 1996. “Engineering Criteria 2000 (EC2000).” See for instance Engineering Change: A Study of the Impact of EC2000 (executive summary). Accessed April 20. www.abet.org/engineering-change/ Algreen-Ussing, H., and N. O. Fruensgaard. 1990. Metode i Projektarbejde. Aalborg: Aalborg University Press. Augustine, N. R. 1994. “Socioengineering (and Augustine’s Second Law Thereof).” The Bridge 24 (3): 3–14. Bankel, J., K.-F. Berggren, K. Blom, E. F. Crawley, I. Wiklund, and S. Östlund. 2003. “The CDIO Syllabus:A Comparative Study of Expected Student Proficiency.” European Journal of Engineering Education 28 (3): 297–315. Barrows, H. S. 1996. “Problem-based Learning in Medicine and Beyond: A Brief Overview.” New Directions for Teaching and Learning 1996 (68): 3–12. doi:10.1002/tl.37219966804 Beddoes, K. D., B. K. Jesiek, and M. Borrego. 2010. “Identifying Opportunities for Collaborations in International Engineering Education Research on Problem- and Project-Based Learning.” Interdisciplinary Journal of Problem- based Learning 4 (2): 7–34. doi:10.7771/1541-5015.1142 Biggs, J., and C. Tang. 2011. Teaching for Quality Learning at University. Maidenhead: McGraw-Hill and Open University Press. Bisagni, C., D. R. Brodeur, J. Bosch, R. Camarero, B. Carlsson, A. Castelli, A. Causi, et al. 2010. “DOCET Final Report. EQF – CDIO: A Reference Model for Engineering Education: A Guide for Developing Comparable Learning Outcomes to Promote International Mobility.” Accessed April 21. www.eqfcdio.org/results Brodeur, D., E. F. Crawley, I. Ingemarsson, J. Malmqvist, and S. Östlund. 2002. “International Collaboration in the Reform of Engineering Education.” Proceedings of the 2002 ASEE/IEEE frontiers in education conference proceedings, November 2002, Boston, MA. Carlsson, C.-J., H. Malmström, and K. Edström. 2010. “Engineering and Communication Integrated Learning – Col- laboration Strategies for Skills and Subject Experts.” Proceedings of the 6th international CDIO conference, École Polytechnique, Montréal, June 15–18. CDIO. 2010. “The CDIO Standards v 2.0 (with customized rubrics). December 8.” Accessed April 21. www.cdio.org/files/document/file/CDIOStdsRubricsv2.0_2010Dec8.doc Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February Crawley, E., K. Edström, J. Malmqvist, D. Soderholm, and S. Östlund. 2005. “Curriculum Design Based on the CDIO Model.” Proceedings of the 2005 international SEFI conference, Ankara, Turkey. Crawley, E., J. Malmqvist, S. Östlund, and D. Brodeur. 2007. Rethinking Engineering Education, The CDIO Approach. New York: Springer. Crawley, E., J. Malmqvist, S. Östlund, D. Brodeur, and K. Edström. 2014. Rethinking Engineering Education, The CDIO Approach. 2nd ed. New York: Springer. Crawley, E. F. 2001. “The CDIO Syllabus: A Statement of Goals for Undergraduate Engineering Education: MIT CDIO Report #1.” Accessed April 21. www.cdio.org/framework-benefits/cdio-syllabus-report Crawley, E. F., J. Malmqvist, W. A. Lucas, D. R. Brodeur. 2011. “The CDIO Syllabus v2.0. An Updated Statement of Goals for Engineering Education.” Proceedings of the 7th international CDIO conference, Technical University of Denmark, Copenhagen, June 20–23. Dochy, F., M. Segers, P. Van den Bossche, and D. Gijbels. 2003. “Effects of Problem-based Learning: A Meta-analysis.” Learning and Instruction 13 (2003): 553–568. Edström, K. 2012. “Student Feedback in Engineering: Overview and Background.” In Enhancing Learning and Teaching through Student Feedback in Engineering, edited by P. Mertova, S. Nair, and A. Patil, 1–23. Cambridge: Woodhead Publishing. Edström, K., I. Froumin, E. F. Crawley, and T. Stanko. 2013. “Engaging Stakeholders in Defining Education for Inno- vation in Russia: Consensus and Tensions.” Paper presented at the EAIR 35th annual forum 2013, Rotterdam, the Netherlands, August 28–31. 554 K. Edström and A. Kolmos
Edström, K., S. Hallström, K. El Gaidi, and J. Kuttenkeuler. 2005. “Integrated Assessment of Disciplinary, Personal and Interpersonal Skills – Student Perceptions of a Novel Learning Experience.” Proceedings of the 2005 13th international symposium improving students learning, London, September 5–7. Oxford: Alden Press. Edström, K., S. Hallström, and J. Kuttenkeuler. 2011. “Workshop: Designing Project-Based Courses for Learning and Cost-Effective Teaching.” Proceedings for the frontiers in education 2011 conference, Rapid City, South Dakota, October 12–15. Edström, K., and A. Kolmos. 2012. “Comparing Two Approaches for Engineering Education Development: PBL and CDIO.” Proceedings of the 8th international CDIO conference, Queensland University of Technology, Brisbane, July 1–4. Edström, K., J. Törnevik, M. Engström, and Å. Wiklund. 2003. “Student involvement in principled change: Understanding the student experience.” Proceedings of the 2003 11th international symposium improving students learning: theory, research and scholarship, Hinckley, September 1–3. Oxford: Alden Press. Enelund, M., M. Knutson Wedel, U. Lundqvist, and J. Malmqvist. 2012. “Integration of Education for Sustainable Development in a Mechanical Engineering Programme.” Proceedings of the 8th international CDIO conference, Queensland University of Technology, Brisbane, July 1–4. Enelund, M., S. Larsson, and J. Malmqvist. 2011. “Integration of Computational Mathematics Education in the Mechanical
2015 Engineering Curriculum.” Proceedings of 7th international CDIO conference, Technical University of Denmark, Copenhagen, June 20–23. Fors, E., E. Faahraeus, A. Hossjer, and C. Sonnerbrandt. 2007. “LIKA - Digital literacy in Teacher Education.” In Proceedings of Society for Information Technology & Teacher Education International Conference 2007, edited by R. Carlsen et al., 1474–1481. Chesapeake, VA: AACE. Gordon, B. M. 1984. “What is an Engineer?” Invited keynote presentation, annual conference of the European Society for Engineering, University of Erlangen-Nurnberg. de Graaff, E., and A. Kolmos. 2003. “Characteristics of Problem-based Learning.” International Journal of Engineering Education 19 (5): 657–662. de Graaff, E., and A. Kolmos. 2007. Management of Change Implementation of Problem-Based and Project-Based Learning in Engineering. Rotterdam: Sense Publishers. Graham, R. 2009. UK Approaches to Engineering Project-Based Learning. White Paper sponsored by the Bernard M. Gor- don, MIT Engineering Leadership Program.Accessed March 29. http://web.mit.edu/gordonelp/ukpjblwhitepaper.pdf Graham, R. 2012. Achieving Excellence in Engineering Education: The Ingredients of Successful Change. London: Royal Academy of Engineering. Hallström, S., J. Kuttenkeuler, and K. Edström. 2007. “The Route Towards a Sustainable Design-implement Course.” Proceedings of the 3rd CDIO conference, MIT, Cambridge, MA, June 11–14. Illeris, K. 1976. Problemorientering og deltagerstyring: oplæg til en alternative didaktik. Copenhagen: Munksgaard. Illeris, K. 2007. How we Learn – Learning and Non-learning in School and Beyond. New York: Routledge. Jonassen, D. H., and W. Hung. 2008. “All Problems are Not Equal: Implications for PBL.” Interdisciplinary Journal of Problem-Based Learning 2 (2): 6–28. doi:10.7771/1541-5015.1080 Kirschner, P. A., J. Sweller, and R. E. Clark. 2006. “Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist, Discovery, Problem-based, Experiential, and Inquiry-based Teaching.” Educational Psychologist 41 (2): 75–86. Knutson Wedel, M., J. Malmqvist, M. Arehag, and M. Svanström. 2008. “Implementing Engineering Education for Environmental Sustainability into CDIO Programs.” Proceedings of the 4th international CDIO conference, Gent, Belgium, June 16–19. Kolmos, A. 1996. “Reflections on Project Work and Problem-based Learning.” European Journal of Engineering Education 21 (2): 141–148. Kolmos,A., and K. Edström. 2011a. “Should we do CDIO or PBL?Yes!” Invited workshop at the annual CDIO conference, DTU, Lyngby, Denmark, June 20–23.
Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February Kolmos, A., and K. Edström. 2011b. “Should we do CDIO or PBL?Yes!” IIDEA workshop at the SEFI annual conference, Lisbon, Portugal, September 27–30. Kolmos, A., and E. de Graaff. 2013. “Problem-based and Project-based Learning in Engineering Education – Merging Models.” In Cambridge Handbook of Engineering Education Research (CHEER), edited by A. Johri and B. M. Olds, 141–160. New York: Cambridge University Press. Kolmos, A., E. de Graaff, and X. Du. 2009. “Diversity of PBL – PBL Learning Principles and Models.” In Research on PBL Practice in Engineering Education, edited by X. Du, E. de Graaff, and A. Kolmos, 9–21. Rotterdam: Sense Publishers. Kontio, J., J. Roslöf, K. Edström, S. Naumann, P. Munkebo Hussmann, K. Schrey-Niemenmaa, and M. Karhu. 2012. “Improving Quality Assurance with CDIO Self-Evaluation: Experiences from a Nordic Project.” International Journal of Quality Assurance in Engineering and Technology Education 2 (2): 55–66. Litzinger, T. A., L. R. Lattuca, R.G. Hadgraft, and W. C. Newstetter. 2011. “Engineering Education and the Development of Expertise.” Journal of Engineering Education 100 (1): 123–150. Malmqvist, J. 2012. “A Comparison of the CDIO and EUR-ACE Quality Assurance Systems.” International Journal of Quality Assurance in Engineering and Technology Education 2 (2): 9–22. Malmqvist, J., and M. Arehag. 2008. “Experiences from Using Integrated Program Descriptions to Support Program Development.” Proceedings of 3rd international CDIO conference, MIT, Cambridge, Massachusetts, June 11–14. Malmqvist, J., K. Edström, S. Gunnarsson, and S. Östlund. 2006. “The Application of CDIO Standards in the Evaluation of Swedish Engineering Degree Programmes.” World Transactions of Engineering and Technology 5 (2): 361–364. European Journal of Engineering Education 555
Malmqvist, J., S. Östlund, and K. Edström. 2006. “Using Integrated Programme Descriptions to Support a CDIO Programme Design Process.” World Transactions on Engineering and Technology Education 5 (2): 259–262. Munkebo Hussmann, P., A. Bisi, J. Malmqvist, B. Carlsson, H. Lysne, and A.-K. Högfeldt. 2012. “Peer Evaluation of Master Programs: Closing the Quality Circle of the CDIO Approach?” International Journal of Quality Assurance in Engineering and Technology Education 2 (2): 67–79. National Academy of Engineering. 2004. The Engineer of 2020 – Visions of Engineering in the New Century. The National Academies Press. www.nap.edu/openbook.php?isbn=0309091624 Neville,A. J., and G. R. Norman. 2007. “PBL in the Undergraduate MD Program at McMaster University: Three Iterations in Three Decades.” Academic Medicine 82 (4): 370–374. Royal Academy of Engineering. 2007. Educating Engineers for 21st Century. www.raeng.org.uk/news/release/pdf/ Educating_Engineers.pdf Savin-Baden, M. 2003. Facilitating Problem-based Learning: Illuminating Perspectives. Maidenhead: Society for Research into Higher Education & Open University Press. Savin-Baden, M., and C. Howell Major. 2004. Foundationsof Problem-based Learning. Maidenhead: Society for Research into Higher Education & Open University Press. Schmidt, H. G., and J. C. Moust. 2000. “Factors Affecting Small-group Tutorial Learning: A Review of Research.” In
2015 Problem-based Learning: A Research Perspective on Learning Interactions, edited by D. H. Evensen and C. E. Hmelo, 19–52. Mahwah, NJ: Lawrence Erlbaum Publishers. Sheppard, S., K. Macatangay, A. Colby, and W. M. Sullivan. 2009. Educating Engineers: Designing for the Future of the Field. San Francisco, CA: Jossey-Bass. Suchman, M. S. 1995. “Managing Legitimacy: Strategic and Institutional Approaches.” Academy of Management Review 20 (3): 571–610. The Boeing Company. 1996. “Desired Attributes of an Engineer.” See J. H. McMasters and N. Komerath. 2005. Boeing – University Relations – A Review and Prospects for the Future. ASEE paper 2005-1293, Portland, WA, June 13–15, 2005. Thomas, J. W. 2000. “A Review of Research on Project-Based Learning.” Autodesk Foundation. Accessed April 20. www.bie.org/research/study/review_of_project_based_learning_2000
About the authors
Kristina Edström is Associate Professor in Engineering Education Development at KTH Royal Institute of Technology in Stockholm, one of the CDIO founding institutions. She contributed to Crawley et al. (2007, 2014) and served on the CDIO Council 2005–2013. During 2012–2013, she was also the Director of Educational Development at Skolkovo Institute of Science and Technology, Moscow, Russia. Anette Kolmos is Professor in Engineering Education and PBL and Chairholder for UNESCO Chair in Problem Based Learning in Engineering Education, Aalborg University, and guest professor at the KTH Royal Institute of Technology, Stockholm. She is active in developing the profile of Engineering Education Research in Europe and internationally. She was President of SEFI 2009–2011. She has published more than 190 articles in various books and journals. Downloaded by [Kungliga Tekniska Hogskola] at 03:13 05 February PDPTDPW
" **+) " )+ (!#( +#(! . -#)( / &)*' (-
?6@A6;.1@A?K:.;1 .8
Kristina Edström Engineer & Educational developer M. Sc. in Engineering, Chalmers PhD in Technology and Learning, KTH Associate Professor in Engineering Education Development at KTH Royal Institute of Technology, Stockholm, Sweden Editor-in-Chief of the European Journal of Engineering Education 700 participants in the 7.5 ECTS course Teaching and Learning in Higher Education, customized for KTH faculty, 2004-2012 Director of Educational Development at Skolkovo Institute of Science and Technology, Moscow, 2012-2013 Strategic educational development, national and international CDIO Initiative for reform of engineering education since 2001 SEFI Administrative Council, 2010-2013 Some publications Edström, K. (forthcoming, 2018). Academic and Professional Values in Engineering Education: Engaging with History to Explore a Persistent Tension. Engineering Studies, 10(1) Edström, K. (2017). Engineering education research: Combining usefulness and scholarliness. European Journal of Engineering Education. Crawley, E.F., Malmqvist, J., Östlund, S., Brodeur, D.R., and Edström, K. (2014) Rethinking Engineering Education: The CDIO Approach, 2nd ed., Springer Verlag Edström, K., & Kolmos, A. (2014). PBL and CDIO: complementary models for engineering education development. European Journal of Engineering Education, 39(5), 539-555 Edström, K. (2008) Doing course evaluation as if learning matters most, Higher Education Research & Development, 27(2), 95 – 106
P PDPTDPW
If you want to learn about a system, try to change it
(attributed to Kurt Lewin)
) ( 1
B4# *!3#0 )"$) .$)"/01 )0/ /#*1' ' .)) 3#5 6# #'(, $## #'7
Q PDPTDPW
) $!'%'(%) +(
40 .)' "%!$.'( )*#)()*#)( )0 .)' G . ($)'5 *10 G . *10 *0# . /1'0/H ,.* //) . /1'0/H
$ ).$ ). *!). # #' # # #' # *) $#*) $#
An education about An education in technology engineering
)( #/ 0B@A<:2?;221@A205;<9<4F 2;A2?=?6@2@A?.A24F?24B9.A6<;@.;1 0<;02=AB.9A205;60.9.;1/B@6;2@@ =9.;@ ,#!(=9.;@1?.D6;4@.;1.94
R PDPTDPW
Disciplinary theory Theory and judgement applied to applied to real problems “problem-solving”
Cross disciplinary boundaries Sit in contexts with societal and business aspects Complex, ill-defined and contain tensions Need interpretations and estimations (‘one right answer’ are exceptions) Require systems view
*)// )<?<0.* '< ?<L <??GQOOUH?2 .5 5,.* ' (/*'2$)"$) )"$) .$)"> //*)/!*. )"$) .$)" 1 0*./? !'$ ! $ '& ! <XTGQH<PRX?
Individual approach Communicative and collaborative approach
Crucial for all engineering work processes Much more than working in project teams with well-defined tasks Engineering is a social activity involving customers, suppliers, colleagues, citizens, authorities, competitors Networking within and across organizational boundaries, over time, in a globalised world
S PDPTDPW
)#'B3 $#)-)
Education set in Educate for the context Engineering science of Engineering
)#' B : $#)-) *,0$*)*!0# ,.$) $,' 0#0,.* 1 0<,.* //< ) /5/0 ('$! 5 ' 2 '*,( )0) ,'*5( )02 ! ( .% . " & "$& 2 . 0# *)0 40 !*. )"$) .$)" 1 0$*)?
Engineers who can engineer!
*), ) ,$( *!).1
"%!$.'( )*#)(
$ ). *!).*!). # #' # *) $#
T PDPTDPW
%',$' # #$,! $ ( %! #'. *#"#)!(
Functional knowledge Not just reproduction of #5) known solutions to known problems 7 ) )7 Conceptual understanding Being able to explain what they do and why 7 $) )7 ;
%((-" !-"
!*.$)/0) 61.<?GPXXVH$ %&$'& ! <) ( .L 1"#0GQOOVH ($% &+ 0
*! ).$()*#)!'# # : (!9 " )/ # !-$#$".
B142 '</2./92A<0?6A60.99F2C.9B.A2:B9A6=92@<9BA6<;@.;1 @2920A.;<=A6:B:@<9BA6<; &<9C2 5.?.0A2?6G2.;.9FG2.;1@F;A52@6G2A<:<129. @F@A2:=? E=9.6; 2./92A<@A.A2A52=?<02@@ I $/ '<??< #$)"01 )0/0**)0$)1 # $. 1 0$*)<$! %!&$! & $% '& ! ! $ <PXWU?J U PDPTDPW )-$#$". : "$'*(*!!(( ) $#( '()'*)*'! # ()'*)*'! *!) ()'*)*'! !) $#! -)#()') I$""/) *''$/<PXWQ= (" 5( **&J Adapting CDIO to Civil Engineering: Investigate – Plan – Design – Construct – Operate and maintain .0$)$'//*).0$$) $'//*) 0.$)0.$) ' .* ' .* 1' )$2 ./$05*! #)*'*"5 1' )$2 ./$05*! #)*'*"5 Kristina Edström KTH Royal Institute of Technology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he strategy of CDIO is integrated learning of knowledge and skills ! PO PDPTDPW )#'D: #) ')*'' *!*" The CDIO strategy is the integrated curriculum where knowledge & skills give each other meaning! #' % & ! !& !) )#' D : #) ') *'' *!*" 1..$ 1'1( /$") 3$0# (101''5 /1,,*.0$)" $/ $,'$).5 *1./ /< 3$0# ) 4,'$ $0 ,') 0* $)0 ".0 , ./*)'< $)0 ., ./*)'< ) ,.* 1 0< ,.* //< ) /5/0 ( 1$' $)" /&$''/? ( !" &! $ % % +'.!'# # -%' #()( !##'!) $#( % Discipline-led learning Problem/practice-led Well-structured knowledge base learning Evidence/theory, Model/reality Integration and application, synthesis Methods to further the knowledge frontier Open-ended problems, ambiguity, trade- offs CONNECTING WITH Context PROBLEM/PRACTICE Professional work processes Deep working understanding = ability to ”Creating that which has never been” apply CONNECTING WITH DISCIPLINARY Seeing the knowledge through the lens of KNOWLEDGE problems, interconnecting the disciplines Discovering how the disciplinary Integrating skills, e.g. communication and knowledge is useful collaboration Reinforcing disciplinary understanding Motivational context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his is about how to get better student learning from the same (finite) teaching resources )#'BA99 # #"#)$ *!). # $"%)# 0$*)/0#0 )#) ! 1'05 *(, 0 ) $) ,.*2$ $)"$)0 ".0 ' .)$)" 4, .$ ) /<$) 1/$)" 0$2 4, .$ )0$'' .)$)"( 0#* /< ) $)// //$)"/01 )0' .)$)"? -"%!(' !!*()') $#($%' # %!( /, $!$ 3$'' !!*()') -"%! #' 0* %' # %!( #(% ' %%! ) $#( E*!()5 $!! . )0&$) /? PV PDPTDPW Educational development strategies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ducational development strategies # ( %! #'.$*'(( #%'$!"8%') 9( #%'$!"8%') 9( $*'(($*'(( (,.*2$)"0# -1'$05*!1) ./0) $)" $)" $)",.* ' (D,. 0$ E / ' .)$)" $)",.* ' (D,. 0$ E / ' .)$)" )*3' " ,. ,. !*.1/ >/ $)"0# "0# 4, .$ ) / 4, .$ ) / &)*3' " 0#.*1"#0# ' )/ *!,.* ' (/* ' (/ - .'5 )"$) .$)" 4, .$ ) .'5 )"$) .$)" 4, .$ ) $'$050* *((1)$ 0 ) *'' *.0 .0 - / -1 ) *! /$")E (,' ( )0/ -1 ) *! /$")E (,' ( )0 4, .$ ) /4, .$ ) / )0 . *)) 0$)"0# $/ $,'$) / (,.*2$)". !' 0$*)) ' .)$)" (,.*2$)". !' 0$*)) ' .)$)" (,.*2$)" */0E !! 0$2 ) //*! (,.*2$)" */0E !! 0$2 ) //*! 0 #$)"0 #$)" ( #9 "%!"#)-%' #( ()*#))"(( ## "%!"#))*!%'$*)(2%'$(((2$'(.()"( #?<720A@A.8216332?2;A3 QP PDPTDPW '# # #( #9 "%!"#)-%' #( "'$"!% % !&&!' & %. '&&! $! ' & % $0$/& 50#0/01 )0/ .$)"0# $. /$")/) /*'10$*)/0*) $%') $#!!.)()!())? *01.),. 0$ ' 4, .$ ) /$)0*' .)$)"</01 )0/. *)0$)1*1/'5 "1$ 0#.*1"#'!) $## -' ((/1,,*.0$)"0# ( 0* 2'10 0# $.3*.&) $ )0$!5,*0 )0$'$(,.*2 ( )0*!. /1'0/ ) ,.* // /? (((("#)# ' # /#*1' . !' 00# -1'$05*!00$) !'# # $*)$"(<.0# .0#)0# ,.* 1 0, .!*.() $)$0/ '! MORE ABOUT THE PROGRAM LEVEL QQ PDPTDPW 2 curriculum principles Individual student choice Predefined curriculum from an offering of courses Program corresponds to degree Bounded by degree requirements requirements (for breadth, Cohorts follow the program depth, etc.) Advising -#)(&& / && +(#(!).- )' ,,- +) # ( #( (!#( +#(! Students must demonstrate: Knowledge and understanding knowledge of the scientific basis and proven experience of their chosen area of engineering, together with insight into current research and development work; and both broad knowledge in their chosen area of engineering, including knowledge of mathematics and natural sciences, and substantially deeper knowledge in certain parts of the field. Skills and abilities an ability, from a holistic perspective, to critically, independently and creatively identify, formulate and deal with complex issues, and to participate in research and development work so as to contribute to the development of knowledge; an ability to create, analyse and critically evaluate different technical solutions; an ability to plan and, using appropriate methods, carry out advanced tasks within specified parameters; an ability to integrate knowledge critically and systematically and to model, simulate, predict and evaluate events even on the basis of limited information; an ability to develop and design products, processes and systems taking into account people’s situations and needs and society’s objectives for economically, socially and ecologically sustainable development; an ability to engage in teamwork and cooperation in groups of varying composition; and an ability to clearly present and discuss their conclusions and the knowledge and arguments behind them, in dialogue with different groups, orally and in writing, in national and international contexts. Judgement and approach an ability to make assessments, taking into account relevant scientific, social and ethical aspects, and demonstrate an awareness of ethical aspects of research and development work; insight into the potential and limitations of technology, its role in society and people’s responsibility for its use, including social and economic aspects, as well as environmental and work environment aspects; and an ability to identify their need of further knowledge and to continuously upgrade their capabilities. QR PDPTDPW ". . -1$. ( )0/ .*".(' .)$)" * % 0$2 / *1./ ' .)$)" * % 0$2 / .)$)" // //E 0$2$0$ / ( )0 Our curriculum system has 2 logical links The strength of the chain – the extent to which graduates will actually meet the program learning objectives – hinges on: the connection between courses and programs that the sum of course learning objectives actually equals the program objectives, and the constructive alignment that each course actually teaches and assesses students according to its learning objectives. QS PDPTDPW ("( #(!*+)!+ ,,#)(-"+).!"-" .++# .&.' )(-+# .-#)(-) #(& & +(#(!).- )' , ''! 1'05!*.(1'0 0# $. *1./ *)'5/$),10D*10,10> #%*)>B# )/01 )0/ *( 0*(5 *1./ 3)00# (0* ' 0*@C *)%*)>B# )/01 )0/' 2 (5 *1./ 0# 53$'' ' 0*@ 1/ 0#$)&0#$/$/) //.5$),10!*. *1./ @C & % )1 1 + #, '' *1./ /. ,. / )0 0#.*1"#$),10) *10,10*)'5> ;./92@2336062;A16@0B@@6<;@ .82@0<;;20A6<;@C6@6/92.@D299.@9.08A52?2<3 6C2@.993.0B9AF.; B?6;4A5216@0B@@6<;@ <0B:2;AD56050 QT PDPTDPW Dimensions of progression Subject content Personal, professional and engineering skills Theoretical maturity – not just ”more” theory, E2?06@23 42+( 2# % #+ ()$( $,) %$,'$%'$ '"%'(%) +3 # !# #' # ) !"'(>B? )(- 1- 205.;60.9;46;22?6;46@.F2.?6;A24?.A21.0529 " *)0 +) -" *+)!+'**+) " 22=6;4A52=?<4?.::2B;63621.9A5 QU PDPTDPW ()$( $,) %$,'$%'$ '"%'(%) +3 # !# #' # ) !"'(>C? )'*.--#)(&'-" '-# , '52.6:6@A<') +(#3 A52:.A52:.A60.90<;A2;AD5692.9@< @A?2;4A52;6;4A52 )(( -#)( /2AD22;2;46;22?6;4.;1:.A52:.A60@ +#( #*& , &AB12;A@;221A<92.?;A<@<9C2:C6@A &AB12;A@@5 ()$( $,) %$,'$%'$ '"%'(%) +3 # !# #' # ) !"'(>D? " #(- +/ (-#)(,#(-" *+)!+'' #(/)&/ !2D/.@60:.A50 &6:69.?9FA526;A24?.A6<;<3@[email protected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el Aviv Academic College of Engineering Astrakhan State University The international CDIO community Bauman Moscow State Technical University ,# North America Cherepovets State University 2676;4 ;@A6ABA2 <3 #2A?<052:60.9 '205;<9<4F Delft University of Technology Arizona State University 2676;4 6. ##*! #)'#) $#! Next: European CDIO Regional meeting $#'# January 2017, Skolkovo, Moscow, Russia QOOT1 )A/)$2 ./$05< $)"/0*)<) 14th International CDIO Conference QOOU$)&+,$)")$2 ./$05<$)&+,$)"< June 2018, Kanazawa, Japan 3 ) 15th International CDIO Conference QOOV *" / #**' )0< )0< '"$1( June 2019, Aarhus, Denmark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xploring the dual nature of engineering education Opportunities and challenges in integrating the academic and professional aspects in the curriculum KRISTINA EDSTRÖM Doctoral thesis in Technology and Learning School of Education and Communication in Engineering Science KTH Royal Institute of Technology Stockholm, Sweden 2017 TRITA-ECE 2017:2 ISBN 978-91-7729-596-9 Academic Dissertation which, with due permission of the KTH Royal Institute of Technology, is submitted for public defence for the Degree of Doctor of Philosophy on Wednesday the 13th December 2017, at 2:00 p.m. in Salongen, KTHB, Osquars backe 31, Stockholm. © Kristina Edström, 2017 Printed by Universitetsservice US AB “You don’t get good work without good ideas, but the ideas come from the work. […] And learning to listen to the work that you’ve already made is really where all the core ideas come from. One work is the mother of the next.” Antony Gormley (CNN, 2015) Contents Acknowledgements ...... 6 Abstract ...... 7 Svensk sammanfattning (Swedish abstract) ...... 9 1. INTRODUCTION TO THE THESIS ...... 11 1.1. Theme and research questions ...... 11 1.1.1. The dual nature of higher engineering education ...... 11 1.1.2. Development as a starting point ...... 11 1.1.3. Research questions and structure of the thesis ...... 12 1.2. Research approach and methodology ...... 13 1.2.1. A problem-led and naturalistic approach ...... 13 1.2.2. Educational development and critical educational research ...... 14 1.2.3. Engagements rather than measurement ...... 16 1.2.4. The insider and outsider perspective ...... 17 2. EFFORTSEFFORTS TOTO INTEGRATEINTEGRATE ACADEMICACADEMIC AND PROFESSIONALPROFESSIONAL AIMSAIMS ...... 1919 2.1. Engineering education development – tthehe CDIO approach ...... 1199 2.1.1. Taking the initiativeinitiative ...... 1199 2.1.2. Programme level development ...... 2222 2.1.3. Course level development ...... 2244 2.1.4. Faculty development ...... 32 2.2. Further development of the CDIO concept and community ...... 34 2.2.1. Comparing CDIO and PBL ...... 34 2.2.2. Connecting CDIO and engineering education research ...... 36 3. A PERENNIAL TENSION ...... 39 3.1. A state of déjà vu ...... 39 3.1.1. Carl Richard Söderberg (1895 – 1979) ...... 39 3.1.2. Comparing the ideals of CDIO and Söderberg ...... 40 3.2. Learning from the past ...... 44 3.2.1. Decommissioning the pendulum metaphor ...... 44 3.2.2. Faculty, not curriculum ...... 44 3.2.3. Using a historical perspective ...... 45 4. MAKING SENSE OF UNSUSTAINABLE CHANGE ...... 47 4.1. Organisational gravity ...... 47 4.1.1. Experiences of unsustainable change ...... 47 4.1.2. Testing the model ...... 48 4.1.3. How things work around here ...... 49 4.2. Implications ...... 50 4.2.1. Two change strategies ...... 50 4.2.2. Educational development as a compensatory activity ...... 51 4.3. Reflections ...... 52 4.3.1. The value of the model ...... 52 4.3.2. Similar concepts ...... 52 4 5. AN ORGANISATIONAL PERSPECTIVE ...... 55 5.1. Understanding organisations and institutions ...... 55 5.1.1. The university as a machine ...... 55 5.1.2. Shattering the machine metaphor ...... 56 5.1.3. The institutional logics perspective ...... 61 5.1.4. Practices and identities in the organisation ...... 62 5.1.5. Organisational culture: values, beliefs and assumptions ...... 65 5.2. Perspectives on change ...... 66 5.2.1. Reform as routine – and as producer of hope ...... 66 5.2.2. Institutions as resources for institutional innovation ...... 68 5.2.3. A note on “change management” literature ...... 70 5.2.4. Change in higher education ...... 71 6. DISCUSSION AND CONCLUSIONS ...... 73 6.1 Seeing the duality in the light of institutional logics ...... 73 6.1.1. Practices ...... 73 6.1.2. Competing logics in engineering education ...... 74 6.1.3. Competing logics in research ...... 75 6.1.4. Interplay between education and research ...... 77 6.2. Seeing CDIO in the light of institutional logics ...... 79 6.2.1 CDIO as integration of the academic and professional logics ...... 79 6.2.2. CDIO as institutional innovation ...... 82 6.3. Wrapping up ...... 83 6.3.1. Conclusions ...... 83 6.3.2. Contribution ...... 86 6.3.3. Future research ...... 86 REFERENCES ...... 88 Papers: I. Edström, K., & Hellström, P.-E. Improving student learning in STEM education: Promoting a deep approach to problem-solving. Manuscript in preparation. II. Edström, K. & Kolmos, A. (2014). PBL and CDIO: complementary models for engineering education development. European Journal of Engineering Education, 39(5), 539-555. III. Edström, K. (2017) The role of CDIO in engineering education research: Combining usefulness and scholarliness. European Journal of Engineering Education. Submitted 4 April 2017, accepted 16 October 2017, in press. IV. Edström, K. (forthcoming, 2018) Academic and professional values in engineering education: Engaging with history to explore a persistent tension. Engineering Studies. Submitted 10 Dec 2016, final acceptance pending minor revisions. 5 Acknowledgements This thesis is dedicated to everyone working to improve engineering education. Special thanks: To KTH for establishing this research area and the group, and for providing me with (nearly) everything that Virginia Woolf (1929) specified as necessary for writing. To my fine supervisors for supporting me in all weathers: Anette Kolmos, global doyenne of the engineering education research field who generously came to work with KTH and me in this endeavour, Lars Geschwind, erudite and level-headed team player who created the perfect research climate, Åsa Lindberg-Sand, compulsively intellectual role model, and informal mentor in academic matters for at least six years before I even enrolled as a PhD student. To those with important involvement in parts of the process: Jonte Bernhard, Lena Gumaelius, Roger Hadgraft, Fredrik Lundell, and Arnold Pears. To the PhD student group for stimulating and enjoyable fellowship, in particular those whose paths most overlapped with mine: Per Fagrell, Malin Henningsson, Sara Karlsson, Marie Magnell, Malin Ryttberg, and Johan Söderlind. To editors and anonymous reviewers, for engaging productively with my work. To those who kindly encouraged and supported me at various stages, including but not limited to: Margareta Bergman, Ed Crawley, Oskar Gedda, Ruth Graham, Jenny Grensman, Stefan Hallström, Mats Hanson, Anna-Karin Högfeldt, Aldert Kamp, Viggo Kann, Jakob Kuttenkeuler, Johan Malmqvist, Per Norström, Björn Pehrson, Peder Roberts, Brit Rönnbäck, Bruce Seely, Karin Svedung, Martin Vigild, Maria Knutson Wedel, and Sören Östlund. To THS and all student representatives, for engaging in improving engineering education through the years. To colleagues in many institutions in many countries, for exposing me to the insides of things, by sharing experiences or inviting me into different kinds of work. To my family: Adam, my brilliant and cheerful spouse, life companion, and proof- reader. Viking and Helga, our much loved descendants. Barbro, superb role model as a teacher and learner, and maker of the cover quilt. Sara, Anders, Alice, Johan, and Ida, enthusiastic supporters. 6 Abstract The theme of this thesis is the dual nature of higher engineering education, meaning that it is simultaneously academic, emphasising theory in a range of subjects, and professional, preparing students for engineering practice. The dual nature ideal is however also a source of tensions. Taking a critical approach and embracing the complexities of the issues, the theme is explored in the context of engineering education development, here represented by the CDIO (Conceive, Design, Implement, Operate) approach, founded in 2000 by MIT, Chalmers, KTH, and Linköping University. Cases on programme and course level illustrate how the dual nature ideal is pursued in the development of the integrated curriculum. CDIO is also compared with PBL (problem/project-based learning), which leads to an investigation of opportunities to further emphasise research in the CDIO community. Two critical investigations are made to deepen the understanding of the theme. First, taking a historical perspective, the CDIO approach is compared with the writings of Carl Richard Söderberg (1895-1979), showing the persistence of the academic- professional tension. Further, many of his ideals, arguments, and proposed strategies are fully recognisable in today’s discussion. Notably, Söderberg and CDIO share the ideal of mutually supporting professional and disciplinary preparation, implying that there need not be a zero-sum game in the curriculum. This leads to a critique of the common swinging pendulum metaphor. Next, another critical retrospection is used to problematize engineering education development. Accounts of unsustainable change leads to a model called organisational gravity, explaining the stability of programmes. The model implies two change strategies, each with different availability, risks, resource demands, and sustainability of results. Another consequence was to conceptualise educational development as compensatory work, promoting such values that are necessary for education but insufficiently represented in the organisation. Both these critical accounts suggest widening the perspective from curriculum development per se, to exploring the organisational conditions. Refuting a rationalist “machine” view on the organisation, an alternative theoretical framework is assembled, based on institutional theory. In particular, an institutional logics perspective is applied, focusing on practices and identities in the organisation, and discussing the scope of institutional innovation in the interplay between the organisation and actors on the field level. In the light of the theoretical framework, a tension between two competing professional logics within engineering education is identified: the logics of the engineering profession that we educate for, with the assumption that education is about teaching future engineers, and the logics of the educators’ academic profession, consistent with the assumption that teaching is about conveying theory. A corresponding tension is identified within the research practice: between the university as academia, seeking knowledge for its own sake, and as public service, 7 seeking useful knowledge. The first is consistent with the logics of the academic profession, while the latter shares many values with the logics of the engineering profession. The analysis suggests a double hegemony where the logics of the academic profession are the strongest in both education and research. The two practices are also strongly interdependent, and therefore the more the research practice is dominated by the academic logics, and the more research dominates over education, the more the balance will be tilted also in education, in favour of teaching theory over (other) professional preparation. Analysing the integrated curriculum strategy, leads to the conclusion that its success on the course level is contingent on educators’ ability to unite theoretical and professional aspects, and the success of the programme level is further contingent on the collegial capacity for coordination, between the programme and the courses, and between courses. Finally, the CDIO initiative is conceptualised as a field-level driver of institutional innovation. Some of the strategies are analysed in the light of the theoretical framework, leading to suggestions for strengthening the approach and the community. 8 Svensk sammanfattning (Swedish abstract) Temat för denna avhandling är ingenjörsutbildningens dubbla natur, som samtidigt akademisk, med teori i många olika ämnen, och professionell, med förberedelse för yrkeslivet. Denna dubbelhet är ett ideal men också en källa till spänningar. Avhandlingen antar ett kritiskt perspektiv och väjer inte för frågans komplexitet. Temat utforskas i ett sammanhang av utveckling av ingenjörsutbildning, här exemplifierat av CDIO-initiativet (Conceive, Design, Implement, Operate), som bildades år 2000 av MIT i USA, Chalmers, KTH och Linköpings universitet. Exempel på program- och kursnivå visar hur den ideala dubbelheten eftersträvas i utvecklingen av ett integrerat curriculum. CDIO jämförs även med PBL (problem- och projektbaserat lärande), vilket även lyfter frågan om möjligheter med ett tydligare forskningsengagemang i inom CDIO. Två kritiska undersökningar leder till en fördjupad förståelse av temat. Den första antar ett historiskt perspektiv. CDIO-konceptet jämförs med Carl Richard Söderbergs (1895-1979) idéer, som de uttrycks i hans skrifter, vilket visar att spänningen mellan det akademiska och professionella är mycket långvarig. Dessutom är Söderbergs ideal, argument och föreslagna strategier fullt igenkännbara i dagens diskussioner om ingenjörsutbildningens utveckling. Speciellt delar han med CDIO idealet om synergi mellan det ämnesmässiga och det yrkesförberedande, vilket innebär att spänningen inte behöver medföra ett nollsummespel i curriculum. Detta leder till en kritik av en vanlig metafor, ”den svängande pendeln”. Därefter följer en annan kritisk återblick för att problematisera pedagogisk utveckling. Berättelser om förändringsprojekt där resultatet inte varit långsiktigt uthålliga leder till en modell kallad organisationens gravitation, avsedd att förklara utbildningsprogrammens stabilitet. Som en följd av modellen identifieras två förändringsstrategier, med olika tillgänglighet, risker, resursbehov och resultatens uthållighet. En annan konsekvens var att se på pedagogisk utveckling som ett kompensatoriskt arbete, för att stärka sådana värden som behövs för utbildningen men är otillräckligt representerade i organisationen. Båda dessa kritiska skildringar talar för att perspektivet behöver vidgas från ett fokus på utveckling av utbildningsprogrammen i sig, till att även undersöka villkoren i organisationen. En rationell ”maskinmässig” syn på organisationen avfärdas som otillräcklig, och i stället sammanställs ett teoretiskt ramverk som bygger på institutionell teori. Speciellt används det begreppsliga ramverket institutionella logiker, med fokus på praktiker och identiteter inom organisationen, och där samspelet mellan organisationen och aktörer på fältnivå skapar utrymme för institutionell innovation. I ljuset av det teoretiska ramverket identifieras två konkurrerande institutionella logiker inom ingenjörsutbildningen, som kan härledas till två professioner: dels ingenjörsprofessionen, som vi utbildar för, som är förenlig med antagandet att utbildningen syftar till att utbilda nästa generations ingenjörer; dels lärarnas 9 akademiska profession, som är förenlig med antagandet att utbildningen syftar till att förmedla teori. En motsvarande spänning identifieras inom forskningen: mellan synen på universitetet som akademi, som söker kunskap för kunskapens egen skull, och synen på universitetet som public service, som söker användbar kunskap. Det förra alternativet hänger samman med den akademiska professionens logik, medan det senare har flera gemensamma värden med ingenjörsprofessionens logik. Analysen visar att den akademiska professionens logik dominerar både inom utbildningen och forskningen. Eftersom utbildningen och forskningen också är starkt samberoende är slutsatsen att ju mer forskningen domineras av den akademiska professionens logik, och ju mer forskningen dominerar över utbildningen, desto mer påverkas balansen i utbildningen att tippa över mot teoriundervisning, på bekostnad av den (övriga) yrkesmässiga förberedelsen. En analys av strategin med integrerat curriculum visar att framgången på kursnivå är beroende av enskilda lärares förmåga att förena de teoretiska och professionella aspekterna, och framgången på programnivå även är beroende av lärarkollegiets kapacitet för koordination, dels mellan kurs och program och dels mellan kurserna. Slutligen skildras CDIO-initiativet som en aktör på fältnivå som driver på institutionell innovation. Några av strategierna diskuteras i ljuset av teorin vilket leder till förslag för att stärka konceptet och de gemensamma aktiviteterna. 10 1.1. Theme and research questions 1.1.1. The dual nature of higher engineering education The overall theme addressed in this thesis is the dual nature of higher engineering education. By dual nature is implied that engineering education is simultaneously academic, emphasising theory in a range of disciplines, and professional, preparing students for engineering practice. Hence, the theoretical and the professional aspects are not merely two separate components that need to be balanced in appropriate proportions, but they should also be in meaningful relationships in the curriculum. While the academic-professional duality is an ideal, it is however also a source of tensions. This is a consequential issue for all stakeholders of engineering education, i.e. students, educators, employers, and society in general. And while this thesis explores the theme from the perspective of engineering education development, the same ideals and tensions are also present in other domains. The academic-professional duality is consistent with the stated aims of most engineering programmes, and conceptualised in policy work such as governance, evaluation, and accreditation of engineering education. Similar issues are also relevant for professional education in other fields, such as medicine (see for instance Bolander Laksov, McGrath, & Josephson, 2014; Christakis, 1995). 1.1.2. Development as a starting point The thesis investigates approaches and strategies deployed within endeavours to develop engineering education towards the dual nature ideal, as well as some of the challenges experienced. The relationship between disciplinary and professional aims is a key issue in several reform initiatives with international communities. In this thesis, engineering education development refers to efforts to improve engineering education, with the CDIO approach as the main case (Crawley, Malmqvist, Östlund, Brodeur, & Edström, 2014). Such work is performed and promoted by people in many different roles, including educators in all subjects in engineering programmes, programme managers and other leaders including university management, student representatives and associations, administrators on all levels, specialised educational developers (like myself), professional representatives and their associations, as well as various international and national interest groups and associations. Hence, educational development refers here to the work itself, not to any particular category of people or role. Not only is educational development the context for this thesis, but it is also taken to imply a critical perspective with focus on tensions and conflicting interests. Already 11 using the term development implies a favourable evaluation a priori, as it usually refers to deliberate change to the better. An even stronger normative statement is implied by improvement, which will here be used as a synonym. Both development and improvement are like vectors in that they have a direction as a part of their definition. The intended direction can also be called an agenda, which means that also agency is implied. In the discussion about what development is desirable, there are many different positions possible, but it is a normative, ideological or political debate, meaning that there is no objective or neutral position available. Barnett (1992, p. 6) puts it bluntly: “The debate over quality in higher education should be seen for what it is: a power struggle where the use of terms reflects a jockeying for position in the attempt to impose own definitions of [the aims of] higher education.” Any discussion about the aims of education takes us to contested grounds, with a whole chorus of stakeholders advocating their particular interests. More than half a century ago, Brown (1962, p. 343) observed: “[The] diversity of needs, desires, and opportunities, both educational and professional, is so great that no single pattern of what an engineering education ought to be will serve.” The thesis is also written from a basis of personal experiences in engineering education development. Hence, my role and identity embrace both that of researcher and developer. As a researcher I study opportunities and challenges for change – as a developer I am advocating, enabling and driving it. In Barnett’s terms I am jockeying for a position. Therefore, to maintain credibility as a researcher, I need to be aware of my own perspective, and be open about it. Hopefully, given the full disclosure, the insider perspective might also bring strengths, because “understanding change is just as much a matter of ‘doing’ reform as it is studying it” (Fullan, 1999). I will not pretend to be neutral and objective, since I hardly believe such a position exists, even for researchers. For instance, to embrace the dual nature of engineering education as an ideal, as I did above by making it sound natural and reasonable, is to take a normative stance. While most people would agree that this is an ideal, there are also other positions possible. The fact that the national qualifications framework supports (even mandates) this ideal does not make it neutral; it is still a value statement. 1.1.3. Research questions and structure of the thesis The aim of the thesis is to explore the dual nature of engineering education, by which is meant the ideal that the academic and professional aspects should be mutually supporting. This ideal is however also a source of tensions. The focus here is in particular to investigate opportunities and challenges in efforts for developing engineering education according to the ideal. The investigation will take three main turns: first through the current strategies promoted in international educational development communities, then taking a historic perspective, then critically 12 considering some of the underlying challenges for this kind of educational change. Finally, the strategies and challenges will be related to organisational matters. The first part is an exploration of present-day models for engineering education reform. Focusing on the CDIO approach in particular, particular focus is placed on the strategies promoted to improve the education, and their underlying ideals and ideas about the relation between disciplinary theory and professional preparation. Then CDIO and PBL, another international community for educational development, are defined and related to each other. This also leads to a special investigation regarding the relationship between educational development and research. This interest is expressed as a sub-question: What approaches and change strategies can be identified in major engineering education development communities? (SQ1) Next, the investigation tries to further deepen the understanding of the tensions between the academic and the professional aims of engineering education. This is done through an excursion into the past, tracing some of the historical roots of the issue. Of particular interest is to compare the arguments and positions used in the past with those that are advocated today, particularly in the CDIO approach. The sub- question is: How has the tension between the academic and professional aspects played out in the past, and what can be learned from comparing past and present ideals and debates? (SQ2) Finally, the thesis critically explores some underlying challenges in curriculum development, in particular that of making change sustainable. This leads to a need for a more sophisticated understanding of the organisation, and the conditions for this kind of change. This corresponds to the third sub-question: What challenges apply to the sustainability of educational development in engineering programmes, and how can we understand those challenges in relation to the university organisation as a context for the change? (SQ3) 1.2. Research approach and methodology This section discusses the research approach adopted for this work, making an argument for building understandings through the engagement in practical problems situated in their natural context. 1.2.1. A problem-led and naturalistic approach The aim of the thesis is to produce more meaningful understandings of the relationship between the academic and the professional values in engineering education, in particular as seen from the perspective of educational development. This work thus takes its starting point in a problem, relevant for the practice of educators, 13 programme leaders, educational developers, and many others. Borrego and Bernhard (2011, p. 30) distinguish between method-led and problem-led research. They explain that the value of problem-led research lies in the “quality of the ideas and insights that are generated” and “the light shed on the problem under consideration”. Following Lincoln and Guba (1985, p. 189), it can also be argued that this issue takes its meaning as much from its context as it does from itself. They note that any observations are inevitably time- and context-dependent, and continue: “No phenomenon can be understood out of relationship to the context that spawned, harboured, and supported it”. Recognising the significance of the context where the tension and its different implications are manifest, it was therefore necessary to study it in its naturalistic setting. Since the issue is present on so many levels in engineering education and in the university organisation, it needed to be viewed from multiple angles and temporal perspectives. Relevant here is for instance how the tension is, has been, and can be enacted in the engineering curriculum, and in activities called engineering education development. Barnett and Coate (2004, p. 27) warn that “in the absence of explicit understandings of the curriculum, we are in danger of being steered towards inadequate or overly narrow conceptualizations of curricula”. The curriculum is not created in a social vacuum either, so I will also go further and locate the issues also in the university organisation, focusing on concrete implications for learning and for power. Hopefully, this research may challenge some taken-for- granted ways of working, in order to offer alternative understandings, which can sometimes inform action. Robinson (1993, p. ix) points out that “when researchers intend their work to contribute to the improvement of practice”, it means that researchers should engage with the theorising of the people involved in the problem situation and focus on making holistic and accessible analyses. But as the aim is not necessarily to “solve” the problem, the research can be more or less intervention-based even when addressing a practical problem. As obviously no solution can do away with the tensions in education once and for all, the ambition here is rather to shed light on the problems that can be attributed to the tension, and discuss strategies for handling them more productively. The objective is then to deepen the understanding of the character of the problem and how it is manifest on different levels, questioning the current situation, and considering possible alternatives. In particular, Alvesson and Sandberg (2013, p. 63) argue for producing alternative assumptions as a way to increase understanding. These should be of interest both from an academic view and for people for whom the problem is real and consequential (see also Alvesson, Gabriel, & Paulsen, 2017). 1.2.2. Educational development and critical educational research The engagement in engineering education, the experience base from which this thesis is written, consists of both development and research, with the ambition to bridge these two worlds. The point here is emphatically not that development is the “doing” 14 while research is the “thinking”, because both activities amalgamate doing and thinking. Due to differing requirements on the end results, however, the priority can be slightly differently balanced. Even when development is made in a reflective and well-informed way, the research mode can afford an additional level of reflexivity and distance. In such a context, the curiosity and confusion can be allowed take the lead, and it is possible to linger in the problematizing mode, as the immediate need for practical solutions is relaxed. Of course this may just as well spawn unproductive detours, and the lack of urgency can also become enervating. The demands created by actively working on consequential problems in real life situations, together with people who urgently need to address pressing issues, should not be underestimated; it creates a special kind of acuity, together with the benefits of immediate field test opportunities. In fact, the efforts to change things can be seen as a form of experimentation, an active probing which sometimes provokes interesting responses – potentially revealing clues to forces that are at play in the system, under the surface. Therefore, regardless of what other results are achieved, whether success or failure, educational development can also produce understanding. When new understandings come as by-products of educational development, it would be unethical not to harvest and make the most out of them, in order to grow wiser and to inform future work. More than just a background, the professional activities in educational development constitute a direct breeding ground for the work reported here. It was these experiences that provided the inspiration for the theme and rationale for the research questions. The object of research is engineering education with its ideals and tensions, in particular as they are revealed in educational development activities, and people with an interest in educational development are, together with the research community, the main intended recipients of the results. But, educational development has also influenced the research approach in a more fundamental sense. As pointed out earlier, development attempts to transform practices into something better, and I chose to allow this critical stance to influence the research approach; perhaps this was even inevitable. There is therefore an affinity with the critical research tradition, in which the purpose is not only to understand, but also to confront, the status quo. Critical theory seeks to “uncover the interests at work in particular situations and to interrogate the legitimacy of those interests” (Cohen, Manion, & Morrison, 2011, p. 26). Following Habermas, Cohen et al. (2011, pp. 28-29) suggest that an emancipatory knowledge interest can be addressed by making sense of the current situation, penetrating its causes and purposes, analysing the power and legitimacy of the interests and ideologies at work, and proposing and testing an agenda for altering the situation. They suggest that the curriculum can be seen as a site for ideology and power: “Ideologies can be treated unpejoratively as sets of beliefs, or, more sharply, as sets of beliefs emanating from powerful groups in society, designed to protect the interests of the dominant. If curricula are value-based then why is it that some values hold more sway than others? The link between values and power is strong. This theme asks not only what knowledge is important but 15 also whose knowledge is important in curricula, what and whose interests such knowledge serves, and how the curriculum and pedagogy serve (or do not serve) differing interests.” Cohen et al. (2011, p. 31) By the same token, taking the consequences of research seriously also means considering whose interests are served by the research. This can happen in subtle ways, for instance as a side effect of focusing on some phenomenon or accepting some circumstance without reflection. Alvesson and Sköldberg (1994, p. 327) suggest that if research should not simply reinforce current elite positions, the independent researcher may strive to formulate such research questions that dominant groups may have little interest in having answered, but that are more pressing for disadvantaged groups. 1.2.3. Engagements rather than measurement Alvesson and Sköldberg identify a problematic circumstance for critical research, that researchers are subject to strong socialisation pressures from the research community to conform to established templates for desirable and legitimate research (1994, p. 329). One of the most common templates, also beyond the research community, is that research is all about producing solid evidence through the rigorous application of accepted procedures for generating, organising and interpreting data (see also Bernhard & Baillie, 2016, p. 2379). Following a system of conventions, data should be reduced to produce a limited sample, a well-defined “dataset”, which can be more completely analysed through a transparent process, available for anyone to scrutinise and reproduce. The point of this rigorous process is to “minimize the influence of the researcher’s individuality” (Bishop, 1992, p. 713). The risk with reducing complexity, however, is to diminish the relevance of the findings for the practical issue in its context (Cohen et al., 2011, p. 19). Here, the choice to study an almost omnipresent problem in its naturalistic setting does not suggest adopting formulaic methods, because the complex issues under investigation would be difficult to capture meaningfully, at least in this explorative stage. Therefore, this thesis does not attempt to follow the conventional template. Instead of restricting the mode of inquiry to any given set of operations and rules, pragmatic choices were made to drill gradually deeper into the problem, as it was understood in that given moment. Relinquishing formal reductionism made it necessary to accept and embrace the complexity and see the project as a search for meaning rather than for objective measurement. This places the thesis in the interpretive research tradition. It must be noted that the hope to reach objective truth is limited anyway, (especially) in matters of social reality. I side with Schwartz and Ogilvy (cited by Lincoln & Guba, 1985, p. 55): “There may, indeed, be an ultimate reality. However, every time we try to discover what it is, our efforts will be partial”. In the absence of given formal procedures, Alvesson and Sköldberg (1994, p. 330) emphasise the importance of interpretation and reasoning, and of seeing phenomena in their broad societal context. They emphasise the role of engagement, and 16 recommend researchers to engage on a considerably wider front, applying their imagination, creativity and critical mind-set in a more varied way, than when engaging with a more limited and controlled empirical section. Nevertheless, ruthless selection must be applied in this situation of limitless opportunities, and here the guiding principle was to follow my own most urgent curiosity. This is not to claim that my excitement is sufficient to make the work interesting to others; it is however a necessary condition for being able to create anything of value at all. To exploit my personal engagement, while also sustaining and feeding it, I actively sought out puzzlements and sore spots in my own understandings. The ambition, then, was to do something similar to the description by Alvesson and Kärreman (2011, p. 43): “[The] process of engagement, in which the languages and theories of the researcher are activated, is central rather than the passive mirroring of reality (e.g. through collecting data and coding, processing, and trying to ‘discover’ what is there). This view is different from most conventional approaches, guided by a wish to order, control, and domesticate what is studied. But the impulse to control – through measuring, codifying, checking, and so on – can be bracketed, and a desire to become challenged, surprised, bewildered, and confused may take centre stage in research.” The thesis can be seen as a series of such engagements, highlighting different aspects of a common theme on different levels and from a range of temporal perspectives. Rather than following an initial grand design for the study as a whole, it was a series of open and explorative investigations. The design was emergent, in that each sub- project informed or even spawned from one another, or from work that was done previously or in parallel with the thesis project. The experience resembles the process described by sculptor Antony Gormley: “You don’t get good work without good ideas, but the ideas come from the work. […] And learning to listen to the work that you’ve already made is really where all the core ideas come from. One work is the mother of the next.” (CNN, 2015) Given the pervasive nature of the issue under investigation, there are numerous other matters that could potentially have been part of the thesis, and some are discussed in the section on future research. 1.2.4. The insider and outsider perspective My professional role in educational development has afforded me a simultaneous insider and outsider perspective. As an engineer I am an insider in engineering education, and also by being securely employed at a technical university for twenty years. In educational development I am an insider, through visits, consultancies, commissions, networking, collaborations and discussions with people from other universities worldwide. This has given me privileged access to many discussions and deliberations related to the very issues I am exploring, in a great variety of contexts. 17 Not least, it has allowed me to notice what was interesting also to others. As Weick (1989, p. 517) points out: “…a theory is judged to be more plausible and of higher quality if it is interesting rather than obvious, irrelevant or absurd, obvious in novel ways, a source of unexpected connections, high in narrative rationality, aesthetically pleasing, or correspondent with presumed realities.” The discussions also provided opportunities to test and refine many of the thoughts in this thesis, making them to some extent already jointly considered and validated, albeit informally. At the same time I am also outside the mainstream, not having taken the normal route to a faculty position, through a PhD education in a technical field. However, my strongest outsider factor comes from the commitment to changing the order of things. Moss Kanter pointed out how the position outside the norm can be sensitising: “The Other has to always be super conscious, whereas the dominant player can take everything for granted because the world just makes room for him. I think that dominant players are often less interested in knowing how the world works, because it is working for them, whereas those who feel like the Other are automatically more interested.” (Puffer & Moss Kanter, 2004) Taking the time needed to write this thesis offered a long-lasting opportunity to partly distance myself also from the role as educational developer, to watch such activities from the outside perspective, and consider my own assumptions, motivations, and identity. To some extent this has supported defamiliarization – a strategy for interpretation by making the familiar seem remarkable and less taken for granted (Alvesson & Sköldberg, 1994). Hence, I’m taking a critical stance not only in relation to the status quo, but also to educational development per se. Perhaps this is also simply a sign of educational development coming of age (compare for instance Boud & Brew, 2013; Gibbs, 2013; Jessop & Bolander Laksov, 2017; Roxå & Mårtensson, 2017; Stensaker, 2017). *** The main engagement in this thesis concerns the concepts and communities for engineering education development. The next chapter aims to illustrate what is here referred to as “engineering education development”. 18 The following chapter explores more precisely the nature of the endeavours referred to as “engineering education development” in this thesis. I have chosen to focus on the CDIO approach as a representative of attempts to integrate the academic and professional aims. It is also an important part of the professional experience that spawned this research. The chapter is structured as follows. First, the CDIO initiative is briefly introduced, followed by an exposition of its strategies for integrating the disciplinary theory and professional aims, in curriculum development on the programme and course level, and in faculty development. Along the way, a few mini- cases are presented as illustrations and some of the literature found useful in this endeavour is reviewed. 2.1. Engineering education development – the CDIO approach 2.1.1. Taking the initiative The CDIO Initiative for engineering education reform started as a project in 2000 by the Massachusetts Institute of Technology (MIT) in the United States, and three Swedish universities: Chalmers, KTH Royal Institute of Technology and Linköping University. The starting point was the recognition that engineering education had become increasingly distanced from engineering practice, as engineering science had replaced engineering practice as the dominant culture among faculty in the past decades (Crawley, 2001). This created a need to “educate students who understand how to Conceive-Design-Implement-Operate (CDIO) complex, value-added engineering systems, within a modern team-based engineering environment”. In the original funding application, the partners stated that by embedding hands-on engineering experience, “education will be improved in two ways: it will give students a deep working knowledge of the fundamentals; and it will simultaneously educate the students in the system development process” (MIT, 2000). Each university chose a pilot programme as project partner: it was the Aeronautics and Astronautics programme at MIT, the Vehicle Engineering programme at KTH, the Mechanical Engineering programme at Chalmers, and the Electrical Engineering and Applied Physics programme at Linköping university. The four partners set out to jointly develop the reform concept methodology, and simultaneously applying it in their respective programmes. Quite soon, other universities showed an interest and were welcomed as collaborators. When the first edition of the book Rethinking Engineering Education: The CDIO approach was written (Crawley, Malmqvist, Östlund, & Brodeur, 2007) some twenty institutions had already joined, by the time of the second edition (Crawley et al., 2014) they had reached one hundred, and to date the CDIO Initiative is a worldwide community with over 140 member institutions. 19 See Figure 2.1 for a world map. The CDIO community holds two international meetings per year, one of which is the annual conference. Most regions, colour-coded in Figure 2.1, also organise annual regional meetings. The organisation has evolved with democratic elections of leaders and council members, whereas the ten first members previously held permanent seats. For more details on the history of CDIO see paper IV (Edström, forthcoming 2018). In the following, the resulting reform concept is described. Figure 2.1. World map of CDIO collaborators, 2017, made with Google My Maps. Retrieved from www.cdio.org, where a complete list of collaborating institutions can also be found. The programme-level scope is a key defining feature of CDIO. Since students experience a programme, it should not be seen “as a set of elements, but as a system in which each element carries both individual and collective learning objects for the program” (Crawley et al., 2007, p. 17). The CDIO curriculum model can essentially be characterised as programme-centric curriculum development with an outcomes- based approach. In essence, the curriculum theory implied in CDIO specifies a number of logical links, with the programme at the centre. The key characteristic of the integrated curriculum is the ideal to integrate the theoretical and the (other) professional aims, in every stage of this system: The starting point is to formulate a vision of what engineers do. What students therefore need to learn is expressed as intended learning outcomes at the programme level. These are apportioned to the course level, as course learning objectives. The course learning objectives are finally reflected in the design of learning activities and assessment of student learning outcomes. In the steady state, these links are continuously improved through cycles of evaluation and development involving the programme stakeholders. 20 It is worth noting that today the outcomes-based approach is mainstreamed in large parts of the world, but at the time when the CDIO initiative was started it was quite novel. This was not least true for the Swedish universities. At the time, the US-based Accreditation Board for Engineering and Technology (ABET) had adopted an outcomes-based accreditation scheme from 1997 (ABET, 1994), so the MIT team were ready to share experiences of formulating and using learning objectives. The Swedish partners could contribute to the curriculum model the ideas of constructive alignment (Biggs, 1999), which provided principles for outcomes-based course design. In 2007, when the same paradigm was implemented in Swedish higher education through the Bologna process (Prop. 2004/05:162), the CDIO collaborators had up to six years experience of outcomes-based curriculum development of their own volition. At the CDIO member universities there was considerable new expertise, which became sought after by colleagues in other programmes and in other universities. Hence the Bologna implementation could to a larger extent be interpreted as a genuine opportunity for meaningful development, and less as a bureaucratic imposition (cf. Aamodt, Frølich, & Stensaker, 2016; Bleiklie, Frølich, Sweetman, & Henkel, 2017; McGrath & Bolander Laksov, 2014). The CDIO model for curriculum development is tightly controlled through the official documents, mainly the CDIO Syllabus and the CDIO Standards, and at the same time completely open source, meaning that one can pick and choose, modify and adapt as desired, even give it a new name. Together with the great diversity among member institutions with their various specific circumstances and needs, this makes implementations considerably different with many “dialects”. What will be presented here is a generic model, as defined by the standards, along with illustrations from implementations at Chalmers and KTH, both technical universities in the Swedish context and original CDIO founders. The following description is structured along the framework of the CDIO Standards. The main objective is to show the attempts to integrate disciplinary theory and professional aims through curriculum development, first on the programme level, then on course level, and finally in faculty development. Here, it is worth reiterating that development is a normative activity; it is directed towards some values. Hence, there can be no such thing such as value-free development. This section will also show the values embedded in the CDIO concept, as well as some of the rhetoric used to promote these values. 21 2.1.2. Programme level development CDIO Standards for programme development Standard 1. The Context Adoption of the principle that product, process, and system lifecycle development and deployment – Conceiving, Designing, Implementing and Operating – are the context for engineering education. Standard 2. Learning Outcomes Specific, detailed learning outcomes for personal and interpersonal skills, and product, process, and system building skills, as well as disciplinary knowledge, consistent with program goals and validated by program stakeholders. Standard 3. Integrated Curriculum A curriculum designed with mutually supporting disciplinary courses, with an explicit plan to integrate personal and interpersonal skills, and product, process, and system building skills. Standard 12. Program Evaluation A system that evaluates programs against these twelve standards, and provides feedback to students, faculty, and other stakeholders for the purposes of continuous improvement. The starting point for curriculum development is to form a vision for the professional competence of graduates (standard 1) and express it as intended learning outcomes for the programme (standard 2). The dual nature ideal is explicit, by stating that the learning objectives should reflect a deep working knowledge of the fundamentals, as well as the professional competences for technology development and deployment. Standard 2 also specifies the need to engage with programme stakeholders. Per standard 3, the programme level objectives are broken down and assigned to the course level, integrating disciplinary fundamentals with professional engineering skills. The result, the integrated curriculum, is often documented by a matrix, showing the responsibility of each course towards the programme learning objectives (Malmqvist, Östlund, & Edström, 2006). Standard 12 devises a continuous programme evaluation system, again involving stakeholders. The integrated curriculum – the case of Mechanical Engineering at Chalmers To illustrate the programme development in CDIO, we turn to the Mechanical Engineering programme at Chalmers, one of the four original project partners. It is a five-year programme, combining a Bachelor and Master of Science in Engineering. Their experiences are documented through a series of publications, not least in CDIO conferences. Though mechanical engineering can be the broadest of fields, the Mechanical Engineering programme has a vision of the work it prepares for, namely: “to participate in and lead the development and design of industrial products, processes and systems for a sustainable society. The programme also prepares for positions in other areas of the society where skills in analysis and processing of complex open-ended problems are of great importance. During the studies, the student shall be able to develop her/his personal qualities and attitudes that will contribute to professional integrity and to a successful professional life” (Malmqvist, Bankel, Enelund, Gustafsson, & Knutson Wedel, 2010, p. 3) 22 The curriculum development is documented in the programme description (Malmqvist et al., 2006). Its function is to communicate the current state of the programme and the rationale, and also the next steps. It makes it easier for the programme team to stay focused and prioritise among new ideas and proposed actions, since these will be discussed in terms of their contribution to the goals of the programme (Malmqvist et al., 2010). The programme description documents how ethics, communication and teamwork skills, etc., are integrated in the course learning objectives, according to standard 3. For this thesis, one of the most interesting developments in the Mechanical Engineering programme has been the integration of computational mathematics, which has strengthened the connection between engineering and mathematics. The rationale was, in short, that students need to learn to solve more general, real-world problems, while they can spend less time “solving oversimplified problems that can be expressed analytically and with solutions that are already known in advance” (Enelund, Larsson, & Malmqvist, 2011). One of the guiding principles was that students should work on the complete problem: from setting up a mathematical model and solving it, to simulation of the system, using visualisation to assess the correctness of the model and the solution, and comparison with physical reality. The interventions in the programme involved new basic math courses including a an introduction to programming in Matlab (a technical computing language and environment), new teaching materials (since most textbooks do not take advantage of the development in computing), integration of relevant mathematics topics in fundamental engineering courses (such as mechanics and control theory), and cross- cutting exercises, assignments and team projects shared between the mechanics and strengths of materials courses and mathematics courses. We can note that instead of seeing this as a task for mathematics teachers to solve within the mathematics courses, a programme-driven approach was applied, where making connections to mathematics in engineering subjects was at least as important as making connections to engineering in mathematics. Just as in the previous example, the integration of sustainable development demonstrates how the programme approach enables systematic integration of important topics in several courses, while maintaining links to overall programme learning outcomes and ensuring progression (Enelund, Knutson Wedel, Lundqvist, & Malmqvist, 2013). Programme learning objectives express the sustainability competences in the Mechanical Engineering program, for instance that students should be able to “describe and estimate the economic, societal and environmental consequences of a product or system through its lifecycle”. Through the programme, sustainability elements are pervasive and adapted to the context. Course learning objectives show how courses carry partial responsibility in relation to these programme objectives, and in progression through the programme. Students first encounter sustainability in the Introduction to Mechanical Engineering (standard 4). It is then integrated into several of the engineering fundamentals courses where it is 23 applicable, e.g. in Thermodynamics, Materials Science, Material and Manufacturing Technology. There are also courses with sustainable development as a main topic, such as Sustainable Product Development. Finally, the specialisations on master level also have various degrees of sustainability focus. A significant aspect of this case is how the education is organised, and here the model developed by the CDIO team in Mechanical Engineering has also had considerable influence across Chalmers. For strategic issues and prioritisations the programme leader is supported by an advisory board, with industry, students, admin and faculty represented. For operational issues, the programme office, with an administrator and a study counsellor, supports the programme leader. Chalmers has a “buyer-seller” model in which the programmes commission courses from the delivering departments. In a yearly cycle, the programme leaders reviews the evaluations for all courses, and negotiates next year’s course offering in a dialogue with the vice head of the delivering department. An agreement is written to document learning objectives, content, pedagogy and budget of the courses delivered by the department. While the agreement process is a collegial dialogue, in the end the programme controls the budget, approves the course syllabus documents, and is the recipient of course evaluations. As a result, this has enabled the programme team to implement the integrated curriculum, keeping the programme unified while still being a composite of courses from several departments and disciplines. As a result, the curriculum can also be further developed through a relatively agile process. In summary, the Mechanical Engineering programme has systematically created conditions for leading, planning and developing the programme, and for constantly setting new goals. It has come out on top of national evaluations, and attracted numerous awards (Malmqvist et al., 2010). Further, this organisational model, with the strong power bases in the programmes, has influenced the education organisation across Chalmers. For the university, it is a mechanism to ensure that the educational resources are spent where they benefit the programmes, as no course is established and offered unless a programme commissions it, and keeps including it in the yearly agreement. 2.1.3. Course level development CDIO Standards for course design Standard 7. Integrated Learning Experiences Integrated learning experiences that lead to the acquisition of disciplinary knowledge, as well as personal and interpersonal skills, and product, process, and system building skills. Standard 8. Active Learning Teaching and learning based on active experiential learning methods. Standard 11. Learning Assessment Assessment of student learning in personal and interpersonal skills, and product, process and system building skills, as well as in disciplinary knowledge. 24 Standard 7, 8 and 11 constitute a course design model corresponding to constructive alignment: the learning objectives, learning activities, and assessment should be aligned. The integration between disciplinary knowledge and professional skills should apply in all these components. In the integrated curriculum (standard 3) each course accepts responsibility for a portion of the programme objectives regarding some professional competence, in addition to the deep working understanding of fundamentals in the subject. This integration should also be reflected in the way the course is taught (standard 7 and 8), and assessed (standard 11). For instance, in the Mechanical Engineering case above, the planning on programme-level (standard 3) went hand in hand with programme-driven course development, to address the learning objectives that were assigned to courses. In the following, two cases are presented to illustrate CDIO educational development on course level. The two cases, one a subject course and the other a design project course, were chosen to represent the dual nature of educational development in CDIO, which recognises the discipline-led as well as the problem- or practice-led components of education. Table 2.1 shows some arguments for why both logics are necessary, and how they can form a productive relationship. Table 2.1. The need for both discipline-led and problem/practice-led learning. Adapted from (Edström & Kolmos, 2014) Discipline-led learning is necessary for: Problem/practice-led learning is necessary for: Creating well-structured knowledge bases Integration and application, synthesis Understanding the relations between Open-ended problems, with ambiguity, trade-offs evidence/theory, and model/reality Problems in context, including human, societal, Methods to further the knowledge frontier ethical, economical, legal, etc. aspects Practicing professional work modes Design – in Theodore von Kármán’s words: …while also connecting with problems and ”Scientists discover the world that exists; practice: engineers create the world that never was” (NSF, Deep working understanding (ability to apply) 2013) Seeing the knowledge through the lens of problems …while also connecting with disciplinary Interconnecting the disciplines knowledge: Integrating skills, e.g. communication and Discovering how disciplinary knowledge is used collaboration Reinforcing disciplinary understanding Creating a motivational context These cases illustrate some of the improvements advocated by the CDIO approach, but they are examples and by no means complete. One reason for selecting them is that they share a common theme, which was to represent cost-effective implementations. Improving student learning in a subject course – a case study Paper I in this thesis exemplifies CDIO development on the course level, in the context of discipline-led learning. The role of this paper is to indicate how a subject 25 course can improve its contribution to professional preparation while at the same time strengthening students’ understanding of the technical fundamentals. Hence, it shows that the ideal of synergy between disciplinary and professional aims can be realised on the course level. Edström, K., & Hellström, P.-E. Improving student learning in STEM education: Promoting a deep approach to problem-solving. Manuscript in preparation. The paper describes and analyses the results of an intervention for improving learning in problem-solving sessions, called student-led exercises. Briefly, the teaching method works as follows: instead of the teacher demonstrating a set of problems on the board (which is considered “normal” or traditional at KTH), students are randomly selected to present their solutions, which they have prepared in advance. The paper describes how this teaching method was implemented at KTH in a course on Semiconductor Devices by the second author, Per-Erik Hellström. Further, Carl Henrik Görbitz applied the same method in the very large first-semester Introduction to Chemistry at the University of Oslo. The paper presents quantitative data in the form of course results, qualitative data in the form of student interviews made mainly for evaluation purposes, and teacher reflections over the experiences. From a methodological perspective it was valuable to have two contrasting implementations in different contexts (a very large, first-semester course vs. one in the third year with a smaller class), because they could provide different insights regarding the potential advantages of the teaching method. While the results of the Semiconductor Devices implementation indicated improved understanding and motivation, the most consequential result in the Introduction to Chemistry was a significant decrease in dropouts. The results demonstrate how even a modest and cost-effective intervention can improve the contribution of subject courses, improving students’ understanding of disciplinary theory while also allowing them to practice communication skills (Standard 7). The point here is to demonstrate that every ordinary subject course should be able to contribute to the integrated curriculum at least on this very modest level. It also shows how the deliberate integration of relevant skills also generates an active learning format (Standard 8). The activity where students prepare, present, and discuss the solutions is far better aligned with professional practice than an activity where they are mainly copying given solutions, for cramming later. Since the intervention increases student understanding of the subject, and is cost-neutral in terms of teacher time, this is a contribution to professional preparation that every subject course should be able to achieve. In fact, even for an educator who is mainly focused on conveying theoretical understanding, the intervention is justified already by considering the improvement in student understanding, and the practicing of communication skills comes as a bonus. 26 To classify the quality of intended learning outcomes the Feisel-Schmitz taxonomy (Feisel, 1986) (see paper I for an explanation) has been found useful in CDIO because it makes a clear distinction between problem-solving with or without understanding. Problem-solving with understanding, labelled “Solve” in the taxonomy, precisely captures the aim referred to in CDIO as deeper working knowledge. Problem-solving without understanding, called “Compute” in the taxonomy, relates to one of the most problematic issues in engineering education: the focus on reproducing given solution procedures for standard types of problems. Therefore, taxonomies that downplay this distinction are unhelpful in the context of engineering education development. In the most widely used taxonomy, by Bloom (1956), the application category is placed, as a whole, on a higher level than understanding. In the revised Bloom’s taxonomy (Krathwohl, 2002), the parallelism between understanding and application is better recognised, and the new two-dimensional model can accommodate the distinction, although in a more complicated scheme than Feisel-Schmitz. As an analytic tool the Feisel-Schmitz taxonomy tends to resonate widely with engineering educators, including also those who are most interested in disciplinary accomplishments. Hence, the taxonomy has helped identifying common ground, by highlighting the importance of disciplinary theory for professional practice. Approaches to learning are used to operationalize the quality of learning processes, given how a deep approach is associated with better learning outcomes than the surface approach (see for instance Marton, Hounsell, & Entwistle, 1984). Most notably this is a conceptual underpinning to constructive alignment (Biggs & Tang, 2011), which implies that learning objectives, learning activities, and assessment should be aligned to invite a deep approach, and discourage a surface approach. Extending the classic deep and surface approaches, Case and Marshall (2004) identified the deep and surface procedural approaches in relation to problem-solving. In paper I, we proposed an amendment to their model, arguing that the deep procedural approach should not only be treated as an intermediate stage towards a more desirable (conceptual) deep approach. While we agree that problem-solving as a learning activity is a means to reach conceptual understanding, it is not only that; it is also about learning to solve problems. This led us to position problem-solving as an aim in its own right, on the same level as understanding concepts and theory. Again, the intention is to find conceptual common ground, acceptable to those who emphasise disciplinary theory as well as those who emphasise what students can do with their understanding. Finally, if the approaches to learning focus on what students do to learn, based on their intentions, the research on epistemological views (Gainsburg, 2015; Perry, 1998) can further explain this by highlighting their views on knowledge. Gainsburg identifies that students with the more sophisticated views increasingly connect mathematical modelling of course problems with the real problems they represent, and with the nature of problems and processes used in engineering practice. 27 CDIO Standards for problem- and project-led learning Standard 4. Introduction to Engineering An introductory course that provides the framework for engineering practice in product, process, and system building, and introduces essential personal and interpersonal skills. Standard 5. Design-Implement Experiences A curriculum that includes two or more design-implement experiences, including one at a basic level and one at an advanced level. Standard 6. Engineering Workspaces Engineering workspaces and laboratories that support and encourage hands-on learning of product, process, and system building, disciplinary knowledge, and social learning. PBL, or problem-based and project-organised learning, is an essential component in the CDIO curriculum model. Here, students can work in the logic of real problems (Jonassen, 2014; Jonassen, Strobel, & Lee, 2006). Standard 4 and 5 can be seen as special cases of standard 7, since both describe two kinds of integrated learning experiences. Standard 4 recommends an introduction to engineering early in the programme, to give students a first contact with engineering practice and the role of engineers. Standard 5 implies a sequence of design–implement experiences, with progression across the curriculum. By design-implement experiences are meant projects in which the students learn through the development and deployment of products, processes or systems, under working modes that resemble engineering practice. A key feature is to take solutions to a testable state, allowing students to evaluate and reflect on their work, with regards to the process and the results. Standard 6 is about creating a learning environment to accommodate such realistic engineering experiences. It is a cornerstone of the CDIO philosophy that the hands-on component should run continuously across the curriculum, starting early and progressing through the programme. This can be seen as a reaction to curricula where the first years are filled with basic theoretical subjects, where students risk losing sight of why they wanted to become engineers in the first place (see for instance Holmegaard, Madsen, & Ulriksen, 2016; Holmegaard, Ulriksen, & Madsen, 2010). Improving student learning in a project course – a case study The following case is based on the experiences in a master level design project course taught by Jakob Kuttenkeuler and Stefan Hallström, from the Vehicle Engineering department at KTH, one of the original founding partners of CDIO. The teachers have involved me in discussing and designing improvements to the teaching and assessment on a regular basis since 2001, and our joint reflections and experiences have been reported (Edström, El Gaidi, Hallström, & Kuttenkeuler, 2005; Edström, Hallström, & Kuttenkeuler, 2011; Hallström, Kuttenkeuler, & Edström, 2007) and in a book chapter (Hallström, Kuttenkeuler, Niewoehner, & Young, 2014). It is not the intention here to explain project courses generally, but to describe the course and experiences sufficiently for illustrating two points: 28 The learning perspective – The case shows a learning-centred design of teaching and assessment. In short, the purpose is not that the students should build things; it is that they should learn from building things. The teaching perspective – The case shows some principles for making this learning activity sustainable from a teaching perspective, as project-based learning is often assumed to be expensive and require high teaching effort. The course mixes students from several programmes, and its name is Naval Design or Lightweight Design depending on which programme a student comes from. The scope is 20 ECTS credits spread over an entire academic year, i.e. allocating one third of students’ time. Students are divided into large groups, typically of 8-15 students, and given an open-ended task to design, manufacture and test a technical system, typically an unorthodox vehicle. Previous groups have built things like a solar powered aircraft, an autonomous underwater glider, a craft that can plane on the surface but also submerge, an electric single-hydrofoil vehicle for play, and a human-powered submarine (for video clips, see Kuttenkeuler, 2017). While the technical challenge is new for every group, the learning objectives are the same every year. See Table 2.2. The course design and teaching philosophy is guided by some key principles. The first principle is that students are directly exposed to real problems in the project work. In other words, teachers do not stand between the students and the problems. Most previous courses follow the cognitive structure of a subject, where textbook problems are used to illustrate theory, and where the teacher knows the right answers in advance. There, it is often clear from the course context what sort of problem it is and what theory should be used. In contrast, the problems that emerge in the project work come without any labels telling students what theory is relevant. Some problems may require students to search for and use theory and methods that are new to them. 29 Table 2.2. Intended learning outcomes of the Naval/Lightweight Design course. Students should be able to: Examples of related challenges: take on technical problems in a Knowing and prioritising the crucial challenges and keys to systems view success. Where to start. Considering the implications of different concepts (solutions). Handling the interfaces between sub-systems. handle technical problems which How to handle interdependent tasks, e.g. idling while just are incompletely stated and subject waiting for data from each other. How can the work be to multiple constraints assigned to individuals but the big picture maintained? develop strategies for systematic Knowing what aspects matter most, and keeping focus on choice and use of available them. Choosing the right level of precision, e.g. start by engineering methods and tools sketching on napkins rather than using supercomputers. make estimations and appreciate Using estimations correctly, revisiting and challenging them. their value and limitations Interpreting results in the light of assumptions. make decisions based on acquired Creating a relevant basis for decisions. Act when the knowledge information is good enough. Documentation and traceability. pursue own ideas and realise them Discussing, arguing, debating, standing up for your standpoint, practically and letting go of darlings. Struggling with real world conditions, e.g. there is no infinitely strong glue. assess quality of own work and Reflecting on different approaches. Seeing where one’s work work by others made a difference. Reflecting on what can be improved. Role modelling. work in a true project setting that Decision-making. Minimising idling. Identifying time-critical effectively utilises available tasks. Professionalism. resources explain mechanisms behind How to interpret and handle problems. Getting true status progress and difficulties in such a overviews and responding appropriately. setting communicate engineering – orally, Using all possible modes of communication in authentic in writing and graphically situations. The course design and teaching philosophy is guided by some key principles. The first principle is that students are directly exposed to real problems in the project work. In other words, teachers do not stand between the students and the problems. Most previous courses follow the cognitive structure of a subject, where textbook problems are used to illustrate theory, and where the teacher knows the right answers in advance. There, it is often clear from the course context what sort of problem it is and what theory should be used. In contrast, the problems that emerge in the project work come without any labels telling students what theory is relevant. Some problems may require students to search for and use theory and methods that are new to them. At the same time, it can be troublesome for the students to recognise even the most fundamental theory in the wild, such as Newton’s Second Law, Ohm’s Law, or Archimedes’ Principle. As students need to learn how to handle unforeseen and poorly defined problems, the teachers avoid stepping in too hastily to “help” with “correct” interpretations. Further, since the problems are open-ended and in no way prepared or adjusted, there are no black-or-white right answers. Instead of using the teachers for convenient affirmation that answers and solutions are “right”, students need to seek different forms of validation and develop their own judgement. In other 30 words, students need to think for themselves, and this can only happen to the extent that the teachers can resist accommodating student expectations, i.e. the desire to get right answers or to avoid the stage of bewilderment. Until students get accustomed to this new order of things, their conceptions of student and teacher roles are often challenged, and so are their epistemological views. There is a tension here, between learning and task achievement. Naturally, if the teacher were constantly “helping” the students, they could build a better boat. Instead, they are allowed to face these highly relevant challenges, because learning is more important (Edström et al., 2005). In order to prioritise learning, the student teams also need to take full responsibility and ownership. This principle is that the students own the project, all aspects of it. If teachers were to start taking initiatives, it could shift students to a more passive role. A major implication is that the teachers’ role is to coach and advise in the engineering process, but not to drive it, and never suggest solutions. Hence, students are not protected from mistakes, contradictions or confusion. As a result of this principle, the project results will reflect the proficiency of the students, not of the teachers. Again, learning is prioritised over the product performance. A related principle is that the project sets the logic, not the teachers. This means that teachers refrain from unnecessarily making decisions in the project. E.g. deadlines are not set by teachers, but by the project plan created by the student team. Teachers do not specify the length of a report; it is inferred by what it needs to achieve in the project. For instance, when the project commissions an investigation by a sub-team, their report should contain precisely the information needed to make the subsequent decision – and the length, and the deadline, follow as consequences of its function. For many students this is the first time they write a document that actually has a function; previously they have mostly written to demonstrate to teachers that they deserve a grade, so their normal mind-set is: “What does the teacher want?” When they let go of the teacher orientation, and start to become project-oriented, their work becomes much more meaningful, and easier. Obviously, when teachers refrain from managing (and micro managing) the project, it also makes the course far more sustainable in terms of teacher time. The assessment system is also designed to generate learning. We note that it is common in project courses to grade group products (or final reports). In our opinion, product grades are loosely related to learning outcomes, and they create disincentives for learning, because when students focus on task achievement they tend to share the task so each of them can do what they already do best. Group grades are also aggregated to hide individual attainment – which is inherently unfair and often creates conflicts due to different levels of ambition. In this course, instead, teachers grade students individually based on the learning outcomes as evidenced in the process. Since students work on many different tasks, the principle is that the students take responsibility for their own learning outcomes, individually. Furthermore, we believe that doing is not sufficient for learning; students need to reflect in order to turn experience into learning. A portfolio assessment system (for details, see Edström et 31 al., 2005) is designed to generate reflection in relation to the learning goals. For a mid-course formative peer feedback round, each student submits a one-page self- evaluation, which is distributed to all members of the team. It is based on the portfolio and structured according to the learning objectives, with any claims substantiated by referencing project documents that are openly available on the project website. Writing feedback to up to 14 teammates is a comprehensive task, but it is justified by the reflection it elicits. The mid-course feedback comes when students still have another semester ahead of them to make adjustments. For instance, students may discover that they need to engage in different tasks in order to reach all the learning objectives. At the end of the course, after a second peer feedback round, grades are set, individually and in relation to the learning objectives, by the two teachers. Each teacher notes preliminary grades independently, based on a holistic assessment of the portfolios and the work referenced, the feedback given and received, as well as continuous observations throughout the course. They then meet to compare and discuss until reaching consensus. From a teaching perspective, the assessment takes no more time than in other (so called normal) courses. The portfolio model reverses the burden of proof; it is up to the students to show evidence of their individual learning. Furthermore, teachers do not take it upon themselves to provide written feedback; when the students give (formative) feedback to each other, they learn from both the act of giving and of receiving. The experiences in these two cases, in Hellström’s subject course and in Hallström’s and Kuttenkeuler’s project course, clearly showed new demands on the teacher competence, regarding what to teach, and how to design the learning activities and learning assessment. Next, we turn to the matter of faculty development. 2.1.4. Faculty development CDIO Standards for faculty development Standard 9. Enhancement of Faculty Competence Actions that enhance faculty competence in personal and interpersonal skills, and product, process, and system building skills. Standard 10. Enhancement of Faculty Teaching Competence Actions that enhance faculty competence in providing integrated learning experiences, in using active experiential learning methods, and in assessing student learning. Standard 9 and 10 both concern enhancement of faculty competence. These are also the standards that are the least discussed, and for which the least progress has been reported by CDIO implementers (Malmqvist, Hugo, & Kjellberg, 2015). The term “enhancement” has often been taken synonymously with activities supporting the further development of the existing faculty, but it may just as well refer to the composition of the faculty, for instance through hiring and promotion criteria (Theodorsdottir, Saemundsdottir, Malmqvist, Turenne, & Rouvrais, 2013). One general challenge with recommending faculty development as part of a programme- 32 centred development concept is that although it is an important condition for success – in fact often the most critical – it is often a domain in which the programme has little influence. This was the case at Chalmers, for instance, where the programme buys courses from departments, but has no (formal) influence on processes ensuring teacher competence, such as hiring and promotion (Malmqvist et al., 2010). Even in systems where a department owns a programme, faculty recruitment and development may prioritise the needs of research over those of education. Cautious steps are taken in many places to strengthen faculty engineering competence and teaching competence. Such policies are most often university-wide. At MIT, a limited number of Professors of the Practice can be hired (de Weck, 2004; MIT, 2017). In every hiring and promotion case at Chalmers, at least one of the external evaluators is a teaching expert focusing on the teaching competence of the candidate. In addition, the Chalmers appointment regulations specify special positions based on professional skills, as well as positions up to Professor (not holding a chair) with emphasis on pedagogical expertise (Chalmers, 2013). Though CDIO Standard 9 is simply named enhancement of “faculty competence”, it really refers to faculty professional engineering competence, expressed as their “personal and interpersonal skills, and product, process, and system building skills”, which are best developed “in contexts of professional engineering practice”. Examples of actions to support faculty engineering competence are: sabbaticals to work in industry (including the public sector), partnerships with industry in research and education projects, valuing engineering practice as a merit in hiring and promotion, allowing and encouraging consultancy work, and professional development activities at the university (Malmqvist, Gunnarsson, & Vigild, 2008). It can be noted that the theoretical and scientific competence of faculty is not even mentioned; this is taken for granted, perhaps reflecting the prevailing academic culture of the research-intensive universities where the CDIO approach was first developed. CDIO Standard 10 concerns enhancement of faculty teaching competence. In the Swedish context, most universities offer courses on teaching and learning to faculty. One reason is that ten weeks of such training was for many years a national eligibility requirement for senior lecturers and professors (Lindberg-Sand et al., 2005). At KTH, the faculty development activity created an opportunity for mainstreaming the CDIO approach to course and programme development. One of the faculty development courses, Teaching and Learning in Higher Education, 7,5 ECTS credits (i.e. half the requirement), was redesigned in 2004 to emphasise matters of course design, inspired by the experiences with CDIO. Some 700 participants took the course during the decade when it was offered. One teaching strategy applied in this course was to engage other faculty members as guest teachers, presenting their own experiences of course development. Those cases were analysed as examples, to derive theoretical principles for guiding practical implementation. The presence of the guest teachers also demonstrated locally developed proofs-of-concept, showing that it works here. 33 The most prominent cases were the cases discussed above, the design project course presented by Jakob Kuttenkeuler and Stefan Hallström, and the student-led exercises (featuring in paper I), presented by Per-Erik Hellström. The cases in this chapter can therefore also to some extent serve as illustrations of standard 10. An increasing emphasis on scholarship in CDIO conferences (discussed in the next chapter) could also be seen as a dimension of faculty development, by generating more systematic and scholarly reflection and documentation. 2.2. Further development of the CDIO concept and community The previous section discussed the development of the CDIO approach (Crawley et al., 2007; Crawley et al., 2014), and showed its strategies for integrating disciplinary and professional learning through curriculum development, and faculty development. CDIO was chosen not because it is the only model available, but because it is representative of an effort to address the tension that is the theme here, and part of the professional engagement that is the background. The rest of this chapter will discuss two engagements to further develop the approach and the community. First, CDIO will be compared with PBL, another educational development concept with a large international community that also addresses the tension between professional and disciplinary aspects. Finally we discuss a present proposal to connect the CDIO community and the field of engineering education research. 2.2.1. Comparing CDIO and PBL In paper II, CDIO is compared with PBL (problem-based/project-organised learning). Both are models for reforming engineering education with organised international communities. This study came about because both authors, rooted in the PBL and CDIO communities respectively, had often been asked, “Should we do PBL or CDIO?” We felt the need to produce a thorough answer with a systematic approach. The resulting publication was: Edström, K. & Kolmos, A. (2014). PBL and CDIO: complementary models for engineering education development. European Journal of Engineering Education, 39(5), 539-555. In this study, the main methodological challenge was to generate a framework for the analysis, and this was done in stages through an inductive, participatory, and iterative approach. The project started by generating a gross list of aspects that could be compared. In three conference workshops, with a total of 70 experienced practitioners as participants, our first iterations of comparisons were presented and discussed. During the workshops some aspects emerged as most salient and productive in generating insights, by revealing similarities, differences, surprises, misconceptions, or unreflected assumptions. The final framework consisted of the following core aspects: history, community, definition, curriculum design, relation to disciplines, engineering projects, and change strategy. Through a correspondence between the 34 authors, complemented with document studies, these aspects were then examined and analysed for PBL and CDIO, respectively, then contrasted. See Table 2.3. Table 2.3. Summary of the PBL and CDIO comparison (Edström & Kolmos, 2014). PBL CDIO The starting point is the learning process. The starting point is a vision of graduates’ Starting Started in reform universities in the 1960s competence expressed as learning outcomes. point and 1970s, in response to critical student Started at MIT in the late 1990s, forming a movements. Applicable in medicine, project with three Swedish universities, in engineering, science and many other fields. response to distancing of engineering education from engineering practice. Implementation of PBL cannot be estimated, About 140 institutions are formally CDIO Communities due to different levels of implementation collaborators in the CDIO Initiative. Extent of from a single course to whole universities. CDIO implementation is difficult to estimate. Several international networks: PBL Global Network, International PBL Symposium, Pan-American Network for PBL. A broad educational approach, focusing on The CDIO Syllabus addresses what students Definitions the learning process, and loosely defined. The learn. The CDIO Standards address strategies principles can be applied on course, for curriculum and faculty development. programme, or institutional level, in different Practices vary among implementing fields of education, and any level from school institutions. to university. Practices vary, with the McMaster/ Maastricht and Aalborg models well documented. Projects are the platform for student learning. The 12 CDIO Standards describe an outcomes- Curriculum based approach for designing the integrated design curriculum. Many hybrid models where at least half the Subject courses a major part of the curriculum Relation to curriculum is subject-based. as integrated learning experiences: students disciplines should master a deeper working knowledge of technical fundamentals and simultaneously develop professionally relevant skills. Discipline projects, students apply theory to Project-based learning features most notably in Projects problems in order to reach mainly a sequence of engineering projects, design- disciplinary learning outcomes. Problem implement experiences, where students projects, where students address real conceive, design, implement and operate problems with contextual and societal products, processes and systems. Progression dimensions; here the problem determines through the programme. The intention is not to what theory is used. replace subject courses as the primary site to systematically learn disciplinary knowledge. Research evidence to show the positive CDIO has its origin in engineering, and is Change effects of PBL. Well-documented institution- created by engineering faculty. Curriculum strategies wide implementations. A change development as engineering design. Working management perspective to handle resistance. within the discipline-based structures. Stakeholder involvement. Centres with researchers specialising in PBL Some literature documenting CDIO, including Relation to evidence. Much literature documenting PBL, a few books, journal articles. Peer review in the research including dozens of books. Two specialised annual conference from 2009, and a research journals: Interdisciplinary Journal of track from 2016. One special issue forthcoming Problem-Based Learning and Journal of PBL in European Journal of Engineering Education. in Higher Education. 807 documents in 278 documents in Scopus (CDIO AND Scopus (PBL AND “engineering education”) “engineering education”) 35 Both communities have important roles as centres for jointly developing, sharing and qualifying a knowledge base, consisting of the approaches in themselves, as well as the collective experiences in applying them. The knowledge base and the communities serve to strengthen local change agents, who are otherwise often isolated with only their own developed strategies, specific experiences, and limited opportunities for critical reflection. Hence, the communities contribute to the identities of practitioners, and helps legitimise their work. This paper also identified some significant differences in how each community conceptualises and handles the relationship between disciplinary fundamentals and professional aspects. Significant differences were seen in the starting point, the proposed role of disciplines, and the scope of the concepts. In PBL the starting point is the learning process, in that a problem- and project-based approach is advocated, for any type of learning outcomes. PBL is implemented in single courses, whole programmes, or whole universities. It is the PBL format in itself that prepares students for professional practice, through its similarity to working life. Despite the fact that some half of the curriculum in PBL universities is discipline-based, this is less addressed. In CDIO the starting point is to align the learning outcomes with professional practice. This led to the ideal of including both discipline-led and problem/project-led approaches in the curriculum, and there are strategies for both developing the contributions of both types of courses. After the comparative paper (Edström & Kolmos, 2014) was published, our joint reflections on the differences between the communities continued. In retrospect, it struck us as remarkable that one potential point of comparison was absent. Although the role of research is clearly a key difference between the two communities, it did not emerge as a separate category from our process of generating the comparative framework. One reason might be that the categories were generated in CDIO and SEFI conferences, but not in a pure PBL conference where research is more emphasised. The difference was however still visible in the results of the study, for instance when comparing the communities and change strategies. This lack of connection between CDIO and the emerging engineering education research (EER) community became a lasting conundrum, and the source of a new engagement. 2.2.2. Connecting CDIO and engineering education research Although much work in CDIO had been documented and published, also in international peer-reviewed journals, it was still seen exclusively as a community for educational development. Why the community has not been engaging more in the emerging engineering education research community was therefore truly puzzling. It was even stated already in the original application to the Wallenberg foundation that “a research program on teaching and learning is embedded in our initiative” (MIT, 2000). This led to a new engagement to organise an arena for educational research within the CDIO community, more specifically by establishing a research track starting in the annual conference in 2016. In opening for engineering education research, the hope is to further improve the knowledge base of the work, and 36 strengthen the legitimacy of practitioners, but there are also risks of losing important values. Therefore it felt important to transparently discuss the rationale for this move, and this became the theme for paper III: Edström, K. In press, 2017. The role of CDIO in engineering education research: Combining usefulness and scholarliness. European Journal of Engineering Education. The objective of this study was to consider the relationship between engineering education development and engineering education research, from the perspective of the CDIO community. It traces the development of engineering education research (EER) and some of the debates that are relevant for the formation of the field. The narrative is informed by observations during several years in various research and development communities, so here it mattered to be an insider. The development of the EER field, limited to the US and Europe, is distilled to a very short summary presented with support from a number of sources, in particular the key journals involved. It discusses the nature of research that might be most relevant for engineering education development and for furthering the community, by highlighting and comparing three concepts related to different aims of research: Boyer’s four scholarships (Boyer, 1990), Mode 1 and 2 (Gibbons et al., 1994), and Pasteur’s Quadrant (Brooks, 1967a; Stokes, 1997). The aim was to provide perspectives to help make sense of the available opportunities in EER, and discuss implications. In particular, Pasteur’s quadrant appeared useful to keep the hopes up that the research mission need not thwart ambitions to improve engineering education. See Figure 2.2. Figure 2.2. Pasteur’s Quadrant (based on Brooks 1967; Söderberg 1967; Stokes 1997), from paper III (Edström, in press, 2017). 37 The ensuing proposal for review criteria in the CDIO conference research track was intended to take a stance in the debates by combining considerations for scholarliness and usefulness. In the 12th International CDIO Conference, held in Turku 2016, the new research track attracted 40 proposals. After the peer review process, 14 full papers were published in this track (Björkqvist et al., 2016). The following year, 40 proposals were submitted for the 13th International CDIO Conference held in Calgary, finally resulting in 11 full papers (Brennan et al., 2017). Further, a special issue on the theme “Scholarly Development of Engineering Education – the CDIO approach” was announced in the European Journal of Engineering Education (2016), with contributions currently in the review process. It is an issue for future research to evaluate the results of this move, whether it will actually provide new understandings that are scholarly or useful, or both, or neither, and what other consequences it may bring about. *** This chapter described the CDIO approach for engineering education development. CDIO was chosen, not because it is the only model available, but to represent what is meant in this thesis by “engineering education development”. The chapter laid out, in some detail, the attempted strategies to integrate disciplinary knowledge and professional aspects in the curriculum, on the programme and course level, and in faculty development. Then, some engagements to increase self-reflection and support the further development of the CDIO approach and community were discussed. The first was a comparison of CDIO and PBL, and the second was a subsequent effort to connect the CDIO community with the emerging engineering education research community. The next two chapters take a more problematizing view on the conditions for engineering education development. We start by uncovering some historical roots of the main theme: the tension between the academic and the professional. 38 Up to this point, this thesis has quite straightforwardly discussed the what, why and how of engineering educational development – which refers to the work performed by educators, leaders of programmes and departments, committees, educational developers, and many other kinds of other change agents. In this section, I will use a historical perspective to problematize our understanding of the theme of this thesis, the tension in engineering education between engineering and engineering science. 3.1. A state of déjà vu Seely (2005) pointed out that when we consider educational reform it is useful to see what has led to the situation that we have now, and to recognise patterns in the history of reform attempts. To be somewhat self-critical about the discourse in engineering education development: we sometimes seem to act as if the problems we work on were discovered in our time, and we seem to devise solutions as if nobody has suggested or tried them before. Perhaps the historical innocence is even a necessary prerequisite for taking on this kind of work with optimism. It is easy to empathise with the historian complaining about a persistent state of déjà vu, caused by the engineering education community, publishing “…study after study that would benefit from expanding the analytical lens from primarily viewing students and curricula to include organizations, vested interests, and the accretion of decades of competing initiatives that structure current reform efforts” (Wisnioski, 2015). By exploring some of the history and comparing past and present discussions, I will show not only how the issue that this thesis focuses on has a long history, but also that many of the arguments and proposed strategies are very similar across time. As a result, the déjà vu will instead become mine. 3.1.1. Carl Richard Söderberg (1895 – 1979) The engagement with the past came about serendipitously, during background research for paper III. Reading a piece by Harvey Brooks (1967) in a report to the US Congress, my eyes fell on a text in the same volume by Carl Richard Söderberg: A Note on Engineering Education. His name was vaguely familiar, thanks to a short mention by Seely (1999). Reading the Note, I was immediately struck by the similarities with today’s discussion on engineering education, in particular the CDIO approach. This discovery led to a historical excursion, resulting in paper IV: Edström, K. Academic and professional values in engineering education: Engaging with history to explore a persistent tension. This study explores how the tension between the theoretical and professional preparation has been discussed for a long time. Seely uses the swinging pendulum as a metaphor to describe the turn from practice to science, when engineering education in 39 the United States was transformed due to a dramatic increase in research that started during World War II. The engineering science endeavour was a strategy for status and a strategy for institutional growth. Over time, this “avalanche” of government research funding changed the character of faculty, and the dominant culture went from engineering practice to engineering science, leading to increasingly theoretical curricula. While science and theory were originally intended to improve professional preparation, it came instead to dominate the education. The interesting life and work of Carl Richard Söderberg is traced against this background, focusing in particular on his views of engineering education. Besides the Note, the most important sources were the report of the Lewis committee (MIT, 1949), of which he was a participant, and his personal memoirs (Söderberg, 1979). Other sources were reports and papers, and correspondence and other documents found in archives in Sweden and the United States. Söderberg was born in 1895 as the first child in a large fishing family on Ulvön, famous for its fermented herring.1 Carl Richard’s aptitude for technical things was revealed when he at the age of ten became the best expert on the first boat engine in the family, with one cylinder. He was sent to technical school and then to study Shipbuilding at Chalmers. After graduation he received a scholarship to study for a year at MIT, after which he stayed in the US for a successful engineering career in industry. In 1938 he was invited back to MIT as professor, making an illustrious academic career, and ending as Dean of Engineering. During all his years at MIT he was always still practicing engineering through extensive industrial consultancies in the US and Sweden. While Söderberg was a proponent of the science-based curriculum, this was because of his experience with real problems in industry and society, and he eventually came to harshly criticise how the education had become distanced from engineering practice (1967). 3.1.2. Comparing the ideals of CDIO and Söderberg The next step in the study was to compare Söderberg’s views on engineering education, based on his writings, with those of CDIO. What follows is a different comparison than the one in paper IV. Here, the structure is reversed, with Söderberg’s views analysed through the lens of CDIO. For each CDIO standard, quotes are used to illustrate the extent to which Söderberg’s views match these principles. (This is included here because it could be of interest to implementers of CDIO. For the complete comparison, see the paper.) 1 Carl Richard Söderberg’s father and uncle were pioneers in packaging the fermented herring (surströmming) in tins. It had previously been distributed in small wooden barrels, with the disadvantages of not quite containing the smell, and also keeping the fishermen in hard work producing hundreds of barrels during winter. 40 Standards 1, 2 and 12 – Aims, Learning outcomes and Evaluation CDIO standard 1 declares that engineering education is a professional education, and “what engineers do” is expressed as conceiving, designing, implementing and operating products, processes, and systems. CDIO standard 2 recognises that learning outcomes should reflect the dual nature of professional education, i.e. both the understanding of disciplinary knowledge and professional competence. In many ways, these standards embrace Söderberg’s ideals, asserting the dignity of useful knowledge and advocating that education should prepare students for the practice of engineering, by acquiring the strengths of the disciplines as well as the abilities to deal with real problems in all their messiness. However, when CDIO describes engineering practice in analytic terms – conceiving, designing, implementing and operating, Söderberg would most likely have used holistic terms: problem-solving and the process of technology. I also believe that he would have mentioned that the ultimate aim is to advance the human condition. His engineering education ideals also emphasise the integration of general education aims more clearly, as a part of becoming a better engineer, but also for personal development. CDIO standard 2 advocates that the intended learning outcomes should be formulated in dialogue with stakeholders, and standard 12 recommends regular collection of feedback through a number of methods, to be reported to all stakeholders, and form the basis for decisions about the programme and its continuous improvement. The idea of consulting with external stakeholders has little equivalent in Söderberg’s writings. While the process of the Lewis committee constitutes an excellent role model for deep and insightful self-evaluation, they consulted only internally within the faculty. One difference, however, is that Söderberg was always an active engineer in industry and government consultancies, and was most likely in an excellent position to consider the needs of industry and society. This is less common in today’s faculty. Standards 3 and 7 – The integrated curriculum CDIO standards 3 and 7 identify the strategy of integrating the disciplinary fundamentals with the development of professional competences. Söderberg (1967, p. 400) comments that, “the purity of discipline orientation does not seem realistic for professional institutions of learning.” Likewise: “The division of a curriculum into isolated subjects each taught by a specialist may be a necessary expedient, but we need to be continually reminded that it is essentially unrealistic and that the ideal lies in the opposite direction” (MIT, 1949, p. 30). When CDIO calls for mutually supporting disciplinary courses it implies that integration cannot be left to the students, but every instructor should have some interest in and knowledge of the contribution of other subjects in the programme. The Lewis committee also urges that the curriculum needs to support integration and that faculty must be role models, “If each subject is narrowly compartmentalized and if the student sees each teacher as a specialist interested only in his own field, then the student’s task of integrating his 41 educational experiences will be nearly hopeless” (MIT, 1949, p. 93). They further advocate integration of professional skills, as instructors “must insist on competence in such subjects and make the students understand that bad human relations or careless writing, for example, can be the cause of failure in an engineering problem in school just as it can be in professional life” (MIT, 1949, p. 31). Standard 4, 5, and 6 – Engineering projects CDIO standards 5 and 6 address students’ need to learn to handle real engineering projects under conditions that are characteristic for professional life. CDIO curriculum development implies a sequence of such experiences, starting early in their education and with progression through the years. Here, students practice working on problems that increasingly include the full contextual aspects, under near- authentic conditions and working modes. Such learning activities can also reinforce students’ disciplinary understanding, through the opportunity to express and apply theory, and increase their motivation for learning new theory. Likewise, Söderberg calls for an educational environment in which professional work is being pursued, and emphasises that “more of the problems chosen for study could be full multidimensional ones, like the real problems of professional life, which rarely respect academic departmental barriers; problems requiring evaluation and judgment as well as calculation, [not ignoring the] social setting and human and even ethical dimensions” (MIT, 1949, p. 30). Moreover, Söderberg (1954, p. 62) appealed for such learning experiences to start early in the curriculum. He cited the need to break the monotonous sequence of fundamentals followed by application, “Many of us have felt that this sequence overlooks the elementary facts of [student] motivation. Their enthusiasm has all but vanished by the time they reach any of these applications.” This sentiment is exactly echoed in CDIO standard 4 (and 7). Söderberg also describes as a novelty what we today would call a capstone course. He reports: “the search for educational experiences in imaginative design [have] given encouraging results”, however he also has concerns about the resource requirements (Söderberg, 1967, p. 411). Half a century later, the CDIO initiative and others have gained much more experience and addressed resource-effective ways to organise design-based education, to some extent refuting Söderberg’s concerns. Standard 8 – Active Learning CDIO standard 8, Active Learning, recommends engaging students directly in appropriate learning activities, manipulating, applying, analysing, and evaluating ideas and taking on roles that emulate engineering practice. Likewise, the Lewis committee (MIT, 1949, p. 29), identified as a key value the “insistence on active participation by the learner rather than passive absorption.” The students must “by their own efforts convert facts into the body of their personal experience” (p. 9). 42 Standards 9 and 10 – Faculty Competence CDIO standards 9 and 10 recognise, just like the Lewis report, that the “performance of an institution is determined in the ultimate by its staff” (MIT, 1949, p. 33). Söderberg came to complain that faculty were increasingly shaped by research, “The teachers are gradually becoming more strongly oriented towards the scientific and theoretical, and the universities are gradually getting less competent for the task of teaching professional engineering” (Söderberg, 1962). He suggests: “the disciplinary specialist must be sufficiently in tune with the professional environment… capable of forming a bridge between the discipline and the profession” (Söderberg, 1967, p. 407). CDIO standard 9 recognises the need to strengthen the professional competence of the faculty, because the strategy of integration skills will be feasible only as far as faculty can support it. A key idea is also the mutually supporting courses, implying that each course has a function in the programme as a whole. This requires, as the Lewis committee suggests, that the instructor, “should have some interest in and knowledge of the contribution of other subjects to the education of his students. As a contributing scholar his responsibility may be confined to a narrow specialty, but as a teacher his responsibility is not limited by the boundaries of a particular subject” (MIT, 1949, p. 31). They continue: “the teacher must be… professionally competent” and “maintain some sort of inspiring contact” with the professional field (p. 32). With regards to faculty teaching competence, CDIO standard 10, the Lewis committee emphasise in particular the criteria for appointment and promotion, “It is our opinion that more emphasis should be placed on searching out and rewarding the great teacher as well as the outstanding scholar” (MIT, 1949, p. 143). In CDIO, work along standard 10 has included activities directly addressing faculty development of teaching skills, such as workshops and courses, but also recognition of teaching merits, and educational scholarship. Standard 11 – Learning Assessment CDIO standard 11 emphasises that effective assessment processes are necessary for evaluating the full range of intended student learning outcomes. The principle of integrated learning extends also to the assessment of learning. What must be practised and assessed is authentic performance where these aspects are interwoven; in short, integrated learning calls for integrated assessment (Edström et al., 2005). Also the Lewis committee noted the importance of assessment, and recommended “vigorous experimentation with testing methods with a view to making the examination a better educational experience and a better instrument for evaluating a student’s judgment and intellectual power rather than the memory”. They also recommend that professional skills be taken into account in assessment, as instructors “must insist on competence in such subjects and make the students understand that bad human relations or careless writing, for example, can be the cause of failure in an engineering problem in school just as it can be in professional life” (MIT, 1949, p. 31). 43 In summary, the agreement between Söderberg’s writings and the CDIO approach is considerable. 3.2. Learning from the past 3.2.1. Decommissioning the pendulum metaphor Reviewing Söderberg’s views together with CDIO confirms Seely’s conclusion that this challenge has “remained remarkably consistent over time” (2005, p. 125). Since the tensions between the academic and professional values are expected to remain, a better understanding is needed to support the productive relationship in the curriculum. The comparison showed remarkably similar positions regarding the dual nature ideal, integrating disciplinary theory and the other aspects of professional preparation. However, there were difficulties in the practical realisation of this ideal, also in the past. An imbalance in resources, in favour of basic research, came to shape the faculty and the organisation, and this soon led observers to lament, already fifty years ago, how education had come to reflect the primacy that engineering science had achieved over engineering practice. While Söderberg wanted to improve a practical education by integrating much needed theoretical aspects, the CDIO Initiative emerged as a reaction to a theoretical education, and wants to improve it by integrating (also other) much needed professional aspects. This means that they arrived at a common ideal from exactly opposite paths. Both Söderberg and CDIO recognise the dual nature of engineering education, and refuse to single out one side over the other. This shows that Seely’s swinging pendulum metaphor fails to challenge the misconception that engineering education must necessarily lean either to the academic or to the professional side. When Söderberg advocated a more theoretical approach, it was to strengthen professional practice. Likewise, when CDIO advocates professional competence, the deeper working understanding of disciplinary fundamentals constitutes a critical preparation for practice. The common ideal identified here is to make the professional and disciplinary preparation mutually supporting. The conclusion is that engineering education would benefit from ending the trench wars over “how much” should be theoretical or practice-oriented, and make more efforts to strengthen the meaningful relationship between these aspects in the curriculum. One conclusion is to let go of the swinging pendulum metaphor. Instead of seeking balance and compromise, as the pendulum imagery would suggest, we should seek syntheses and synergies. 3.2.2. Faculty, not curriculum In my opinion, the ideal of mutually supporting disciplinary theory and professional preparation is not unrealistic; it is fully feasible to devise such curricula, using for 44 instance the strategies devised in the CDIO approach. But, Harvey Brooks2 pointed out, “the main fault of engineering education is the excessive preoccupation with curriculum.” Instead, according to him, “the heart of the problem lies in the character and orientation of the engineering faculty. In the long run the courses and curriculum, and the knowledge and motivations of the students, are bound to reflect the research interests, the consulting experience, and the values of the faculty.” (Brooks, 1967b) The challenge, then, is to shape a faculty whose collective competence can support curricula closer to the ideal. The focus on Söderberg as a person also makes a point because he so clearly combined the practical and theoretical interest, himself embodying the dual nature of engineering. This could suggest that to achieve the duality, we need enough people in the faculty who can simultaneously defend both the academic and the professional values. The binary view, with the pendulum image and the trench wars over the curriculum, may be unavoidable if too many people favour one side with little consideration for the other. In fact, engineering faculty need competence in three areas: theoretical-scientific expertise, professional competence, and teaching competence. If this seems daunting, we can look around our faculty and say, quoting the Lewis committee: “We have such people; we can have more” (MIT, 1949, p. 93). 3.2.3. Using a historical perspective Engaging in the past certainly gave a sense of déjà vu. While the history was already to some extent familiar through Seely’s work, including the understanding that the practice vs. theory conflict is as old as engineering education itself, the deeper engagement made a difference. Despite the warning that “while the debate may seem the same, the content has changed radically” (Jørgensen, 2014), it was exciting to hear the echo, across half a century, of almost exactly the same ideas and words that we are using today, in CDIO and in other engineering education communities. Perhaps the similarities are not so remarkable after all, because it was the ideals of the past and the present that were compared, not the actual curricula, and it is always easier to agree and resolve tensions on the abstract level. The results of this engagement into the past can certainly prompt critical reflection. Just losing our historical innocence and knowing more about how deeply this issue runs makes a difference. It may increase the awareness of how necessary it is to defend these ideals in practice. Above all, a historical sensibility is part of knowing the issues in depth. As I demonstrated through this triangulation of the past and 2 Harvey Brooks, 20 years younger than Söderberg, became Dean of the Division of Engineering and Applied Sciences at Harvard when Söderberg was Dean of Engineering at MIT. 45 present, it is a perspective that can strengthen our understanding and our arguments, as well as our reflexivity in our work. *** This chapter took a long-term perspective to inspire new ways to understand the tension between academic and professional preparation, by comparing how it is conceptualised in past and current initiatives for engineering education development. Taking us back much nearer the present time, the next chapter aims to provide another critical perspective, by considering unsustainable change. 46 Taking the starting point in practical experiences when change has been unsustainable, this chapter makes a new interpretation of educational development. 4.1. Organisational gravity 4.1.1. Experiences of unsustainable change In 2011, I had been discussing engineering education development with educators, programme managers, deans, and educational developers for over a decade. They were based all over the world, and their change projects were of various types, not solely CDIO and PBL implementations. Over time a pattern began to emerge. Some of the colleagues confided that even projects that were considered highly successful had achieved smaller results than intended, and further, that change was not sustainable, in that engineering programmes tended to revert “back to normal”. My informants reported that they felt a need to constantly keep working hard just to sustain new practices in the programme. Otherwise, as soon as their attention turned to other matters, the new practices would wither away and the programme revert. Out of pure curiosity I began to informally ask more questions about their experiences, opportunistically during coffee breaks and conference dinners, in corridors and aisles, trains and airplanes, and bars. There was a common theme in their stories, some remarkable phenomenon, but what was it and how could it be understood? It also felt novel, in the sense that it was not part of the normal discourse about change in these circles. The poor sustainability of change had evidently come as a surprise to my colleagues; we did not have concepts to describe it. This indicated a need for a new way to think about educational development. Several new questions emerged: What makes programmes revert? What do they revert to – is there a particular ground state for a programme? If so, what defines or shapes it? Why is it more stable than other states? The result of my reflections was a new theoretical model that connected the educational programme with the organisational characteristics of the organisation in which it is situated. Being an engineer and communicating with engineering educators, it was natural to use a technical metaphor: Organisational gravity is a force acting on education programmes, causing them to reflect the inherent characteristics of the organisation providing it. The most stable state (lowest energy state) for a programme is thus to reflect the institution. This is the ground state. Every other state requires that some kind of energy is introduced into the system to counteract the gravity and ‘lift’ the programme to an alternative, more desired, state. Such energy can be applied in many different forms, for instance through money, leadership, attention, and other resources, in projects and interventions. But since the organisational 47 gravity keeps exerting its force on the programme, we must continuously add resources to keep it from reverting to the ground state. (Edström, 2011) 4.1.2. Testing the model The organisational gravity model postulated that the ground state for educational programmes is where it simply reflects the characteristics of the organisation. If so, it should be possible to analyse what type of educational development could be harder to achieve and sustain, and what types should be easier. I will explore this using two examples. Example 1. Why do engineering curricula often consist of courses that reflect the organisational boundaries of the university? Even when cross-disciplinary learning activities are considered desirable for the education it seems hard to form and sustain collaboration across organisational boundaries. The barriers are not only the different traditions of the subjects, but there are also considerable practical concerns. Several mundane issues have to be resolved to handle a course over different cost centres, administrative classifications, and staff territories. Relations between faculty members may be weaker; they may not even know each other in the first place. The physical setting often reflects organisational boundaries, with one building for scientists working on a class of problems called control theory, and other buildings for people working on programming, and so on. Even when we succeed in creating crosscutting collaborations, they tend to involve extra work to establish and maintain, and they are extra vulnerable to discontinuation since they often rely on personal connections and trust. It is consistent with the model that programmes consist mainly of courses corresponding to the administrative territories of the organisational chart. As most universities are organised along the disciplines, these disciplinary boundaries tend to be reflected in the courses of the programme. Example 2. Why is it hard to integrate learning outcomes related to professional practice? The degree requirements for engineering education in most higher education systems contain learning outcomes related to professional skills, including the ability to integrate knowledge and apply it to ‘real’ engineering problems. But real problems do not respect disciplinary boundaries, and solving them often requires integration of knowledge – including old knowledge, far from the research frontier. They further require interpretation of the context, and judgement and creativity in conceiving and implementing solutions. When hiring and promoting faculty, disciplinary research merits are more valued than strengths related to integration, application and professional practice. Therefore the faculty, at a collective level, has relatively little professional engineering experience. As a consequence, problems that are generated from within the disciplines are considered interesting in academia, while problems that do not map to the disciplines fall outside the perceived responsibility of most 48 researchers. To provide professional preparation, the university needs strengths related to integration and to application of knowledge. This is not easily provided in the discipline-based organisation, whose logic follows the research frontier: reduction, analysis, specialisation, and branching off new fields and sub-disciplines. Therefore it takes special effort in the programmes to address learning outcomes related to professional engineering practice. It is consistent with the model that some learning outcomes are more difficult to address in the education, because their representation in the organisation is too weak. This applies to learning outcomes related to real problems, to integration and application, and to the aims of engineering practice, i.e. innovation and entrepreneurship, or sustainable development. The conclusion was that when values are not sufficiently represented in the setup of the organisation, they are harder to implement sustainably in programmes. Unfortunately, this applies to some of the most important learning outcomes in engineering education. No organisation can optimise on all goals simultaneously (Birnbaum, 1988, pp. 59-63), so if the organisation is optimised for the production of disciplinary knowledge, it can more easily provide the disciplinary theory, while it is more difficult to address the professional aspects. 4.1.3. How things work around here The organisational gravity model was an attempt to describe how the inherent characteristics of the organisation shape the education programme as an image of the organisation, unless resources are constantly applied to keep it in a more desirable state. Organisational characteristics are interpreted in a very wide sense – a simple working definition would be “how things work around here”. This is determined by formal matters such as structure and policy, but also by what values are embedded in the organisation. Table 4.1 lists some possible shaping mechanisms, classifying them as hard or soft and internal or external. Table 4.1. Mechanisms shaping the organisational characteristics (Edström, 2011). Organisational structure, policies and Funding systems, evaluations, accreditation, regulations, hiring and promotion national policies, national and international Hard processes, allocation of power and collaborations, ranking, awards etc. resources etc. Language and conceptions, status, Image, status, personal networks etc. Soft symbolic events, rituals, self-image etc. Despite the attempt to classify the factors, it is easy to see that they interplay and influence each other, and in particular, some factors will enable or limit change in others. For instance, many internal factors are influenced by the external. The internal resource allocation is often aligned with how resources are acquired by the institution, an external factor. Not only the volume of resources shapes the organisation but also 49 the logic of allocation. Incentive effects can be seen when resource grants are awarded in competition, based on (some operationalization of) excellence, while at the same time compensation for education is distributed according to quantity, rewarding all teaching equally, irrespective of quality. Further, soft and hard factors shape each other – ”the symbolic takes part in creating the real” (Dahler-Larsen, 1998, p. 54). For instance, the status of the title ”professor” depends on whether the procedure for appointment is perceived as rigorous and related to socially shared values, and whether the title is associated with substantial rewards, such as independence and access to funding and power (hard factors). The design of such policies and procedures are in turn limited or enabled by the prevailing views on roles and identities in the organisation (soft factors). 4.2. Implications 4.2.1. Two change strategies A direct implication of the model is that there are in principle two kinds of change strategies available for developing educational programmes: the force strategy and the system strategy. The force strategy is to add some kind of extra energy to move the programme to a more desirable state (a higher energy level). This energy can take many forms: funding, leadership, attention, alliances, evaluations, lobbying, personal energy, etc. One advantage is that the force strategy is available to all actors; everyone can just go ahead and make their contribution – using their force – top-down or bottom-up. The disadvantage is that force is continuously needed to counteract the organisational gravity and prevent the programme from reverting back to a lower energy level. The force strategy is therefore potentially not resource-efficient. This is not to say that it does not work, but it works with an agricultural logic: new seeds must be sown every year. Being aware of this logic will help set more correct expectations regarding the results and their sustainability, and make it possible to plan for a continuous supply of resources. In particular, the force strategy risks putting high strain on people, partly because of the high effort it takes to achieve results and to sustain them, and partly because their efforts are likely under-rewarded (more about that below). The system strategy is to change the characteristics of the organisation to enable a more desired stable state for the education. Such change goes deeper; it is not only to change what we do in the education, but also who we are as an organisation. This can be labelled as a self-transformation (Alvesson & Sveningsson, 2008, pp. 238-243). The values needed for the educational mission must then be present in the organisation. In other words: to sustainably change the education it is not sufficient to just change the education. This is also about the nature of research, since it will limit and enable education. Hence, in addition to the production of knowledge, the values 50 related to integration and application are necessary to create an organisation that can also accommodate professional education. Precisely because of its constitutional function, the system strategy is less available. Fewer actors have access to the most important shaping mechanisms, e.g. career systems, funding systems, and they also change rather seldom. The advantage is that even small changes, for instance in the appointment and promotions structure, can have considerable and lasting effects. The ideal is to align the university, as a system, with both its research and educational missions. Then, in theory, organisational gravity could become a positive force, pulling the programme in the right direction. Thus, both strategies have their uses, as they have different strengths, weaknesses, availability, limitations, risks, and implications for resource-effectiveness and sustainability of results. Even if the force strategy seems like an unwise choice at first sight, the Sisyphean labour may actually be justified. It is also understandable if university leaders hesitate to use the system strategy. If it is mainly research-related indicators that will de facto determine the long-term survival and prosperity of the institution, there could be risks associated with creating an organisation that can accommodate good research and good education. 4.2.2. Educational development as a compensatory activity One final observation is that the model can shed new light on educational development. With the exception of new issues such as digitalisation, the label educational development seems to denote a remarkably stable set of issues. They often relate to the professional side of education (e.g. development of professional skills and judgement, integration and application of disciplinary knowledge, problem- based and active learning). Therefore I suggest that the role of many educational development activities is to compensate for such values that are necessary for education but not sufficiently represented in the organisation. In the words of the model, educational development is about moving the education programme to a more desired state, and keeping it there; the work is about counteracting the organisational gravity. One implication could be that educational development consists of precisely such work that the organisation is not set up for, which is why such compensation was needed. For anyone engaged in educational development, this is a sobering thought. If it is true that educational development is about compensating for the values that are not sufficiently embedded in the setup of the organisation, then we can suspect that merits related to educational development constitute by definition such achievements that do not build a career in the university. 51 4.3. Reflections 4.3.1. The value of the model The organisational gravity model and the accompanying discussion were created to capture in abstract form what practitioners (educators, deans, programme leaders, educational developers, and administrators) had reported about their flesh and blood experiences. Despite the technical imagery it should not be confused with a law. Instead the model is an explanatory device: “Both [models and theories] may be seen as explanatory devices or schemes having a broadly conceptual framework, though models are often characterised by the use of analogies to give a more graphic or visual representation of a particular phenomenon” (Cohen et al., 2011, p. 13). Hence, this representation of an empirical phenomenon is offered to practitioners as a lens for interpreting and discussing their experiences. This is theorizing in the context of discovery rather than the context of justification (Swedberg, 2012, 2016). The justification, finding out whether the effect can be observed, is left for empirical investigation. In other words, the point here is not to “prove” that organisational gravity “exists”. Nor is it implied that educational development projects in general will revert; in fact, the cases described in chapter 2 have persisted for 10 to 15 years and counting, which contributes to the opposite picture. The message is not that change will be unsustainable (predictive power), but that if and when it is, there is a language for it and we can perceive and discuss these experiences (sense-making function). Weick (1976) aptly pointed out how concepts can help perception: “The guiding principle is a reversal of the common assertion ‘I'll believe it when I see it’ and presumes an epistemology that asserts ‘I’ll see it when I believe it’”. At this stage, the value of the model rests not in its “truth”, but in whether practitioners find it compatible with empirical reality “in terms of resonance, understanding, explanation, or utility” (Walther et al., 2017, p. 403). If it resonates with practitioners’ experience it has face validity (Cohen et al., 2011, p. 163). It could also have catalytic validity, which refers to the degree to which it helps practitioners understand reality in order to transform it (Marshall & Rossman, 2010, p. 42). 4.3.2. Similar concepts I included the organisational gravity model here just as I formulated it in 2011. This original version also documents my reflections, as an educational developer during that period. Later, I have come across similar ideas expressed by several authors, which can confirm that others have seen similar phenomena. Accounts of unsustainable change are common. Meyer, Scott, and Deal (1983) mention how educational organisations “adopt and slough off” changes (p. 56), and how changes “sweep through and die out”, and are “unlikely to persist” (p.60). 52 Graham also identifies the reversion process in her study on engineering education reform3, where she observed: “significant challenges associated with sustaining change, with the majority of reform endeavours reverting to the status quo ante in the years following implementation. […] Most experience a gradual course-by-course ‘drift’ back to a more traditional curriculum” (Graham, 2012, p. 3). Reflecting on a comprehensive international study of outstanding teaching environments in eleven research-intensive universities, Knapper (2016) says: “in many cases the momentum and enthusiasm for the new approach diminished over time and there appeared to be what we termed a ‘recidivism to the traditional’ […] it shows radical change is difficult to sustain without constant pressure” (p. 110). Elmore and McLaughlin (1988) studied school reforms in the United States, calling it steady work, similar to what I called the force strategy: “measured by substantial changes in what is taught and how, the rewards are puny; but the work is steady, because of the seemingly limitless supply of new ideas for how schools should be changed and no shortage of political and social pressure to force those ideas onto the political agenda.” By the same token, Christensen and Ernø-Kjølhede (2012) call the issue of integrating contextual issues and socio-technical competencies a never-ending story: “There appears to be no optimal and final solutions in implementing theory of science in engineering education but only temporary solutions reached by negotiation and compromise” (p. 16). The need for system-level change is also identified by Miles and Ekholm (1991), who depict how schools revert changes, unless they are made “normal”: “Schools, like other organizations, have a way of weathering down changes, or subtly ejecting them, unless they are built in to the school, become embedded, a part and parcel of ‘normal life’.” They further mention that institutionalisation is directed toward “saving energy in the organization”. It is also easy to draw parallels to the work of Argyris and Schön (1996), who distinguish between two forms of organisational learning. Single-loop learning refers to the adaptive behaviour of a stable system. In contrast, double-loop learning means that the system itself changes by modifying values, strategies, and assumptions. Argyris and Schön comment that “most organizational ‘fixes’ are of the single-loop variety”, but such “reforms and fixes tend to be subverted, over time”. They identify defensive patterns that hinder double-loop learning, in particular the preference to locate problems only in “discussable domains”, and limit solutions “to deal with discussable features of the problems” (p. 281). 3 Disclosure: I was one of Graham’s 187 informants in this study and shared the organisational gravity model in the interview. 53 *** This chapter, and the previous one, have in different ways problematized engineering education development. The following section will follow up on the suspicions generated here, that the crux of the matter can be found in the relation between the nature of change and the setup of the organisation. Whether it is in discussable domains or not, we need to consider how things work around here. 54 The previous chapter considered the challenges of unsustainable educational development. One key hypothesis was that the nature of change matters, which makes it harder to accommodate some of the most desired goals of professional engineering education. My hypothesis in the previous chapter was that whether change was successful or not had to do with how well the nature of change matches the values inherent in the setup of the organisation – or what I called how things work around here. This line of reasoning found support also in the historical case of MIT, which showed how research funding changed the priorities of the faculty and the university, with the consequence that the education became more theoretical and lost some of its professional character. The focus of this chapter is therefore to take an organisational perspective on the tension between the academic and professional aims in engineering education. 5.1. Understanding organisations and institutions 5.1.1. The university as a machine To understand change it is necessary to begin with an understanding of how things work in the first place. This chapter constructs a theoretical framework to support a more sophisticated understanding of the university as an organisation, in particular as an environment for the kind of educational development that is in focus here. It might appear counterintuitive to focus on the status quo when change is wanted, but the ability to develop the curriculum depends on how well the conditions for change are understood. Whether they are implicit or explicit, it matters what models, concepts, or theory we have. As mentioned previously, concepts can function as lenses for discerning things that may otherwise have gone unnoticed. Similarly, they may also limit our view, because to highlight some aspects is also to relegate others to the background. Thus, models and theories can just as well blunt our perception and misguide our interpretation. In the following, one common view of the university organisation will be severely challenged. One way of seeing the university is as a factory or machine for education and research. The metaphor suggests an organisation optimised for effective operation, structured along the organisational chart, and designed to coordinate its activities “in a routinized, efficient, reliable and predictable way” (Morgan, 2006). Individual actors are expected to behave in a rational manner, with polices and decisions coordinating and controlling the rational production of services according to stated goals. Perhaps this view is particularly natural to adopt among engineers, due to our strong tendency to see functionalist rationality as the natural guideline for action (Picon, 2004, p. 429). The problem is not that the machine model is necessarily wrong, but that it lacks explanatory power for many aspects of university life, including the experiences 55 described in the previous chapter. Weick (1976, p. 1) observed that people often find that their experiences in educational organisations “prove intractable to analysis through rational assumptions”, as the rational view simply does not “explain much of what goes on within the organisation”. In particular, the machine model is unproductive when it comes to formulating models for change. In fact, the change strategy that can be derived from this organisational understanding is that change should be mandated from the top and aimed at improving the outputs. But the previous chapter showed precisely the difficulty associated with achieving sustainable change even given top-level decisions, access to resources, and the best intentions with respect to the outcomes of education. These were not sufficient conditions for sustainable change. Thus, there is reason to suspect that a top-down and function- oriented model of the organisation is not useful as a means to inform development, or to make sense of experiences. An alternative framework will be needed, more appropriate to analyse the university as an organisation and assessing the implications that follow for educational development. In the following, I will construct a theoretical framework describing organisations as embedded in and infused by institutions, meaning social orders, or systems of social beliefs and rules. The institutional perspective was chosen because it helps explain organisational behaviours that defy rational assumptions. The focus on discovering other rationalities, or rather uncovering them, matches the critical research approach of this thesis (cf. Suddaby, 2015). If the machine metaphor focuses on the formal and visible structures, resources, activities and outputs, institutional theory also emphasises the subtler roles played by norms and values, and by culture and identities. Before introducing institutional theory, it is necessary to disambiguate the term institution. In education it often refers to a higher education institution, a single university or university college. (To make it more complicated for Swedish readers, the word for academic department in our language is also “institution”.) Further, to institutionalise often means making practices routinized and embedded, or institutionalised, in an organisation. This is closer to the meaning in institutional theory. 5.1.2. Shattering the machine metaphor The (neo-)institutional view on organisations challenges the functionalist conception of rationality. In their landmark paper, Meyer and Rowan (1977) argued that: “organizational success depends on factors other than efficient coordination and control of productive activities”. More important for success and survival, they said, is the ability of the organisation to conform to, and become legitimated by, institutions in the environment. It is therefore not merely the production and activities of the organisation that shape formal structures, assessment criteria, vocabularies, and practices, but also the rules and norms of its institutional environment. Two 56 fundamental problems follow for the organisation, first that institutional rules and norms are often in conflict with each other as they arise from different parts of the environment, and second, that conformity to the institutional rules and norms may be in conflict with efficiency of the production of outputs. In particular when the outputs are difficult to appraise – as they certainly are for universities – legitimacy will depend mainly on conformity with institutional rules. The logic of efficiency is then less crucial than the logic of appropriateness to institutional rules and norms. (While Meyer & Rowan did not use the exact term ‘logic of appropriateness’, they describe the phenomenon (see also March, 1982; March & Olsen, 2004).) Meyer and Rowan further made the point that what really matters is to maintain appearances; the institutional rules are often enacted ritually or ceremonially. Therefore the problem can be stated: “the organisation must struggle to link the requirements of ceremonial elements to technical activities and to link inconsistent ceremonial elements to each other”. Some of the organisational strategies used for absorbing these tensions within organisations are decoupling (minimising disputes and conflicts by avoiding too close integration, more about loose coupling below), preserving ambiguity around goals and outputs (making it hard to evaluate performance), delegation to professionals (individuals are left to work out inconsistencies informally) and maintaining face (through displays of confidence and good faith). At this point, three key concepts will be elaborated in some detail: loose coupling, isomorphism and legitimacy. Loose coupling Meyer and Rowan (1977) use the term loose coupling to describe how organisations can cope with the two conflicts, between inconsistent rules coming from different parts of the institutional environment, and between institutional rules and the efficiency of the practical activity. These tensions can be reduced through decoupling. Weick (1976) described loose coupling in educational organisations: “By loose coupling, the author intends to convey the image that coupled events are responsive, but that each event also preserves its own identity and some evidence of its physical or logical separateness… their attachment may be circumscribed, infrequent, weak in its mutual affects, unimportant, and/or slow”. Noticing that many see loose coupling as a flaw in the organisation, Weick instead takes “a neutral, if not mildly affectionate, stance toward the concept”, and he shows that although it makes matters “modestly predictable”, loose coupling has many advantages. It allows people some autonomy and creativity, makes room for experimentation and development of localised or context-specific solutions. The organisation spends a minimum of resources on coordination and control. Further, the organisation is protected from responding to every change in the environment, and a breakdown in one area need not hit all parts of the organisation. An engineer may make an association to the risks of over-determined systems, and the use of spring- 57 dampers and firewalls. The integrated curriculum strategy in CDIO can be understood as an effort to achieve a tighter coupling, with coordination between the programme and the courses, and horizontally between courses. This was exemplified by the Mechanical Engineering case at Chalmers described in chapter 2. We turn to Brunsson (1989) for an example of how loose coupling can be used to uncover organisational behaviour. He addresses the tensions posited by Meyer and Rowan on the concrete level of organisational behaviour. Brunsson makes the case that in order to cope with inconsistent environments, organisations have two structures – one (informal) for producing action and results, and one political for producing ideology for the purpose of legitimation. Brunsson argues that the tensions between these functions make it necessary for the organisation to keep its talk, decisions and actions apart, in other words, to keep them loosely coupled. His point is that it could be difficult or prohibitively expensive to actually satisfy inconsistent demands in the actions and production of outputs. Therefore, ideologies are much better suited for handling inconsistencies in the multiple ideas about the organisation, about its environment and about what it should do. In fact, as some problems are insoluble in practice, they are accessible to ideological handling only (pp. 26, 233). The result is double talk where the political organisation deals mainly with ideas, intentions, and wishful thinking, debates insoluble problems, launches reform to inspire hope, and focuses on the future. Brunsson makes the important point that when the political organisation deals with the inconsistencies, it protects the action organisation from those tensions. It is far easier to explain and legitimise action and products, i.e. using ideology to make them more appetising, than to turn the ideas into practice: “When ideas are to control action, they have to be reduced to the concrete level of action; but when they are to explain action, they can distance themselves from the concrete and become abstract and more inconsistent” (p. 171). Rather than seeing this “organised hypocrisy” as a problem in the organisation, Brunsson concludes that it is actually a solution for handling the tensions. Nevertheless, he also shows empathy for those who have a functionalist conception of rationality: “it must be very frustrating to believe in the action model” (p.204). Isomorphism Isomorphism is interesting for this thesis because, depending on the nature of change, it can be expected to drive or limit change. Meyer and Rowan (1977) state: “Independent of their productive efficiency, organizations which exist in highly elaborated institutional environments and succeed in becoming isomorphic with these environments gain the legitimacy and resources needed to survive. […] In such a context an organization can be locked into 58 isomorphism, ceremonially reflecting the institutional environment in its structure, functionaries, and procedures.” When DiMaggio and Powell (1983) took a closer look at isomorphism, they referred to the structuration of an organisational field, meaning all relevant organisations that together constitute a recognised area of institutional life (for instance higher education). Whether organisations are in contact or not, and whether they are competing or not, they will shape themselves in response to this common environment, and thereby become more alike (isomorphic). They describe three types of isomorphism pressures: Coercive processes are those that make an organisation to adapt to rules or structures coming from other organisations on which the organisation depends. For instance, universities within the same legal environment will become more alike. Mimetic processes are those that make an organisation imitate others, in particular those perceived as most successful and legitimate. This often happens when there is environmental uncertainty, when the relation between means and ends are uncertain, and when goals are ambiguous. For instance, given the uncertainty around online provision of education many universities around the world model their own strategies on those of high-status universities. Normative processes should be related to pressures from norms and values in society (Deephouse & Suchman, 2008, p. 53) and in particular, in the institutional context. DiMaggio and Powell mention the role of professions in the organisation. The similar background through formal education and continued socialisation in professional networks create categories of individuals who according to DiMaggio and Powell (1983, p. 153) “will tend to view problems in a similar fashion, see the same policies, procedures and structures as normatively sanctioned and legitimated, and approach decisions in much the same way”. While the authors had business managers in mind, they might as well be describing university faculty when they continue: “The exchange of information among professionals helps contribute to a commonly recognized hierarchy of status, of center and periphery, that becomes a matrix for information flows and personnel movement across organizations. This status ordering occurs through both formal and informal means.” It is important to note that while isomorphism makes organisations more alike, this does not necessarily make them more effective. Here, Powell and DiMaggio clearly side up with Meyer and Rowan in often pointing out how pressures are addressed ritually or ceremonially. Practices are adopted whether they really work or not, but if they make organisations appear more legitimate and prestigious they make sense and bring rewards. Again, the logic of appropriateness trumps the logic of effectiveness. 59 Legitimacy Meyer and Rowan (1977) showed how organisations must conform to their institutional environment to achieve the legitimacy and resources they need to survive. Legitimacy can also be applied to groups or individuals within organisations. Suchman (1995) reviewed the literature on organisational legitimacy and offered the definition: “Legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions” (p. 574). Some traits captured in this definition are that observed behaviour is evaluated over time (so specific events can be ignored) in relation to the norm system of some social group. Thus there is no such thing as legitimacy in general, it is always viewed through the eyes of some constituents, and further, in relation to some role or practice (legitimacy for what). Four main types of legitimacy can be distinguished (Deephouse & Suchman, 2008; Suchman, 1995): Pragmatic legitimacy comes from an evaluation based on self-interest, or “what’s in it for me”. The organisation is perceived to produce something of economic, social or political value to the observers. Moral or normative legitimacy is based on an evaluation whether something is the appropriate thing to do. This can apply to outputs and consequences, techniques and procedures, categories and structures, and whether leaders and representatives are perceived as the right people to do it. Cognitive legitimacy entails less of an evaluation – it is instead related to whether the organisation is seen as comprehensible and predictable, or even taken-for-granted as necessary or inevitable. This is the most powerful form of legitimacy because it is conferred in the absence of questioning; alternatives become unthinkable. Professional legitimacy is based on belonging to, or approval by, a professional group, be it on pragmatic, normative or cognitive grounds. These forms of legitimacy come with slightly different implications. For instance, Suchman (1995) makes a difference between seeking active or passive support, and says: “To avoid questioning, an organization need only ‘make sense.’ To mobilize affirmative commitments, however, it must also ‘have value’- either substantively, or as a crucial safeguard against impending nonsense.” Further, he analyses strategies for gaining, maintaining and repairing legitimacy. For this thesis, gaining legitimacy is of most interest. The main strategies for gaining legitimacy are (a) to conform to demands in the current environment, (b) to select among environments to find a supportive audience, and (c) to manipulate the environment to create new audiences and new legitimating beliefs (p. 587). 60 Other related concepts for social evaluations are status and reputation. Deephouse and Suchman (2008, pp. 60-62) sort out the differences: Legitimacy reflects conformity to social guidelines, and is non-rival. Status reflects the relative position of social groups, with more rivalry between than within groups, and individuals who belong can enjoy the privileges of the group. Reputation is mostly personal, earned based on previous performance, and fundamentally rival (e.g. ranking). With their contribution, Meyer and Rowan made it impossible to ever see organisations merely as machines again. After this first excursion into the early foundations of institutional theory follows a later and more comprehensive theoretical framework, called institutional logics. 5.1.3. The institutional logics perspective While the focus of this thesis is mainly on the practices within an organisation, it is obvious by now that the external environment of the organisation plays an important role in shaping the internal conditions. Universities are participants in the higher education field, and under influence from the institutional orders in society. In other words, it is necessary to bring society back into the analysis (Friedland & Alford, 1991). The institutional view of organisations has been elaborated and extended into a framework called the institutional logics perspective (Friedland & Alford, 1991; Thornton & Ocasio, 2008; Thornton, Ocasio, & Lounsbury, 2012). Institutional logics can be succinctly expressed as “the way a particular social world works” (Jackall, cited by Thornton et al., 2012, p. 46), which is remarkably similar to my working definition of organisation in the previous chapter. The more comprehensive definition reads: “the socially constructed, historical patterns of material practices, assumptions, values, beliefs, and rules by which individuals produce and reproduce their material subsistence, organize time and space, and provide meaning to their social reality” (Thornton & Ocasio, 2008, p. 101). From this definition, it is clear that the institutional logics perspective incorporates both the material and normative dimensions. While the early papers (in particular Meyer & Rowan, 1977) provided much-needed insights into the powerful workings of the norm systems, any analysis that neglects the resource environment is incomplete, and here the institutional logics perspective takes a balanced view. On the highest level, Thornton et al. (2012, p. 73) list seven institutional orders in society: state, market, community, profession, corporation, family and religion. These are the ideal types of institutional logics, each with their own set of norms and sources of legitimacy and authority. On the next level is the institutional field, populated by organisations. The field-level institutional logics often combine several of the societal-level institutions, and different organisations in the field can have different balance between these logics. For instance, the higher education sector is an 61 institutional field, where some processes are shaped by professional logics (e.g. peer review), and other aspects are shaped by market logics (e.g. technology transfer) or state logics (e.g. degree frameworks). An individual university can have its own instantiation of the field-level logics: “We assume each institutional field consists of one or more available logics, as well as an array of appropriate collective organizational identities and practices from which individual organizations assemble their particular identities and practices” Thornton et al. (2012, p. 135). The important implications here are that a university is part of a complex institutional environment that contains contradictions between different societal logics, and that the logics embedded within any particular university will also reflect tensions. This has consequences for practices and identities inside the organisation. In the following some of these aspects are elaborated upon and connected to each other. 5.1.4. Practices and identities in the organisation By connecting the institutional logics to practices, what we do, and identities, how we see ourselves, the theory now becomes very relevant for this thesis. The development of engineering education is precisely an attempt to change one of the practices of the university. It is therefore a key concern how the “old” or “new” curriculum models are related to the institutional logics, how they are related to other practices in the organisation (e.g. research), and how they are related to identities in the organisation. Practices Practices are intimately connected to the institutional logics of the organisation. Thornton et al. (2012) describe “a fundamental duality between logics and practice, where constellations of relatively stable practices provide core manifestations of institutional logics”. All practices will not reflect the institutional logics in the same way, as they may align with different parts of the institutional environment, for instance different sets of uncoordinated constituents. In a complex institutional environment with incoherent demands, we can expect a need to accommodate tensions between practices, as well as within practices, and between institutional rules and the effectiveness of the practice. Some practices are “linked to different logics that coexist relatively independently” (Thornton et al., 2012, p. 139), while others can express hybrids of different logics. This thesis will in particular focus on how the institutional logics of two professions are expressed in the curriculum: the academic profession of the educators, and the engineering profession, which we educate for. Thornton et al. further suggest that practices may be conceptualised as interdependent, so that changes in one practice may have ramifications for other practices in the organisation (p. 141). Here, it is mainly the interdependence of education and research that will be in focus. In the context of organisational change in higher education, Oliver (1991) combined the institutional and the resource dependence perspectives. Clearly, both normative 62 and economic rationalities affect the organisation, its practices and identities. Sometimes these forces will coincide so that a practice gains strong support both normatively and materially, while others have more variegated patterns. As already noted in chapter 4, education and research are two practices with quite different resource environments. The resource environment has profound consequences on the inside of organisations. As Pfeffer and Salancik (1978, p. 71) suggested, “organizational environments come to affect organizational actions partly by affecting the distribution of power and influence within the organisation”. We are reminded that both the material and normative dimensions are included in the definition of institutional logics. Thornton et al. (2012, p. 157) note that practices are embedded in economic systems, so that “material resources influence the generation of practices and have partial autonomy from culture and institutions”. They further point out that resource environments are also in turn shaped by institutional logics, e.g. at the societal level. Identities Organisational identity refers to the collective identity of an organisation, as perceived externally or internally. Stensaker (2015) finds that the organisational identity of a university seamlessly connects the past and the future, tradition and modernity, and highlights the values and norms. He also notes that the organisational identity may indicate internal power structures of the university. Given the focus of this thesis, it is however more interesting to consider identities within the organisation. Here, Alvesson distinguishes between the social identity of sub-groups, and self-identity, i.e. how individuals see themselves. Self-identity is linked both to the individual and the collective, and “typically includes elements of social identities, but also more than that, and gives the individual categories more of an individual and rich meaning” (Alvesson, 2012, p. 35). Self-identity can also refer to the identity of the organisation as a whole and to available standardised social identities. Henkel (2005) refers to Bernstein in saying: “identities are strongest and most stable within the context of strong classification, the maintenance of strong boundaries protecting the space between groups, disciplines or discourses”. The classifications of individuals are important in higher education, and university organisations pay much attention to it. Education can be seen as a process where schools and universities handle students through a series of stages, carefully controlling every transition (e.g. admission, examination, degrees), and the classification of academics is no less important (just think of disciplines, titles, appointments and promotions). Meyer and Rowan (1983) find that while educational organisations tend to only loosely coordinate and control their work activity, they do in fact very tightly control the classifications of people. They say: “Education rests on and obtains enormous resources from central institutional rules about what valid education is. These rules define the ritual categories of teacher, student, curricular topic, and type of school” (p. 76). Thus, at least one part of the identity, the belonging in categories, is tightly controlled, as a precondition for awarding freedom in the work they do. 63 As identities are connected to legitimacy, status, and freedom, it is no wonder that it is often actively adopted and managed. Individuals and groups can engage in identity work, attempting to “rework or alter their identities to make sense of or resolve the tensions they face from competing institutional logics” or to “promote a specific understanding about an identity, link this understanding to specific logics and practices, and work to attract potential adherents to the identity” (Thornton et al., 2012, p. 130). One example, discussed in paper III (Edström, in press, 2017), is the engineering education research movement. The collective efforts to create a recognised discipline can be seen as an active transformation into a more legitimate practice, whose institutional logics offers stronger support than that of education or educational development. Interplay between identity and practice There is a close relationship between practices and identity; we can say that they are co-produced. The tight link is also evident when we consider how status is attached both to identity and practice. In the same way that complex institutional environments can generate conflicting rules, they can generate patterns of differentiated status between organisations, and between different practices and groups within the organisations. Status also affects the relationships with the resource environment. Kodeih and Greenwood (2014, p. 10) note that the high-status actors can be expected to have priority access to the most valuable resources. This applies both to the organisation as a whole (such as when high-status universities attract funding, and the most talented researchers and students) as well as to groups and individuals on the inside. Status can play a role in change, since those that are perceived as successful and legitimate are likely to be imitated by peers – and this applies to organisations (DiMaggio & Powell, 1983) as well as to individuals and groups within the organisation (Deephouse & Suchman, 2008, p. 61). This means that role modelling by high-status actors can be a driver of change. It also means that change may be strongly resisted if it is perceived as a threat to the status of organisations, groups, or individuals. For this thesis, it is relevant to consider status in relation to different practices and identities in the university. The status difference between teaching and research comes to mind, but there are also implications for the relative standing of other practices, e.g. professional practice, educational development, and educational research. Further, if we now consider the curriculum as an expression of educators’ identity, it is clear that changes will be seen as more or less valuable and meaningful depending on how well they resonate with faculty identity. For instance, some educators will embrace the engineering profession as a part of their identity, and they may therefore have a more positive attitude to integrating professional aspects. Seeing the curriculum as an expression of faculty identity, it is also obvious how the mere proposal for changes can be experienced as an intrusion. In her influential study of 64 academic identities, Henkel concluded that the discipline and academic freedom were the two things that mattered most, “in many cases the sources of meaning and self- esteem, as well as being what was most valued” (Henkel, 2005, p. 166). The implication is that any change in practices and structures will be strongly resisted if it is perceived to entrench these values. If a reform instead can respect (or even draw on) the core values in faculty identities, it may have a better chance. As Stensaker puts it: “Academic identities can have multiple starting points for either supporting or resisting changes – or both of them simultaneously – in higher education, depending on how the reforms are defined and what their objectives are” (2012, p. 9). 5.1.5. Organisational culture: values, beliefs and assumptions This theoretical excursion followed Thornton et al. (2012, p. 135) in considering how institutional logics are visible in practices and identities. They also influence organisational culture, which according to Alvesson (2012, p. 4) covers aspects like assumptions, beliefs, ideas, rites, rituals, myths, and values. Culture is by no means separate from the material structures and practices in the organisation. It is insufficient to see material structures (such as budget or evaluation systems) as purely “technical” arrangements. Culture is an integrated part of how the social and material structure is constituted through practice (Stensaker et al., 2012, p. 13, citing Knorr Cetina), and the converse must also be true, that culture (and its components) are influenced by the structures and material practices. Culture is problematic conceptually, because it rests much on taken-for-granted assumptions and is therefore only semi-conscious, and because it can be used to refer to everything and consequently nothing (Alvesson, 2012, p. 39). Merely adding culture as a mystery layer on top of the machine metaphor is simply unhelpful, especially if we embrace the view that culture can be managed as a tool – that things will go our way if we only “set” the right culture (Alvesson, 2012; Stensaker et al., 2012). As Graham noted in her survey of literature on change in engineering education: “although ‘changing the culture’ is a phrase used in many recent reports on engineering education, the prevailing culture is rarely defined and suggested strategies for cultural change are limited” (Graham, 2012, p. 11). Similarly, Kogan (1999) laments: “’Culture’ has been used to explain the inexplicable... It is a word unreflectively used by organisational reformers who have run out of specifics; they perceive that rationalistic structuring and engineering of process are not enough to move hearts and minds, and that something must be constructed in the areas of values and affect if members of the system are to produce the desired goods”. Instead of culture, this thesis will use more limited concepts like assumptions, beliefs, and values. Already in chapter 4, values was used to discuss how the organisation was 65 set up, and to make the point that the capacity for practices requires the presence of matching values in the organisation. The choice seems defensible also in the light of the institutional logics theory. For instance, Friedland (2013, pp. 30-37) refers to Weber’s “value spheres” when he coins the term institutional substances, “the unobservable, but essential, ‘value’ anchoring an institutional logic”. Also, like identity, values bring meanings of internalisation and passion: “[actors are] passionately and expressively oriented to unobservable values they have internalized and in whose unconditional requirements they believe” (ibid.). Values signals that the logic of appropriateness is in play, more than the logic of effectiveness. 5.2. Perspectives on change 5.2.1. Reform as routine – and as producer of hope We are already familiar with Brunsson, who highlighted that talk and action are decoupled, in order to protect the action organisation from having to deal with inconsistent demands. He continues to discuss the role and function of reform (2009), noting that organisations are often under constant reform, or attempts to change to a desired state through “reorganizations, change projects, rationalization, restructuring… or administrative reforms” (p. 6). The pressures to reform are so strong that it is difficult to maintain autonomy; it takes stronger leadership to avoid reform than to do it. Brunsson emphasises that reform is an attempt to change: “Reform is not equivalent to change. An organization may undergo several reforms and emerge with little change” (p. 6). He argues that reforms are driven by the supply of problems, of solutions, and of forgetfulness: “Without problems, reforms are difficult to justify; without solutions they cannot be formulated; and without forgetfulness there is a risk that people will be discouraged by the fact that similar reforms have been tried and failed in the past” (p. 14). These factors are discussed below. Problems are in plentiful supply. Brunsson’s earlier work (1989) identified the tension between how the organisation presents itself, the talk, and the way it actually works, the action. This “provides an incentive for repentance, for reform” (Brunsson, 2009, p. 93). Another source of problems is actually previous reforms. Sometimes new reforms are launched to target the same set of values because earlier reforms were better at “raising the level of aspiration than at improving the situation” (p. 95). Since organisations are subject to conflicting demands, it may also be “impossible in practical terms to find any balance that could readily be regarded as the right one”. Therefore every solution “is susceptible to criticism for failing to satisfy one or other – or both – of the needs sufficiently”. Thus a reform furthering one side of such a balance may soon create impetus for new reform in the other direction (Brunsson, 2009, p. 95). Sometimes the organisation ends up just oscillating between different 66 solutions. Incidentally, this phenomenon has been empirically identified for the organisational location of educational development units, which are therefore objects of frequent reorganisations (Edström, 2010, p. 11). To start reforms, reformers need solutions that are perceived as attractive in comparison to the current reality of the organisation. Coming up with promising ideas is easy: “If we set a simple, clear, and good reform idea against our knowledge of the current situation with all its slack, ad hoc solutions; and its uncertainties, inconsistencies, conflicts, compromises, and complicated relationships, then there is a good chance that the new solution will appear better” (Brunsson, 2009, p. 97). Common sources of solutions are consultants and professional reformers, as well as other successful organisations (mimetic isomorphism). Given that reforms are often repetitions of earlier reforms, organisations need a supply of forgetfulness – it is necessary to forget the content and outcome of previous reforms: “Reforms focus interest on the future rather than the present, and forgetfulness prevents the past disturbing the future” (Brunsson, 2009, p. 99). Personnel turnover promotes forgetfulness, as does the use of consultants, who “can see the organisation with fresh eyes and can thus more easily repeat old mistakes” (p. 99). Just like ideas can be more attractive than reality, reforms tend to look very different before and after they are implemented. What started as simple and attractive principles becomes more complex when it is applied to organisational reality with all its conflicts and practical problems, and “making [the ideas] practicable also makes them much less beautiful” (p. 99). An old reform will therefore seem far less attractive than one that is new and still untried, and it may go unnoticed that they are in essence the same. Reformers will reinforce this, as it is certainly in their interest to sell their reform concept as new and different (p. 149). It is the combination of failed reforms and retained hope that drives new reforms. Brunsson quips: “Reforms tend not to deliver what they promise. But their promises are so good that people are easily lured into trying again” (p. 100). Unsurprisingly, Brunsson dismisses results of reforms in terms of changing organisational practice – he even mentions how presenting controversial changes as a reform can prevent their implementation. But reform has other uses. Brunsson views hope as a “fundamental, cultural factor in explaining the production and reproduction of reform” (p. 17). Organisations need its members to be hopeful; it certainly works better than despair and apathy (p. 153). Therefore, perhaps the most important function of reform is to create, maintain, and express the hope “that it will be possible to transform ideas and principles into practice” (p. 140). This makes it dangerous to learn from discouraging experiences, since it risks destroying hope. Accordingly, Brunsson identifies some strategies to avoid facing failure. One is to avoid confrontation between the principles and practice – by staying on the level of talk 67 (e.g. spending all time developing the reform model in itself) and avoiding practice (e.g. evaluation focuses on the effects on people’s thinking rather than on their practices). Another strategy is to focus on the future instead of the present, to postpone every confrontation with practice (p. 144-146). In addition to hope, other advantages with reform is to improve the image of the organisation in the eyes of external audiences, also, the reformers themselves can develop their ability to represent the organisation. But results can also be negative, when attention is drawn to problems and reinforcing the perception of failure. And while hope is necessary, it is precisely by inspiring hope that, “reforms risk increasing the importance of the very goals which the organization has greatest difficulty in achieving” (p. 101). Brunsson cautions: “From the point of view of the individual organization there is reason to avoid these effects, if survival is considered more important than the meeting of high standards” (ibid.). 5.2.2. Institutions as resources for institutional innovation It is easy to think of institutional norms and rules mostly as limiting the autonomy of organisations and individual actors, but an important theme throughout literature is that social systems can be seen as simultaneously constraining and enabling. This is what Giddens (1984, pp. 25-28) calls the duality of structure. Meyer and Rowan (1977) made the point that while it is often most beneficial to conform, organisations can also play active roles in shaping rules and expectations (p. 348). Further, Oliver (1991) suggested an interesting typology of available coping strategies (table 5.1). Table 5.1. Strategic Responses to Institutional Processes (Oliver, 1991, p. 152). Strategies Tactics Examples Acquiesce Habit Following invisible, taken-for-granted norms Imitate Mimicking institutional models Comply Obeying rules and accepting norms Compromise Balance Balancing the expectations of multiple constituents Pacify Placating and accommodating institutional elements Bargain Negotiating with institutional stakeholders Avoid Conceal Disguising nonconformity Buffer Loosening institutional attachments Escape Changing goals, activities, or domains Defy Dismiss Ignoring explicit norms and values Challenge Contesting rules and requirements Attack Assaulting the sources of institutional pressure Manipulate Co-opt Importing influential constituents Influence Shaping values and criteria Control Dominating institutional constituents and processes By focusing on the opportunities, institutional logics can be seen as resources that can be invoked for the legitimacy of identities and practices. Especially in organisations with multiple institutional logics, there are opportunities for individuals and organisations to actively exploit any inconsistencies and contradictions (Thornton 68 et al., 2012). Studying an organisation with two competing logics, Binder (2007, p. 568) concludes: “Logics are not purely top-down: real people, in real contexts, with consequential past experiences of their own, play with them, question them, combine them with institutional logics from other domains, take what they can from them, and make them fit their needs.” Such possibilities are available at all levels, for individuals, for a particular activity or for a whole organisation. Actors and sub-groups can, and do, utilise such opportunities selectively, making the organisation a mosaic of groups, with more or less potential for enabling or resisting change (Greenwood & Hinings, 1996). Organisations, sub-groups and individuals can draw on institutional innovation in the institutional field. For instance, Thornton et al. (2012) mention how people with experience from different institutional contexts are less likely to take things for granted in their local organisation, and may have capacity to create institutional change (p. 110). While the organisational field is a source of values and norms, it can also supply concrete exemplars of structures and practices available to organisations. As a version of isomorphism, Greenwood and Hinings (1996) discuss how templates, or archetypal patterns, reflect ideas and values in the institutional field, shaping structures and practices. Collective sense-making can support some factors necessary for change in organisations: recognising the weakness of existing arrangements and building the capacity for action, which means having sufficient understanding of the new conceptual destination, the skills and competencies required to function in that new destination, and the ability to manage how to get to that destination (Greenwood & Hinings, 1996, pp. 1039-1040). Greenwood, Suddaby, and Hinings (2002) showed the role of professional associations in legitimating change, by hosting debate, justifying and endorsing new practices. They call this theorization, “the process whereby organizational failings are conceptualized and linked to potential solutions” (p. 58). Associations can also be a suitable locus of theorisation and diffusion. This study also stresses how pragmatic legitimacy matters relatively little in professional contexts, compared to normative legitimacy: “What matters within a professional context is the demonstrated conformity of innovation with the values embedded in traditional beliefs. It is only when ideas are couched in such a way that they are perceived to be consistent with prevailing values that they appear compelling and legitimate for adoption” (p. 75). Greenwood, Suddaby, and Hinings note that gaining a shared understanding is a long process: “Models must make the transition from theoretical formulation to social movement to institutional imperative” (ibid, p.60, citing Strang and Meyer). So how can such transitions happen? Thornton et al. (2012, p. 159) refer to field-level vocabularies of practice, defined as “systems of labelled categories used by members of a social collective to make sense of and construct organizing practices”. They explain: “Vocabularies of practice guide attention, decision making, and mobilization, 69 and provide members of social groups with a sense of their collective identity.” This creates common ground, facilitating sense-making and communication, coordination and collective action. Thornton et al. depict a process involving cultural entrepreneurs, narratives, and reification: “While cultural entrepreneurs following their interests are involved in the formation of narratives that lead to vocabularies of practice, the categories must become reified for the institutional logics to move beyond the level of theory or ideology” (p. 160). Reification implies that practitioners “must perceive over time that the categories and systems of practices are generated not as direct products of human intervention and agency, but as the natural order of things” (p. 160). In other words, they must achieve cognitive legitimacy – the most subtle and powerful form of legitimacy (Suchman, 1995). 5.2.3. A note on “change management” literature Much of the so called change management literature is of limited interest here since it is written within the logic of corporations, entailing an image that power is hierarchically structured and profit is the measure of success – even the raison d’être – for organisations as well as individuals. Numerous books seem written for managers looking for tactics to push, evangelise, and manipulate their subordinates to overcome their supposedly illegitimate resistance (for a similar critique, see Alvesson, 2012, pp. 157-158). In one of the most cited works, Kotter (1996) lists eight necessary phases in a change process and discusses potential failures in relation to each. It is a linear model aimed at senior management to firmly push change top-down in their organisations. Efforts have been made to translate some of the strategies into the university setting (see for instance Froyd, Penberthy, & Watson, 2000), however the model fails to capture the institutional nature of university organisations. Clark (1986, p. 275) notes that we cannot understand educational organisations if we analyse them as companies: “We insist in peering at higher education through glasses that distort, producing images that render more confusing a terrain that is naturally difficult”. Change models underpinned by such borrowed assumptions are likely ineffective for analysing educational organisations, for formulating strategies, and for engaging the people in them (Kezar, 2001, p. 74). Further, change models intended for corporations will likely interpret the specificities of universities only as obstacles and barriers, failing to draw on them as strengths and resources (Musselin, 2007). Moreover, much of the change management literature comes across as based on a worldview of “us” and “them”. Referring to technological change, Williams (2002) remarks: “The rhetoric of change management has the effect of trivializing the whole concept of change. […] There are many fault lines in history today, and many of them run right through us. The change agent and the change resister are often the same person” (p. 19). 70 5.2.4. Change in higher education In a review of journal articles on how to promote educational change in STEM, Henderson, Beach, and Finkelstein (2011) categorised change strategies based on whether the focus is on individual educators or organisation, and whether the results are predefined or emergent. While they criticize the general level of scholarship of the studies with regards to the use of theory and the robustness of evidence, they still concluded that effective change strategies must (1) be aligned with or seek to change the beliefs of the individuals involved; (2) involve long-term interventions; and (3) take into account the complexity of universities, meaning that it is necessary to first understand the system and then design a compatible strategy. Later, Borrego and Henderson (2014) suggest that the relevant literature on change “is not necessarily accessible to those who need to apply it”, and proceed to present examples of curriculum change strategies according to the framework. In a large-scale study involving six case studies and in total 187 interviews, Ruth Graham (2012) finds that successful reform is often associated with threats, for instance problems with recruitment, retention, or employability. She also mentions that it often involves faculty with industry experience and/or newly recruited faculty, less invested in the status quo. Such reforms start with a curriculum-wide assessment of goals and a high-level realignment of the curriculum, and Graham finds long-term success associated with “the extent to which the change is embedded into a coherent and interconnected curriculum structure”. Much literature on change in higher education lists strategies and/or necessary conditions for change. For instance, Ambrose (1987) suggests the critical need for five conditions: 1) vision, 2) skills, 3) incentives, 4) resources and 5) action plan. This model also interestingly identifies the states when any of these elements is missing: 1) confusion, 2) anxiety, 3) resistance, 4) frustration and 5) false starts, respectively. This model is often cited in the context of educational development (de Graaff & Kolmos, 2007; Knoster, Villa, & Thousand, 2000). Other authors on change in higher education emphasise possibilities of influencing norms and understandings. Dill recommends change strategies focusing less on what we do in higher education, encouraging universities to instead actively influence culture by managing meaning and social integration. The management of meaning includes “the nurturance of myth, the identification of unifying symbols, the ritual observance of symbols, the canonization of exemplars, and the formation of guilds” (Dill, 1982, p. 316). He notes, “if the new staff is socialized at all, it is within the department or field”. The management of social integration is therefore important to counteract faculty orientation toward the individual, discipline-based, career, and build loyalty to the academic enterprise as a whole. To Dill, influencing culture is not only a concern for the top management and administration; instead he attaches importance to peers and collectively organised socialisation mechanisms. Dill later 71 remarked that attempts to manage the symbolic side of universities are often related to branding, to becoming “world-class” or “entrepreneurial”. These are efforts for external marketing rather than to create “an academic culture that defines, communicates and helps embed the values essential to effective teaching, scholarship and research within universities” (Dill, 2012, p. 229). *** This theoretical framework will now be set to work, to analyse the tensions in the organisation between the academic and the professional aspects of engineering education. This is followed by an analysis of CDIO as institutional innovation. 72 The overall aim of the thesis is to explore the dual nature of engineering education, simultaneously an ideal and a source of tensions. The objective, then, was in particular to investigate opportunities and challenges in efforts for developing engineering education in relation to this ideal. This section begins with a depiction of the academic-professional duality and its tensions, drawing on the theoretical framework from the previous chapter. Next follows a discussion of efforts to work towards realising the ideal, in particular in the strategies on different levels in the CDIO initiative. Here, I will also makes some critical reflections regarding the limitations of the CDIO concept. Finally, I reflect on some of the insights that are particularly challenging for the educational development enterprise. 6.1 Seeing the duality in the light of institutional logics 6.1.1. Practices Using the institutional logics theory we will now consider the academic-professional duality in engineering education, the ideal as well as the tensions. We saw that institutional logics – or patterns of material practices, assumptions, values, beliefs and rules – are embedded in the practices within the university, and in the co-produced identities. Different practices may express the institutional logics differently since they align to different parts of the institutional environment, for instance resource environments or uncoordinated stakeholders. The logics embedded within a particular practice can also contain such contradictions. The dominant practices in higher education – in principle the practices that define a university – are research and education, with intimately related identities for faculty, as researchers and educators. See figure 6.1. Figure 6.1. The major practices in higher education and the co-produced faculty identities. 73 6.1.2. Competing logics in engineering education We begin with the practice that is of primary interest in this thesis, engineering education. My interpretation is that the institutional logics of engineering education expresses two professional logics: the logics of the engineering profession that we educate for, and the logics of the academic profession of the educators. These logics come with slightly different assumptions, beliefs and values regarding the educational mission and the role of the educators. The logic of the engineering profession must reasonably harbour the assumption that the educational mission is about teaching the next generation of engineering professionals. In the logic of the academic profession it could instead be reasonable to see the teaching mission as conveying the theory of their discipline. See Figure 6.2. Figure 6.2. Education, a practice expressing two professional logics. To elaborate some aspects in which the institutional logics of the two professions differ: In the logics of the engineering profession, the educator teaches future professionals. The engineering identity of faculty is strengthened by the fact that most have an engineering degree, and some also have experience of professional practice. Knowledge is relevant if it is useful for engineering practice, and so are the problems and questions that matter in industry and society. Engineering students should prepare for professional practice, i.e. working on real problems in real contexts, which includes a deep working understanding of theory, and the ability to integrate and apply it. In the logics of the academic profession, the educator teaches disciplinary theory. The academic identity of educators is strengthened by the fact that the normal route to a faculty position is through a PhD in one of the disciplines, and a research career. Knowledge is relevant if it is part of the disciplinary canon, and problems and questions are interesting if they have a potential to lead to new discoveries furthering the disciplinary frontier. Engineering students should learn disciplinary theory, and prepare for research education. These factors are summarised in Table 6.1. 74 Table 6.1. Analysis of the institutional logics of the engineering profession and the academic profession, respectively. Institutional The engineering profession The academic profession logics that we educate for of the educators The role of the Teaching future engineers Teaching theory educator Relevant Knowledge useful for engineering The disciplinary fundamentals knowledge practice Interesting Real problems, consequential issues in Pure problems, close to the disciplinary problems and industry and society frontier questions Students are Engineering practice – through deep Engineering practice – through prepared for working knowledge and professional theoretical knowledge competences Research education – disciplinary depth This analytic scheme is not meant to set these two sides of education against each other. Instead the point here is that both sides are necessary, and the ideal is that they should be in a meaningful relationship. This is the dual nature of engineering education, the theme of the thesis. That it is an ideal does not however prevent manifestations of the contradictions and tensions between the logics. For instance, the problem discussed in chapter 4 was that some of the values necessary for engineering education are weakly represented in the organisation, e.g. integration, application, and real problems in context. To reformulate it in the language of institutional logics: if the professional logics are weakly represented among the faculty, it is more difficult to satisfy the related aspects in the curriculum. Simply put, the capacity to teach disciplinary theory is strengthened by the academic logics, while the professional logics create capacity for addressing also the other necessary aims of the curriculum. 6.1.3. Competing logics in research I will now proceed to discuss how the research practice can be characterised by a similar tension within its institutional logics. Here, I also draw on the discussion in paper III (Edström, in press, 2017), although it was there related to the aims of the engineering education research field. My suggestion is that two beliefs about the aims of research exist simultaneously: one that research aims to further the discipline, often called knowledge for its own sake, and one that research is guided by a consideration for usefulness in society. See Figure 6.3. The first belief can be expressed as the university as academia, because in my view, knowledge “for its own sake” quickly translates to the same thing as furthering a discipline. This is because the academic career depends on peer recognition, making disciplines the site that controls the necessary resources for survival. Peer recognition is a sine qua non, since those whose work does not pass this disciplinary quality control will soon be marginalised by a lack of resources. Quite aptly, Gibbons et al. (1994) called disciplines the “homes to which scientists must return for recognition or rewards”. Academic capital comes in hard currencies such as being accepted for 75 publication, passing a thesis defense, receiving grants and prizes, being appointed and promoted, and selected for commissions. Many of those decisions concern the classification of individuals, which was identified in the theoretical framework as a particularly important component of identity. This helps explain the strong socialisation of faculty into the discipline-based identity and beliefs. The academic pursuit can become very personal: “A key element for many academics is the narcissism involved in doing and publishing research. The self is invested in the work and research publications function as reinforcers and stabilizers of a sense of self susceptible to the insecurities and vulnerabilities of a profession constantly exposed to assessment and a level of competition where failures greatly outscore successes for most people…” (Alvesson et al., 2017) Figure 6.3. Two aims of research, with corresponding beliefs. The second belief, the university as public service implies that research is guided by consideration for use. The crucial matter becomes how to evaluate the usefulness dimension of the work, and who should be seen as the legitimate judge. It is quite telling that even funding for highly applied research is often dispensed using academic peer review. While there are efforts to make societal impact matter more in the academic career evaluation, it seems seldom conceptualised as a main consideration within research and education, but as a separate third task, service, and the discourse often has a distinctly commercial character (cf. Dill, 2012). According to the Pasteur’s Quadrant model (see figure 2.2), the two beliefs are not mutually exclusive, because research can simultaneously be directed toward applied goals and lead to significant new understandings. Not least since the resources under academic control are so vital, I suggest that the university as academia belief has stronger support in the institutional logics than does the university as public service. The reason why this matters here is, as I see it, the university as academia is highly consistent with the logics of the academic profession, while the belief in the university as public service has strong similarities with the logics of the engineering profession, for instance the values attached to integration, application, the interest in real problems that are consequential in society and industry, and their real solutions. The core distinction is similar to the description by Williams (2002): “In science, the 76 fundamental unit of accomplishment remains the discovery; in engineering, the fundamental unit of accomplishment is problem-solving” (p. 44). My conclusion here is that in the research practice, the logics of the academic profession enjoy the strongest support in the institutional environment, both normatively and materially. 6.1.4. Interplay between education and research We have discussed two practices separately, education and research, focusing on some tensions within each practice due to inconsistent demands in the embedded logics. What remains is to consider the interdependence between education and research. There is much educational research and development addressing the relationship – often called the nexus – between research and teaching (for a recent overview see Geschwind, 2015). There are many dimensions to this interdependence, for instance we know that faculty draw on their disciplinary networks in educational matters (Mårtensson, 2014, p. 58). Here, guided by the constraints of my research questions and following the theoretical framework, I will mainly focus on the different conditions for the practices, and the influence by research on the dual nature of engineering education. Now the two figures can be merged, see figure 6.4. Figure 6.4. Competing institutional logics in education and research. Due to inconsistent institutional demands we can, according to the theory, expect tensions between practices, and between institutional rules and the effectiveness of the practice. Further, we can expect patterns of differentiated status between these practices and groups within the organisations. Seen from outside the university, both education and research enjoy high status. Engineering education is a prominent source of legitimacy for a university, as a supplier of elite professionals to society and industry. The research activity corresponds to the role of the university as producer of new knowledge and is an important source of status and identity, not least for the university international reputation and brand. However, within the university, while there is certainly also status in excellent teaching, the status of research is generally higher than for education. One reason is the career system, where research merits dominate every step (see for instance Geschwind & Broström, 2015). We are reminded of the 77 imperatives created by the “university as academia” described above. While teaching merits feature increasingly in the hiring and promotion criteria, it seems sufficient (from a career point of view) to be above a threshold level (Graham, 2015). Another reason is the difference in the associated resource environments. Funding for education is distributed internally, most often based on quantitative factors without reward for quality. Research funding varies considerably, between research fields, in terms of availability, and whether the funds afford freedom or come with strings attached. But in contrast to education, research funding is to a large extent sought externally, in competition based on peer review; the rewards for excellence are considerable in terms of resources and prestige. In short, the socialisation and reproduction of the faculty, and the incentives of the resource environment result in a dominance of research. My conclusion is that research has stronger institutional support than education, both normatively and materially. This affects the conditions for education generally, including related matters such as the attention paid to teaching competence, teaching quality, and educational development. While the imbalance between education and research is important, the specific focus here is the dual nature of engineering education. The duality was conceptualised above as competing logics within the education practice: teaching theory and teaching professionals. Because of the role played by research in shaping the faculty, it will limit and enable what is possible in education. While some research and researchers focus on matters of mainly academic interest, the furthering of a discipline, there are also researchers who work on matters with a more direct consideration for use. Many applied and cross-disciplinary fields, and for instance design related subjects, are closer to professional practice. We can presume that many researchers with such interests have more engineering capital, for instance closer contact with industry (including the public sector), and that they in their role as educators might then find it more natural to take on the role of educating professionals (see also table 6.1). What I would like to suggest here is that the institutional logics of research, being the dominant practice, strongly influences the institutional logics of the education. Hence, the more the research practice is dominated by the academic logics, over the consideration for use, the more it will tilt the balance in education, in favour of teaching theory, rather than teaching professionals. When the balance is heavily tilted, it will also be difficult to achieve the ideal of a productive relationship between the academic and professional aims. I have painted a picture here in which research has the primary position in the university organisation, positioning education as a secondary practice. I also argued that the institutional logics of the academic profession have the upper hand not only in research, where disciplinary interests takes priority over considerations for use, but also in education, where teaching theory takes priority over the other aspects of professional preparation. No wonder then that is difficult to make certain kinds of educational changes sustainable, when the primary practice exerts its constant 78 influence and imprint. This happens through the faculty, whose academic identity is stronger than their engineering identity, because research is the birthplace of new faculty, and it holds the keys to continued survival and success. While the organisation naturally needs to spend considerable attention to its own academic reproduction processes, one may wonder if it has not taken a life of its own, to the point where it fully takes precedence over the educational mission of the university. 6.2. Seeing CDIO in the light of institutional logics 6.2.1 CDIO as integration of the academic and professional logics Above, the dual nature ideal was analysed using an institutional logics perspective, describing how the logics of both the engineering profession and the academic profession are embedded in the engineering education practice, with the associated assumptions about “teaching future professionals” and “teaching theory”, respectively. Despite focusing on them separately, the message was that both are necessary and that the ideal was a productive relationship. In chapter 2, the efforts of the CDIO initiative illustrated the integrated curriculum strategy for realising the ideal on the programme and course level, and in faculty development. Here, these strategies will be discussed in turn. Integrated learning on the course level From a perspective of teaching and learning I will argue that there need not be much tension between the disciplinary theory and the (other) professional competencies. They belong together and give each other meaning – as argued in table 2.1 – and we can devise ways to integrate them in subject courses (discipline-led) as well as project-based courses (problem/project-led). One could say that on a course level the integration strategy works. However, it works under one necessary condition. It works depending on individual faculty and their willingness and ability to unite the theoretical and the professional. It works as long as they are prepared to pay attention also to professionally relevant aspects that are not necessarily part of the teaching traditions of the subject. Some educators, but not all, have the inclination and the competence to do it. Since new faculty members are inducted through the research disciplines, as discussed above, we can expect strong orientation towards disciplinary traditions; this is what they believe education is (cf. Roberts, 1982). We remember that Henkel (2005) identified the discipline and academic freedom as the most important values for faculty, even as keys to meaning and self-esteem. Further, the imbalance between education and research makes it even more precarious to depend on individual faculty. For instance, those who perceive that course development does not further their career may choose the path of least resistance, which means following traditions. Integration can be successful, but as long as such courses are seen as exceptions to the predominant norms, the success is tied to individuals and 79 hence vulnerable and temporary. I can conclude that the strategy of integration can work on the course level, but only as far as faculty members can support it. The integrated curriculum The defining aspects of CDIO, the standards, describe a process for establishing structures holding the curriculum together, making the programme a joint collegial project, where every course has a function towards the overall goals. This is a reaction to curricula “consisting of disciplinary courses disconnected from each other, and as a whole, loosely coupled to espoused programme goals, professional practice, and student motivation” (Edström & Kolmos, 2014). CDIO development is a way to establish a power base around the programme, necessary to create and uphold the coherent and interconnected curriculum structures identified by Graham (2012) as a key factor associated with successful and sustainable change. The Mechanical Engineering case at Chalmers, in chapter 2, showed the full strength of the integrated curriculum. For instance the modern computational mathematics could be weaved through the programme, strengthening connections between the mathematics and the engineering in several courses where it was appropriate and meaningful. This programme was developed and refined over a long time, by a team of faculty with high legitimacy and the resource system in their hands. It could be seen as a case of tighter coupling between the programme and its courses, creating an uncommon structural capital and agility, which allowed the programme team to proceed to set and reach new goals. There is no doubt that other programmes, despite similar intentions, have failed to achieve this development or to make it sustainable, as mentioned in chapter 4. One reason is that coordination takes some effort. While faculty members expect to work extremely hard to succeed in research, there is a risk that curriculum development initiatives fail to engage the relevant people, as such work is perceived to go unrewarded. When “stronger leadership” is proposed as a solution, it is often based on rationalist assumptions and refers to stronger line management, which fails to recognise the highly institutionalised context. Given that integration and coordination represent almost the counterweight to the core values of discipline and academic freedom, it is not obvious that the faculty is up for any joint efforts in education at all. On the programme level, I conclude that the integrated curriculum can work, but only as far as the faculty capacity for coordination can support it. Faculty development level Could Harvey Brooks have seen the CDIO standards today, he would likely react just like in 1967, when he criticised engineering educators for being obsessed with the curriculum: “the heart of the problem lies in the character and orientation of the engineering faculty. In the long run the courses and curriculum, and the knowledge and motivations of the students, are bound to reflect the research 80 interests, the consulting experience, and the values of the faculty” Brooks (1967b). While the importance of faculty was recognised in the standards on enhancing faculty competence, as a programme development scheme CDIO has had limited influence over faculty appointment and development. There has still been much development in this area, at least in Sweden, where courses on teaching and learning were required together with more attention paid to the evaluation of teaching competence. If it was perceived as provocative that the needs of education are increasingly taken into account in the appointment and promotion of faculty, it seems even more daring to make suggestions about the research. But the interdependence of education and research does raise the questions about what kind of research could support the educational mission in a university, since the shaping of the faculty is largely under the auspices of the research enterprise, at least in research-led institutions. It could at least be interesting for funding agencies to consider how the research they support helps or hurts the conditions for engineering education. The analysis of institutional logics suggested that research with a consideration for use shares some key aspects with the engineering profession, e.g. the values attached to integration, application, and real problems in context. As Brooks also hinted, and we saw in the life and work of Söderberg (Edström, forthcoming 2018), another promising way to enhance faculty professional engineering competence is through their consultancies. In the faculty development standards, CDIO did not achieve the same balance in the academic and professional duality as we saw on the course and programme level. A complete conceptualisation would need three standards, one related to teaching competence, and two related to the object of teaching – aligned to the academic and the professional aims, respectively. If CDIO had three standards for faculty competence, related to subject, teaching, and engineering competence here is a parallel in the medical field, where there is an understanding that faculty need scientific, teaching, and clinical competence (see for instance Karolinska Institutet, 2011). The missing faculty competence standard concerns enhancement of faculty competence in the subject. However, this should not be equated only with research competence. It is not enough to know the subject for oneself, it is also necessary to be able to guide others into it, so to technical and scientific knowledge I propose to add the concept pedagogical content knowledge (Shulman, 1987). It refers to “the blending of content and pedagogy into an understanding of how particular topics, problems, or issues are organized, represented, and adapted to the diverse interests and abilities of learners, and presented for instruction” (p. 8). When CDIO addresses the need to address what we teach, it constitutes a critique of educational development initiatives, and ways to assess teaching competence in appointments, which are narrowly conceptualised to concern only how we teach. For instance, when the Scholarship of Teaching and Learning movement refers to Boyer’s scholarship of teaching, it is often seen in isolation from the other facets of 81 scholarship: discovery, integration and application (see paper III, Edström, in press, 2017). If we consider the problems raised in this thesis, it would not be sufficient if all educators overnight became twice as good with regards to the methods of instruction. The education would be better, but if the teaching theory view on education still dominates over teaching professionals, some of the most important aims would still be inadequately addressed. Faculty development is a challenge that needs to consider a wide range of possibilities for shaping a faculty, which collectively has the necessary strengths related to what they teach as well as how they teach. The CDIO concepts focus on the programme and course level development, but as we saw here the only principles applying to the organisational conditions are those concerning faculty development. This is a limitation, and one possible extension of the CDIO concept would be to focus on the enabling conditions in the organisation, drawing on institutional theory. 6.2.2. CDIO as institutional innovation Wearing the theoretical lenses crafted in the previous chapter, we can see the CDIO initiative as a field-level driver of institutional innovation. CDIO as a field level site for institutional innovation The CDIO community is situated in the organisational field, promoting new logics through collective mobilisation. It can be seen as a site where narratives are crafted and shared, making up certain vocabularies of practice. When the CDIO community shares experiences from different institutional contexts, it also exposes individuals to a wider repertoire of institutional values, practices and identities, which can then also make them less likely to take things for granted in their home environment. Local innovators can invoke CDIO as a legitimate template, to strengthen the legitimacy of their local work, and as a part of their identity. The legitimacy of CDIO is partly mimetic, due to high-status universities among the founders and adopters. Whether it also achieves normative or cognitive legitimacy depends on how well the values and norms that are invoked in CDIO conform to the institutional environment, and the support will likely differ in different parts, for instance in different types of schools. CDIO as a compromise strategy The legitimacy of CDIO will to a large extend depend on the values and norms that are embedded in the analyses connecting problems with solutions, the methods and instruments for action, and communicated in the activities of the CDIO community. Paper III (Edström, in press, 2017) highlighted some significant differences in how the CDIO and PBL communities conceptualise and handle the relationship between disciplinary theory and professional aspects. Although even in PBL universities only half of the curriculum is problem/project-based, the PBL movement is still perceived to fundamentally challenge the role of the disciplines. CDIO takes another strategy, embracing discipline-led teaching as the major part of the integrated curriculum. 82 When CDIO stresses the need for a deeper working understanding of disciplinary theory, it is a plea fully consistent with the values of both academic and engineering professional logics. The motive is to make the innovation more legitimate, in the sense that it is understandable, consistent with prevailing norms and practices, and less threatening for traditional identities. This is a compromise strategy (cf. Oliver, 1991), balancing different expectations. For instance, CDIO obviously challenges programmes that consist of loosely coupled theoretical courses, but the proposed intervention is not a pure PBL approach. Instead, CDIO takes a middle road, making PBL, with a distinct engineering flavour, a part of the integrated curriculum. As a concept invented by engineering educators specifically for engineering education, CDIO can also to a larger extent invoke an engineering insider identity, as a resource in its institutional entrepreneurship. Engineering education development communities and research Another difference between CDIO and PBL identified in paper II (Edström & Kolmos, 2014) was the role of educational research. There are mature international research communities focused on PBL, with centres and professorships. Research is used to advance the philosophy and create legitimacy for the educational model. One particular emphasis has been to create empirical evidence demonstrating the results obtained through PBL, for instance the quality of the graduates from PBL universities (Kolmos & Bylov, 2016; Kolmos & Koretke, 2017a, 2017b). In contrast, the CDIO community had paid considerably less attention to research, despite a significant amount of published work. Paper III (Edström, in press, 2017) discusses how engineering education research can become more central to CDIO, an idea that arose after identifying the opportunity in the comparison with PBL. Before making such a move, it was necessary to understand the dynamics of the EER landscape, and find a position that does not abandon the development agenda or compromise the CDIO ethos. Seen through the lens of institutional theory, this can be interpreted as a strategic effort to conform to the values in the institutional environment, to improve legitimacy and access to resources, not least by strengthening the academic identities of the people involved. Taking a cautious approach I would venture so far as to suggest that it is mainly the logic of appropriateness at work, and it remains to be seen if the logic of effectiveness will also apply. 6.3. Wrapping up 6.3.1. Conclusions This thesis argues that the problems with sustainably realising the dual nature ideal in engineering education, i.e. the meaningful relationship between disciplinary theory and professional aims, lies not at the level of teaching and learning. Both types of 83 learning outcomes are present in the stated aims of engineering curricula, and they can be productively integrated into both subject-based and project-led courses. There are also educational development concepts that readily support such implementation in the curriculum, for instance the CDIO approach. However, whether the productive relationships can be realised and sustained in courses is dependent on individual faculty members. Whether the integration can be realised and sustained on the programme level is further dependent on structures to coordinate the courses, or in other words to coordinate the work of faculty members. This can make implementations vulnerable. In the words of Alvesson and Kärreman (2011), “the desire to become challenged, surprised, bewildered, and confused may take centre stage in research”. There are in particular two such instances in this thesis. One was the historical excursion, demonstrating the perennial character of the tension. Examining the process by which the engineering curriculum became science-based in the United States showed how change was slow until an “avalanche” of research funding – and the attached status and prestige, both for the university and for the faculty members – changed the character of the faculty, who then in turn changed education to become more theoretical, and in the process, professional values were weakened. This is an interesting case showing the dynamic shift in institutional logics (cf. Thornton & Ocasio, 1999). It also highlights the importance of the resource environment. There is no doubt that the material support for research is still strongest, not least because the resources under academic control are tightly coupled to survival and status for faculty. At least this is true in research-intensive universities, but my impression is that also schools with far less research are to some extent influenced by the same logics, through values, beliefs, assumptions, identities, and status concerns. By triangulating the ideal that Carl Richard Söderberg and CDIO had in common, I concluded that we should let go of the swinging pendulum metaphor, because it depicts a zero-sum balance. Instead of fighting over the amount of disciplinary theory and practice- oriented learning, we should focus on strengthening the meaningful relationships between them in the curriculum. Söderberg as a person combined the theoretical and practical interest, himself embodying the dual nature ideal. This inspired a suggestion that to achieve the duality, we need enough people in the faculty who can simultaneously defend both the academic and the professional values. If too many people stand on one side only, then the zero-sum view, and the ensuing trench wars, may be hard to avoid. The second bewildering moment came from the accounts of unsustainable change in engineering curricula, leading me to formulate the organisational gravity model. The message was not so much that it is difficult to achieve change, but that even successful change may not be sustainable. (Again: it makes no prediction about the sustainability of change. Instead, it is a sensitising and explanatory model to assist perception and provide conceptual tools with which to discuss experiences.) One important debate within institutional theory has been how to conceptualise both 84 stability and change. The organisational gravity model depicts a flexible response that can also be seen as a protective device, a spring-damper. Curriculum change can be successful, temporarily, but I suggested that certain types of changes – those with weaker support in the institutional logics – are gradually undone when the dominant process makes its continuous imprint on education through its dominance over faculty reproduction, socialisation, resources and incentives, and leadership recruitment. The contribution of this discussion could be to suggest a more dynamic image of stability and change. The organisational gravity model suggested widening the perspective from curriculum development in itself, to the organisational level. This was also consistent with the historical study. Drawing on an institutional logics perspective, this thesis suggests that the logics of the academic profession seem to have the upper hand in research and, through the faculty, also in education. The result is a double hegemony where the logics of the academic profession are the strongest in both education and research. This weakens the professional preparation, and at the same time the education as a whole – both the theoretical and the professional preparation – is affected by the general disadvantages of education relative to research. The resulting impression is that the academic profession of the educators takes precedence over the engineering profession that we educate for, and this has consequences for the effectiveness of the engineering curricula. I also suggested that we could understand educational development as efforts to compensate for such values that are obviously necessary for education, but inadequately represented in the organisation. Brunsson made the point that decoupling talk and action can serve to protect the production organisation from incompatible demands. Accordingly, we may consider to what extent educational development belongs to the production organisation or the political organisation, the latter handling insolvable problems mainly through ideology production. I have argued that the academic-professional tension is inherent to engineering education, an insolvable problem caused by conflicting institutional demands. Precisely therefore, it is necessary to have qualified venues for discussing ideals and tensions, and for identifying opportunities for handling the tensions productively in practical situations. Supporting practical development is key, because the tensions will be expressed in the curriculum, one way or the other, whether the implications are understood or not. And even those who subscribe to Brunsson’s refreshing take on reform will understand the value of hope, in helping the organisation carry on in good faith instead of being paralysed by conflict. Applying the institutional logics analysis, some suggestions were made for strengthening the legitimacy of educational development initiatives. One strategy identified here is to find support in the institutional environment by shaping narratives, conceptual tools, and instruments to define and emphasise common ground. Another strategy is to integrate engineering education development and 85 research, producing work that combines scholarliness with usefulness and relevance for improving engineering education. In this thesis I wanted to bridge engineering education development and research, but without implying that development is the optimistic doing and research the critical thinking. I have taken a critical approach, questioning dominant interests and asking what groups are privileged by the way things work around here, and what groups get short-changed. At the same time I have shown that hopefulness is possible, necessary, and justified. 6.3.2. Contribution The thesis is essentially a case study, using the CDIO approach to represent a particular kind of educational development addressing one of the major issues in professional engineering education, the academic-professional dual nature. It contributes a rich description of this case from several perspectives: analysing strategies for curriculum development and faculty development; comparison with PBL; connection to engineering education research; considerations of unsustainable change; and a historical comparison. This account could stimulate further reflection those who are interested in the nature of professional education, for CDIO practitioners, and for anyone considering reform attempts of any kind. The main theoretical contribution was to combine a critical approach with a theoretical framework based on institutional theory, to analyse the academic and professional nature of engineering education, and interpreting engineering education development as a case of institutional innovation. The research outcomes helped highlight both challenges and opportunities for engineering education development, uncovering them in a novel way that may reinvigorate such discussions. The research approach is also a contribution to a more critical discussion about engineering education, about engineering education development, and about engineering education research. Some other fruit were harvested along the way. The thesis contains an attempt to formulate a research agenda in engineering education research, combining usefulness and scholarliness. Further, the organisational gravity model offers us a new view on unsustainable change, and on educational development. Finally, the biography of Carl Richard Söderberg was a contribution in its own right as an historical account, and the comparison of past and present discussions was also an experiment in engineering education research that engages with the past. 6.3.3. Future research Obvious avenues for future research would be to study empirical cases of programme development, in particular using a longitudinal approach to investigate what happens after a reform is considered completed. In the cases of sustainable or unsustainable implementation in curricula – what makes them so? What processes are involved 86 more specifically to sustain, develop or abandon new practices? How do normative and material aspects interplay? It would also be relevant to study how people in the organisation experience and interpret the tension between academic and professional values. How is the tension present in educators’ own views on engineering education, or in their own teaching practice, and in their competence? It would also be interesting to trace how the tension is enacted in key organisational practices. For instance, how are merits related to the academic and professional aspects valued in recruiting and promoting faculty? To further refine this research, it would also be necessary to study differences between different types of institutions, for instance research-intensive universities and those with less research. While the experiences referred to in this thesis mainly come from higher education institutions characterised by significant research, according to institutional theory the norms, values, and patterns on the field level are likely to influence also institutions where there is relatively little research. To extend these findings, a promising alternative is to compare engineering education with professional education in other fields. What is the relationship between theoretical knowledge and professional competence in other fields? Do they have similar ideals and tensions, and how are they conceptualised, debated and handled? What would a historical comparison show? It would be interesting to see if it matters how recently the education has come into the university environment, whether parts of the education takes place in professional environments (such as internships in schools or teaching hospitals), the professional background and experience of faculty, how strong the image is of the professional role and destination, and what role is played by professional organisations and other stakeholders. 87 Aamodt, P. O., Frølich, N., & Stensaker, B. (2016). Learning outcomes – a useful tool in quality assurance? Views from academic staff. Studies in higher education, 1-11. ABET (Accreditation Board for Engineering and Technology). (1994). Engineering Criteria 2000. Baltimore, MD. Alvesson, M. (2012). Understanding organizational culture. London: Sage. Alvesson, M., Gabriel, Y., & Paulsen, R. (2017). Return to Meaning: A Social Science with Something to Say. Oxford: Oxford University Press. Alvesson, M., & Kärreman, D. (2011). Qualitative Research and Theory Development: Mystery as Method. London: Sage. Alvesson, M., & Sandberg, J. (2013). Constructing research questions: Doing interesting research. London: Sage. Alvesson, M., & Sköldberg, K. (1994). Tolkning och reflektion: vetenskapsfilosofi och kvalitativ metod [Interpretation and reflection: philosophy of science and qualitative method]. Lund: Studentlitteratur. Alvesson, M., & Sveningsson, S. (2008). Förändringsarbete i organisationer - om att utveckla företagskulturer [Change work in organisations - on developing company cultures]. Malmö: Liber. Ambrose, D. (1987). Managing complex change. Pittsburgh: The Enterprise Group. Argyris, C., & Schön, D. A. (1996). Organizational learning II: Theory, Method and Practice. Reading: Addison–Wesley. Barnett, R. (1992). Improving Higher Education: Total Quality Care. Buckingham: SRHE and Open University Press. Barnett, R., & Coate, K. (2004). Engaging the curriculum: McGraw-Hill International. Bernhard, J., & Baillie, C. (2016). Standards for Quality of Research in Engineering Education. International Journal of Engineering Education, 32(6), 2378– 2394. Biggs, J., & Tang, C. (2011). Teaching for quality learning at university. Maidenhead: McGraw-Hill. Binder, A. (2007). For love and money: Organizations’ creative responses to multiple environmental logics. Theory and society, 36(6), 547-571. Birnbaum, R. (1988). How colleges work: The Cybernetics of Academic Organization and Leadership. San Francisco: Jossey-Bass. Bishop, A. J. (1992). International perspectives on research in mathematics education. In D. A. Grouws (Ed.), Handbook of research on mathematics teaching and learning (pp. 710–723). New York: Macmillan. Björkqvist, J., Edström, K., Hugo, R. J., Kontio, J., Roslöf, J., Sellens, R., & Virtanen, S. (Eds.). (2016). Proceedings of the 12th International CDIO Conference. Turku: Turku University of Applied Sciences. Bleiklie, I., Frølich, N., Sweetman, R., & Henkel, M. (2017). Academic Institutions, Ambiguity and Learning Outcomes as Management Tools. European Journal of Education, 52(1), 68-79. Bloom, B. S. (1956). Taxonomy of educational objectives. Vol. 1: Cognitive domain. New York: McKay. Bolander Laksov, K., McGrath, C., & Josephson, A. (2014). Let’s talk about integration: a study of students’ understandings of integration. Advances in Health Sciences Education, 19(5), 709-720. 88 Borrego, M., & Bernhard, J. (2011). The emergence of engineering education research as an internationally connected field of inquiry. Journal of Engineering Education, 100(1), 14-47. Borrego, M., & Henderson, C. (2014). Increasing the use of evidencebased teaching in STEM higher education: A comparison of eight change strategies. Journal of Engineering Education, 103(2), 220-252. Boud, D., & Brew, A. (2013). Reconceptualising academic work as professional practice: implications for academic development. International Journal for Academic Development, 18(3), 208-221. Boyer, E. L. (1990). Scholarship reconsidered : priorities of the professoriate. Princeton: The Carnegie Foundation for the Advancement of Teaching. Brennan, R., Edström, K., Hugo, R., Roslöf, J., Songer, R., & Spooner, D. (Eds.). (2017). Proceedings of the 13th International CDIO Conference. Calgary: University of Calgary. Brooks, H. (1967a). Applied Research, Definitions, Concepts, Themes Applied science and technological progress: A report to the Committee on Science and Astronautics, U.S. House of Representatives (pp. 21-55). Washington, DC.: National Academy of Sciences. Brooks, H. (1967b). Dilemmas of engineering education. IEEE spectrum, 2(4), 89-91. Brown, G. S. (1962). New horizons in engineering education. Daedalus, 341-361. Brunsson, N. (1989). The organization of hypocrisy: talk, decisions and actions in organizations. Chichester: John Wiley & Sons. Brunsson, N. (2009). Reform as routine: Organizational change and stability in the modern world. Oxford: Oxford University Press. Case, J., & Marshall, D. (2004). Between deep and surface: procedural approaches to learning in engineering education contexts. Studies in higher education, 29(5), 605-615. Chalmers. (2013). University appointment regulations for teaching and research faculty at Chalmers. Göteborg: Chalmers. Christakis, N. A. (1995). The similarity and frequency of proposals to reform US medical education: constant concerns. Jama, 274(9), 706-711. Christensen, S. H., & Ernø-Kjølhede, E. (2012). Socio-technical Integration in Engineering Education: A Never-Ending Story. In S. Hyldgaard Christensen, C. Mitcham, B. Li, & Y. An (Eds.), Engineering, Development and Philosophy (pp. 197-213). Dordrecht: Springer. Clark, B. R. (1986). The higher education system: Academic organization in cross- national perspective. Berkeley: Univ of California Press. CNN (Producer). (2015, March 13). CNN Ones to Watch. Inside Turner Prize winning artist's studio. Retrieved from http://edition.cnn.com/videos/tv/2015/03/13/spc-ones-to-watch-sculpture-a- block.cnn Cohen, L., Manion, L., & Morrison, K. (2011). Research Methods in Education. London and New York: Routledge. Crawley, E. F. (2001). The CDIO Syllabus: A statement of goals for undergraduate engineering education (CDIO Report #1). Cambridge, MA.: Massachusetts Institute of Technology. Crawley, E. F., Malmqvist, J., Östlund, S., & Brodeur, D. (2007). Rethinking Engineering Education: The CDIO Approach. Boston: Springer. 89 Crawley, E. F., Malmqvist, J., Östlund, S., Brodeur, D. R., & Edström, K. (2014). Rethinking Engineering Education: The CDIO Approach. Cham: Springer International Publishing. Dahler-Larsen, P. (1998). Den rituelle reflektion - om evaluering i organisationer [The ritual reflection - on evaluation in organisations]. Odense: Syddansk universitetsforlag. de Graaff, E., & Kolmos, A. (Eds.). (2007). Management of change: Implementation of Problem-Based and Project-Based Learning in Engineering. Rotterdam: Sense Publishers. de Weck, O. (2004). Professors of the Practice: Bringing the Real World to MIT. MIT Faculty Newsletter, XVII (2). Retrieved from http://web.mit.edu/fnl/vol/172/deweck.htm Deephouse, D. L., & Suchman, M. (2008). Legitimacy in organizational institutionalism. In R. Greenwood, C. Oliver, R. Suddaby, & K. Sahlin- Andersson (Eds.), The Sage handbook of organizational institutionalism (pp. 49-77). London: Sage. Dill, D. D. (1982). The management of academic culture: Notes on the management of meaning and social integration. Higher education, 11(3), 303-320. Dill, D. D. (2012). The management of academic culture revisited: integrating universities in an entrepreneurial age. In B. Stensaker, J. Välima, & C. S. Sarrico (Eds.), Managing Reform in Universities: The Dynamics of Culture, Identity and Organisational Change (pp. 222-237). London: Palgrave Macmillan. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American sociological review, 48(2), 147-160. Edström, K. (2010). Utvärdering av Akademiskt Lärarskap – den högskolepedagogiska verksamheten vid Malmö högskola [Evaluation of Academic Teachership - the teaching and learning unit at Malmö University]. Retrieved from http://mah.se/upload/Medarbetare/akademisktlararskap/dokument/Edstrom201 0UtvarderingAKL.pdf Edström, K. (2011). Organisationens gravitation – om uthållighet i utveckling av utbildning [Organizational Gravity - on sustainability in educational development]. Paper presented at the Att leda högre utbildning [Leading Higher Education], Sveriges universitets- och högskoleförbund (SUHF), 14-15 november 2011, Karolinska Institutet. Edström, K. (forthcoming 2018). Academic and Professional Values in Engineering Education: Engaging with History to Explore a Persistent Tension. Engineering Studies. Edström, K. (in press, 2017). The role of CDIO in engineering education research: combining usefulness and scholarliness. European Journal of Engineering Education. Edström, K., El Gaidi, K., Hallström, S., & Kuttenkeuler, J. (2005). Integrated assessment of disciplinary, personal and interpersonal skills-student perceptions of a novel learning experience Proceedings of the13th Improving Student Learning Conference. Oxford: OCSLD. Edström, K., Hallström, S., & Kuttenkeuler, J. (2011). Mini workshop—Designing project-based courses for learning and cost-effective teaching. Paper presented at the Frontiers in Education Conference (FIE), 2011. 90 Edström, K., & Kolmos, A. (2014). PBL and CDIO: complementary models for engineering education development. European Journal of Engineering Education, 39(5), 539-555. doi:10.1080/03043797.2014.895703 Elmore, R. F., & McLaughlin, M. W. (1988). Steady Work. Policy, Practice, and the Reform of American Education. Santa Monica: Rand Corp. Enelund, M., Knutson Wedel, M., Lundqvist, U., & Malmqvist, J. (2013). Integration of education for sustainable development in the mechanical engineering curriculum. Australasian Journal of Engineering Education, 19(1), 51-62. Enelund, M., Larsson, S., & Malmqvist, J. (2011). Integration of Computational Mathematics Education in the Mechanical Engineering Curriculum. Paper presented at the The 7th International CDIO Conference, Copenhagen, Denmark. European Journal of Engineering Education. (2016). "Call for Paper: Scholarly Development of Engineering Education – the CDIO approach" Guest Editors: K. Edström, J. Malmqvist, and J. Roslöf. Feisel, L. D. (1986). Teaching Students to Continue Their Education. Paper presented at the Frontiers in Education, Arlington, Texas. Friedland, R. (2013). God, love and other good reasons for practice: Thinking through institutional logics. Institutional Logics in Action: Research in the Sociology of Organizations, 39, 25-50. Friedland, R., & Alford, R. R. (1991). Bringing society back in: Symbols, practices and institutional contradictions. In P. J. DiMaggio, & Powell, W. W. (Ed.), The new institutionalism in organizational analysis. Chicago: University of Chicago Press. Froyd, J., Penberthy, D., & Watson, K. (2000). Good educational experiments are not necessarily good change processes. Paper presented at the 30th Annual Frontiers in Education Conference, 2000. Fullan, M. (1999). Change forces: The sequel. London: Falmer Press. Gainsburg, J. (2015). Engineering students' epistemological views on mathematical methods in engineering. Journal of Engineering Education, 104(2), 139-166. Geschwind, L. (2015). Research ties: sine qua non of higher education? In J. Björkman & B. Fjæstad (Eds.), Thinking Ahead. Research, Funding and the Future. RJ Yearbook 2015/2016 (pp. 91-106). Göteborg/Stockhom: Makadam förlag. Geschwind, L., & Broström, A. (2015). Managing the teaching–research nexus: Ideals and practice in research-oriented universities. Higher education research & development, 34(1), 60-73. Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P., & Trow, M. (1994). The new production of knowledge: The dynamics of science and research in contemporary societies. London: Sage. Gibbs, G. (2013). Reflections on the changing nature of educational development. International Journal for Academic Development, 18(1), 4-14. doi:10.1080/1360144X.2013.751691 Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. Berkeley and Los Angeles: Univ of California Press. Graham, R. (2012). Achieving excellence in engineering education: the ingredients of successful change. London: The Royal Academy of Engineering. Graham, R. (2015). Does Teaching Advance Your Academic Career?: Perspectives of Promotion Procedures in UK Higher Education. London: Royal Academy of Engineering 2. 91 Greenwood, R., & Hinings, C. R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21(4), 1022-1054. Greenwood, R., Suddaby, R., & Hinings, C. R. (2002). Theorizing change: The role of professional associations in the transformation of institutionalized fields. Academy of management journal, 45(1), 58-80. Hallström, S., Kuttenkeuler, J., & Edström, K. (2007). The route towards a sustainable design-implement course. Proceedings of the 3rd International CDIO Conference, MIT, Cambridge, Massachusetts, USA. Hallström, S., Kuttenkeuler, J., Niewoehner, R., & Young, P. W. (2014). Design- Implement Experiences and Engineering Workspaces. In E. F. Crawley, J. Malmqvist, S. Östlund, D. R. Brodeur, & K. Edström (Eds.), Rethinking Engineering Education: The CDIO Approach. 2nd Ed. (pp. 117-142). Cham: Springer. Henderson, C., Beach, A., & Finkelstein, N. (2011). Facilitating change in undergraduate STEM instructional practices: An analytic review of the literature. Journal of research in science teaching, 48(8), 952-984. Henkel, M. (2005). Academic identity and autonomy in a changing policy environment. Higher education, 49(1-2), 155-176. Holmegaard, H. T., Madsen, L. M., & Ulriksen, L. (2016). Where is the engineering I applied for? A longitudinal study of students' transition into higher education engineering, and their considerations of staying or leaving. European Journal of Engineering Education, 41(2), 154-171. Holmegaard, H. T., Ulriksen, L., & Madsen, L. M. (2010). Why students choose (not) to study engineering. Paper presented at the Proc. of the Joint International IGIP-SEFI Annual Conference. Jessop, T., & Bolander Laksov, K. (2017). Moving beyond orthodoxies in academic development. International Journal for Academic Development, 22(4), 275- 277. Jonassen, D. H. (2014). Engineers as Problem Solvers. In A. Johri & B. M. Olds (Eds.), Cambridge handbook of engineering education research (pp. 103- 118). New York, NY.: Cambridge University Press. Jonassen, D. H., Strobel, J., & Lee, C. B. (2006). Everyday problem solving in engineering: Lessons for engineering educators. Journal of Engineering Education, 95(2), 139-151. Jørgensen, U. (2014). Historical Accounts of Engineering Education. In E. F. Crawley, J. Malmqvist, S. Östlund, D. R. Brodeur, & K. Edström (Eds.), Rethinking Engineering Education: The CDIO Approach (pp. 231-255). Cham: Springer. Karolinska Institutet. (2011). Qualifications Portfolio for Teachers and Researchers at Karolinska Institutet. Retrieved from http://ki.se/sites/default/files/qualifications_portfolio_-_instructions.pdf Kezar, A. (2001). Understanding and facilitating organizational change in the 21st century. ASHE-ERIC higher education report, 28(4), 147. Knapper, C. (2016). Does educational development matter? International Journal for Academic Development, 21(2), 105-115. doi:10.1080/1360144X.2016.1170098 Knoster, T., Villa, R., & Thousand, J. (2000). A framework for thinking about systems change. In R. Villa & J. Thousand (Eds.), Restructuring for caring and effective education: Piecing the puzzle together (pp. 93-128). 92 Kodeih, F., & Greenwood, R. (2014). Responding to institutional complexity: The role of identity. Organization Studies, 35(1), 7-39. Kogan, M. (1999). The culture of academe. Minerva, 37(1), 63-74. Kolmos, A., & Bylov, S. M. (2016). Ingeniørstuderendes forventning og parathed til det kommende arbejdsliv: Arbejdsrapport no. 1 [Engineering students expectations and transition to working life. Working report No. 1]. Aalborg: Aalborg Centre for Problem Based Learning in Engineering Science and Sustainability. Kolmos, A., & Koretke, R. B. (2017a). AAU teknisk-naturvidenskabelige studerendes forventning og parathed til det kommende arbejdsliv: Arbejdsrapport no. 2 [Aalborg University technology and science students' expectations and transition to working life. Working report No. 2]. Aalborg: Aalborg Centre for Problem Based Learning in Engineering Science and Sustainability. Kolmos, A., & Koretke, R. B. (2017b). Nyuddannede ingeniørers erfaring med overgang fra uddannelse til arbejdsliv: Arbejdsrapport no. 3 [Recent engineering graduates' experience of transition from education to working life. Working report No. 3]. Aalborg: Aalborg Centre for Problem Based Learning in Engineering Science and Sustainability. Kotter, J. P. (1996). Leading change. Boston: Harvard Business Press. Krathwohl, D. R. (2002). A Revision of Bloom's Taxonomy: An Overview. Theory Into Practice, 41(4), 212-218. Kuttenkeuler, J. (2017). Student project videos: Evolo, Infernus, Solar Power Aircraft. Stockholm: KTH Royal Institute of Technology. Retrieved from www.kth.se/profile/jakob/page/evolo Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Beverly Hills: Sage Publications. Lindberg-Sand, Å., Sonesson, A., Lörstad, B., Gran, B., Gustafsson, N., Järnefelt, I., & Lundkvist, H. (2005). Pedagogisk utbildning för högskolans lärare: Pilotprojektet vid Lunds universitet 2002-2005. Resultat, förslag och sedan? [Education in teaching and learning for faculty in higher education: The Pilot Project at Lund University 2002-2005. Results, proposals and then?]. In J. Järnefelt (Ed.), Proceedings från Utvecklingskonferensen för högre utbildning [Proceedings of the Higher Education Development conference], 2005 (pp. 183-192). Karlstad: Karlstad Universitet. Malmqvist, J., Bankel, J., Enelund, M., Gustafsson, G., & Knutson Wedel, M. (2010). Ten Years of CDIO - Experiences from a Long-term Education Development Process Proceedings of the 6th International CDIO Conference. École Polytechnique de Montréal, Québec, Canada. Malmqvist, J., Gunnarsson, S., & Vigild, M. (2008). Faculty professional competence development programs-comparing approaches from three universities Proceedings of the 4th International CDIO Conference. Hogeschool Gent, Gent, Belgium. Malmqvist, J., Hugo, R., & Kjellberg, M. (2015). A survey of CDIO implementation globally–effects on educational quality. Paper presented at the Proc. 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, PR China. Malmqvist, J., Östlund, S., & Edström, K. (2006). Integrated program descriptions–A tool for communicating goals and design of CDIO programs Proceedings of the 2nd International CDIO Conference. Linköping University, Linköping. March, J. G. (1982). Theories of choice and making decisions. Society, 20(1), 29-39. 93 March, J. G., & Olsen, J. P. (2004). The logic of appropriateness. Oslo: Arena. Marshall, C., & Rossman, G. B. (2010). Designing qualitative research. Thousand Oaks: Sage. Mårtensson, K. (2014). Influencing teaching and learning microcultures. Academic development in a research-intensive university. Lund: Lund University. Marton, F., Hounsell, D., & Entwistle, N. J. (1984). The experience of learning. Edinburgh: Scottish Academic Press. McGrath, C., & Bolander Laksov, K. (2014). Laying bare educational crosstalk: a study of discursive repertoires in the wake of educational reform. International Journal for Academic Development, 19(2), 139-149. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American journal of sociology, 340-363. Meyer, J. W., & Rowan, B. (1983). The Structure of Educational Organizations Organizational environments: Ritual and rationality (pp. 71-97). Beverly Hills: Sage. Meyer, J. W., Scott, W. R., & Deal, T. E. (1983). Institutional and Technical Sources of Organizational Structure: Explaining the Structure of Educational Organizations. In J. W. Meyer, W. R. Scott, B. Rowan, & T. E. Deal (Eds.), Organizational environments: Ritual and rationality (pp. 45-67). Beverly Hills: Sage. Miles, M. B., & Ekholm, M. (1991). Will New Structures Stay Restructured? Paper presented at the Annual Meeting of tne American Educational Research Association, Chicago, IL, April 3-7, 1991. MIT. (1949). Committee on Educational Survey. Report to the Faculty of the Massachusetts Institute of Technology. (The Lewis Report). Cambridge, MA.: Technology Press. MIT. (2000). Improved Engineering Education: Changing the Focus towards Active Learning in a CDIO Context. A Proposal for the Knut and Alice Wallenberg Foundation. In cooperation with Chalmers University of Technology, KTH Royal Institute of Technology, and Linköping Institute of Technology. MIT. (2017). MIT Policies & Procedures (2.3.2 - 2.3.3). Retrieved from https://policies-procedures.mit.edu/node/30/#staff2 Morgan, G. (2006). Images of organization. Thousand Oaks: Sage. Musselin, C. (2007). Are universities specific organisations. In G. Krücken, A. Kosmützky, & M. Torka (Eds.), Towards a Multiversity ? Universities between Global Trends and national Traditions (pp. 63-84). Bielefeld: Transcript Verlag. NSF. (2013). Theodore von Kármán (1881-1963), National Medal of Science 50th Anniversary. Retrieved from www.nsf.gov/news/special_reports/medalofscience50/vonkarman.jsp Oliver, C. (1991). Strategic responses to institutional processes. Academy of Management Review, 16(1), 145-179. Perry, W. G., Jr. (1998). Forms of Intellectual and Ethical Development in the College Years: A Scheme. San Francisco: Jossey-Bass. Pfeffer, J., & Salancik, G. R. (1978). The external control of organizations: A resource dependence perspective. New York: Harper and Row Publishers. Picon, A. (2004). Engineers and engineering history: Problems and perspectives. History and Technology, 20(4), 421-436. Prop. 2004/05:162. (2005). Ny värld–ny högskola. Stockholm: Sveriges Riksdag. 94 Puffer, S. M., & Moss Kanter, R. (2004). An Interview with Rosabeth Moss Kanter. The Academy of Management Executive (1993-2005), 18(2), 96-105. Roberts, D. A. (1982). Developing the concept of “curriculum emphases” in science education. Science education, 66(2), 243-260. Robinson, V. M. (1993). Problem-based methodology: Research for the improvement of practice. Oxford: Pergamon Press. Roxå, T., & Mårtensson, K. (2017). Agency and structure in academic development practices: are we liberating academic teachers or are we part of a machinery supressing them? International Journal for Academic Development, 22(2), 95- 105. doi:10.1080/1360144X.2016.1218883 Seely, B. (1999). The Other Re-engineering of Engineering Education, 1900–1965. Journal of Engineering Education, 88(3), 285-294. Seely, B. (2005). Patterns in the history of engineering education reform: A brief essay Educating the engineer of 2020: Adapting engineering education to the new century (pp. 114-130). Washington, DC: The National Academies Press. Shulman, L. (1987). Knowledge and teaching: Foundations of the new reform. Harvard educational review, 57(1), 1-23. Stensaker, B. (2015). Organizational identity as a concept for understanding university dynamics. Higher education, 69(1), 103-115. Stensaker, B. (2017). Academic development as cultural work: responding to the organizational complexity of modern higher education institutions. International Journal for Academic Development, 1-12. Stensaker, B., Välimaa, J., & Sarrico, C. (2012). Managing Reform in Universities: The Dynamics of Culture, Identity and Organisational Change. London: Palgrave Macmillan. Stokes, D. E. (1997). Pasteur's quadrant: Basic science and technological innovation. Washington, DC: Brookings Institution Press. Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571-610. Suddaby, R. (2015). Can Institutional Theory Be Critical? Journal of Management Inquiry, 24(1), 93-95. Swedberg, R. (2012). Theorizing in sociology and social science: Turning to the context of discovery. Theory and society, 41(1), 1-40. Swedberg, R. (2016). Before theory comes theorizing or how to make social science more interesting. The British journal of sociology, 67(1), 5-22. Söderberg, C. R. (1954). President’s Report, MIT (pp. 61-77). Cambridge: Massachusetts Institute of Technology. Söderberg, C. R. (1962). The Trends in Engineering Education: Do They Concern the Power Industry? In Proceedings of the American Power Conference. Chicago: Illinois Institute of Technology. Söderberg, C. R. (1967). A Note on Engineering Education. In Applied science and technological progress: A report to the Committee on Science and Astronautics, U.S. House of Representatives (pp. 399-413). Washington, DC: National Academy of Sciences. Söderberg, C. R. (1979). My Life. Theodorsdottir, A. H., Saemundsdottir, I., Malmqvist, J., Turenne, S., & Rouvrais, S. (2013). Comparison of Hiring and Promotion Criteria Linked to Teaching, Educational Development and Professional Engineering Skills Prodeedings of the 9th International CDIO Conference. Cambridge: MIT & Harvard School of Engineering and Applied Sciences. 95 Thornton, P. H., & Ocasio, W. (1999). Institutional Logics and the Historical Contingency of Power in Organizations: Executive Succession in the Higher Education Publishing Industry, 1958-1990. American journal of sociology, 105(3), 801-843. Thornton, P. H., & Ocasio, W. (2008). Institutional logics. In R. Greenwood, C. Oliver, R. Suddaby, & K. Sahlin-Andersson (Eds.), The Sage handbook of organizational institutionalism (pp. 99-129). Thousand Oaks: Sage. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure, and process. Oxford: Oxford University Press. Walther, J., Sochacka, N. W., Benson, L. C., Bumbaco, A. E., Kellam, N., Pawley, A. L., & Phillips, C. M. L. (2017). Qualitative Research Quality: A Collaborative Inquiry Across Multiple Methodological Perspectives. Journal of Engineering Education, 106(3), 398-430. Weick, K. E. (1976). Educational organizations as loosely coupled systems. Administrative science quarterly, 1-19. Weick, K. E. (1989). Theory construction as disciplined imagination. Academy of Management Review, 14(4), 516-531. Williams, R. (2002). Retooling: A historian confronts technological change. Cambridge, MA.: MIT Press. Wisnioski, M. (2015). What's the Use? History and Engineering Education Research. Journal of Engineering Education, 104(3), 244-251. Woolf, V. (1929). A Room of One's Own. Richmond: Hogarth Press. 96 A SURVEY OF CDIO IMPLEMENTATION GLOBALLY – EFFECTS ON EDUCATIONAL QUALITY Johan Malmqvist Department of Product and Production Development Chalmers University of Technology, Gothenburg, SWEDEN Ron Hugo Department of Mechanical & Manufacturing Engineering University of Calgary, Calgary, CANADA Malin Kjellberg Division of Engineering Education Research Chalmers University of Technology, Gothenburg, SWEDEN ABSTRACT The CDIO approach to engineering education was introduced in the early 2000’s. Some universities have gained considerable long-term experience in applying the approach, and consequently it seems timely to summarize and evaluate those experiences. This paper thus reports the results of a survey distributed to all members of the CDIO Initiative in October 2014. The aims of the survey were to: x Map out where and in what programs/disciplines CDIO is currently applied x Evaluate the effects on outcomes, the perceived benefits, the limitations, any barriers to implementation, and ascertain future development needs Forty-seven universities from twenty-two countries participated in the survey. The main findings from the survey include the following: x The most common engineering disciplines in which CDIO is implemented are Mechanical, Electrical, and Computer Engineering. However, many CDIO schools have also implemented CDIO in Industrial, Civil and Chemical Engineering. x The main motives for choosing to adapt CDIO are; ambitions to make engineering education more authentic; the need for a systematic methodology for educational design; and the desire to include more design and innovation in curricula. x Most CDIO implementations successfully achieve both goals for learning and for external recognition of educational quality. KEYWORDS CDIO implementation, Survey, Success factors, CDIO Standards 1-12 Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. INTRODUCTION The Conceive-Design-implement-Operate (CDIO) approach (Crawley et al., 2014) to engineering education was introduced in the early 2000’s. The goals of CDIO include educating graduates with a deep and working knowledge of engineering fundamentals, who can lead in the development and operation of complex technical systems, and who have a strategic understanding of the role and impact of technology in society. The goals should be achieved within the constraints of fixed resources in terms of student and faculty time, the size of student workspaces, and budgetary restraints. CDIO proposes that an educational design that meets these goals is characterized by learning outcomes established in close contact with stakeholders, by design-implement experiences, and by integrating learning of the discipline with the development of professional skills. CDIO further features a systematic approach for designing and continuously improving education. The CDIO approach is formalized in the CDIO framework, consisting of the CDIO syllabus and the CDIO standards. The large number of universities worldwide that have adapted the CDIO approach is a sign that the CDIO approach has promise and seems plausible to many. However, there are certainly valid questions to ask: What is the evidence that adaptation of CDIO leads to improved student learning? Will the changes to engineering education suggested by CDIO result in gains in certain knowledge and skills areas but reductions in other areas, such as mathematics and science? How resource-demanding is CDIO in comparison to the status- quo approach to engineering education? How successful are universities in achieving the goals of CDIO? What needs to be considered in order to successfully implement CDIO? The overall purpose of this paper is to investigate these issues. We are taking advantage of the circumstance that a large number of universities have implemented CDIO to enable a quantitative and survey-based research design. We are further examining the different levels of CDIO experiences amongst CDIO Initiative members to explore the progress of CDIO implementation. Questions pertaining to implementation include: How long does it take to implement CDIO? Which areas are challenging or easy in order to implement CDIO? This paper thus reports the results of a survey conducted in October 2014 with all members of the CDIO Initiative, which currently stands at more than 120 universities from around the world. Specifically, the aims of the paper are to: x Map out where and in what programs/disciplines CDIO is currently applied; x Evaluate the effects on outcomes, the perceived benefits, the limitations, and any barriers to implementation; and x Ascertain future development needs. The remainder of the paper is structured as follows. We first review earlier work that has aimed to categorize and follow up on CDIO implementation efforts, focusing on the university and program level. We then account for the design of the survey. A presentation and discussion of the findings follow. Finally, conclusions are listed. Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. EARLIER WORK Gray (2008) conducted a CDIO status survey, in which 23 of the then 27 members of the CDIO Initiative participated. Gray’s survey focused on the use of the CDIO standards as a quality enhancement tool and on the progress of CDIO member universities with respect to the standards. The CDIO members were categorized as “new” (ζ 2 years CDIO experience), “intermediate” (3-4 years experience) and “senior” (η 5 years experience). Gray’s survey showed that for new CDIO members, Standard 5 (Design-implement experiences) was rated highest, while Standard 9 (Faculty Professional Skills) and Standard 12 (Program Evaluation) were rated lowest. Amongst the senior CDIO members, CDIO standards 1, 2, 4, 5 and 12 were rated highest and standards 9 and 10 lowest. Gray’s data suggested that many schools had joined CDIO with an already existing interest and experiences in design-implement, but also that the standards related to faculty competence (9, 10) are the most difficult to improve on. A number of CDIO programs have enough experience of CDIO to be able to evaluate long- term effects of CDIO implementation: An early study was conducted at Linköping University comparing student cohorts who had started their studies before the CDIO introduction with students that had followed a CDIO program from the start. They found that the CDIO students considered themselves as significantly better at teamwork, and that they valued the CDIO courses/project as the most valuable learning experience during their studies. The Linköping students specifically identified the skills of problem solving, of critical thinking, of handling heavy workloads, and of project management as the most transferable to professional work situations (Edvardsson Stiwne & Jungert, 2007). However, they were not able to discern any difference between pre- CDIO and CDIO graduates concerning employability (high already before) and student retention (still low) and student recruitment (continued to drop). Malmqvist et al. (2010) presented a ten-year follow up of Chalmers’ CDIO implementation in mechanical engineering. Chalmers alumni survey data shows that mechanical engineering graduates self-assess their design, communication and teamwork skills significantly higher than graduates from other programs at Chalmers. The mechanical engineering program at Chalmers has won several national (Sweden) awards for high quality education. The paper further shows that the CDIO implementation required substantial investment costs, but that the operating costs for the education are manageable. Evaluations of the chemical engineering program at Singapore Polytechnic (Cheah et al. (2013); Ng (2014)) have identified positive effects related to student retention and alumni self-assessment of communications, systems thinking and creative skills. However, the same studies also found it hard to assess conclusively effects on graduate employment rate, salary and course satisfaction ratings. ISEP Porto (Martins et al., 2013) reported that the CDIO implementation of its computer engineering program had lead to a national (Portugal) ranking as a leading computer engineering program, and improved student retention, employer satisfaction and quality of final degree projects. Duy Tan University (Nguyen et al., 2014) describe how CDIO was used for the successful ABET accreditation of their programs. CDIO is said to have helped identify weaknesses with Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. regards to ABET criteria, and it is argued that CDIO is the best tool for ABET accreditation preparation. In conclusion, the demographics and progress of CDIO implementations have not been surveyed since 2008. Since then, the CDIO Initiative has grown from 27 to 118 members. Effects, barriers and success factors of/for CDIO implementation have so far, to our knowledge, only been surveyed by individual programs. These program-specific evaluations identify typical benefits of CDIO, but also some intended goals that are more difficult to conclude on. The work reported here aims address to this gap, by investigating if reported benefits and success factors are valid across a larger sample of implementations. SURVEY DESIGN The survey was comprised of approximately 50 questions in the following categories: x University categorization and CDIO use; x Level of a university’s CDIO implementation; x Statements about the effects on input, resource and output metrics; x Barriers and success factors; and x Open-ended questions. The university categorization questions considered basic university demographics such as size, location, QS ranking, and faculty-to-student ratio. The state of the university’s CDIO implementation section included questions on the disciplines to which CDIO was applied, motives for joining CDIO, CDIO-like experience prior to joining the CDIO Initiative, and participation in CDIO Initiative activities. In this section, the respondents were asked to complete a self-evaluation regarding CDIO standards at the initial state of implementation (defined as when the university joined the CDIO Initiative) and the current state. The third section of the survey aimed to map out the effects of the CDIO implementation on metrics for educational input/output, learning and support processes, and control and resource elements. A number of statements were presented for the respondents, who were asked to rate their agreement with the statement on a 1-10 scale, ranging from “totally disagree” (1) to “totally agree” (10). The fourth section had a similar design to the third, but with statements related to barriers and success factors for CDIO implementation. The final section of the survey comprised free-text response questions regarding customizations of the CDIO framework, on development needs for the CDIO framework, and for the CDIO Initiative. The survey was sent to the CDIO school representatives, “CDIO leaders”, one person at each member institution. The CDIO leaders were recommended to form a small team of faculty from their institution to discuss the questions and their answers before responding. Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. FINDINGS The survey was distributed to 119 potential respondents. 47 responses (39.5%) from 46 universities were received. One university submitted two responses that related to two different programs. The responses were from 22 countries and from 7 CDIO regions. Demographics of participating institutions The majority of the institutions that participated in the survey reported having 15,000 or more students, as noted in Figure 1. Teaching resources are presented in Figure 2, with the largest percentage of institutions having between 2 to 5 graduating students (Bachelor’s or Bachelor’s + Master’s) each year per full-time equivalent (FTE) faculty member. Research intensity is presented in Figure 3 where almost 40 % of the institutions reported having between 1 and 4 graduate students per faculty member. An equal percentage of institutions reported having 1 or fewer graduate students per FTE. > 15,000 students 5,000-15,000 students 1,000-5,000 students 25.5% 53.2% 21.3% Figure 1: Size of Institutions Participating in Survey > 8 5-8 2-5 < 2 11.4% 17.1% 42.9% 28.6% Figure 2: Teaching Resources - Undergraduate Degrees Awarded / Full-Time Equivalent (FTE) Faculty Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. > 4 1-4 0.2-1 < 0.2 19.4% 38.7% 19.4% 22.6% Figure 3: Research Intensity - Graduate Students / Full-Time Equivalent (FTE) Faculty > 6 years 31.9 3-6 years 17 1-3 years 36.2 0-1 years 14.9 0% 10% 20% 30% 40% Percentage of Responses Figure 4: For how long have you applied CDIO? Of the universities completing the survey, the majority have been involved with the CDIO Initiative for 1 to 3 years, as shown in Figure 4. The second largest cluster is for institutions with more than 6 years of experience. This is reflective of the distribution of institutions within the CDIO Initiative. In 2009, there were 35-40 collaborators, and these were institutions with 6 or more years of experience. Given the 119 collaborators at the time the survey was conducted in 2014, the percentage of institutions involved with 6 or more years of experience would be between 30% and 35%, as reflected by the survey participant results in Figure 4. Motives for joining CDIO, prior experience and application to what disciplines The motivating factors for joining CDIO are outlined in Figure 5, with nearly three quarters of the respondents indicating the positive aspects of a systematic approach for education reform and methods for making education more authentic. Approximately one-third reported employer feedback about the lack of certain skills in graduates, and only 10% reported student recruitment, retention, or satisfaction as reasons for applying CDIO. Approximately 60% of the institutions reported having only applied a small number of the concepts of CDIO prior to their programs joining, as shown in Figure 6. Less than 10% reported having already applied the concepts extensively prior to joining. The disciplines to which CDIO is applied by institutions is shown to the left in Figure 7, with Electrical and Mechanical Engineering programs being the most common, followed by Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. Computer Science, Industrial, Civil, Chemical, Aeronautics and Aerospace, and Bioengineering. This data is compared to data published (Yoder, 2014) by the American Society of Engineering Education (ASEE) to the right in Figure 7. The ASEE data shows the percentage of North American graduating Bachelor’s by discipline. Although the ASEE data considers only North American Engineering programs, it does provide an indication on the relative size of disciplines. With this it is possible to better understand why application of CDIO (Bioengineering, for example) is lower than application to another discipline (Mechanical, for example). Poor alumni satisfaction 2.2 Other, specify 6.5 Poor employability of graduates 6.5 Poor student satisfaction 10.9 Poor student retention 10.9 Poor student recruitment 10.9 Accreditation requirements 15.2 Employer complaints of lacking skills among graduates 32.6 Leading universities were doing CDIO 45.7 Internationalization of education 47.8 Needed approach to develop generic skills (teamwork, communication, ethics) 47.8 Community for collaboration 52.2 Wanted to include more design and innovation in education 58.7 Needed a systematic methodology for educational development 71.7 Ambition to make engineering education more authentic 71.7 020406080 Percentage Choosing Reason Figure 5: Motivation for Applying CDIO Little or not at all 21.3 We had one or a few CDIO learning experiences 38.3 We had a good amount of CDIO learning experiences already 31.9 Comprehensively 8.5 Do not know 0 0% 10% 20% 30% 40% Percentage of Responses Figure 6: Extent of applying CDIO prior to joining the CDIO Initiative Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. ASEE - Percentage of NA Graduates by Discipline Engineering mathematics 2.1 0 5 10 15 20 25 30 Applied physics 4.3 Applied physics Non--engineering disciplines, specify 10.6 Non--engineering disciplines, specify CDIO Bioengineering 17 Bioengineering ASEE Aeronautics & aerospace engineering 17 Aeronautics & aerospace engineering Chemical engineering 27.7 Chemical engineering Civil engineering 31.9 Civil engineering Industrial engineering 36.2 Industrial engineering Other engineering disciplines, specifiy 40.4 Other engineering disciplines, specifiy Computer science and engineering 46.8 Computer science and engineering Mechanical engineering 53.2 Mechanical engineering Electrical engineering 53.2 Electrical engineering 0 1020304050 0204060 Percentage Reporting Discipline CDIO - Percentage Reporting Application to Discipline Figure 7: To what disciplines have you applied CDIO? Progress of CDIO implementation The data provided in Table 1 provides an indication of how institutions have self-reported their progress on implementation of the twelve CDIO Standards, self-assessing both their Initial and Current state. The data presented in the Table is an average of all reporting institutions, irrespective of the amount of time spent on implementing CDIO. Figure 8 presents the Table 1 data set, only now segregated based on the number of years since first adopting CDIO. The figure reports the Cumulative Increase, with Increase being defined as the difference between the Current and the Initial self-assessment value for each of the CDIO Standards. The figure reveals that programs continue to move higher on the 0-5-points self-assessment scale for most of the CDIO Standards. One exception is Standard 3 where an uncharacteristic value is reported in the “0-1 years” grouping. Standard 3 refers to Integrated Curriculum, and data for the “0-1 years” category disagrees with what otherwise would be monotonic growth. Examining the survey data, only 7 responses were received in the “0-1 years” grouping for Standard 3, and three of these seven chose “No Response.” Consequently the “0-1 years” data for Standard 3 is viewed as statistically unreliable. Table 1: Progress of CDIO Standards Implementation Standard Initial Current Average Std dev Average Std dev 1 CDIO Context 1.73 1.14 3.55 0.97 2 CDIO Learning outcomes 1.88 1.26 3.77 0.87 3 Integrated curriculum 1.66 1.19 3.37 1.04 4 Introduction to engineering 1.92 1.42 3.78 1.28 5 Design-implement experiences 2.21 1.51 3.88 1.17 6 Engineering workspaces 1.97 1.18 3.34 0.96 7 Integrated learning experiences 1.74 1.36 3.29 1.11 8 Active learning 1.65 1.07 3.15 0.99 9 Enhancement of faculty engineering competence 1.36 1.19 2.64 1.18 10 Enhancement of faculty teaching competence 1.64 1.11 2.95 0.99 11 CDIO skills learning assessment 1.58 1.16 3.05 0.88 12 Program evaluation 1.23 1.12 2.69 1.26 Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. Other exceptions to growth with time exist for Standard 9 (Enhancement of Faculty Competence) and Standard 10 (Enhancement of Faculty Teaching Competence) where the vertical distance between lines for these Standards do not increase significantly with years of CDIO application. This is further explored in Figure 9 where the Cumulative Increase in Standards 9 and 10 (plotted on the left) reveal modest growth with time while the remaining ten Standards (shown on the right) reveal more substantial growth. Standards 9 and 10 refer to a change in behavior of faculty members, something that is more difficult to influence than Curriculum (Standards 1-5, 7-8), Evaluation (Standards 11-12), or Workspaces (Standard 6). A recent study conducted in the UK (Graham, 2015) reached a similar conclusion. Graham found that despite a growing recognition of the importance of teaching quality by university leaders, faculty are still not convinced that teaching achievements above an acceptable level will be counted in promotion cases. As long as this perception dominates, the likelihood for significant improvement of faculty teaching skills will be low. 25 12 Program Evaluation 20 11 Learning Assessment 10 Enh of Faculty Teaching Competence 9 Enhancement of Faculty Competence 8 Active Learning 15 7 Integrated Learning Experiences 6 Engineering Workspaces 10 5 Design-Implement Experiences Cumulative Increase 4 Introduction to Engineering 3 Integrated Curriculum 5 2 Learning Outcomes 1 The Context 0 0-1 years 1-3 years 3-6 years > 6 years Years Since Adopting CDIO Figure 8: Cumulative Increase in CDIO Standards by Year Since Adoption 25 25 20 20 12 Program Evaluation 11 Learning Assessment 8 Active Learning 15 15 7 Integrated Learning Experiences 6 Engineering Workspaces 10 10 5 Design-Implement Experiences Cumulative Increase Cumulative Increase 4 Introduction to Engineering 3 Integrated Curriculum 5 5 2 Learning Outcomes 10 Enh of Faculty Teaching Competence 1 The Context 9 Enhancement of Faculty Competence 0 0 0-1 years 1-3 years 3-6 years > 6 years 0-1 years 1-3 years 3-6 years > 6 years Years Since Adopting CDIO Years Since Adopting CDIO Figure 9: Cumulative Increase in Select CDIO Standards by Year Since Adoption Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. The average increase by Standard for all programs is shown in Figure 10. The graph shows that the increase (Current State – Initial State) is greatest for Standard 2 (Learning Outcomes) and lowest for Standard 9 (Enhancement of Faculty Competence). In order to determine if the increase was related to the Initial State, a stacked bar graph was also considered, as shown in Figure 11. Figure 11 shows that Standard 5 (Design-Implement Experiences) is rated highest overall; however, the increase in this Standard is only fifth highest (Figure 10), perhaps due to the fact that the Initial State for Standard 5 was already the largest to begin with, leaving less room for incremental improvement. Despite these differences in relative improvement between standards, however, the survey respondents’ universities/program have clearly made significant advancements in all CDIO standards. It also appears that survey respondents have adapted CDIO more as a whole than by cherry-picking specific areas. These data are similar to Gray’s (2008) findings. Gray also found that Standard 5 (Design- Implement Experiences) tended to have the highest rating amongst new universities, and that Standard 2 (Learning Outcomes) had a high increase. However, in Gray’s study the highest gain was for Standard 12 (Program evaluation). 9 Enhanced Faculty Competence 10 Enhanced Faculty Teaching 6 Engineering Workspaces 12 Program Evaluation 11 Learning Assessment 8 Active Learning 7 Integrated Learning Experiences 5 Design Implement Experiences 3 Integrated Curriculum 1 The Context 4 Introduction to Engineering 2 Learning Outcomes 0.0 0.2 0.4 1.2 1.4 1.6 1.8 2.0 Current State - Initial State Figure 10: Increase by CDIO Standard Increase Initial State 9 Enhanced Faculty Competence 10 Enhanced Faculty Teaching 6 Engineering Workspaces 12 Program Evaluation 11 Learning Assessment 8 Active Learning 7 Integrated Learning Experiences 5 Design-Imp Experiences 3 Integrated Curriculum 1 The Context 4 Introduction to Engineering 2 Learning Outcomes 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Self Evaluation Figure 11: Current State in CDIO Standard = Initial State + Increase (Largest Increase – Std 2; Smallest Increase – Std 9) Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. CDIO implementation effects The survey further aimed to investigate how the survey participants rated the effects that CDIO implementation had with reference to specific statements indicating positive effects on educational quality, such as improved graduate knowledge and skills, improved alumni satisfaction, or graduate employability. A few statements of negative effects such as increased costs and decreased math and science knowledge were also included. The sources for the statements included Crawley et al. (2014) and the EUR-ACE framework standards (ENAEE, 2008). Survey participants were asked to respond to each statement on a ten point scale, ranging from “totally disagree” to “totally agree,” including a provision for “cannot assess.” Table 2 shows the statements and their ratings. It can be observed that there is very strong agreement for statements related to CDIO’s main goals for learning (improved conceive- design-implement-operate, personal and interpersonal skills). Further, there is also strong agreement for statements related to external recognition (accreditation, government awards) and collaboration with other universities. We note that the statements for which there is strongest agreement are well aligned with the motives for applying CDIO (Figure 5). One respondent commented: ““CDIO gave a framework for our students to develop these skills in a 'structured' manner, and more holistically. Prior to CDIO, student experiences were more skewed towards 'implementing and operating'. Now, with inclusion of design thinking, students develop good conceiving skills as well.” Further, there is strong agreement related to student and alumni satisfaction and recognition (course rating, alumni satisfaction and employability “We have positive signals from employers that we have strengthened our student's employability”, student awards “Students are doing better in student competition teams such as FSAE, Mini Baja, Solar Car, Solar Decathlon, Petro Bowl, iGem, etc.”). However, it can also be noted that relatively many participants replied “cannot assess” to statements related to alumni. It seems that many are lacking a systematic mechanism for surveying their alumni. Statements with a tendency to neutral agreement include student recruitment, retention and higher pay for graduates. Such goals are often common to education reform efforts (Graham, 2012), including CDIO (Crawley et al., 2014), but the effects of CDIO implementation on these areas seem to be minor or difficult to discern. This finding confirms results from Ng’s (2014) follow up of Singapore Polytechnic’s chemical engineering program. A number of statements expressed possible trade-offs or risks related to CDIO implementation (“CDIO implementation required significant investments in education infrastructure”, “CDIO implementation has led to increased operating costs”, “Graduates have less knowledge of math and science”). These statements have been valued as neutral or weak, i.e. there is little support for these statements amongst the survey participants: “Our core engineering science and maths is still present to the same degree as it always has been. Contextualizing the work through projects should help to embed and allow students to apply core knowledge.” However, there is a relatively high spread, indicating that in some cases CDIO has been associated with both significant investments and higher operating costs. Nevertheless, even considering the spread, there seems to be little support for the statement that CDIO implementation had led to less knowledge of mathematics and science. Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. Table 2: Effects on education input, resources and output Statement Avg Std dev Strong agreement (average > 6.5) Graduates have improved conceive-design-implement-operate skills 7.9 1.8 CDIO implementation has supported accreditation 7.9 1.7 Graduates have improved interpersonal skills 7.8 1.6 We have received recognition for high quality in education (for example awards from 7.5 2.1 government agencies) We have increased collaboration with other universities for educational development 7.5 1.9 Graduates have improved personal skills 7.5 1.6 We have an increased number of published papers on educational development 7.2 2.3 Faculty teaching competence has improved 7.1 1.8 Quality of final degree reports/capstone design projects have improved 7.0 1.9 Alumni satisfaction has increased 6.9 1.7 Course satisfaction ratings have improved 6.9 1.9 Graduate employability has improved 6.6 2.1 Our graduates have received more awards (for example prizes for projects or won 6.6 2.4 student competitions) Neutral agreement (average 3.5 – 6.5) Faculty engineering professional competence has improved 6.5 1.8 Student recruitment has improved 6.4 1.9 Student retention has improved 6.3 2.0 CDIO implementation required significant investments in education infrastructure 6.2 2.1 More alumni are starting new companies 5.8 1.6 Graduates entry salaries are higher than for nearby universities who have not 5.8 1.0 implemented CDIO CDIO implementation has led to increased operating costs 5.5 2.5 Weak agreement (average < 3.5) Graduates have less knowledge of math and science 3.4 2.2 We can note that some statements have strong agreement on effects, despite typically low amount of increase in terms of CDIO Standards (Figure 10): “CDIO implementation has supported accreditation” is ranked relatively high at 7.9, yet this is the area that shows the fourth least amount of increase in terms of the CDIO Standards, as reflected by Standard 12 – Program Evaluation. We can note that “Faculty teaching competence has improved” is ranked relatively high at 7.1, yet this is the area that shows the second least amount of increase in terms of the CDIO Standards, as reflected by Standard 10 – Enhancement of Faculty Teaching Competence in Figure 10. “Faculty engineering professional competence has improved” is rated somewhat lower at 6.5, yet this is the CDIO Standard that shows the smallest increase – Standard 9 – Enhancement of Faculty Competence. This can be interpreted as that although it seems difficult to achieve a high self-evaluation rating with respect to these CDIO Standards, even moderate increases may still have noticeable positive effects on education quality. Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. Table 3: Barriers and success factors for CDIO implementation Statement Avg Std dev Strong agreement (average > 6.5) CDIO is well aligned with the vision and strategy of our department/university 8.9 1.5 University management strongly supported our CDIO implementation 8.1 2.0 The CDIO implementation was associated with higher ambitions for our education 8.1 1.5 We had clear visions and goals for what we wanted to achieve by the CDIO 7.9 1.6 implementation It was easy to customize the CDIO framework to fit our local context 7.3 2.2 If the main CDIO proponent at your university was to retire tomorrow, the changes 7.2 2.1 that have been made to date would remain five years from now CDIO has created attention for education in our university 7.2 2.3 We had sufficient financial resources to implement CDIO 6.6 2.4 Neutral agreement (average 3.5 – 6.5) Faculty were incentivized and recognized for CDIO implementation efforts 5.8 2.5 We measured the impact of our CDIO implementation with suitable indicators 5.7 2.1 Faculty teaching competence was a barrier to CDIO implementation 5.2 2.1 Faculty were resistant to CDIO 4.9 1.9 Faculty engineering professional competence was a barrier to CDIO implementation 4.9 2.4 Weak agreement (average < 3.5) None -- Barriers and success factors for CDIO implementation One part of the survey aimed to investigate how the survey participants rated the importance of published success factors and barriers for sustainable education reform, in context of their CDIO implementation effort. The sources for the statements included Crawley et al. (2014), Graham (2012) and Malmqvist et al. (2010). Table 3 shows the statements and their ratings. It is evident that the survey participants’ CDIO projects generally fit well with university visions and strategies, had strong support from management, had clear goals and visions and were associated with higher ambitions for their education. Table 3 suggests that these higher ambitions would tend to be related to specific goals for learning (CDIO skills, interpersonal skills) and/or external recognition (accreditation, government awards). All these statements have a very strong agreement and relatively low spread. The responses further show that the CDIO framework can be purposefully adapted to local contexts, are embodied by more than single individuals and create attention for education at universities. However, the spread of the responses indicate a higher fraction of neutral agreements with these statements. The sufficiency of financial resources is rated just above the border to neutral agreement and with a relatively higher spread. The respondents free-text comments range from “No extra resources allocated” to “We have special funds for CDIO” or “… Significant external funding for CDIO.” It is apparent that the available financial resources have varied significantly. Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. However, even schools with low or no additional funding report positive effects of CDIO implementation. Finally, it can be noted that statements related to faculty resistance and competence were neutrally rated and thus not experienced as strong barriers to the CDIO reform. Development needs for the CDIO Framework A free response question asked “What development or change needs do you see for the CDIO framework”. The answers can roughly be grouped into three categories: renewal of the CDIO vision, specific revisions of the CDIO framework and CDIO implementation guidelines. Renewal of the CDIO vision. The vision for a CDIO-based education (Crawley et al., 2014) proposes that an engineering education should stress the fundamentals while set in the context of conceiving-design-implement-operation. Further, that salient features include stakeholder-based program goals, an integrated approach to learning disciplinary knowledge and professional skills, and a multitude of design-implement learning experiences. However, one respondent argued that CDIO could be more active in disseminating novel ideas and concepts, such as on-line education, virtual and remote labs. Another respondent suggested that CDIO “should develop a clear vision on what knowledge and skills the engineer of the future (2030) will need. What are the main differences from the engineer of today? How can the CDIO framework be adapted in order to meet these new demands?” Specific revisions of the CDIO framework. Suggestions in this category include revisions of the rubrics for CDIO standards, self-evaluation, stronger attention to internal motivation as a key factor for learning and for successful practice, and explicit consideration of gender and sexual diversity. It was also proposed to develop and specify CDIO framework components for Master and PhD programs. CDIO implementation guidelines. A number of respondents requested guidelines or instructions for CDIO implementation, including “how to gather evidence of effects at different levels”, “a step-by-step how-to-implement book”, “more guidance of teaching professional skills”. It seems that the implementation advice that is available on the CDIO website is either not fully adequate or too difficult to access. Development needs for the CDIO Initiative A free response question asked “What development or change needs do you see for the CDIO Initiative?”. The answers can roughly be grouped into internal collaboration and external collaboration. Internal collaboration. Many respondents expressed an interest in expanding collaborative mechanisms such as student and faculty mobility and joint student projects. Some respondents further suggested activities targeted at different CDIO experience levels (novice, intermediate, experienced): “enhanced focus on outcomes for experienced members”, “it would be nice to provide certification of compliance with the CDIO standards”, “differentiate between probationary and proper CDIO members”. External collaboration. A few respondents suggested that the CDIO Initiative should increase its external collaboration efforts, including liaising more closely with other professional engineering education associations such as ASEE, CEEA, and SEFI, and with Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. national and regional university management networks. An international journal of CDIO was also proposed. Finally, one respondent pointed out the potential tension between scholarship and practice in engineering education development, arguing that the CDIO community should lean towards the latter: “I would like to see the organization shift into more of a sharing and how-to mode. Focus on practice of engineering rather than scholarship”. CONCLUSIONS The CDIO status survey aimed to investigate the use of CDIO, its progress, benefits and barriers/success factors for implementation. About 40 % of the CDIO Initiative members participated in the survey. The disciplines in which CDIO is most commonly applied are mechanical, electrical and computer engineering. However, many CDIO members also report CDIO applications in industrial, civil and chemical engineering. The participants also listed a large number of other engineering disciplines in which CDIO had been applied. The most common motives for applying CDIO were ambitions to make engineering education more authentic, needs for a systematic methodology for educational development, wishes to include more design and innovation in the education and to find an international community to support education knowledge sharing and collaboration. The survey results show that the participant universities are successfully improving graduate conceive-design-implement-operate, personal, and interpersonal skills. The participants further confirm that they have obtained external recognition for educational quality and established collaboration with other universities. The participant’s responses provide little support for statements related to negative effects on learning or on resource consumption for education. Intended CDIO implementation effects also include improved student recruitment, retention and graduate salaries. However, the survey shows a neutral agreement with these statements. CDIO implementations viewed as university development projects seem to be strongly supported by university management, aligning well with university visions and strategies, and being sufficiently funded. The survey was not distributed to individual faculty members; hence individual faculty members’ views on CDIO are not studied. However, it can be concluded that the university leaders that responded did not experience (the potential) faculty resistance as a barrier to successful implementation. Identified development needs for the CDIO framework include a renewal of the educational vision considering the needs of 2030 and the opportunities of digital education tools, revisions of certain CDIO standards rubrics and more specific implementation guidelines. Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. REFERENCES Cheah, S.-M., Phua, S. T. Ng C. H. T. (2013) The Chemical Engineering CDIO Experience after 5 Years of Implementation, Proceedings of 2013 International CDIO Conference, Cambridge, MA, USA. Crawley, E., Malmqvist, J., Östlund, S., Brodeur, D, Edström, K. (2014) Rethinking Engineering Education – The CDIO Approach, 2nd ed, Springer Verlag, New York. Edvardsson Stiwne, E., Jungert, T. (2007) Engineering Students Experiences of The Transition From Study To Work, Proceedings of 2007 International CDIO Conference, Cambridge, MA, USA. ENAEE. (2008) EUR-ACE Framework Standards for the Accreditation of Engineering Programmes, http://www.enaee.eu/wp-content/uploads/2012/01/EUR-ACE_Framework-Standards_2008-11- 0511.pdf, accessed on February 25, 2015. Graham, R. (2012) Achieving Excellence in Engineering Education: The Ingredients of Successful Change, The Royal Academy of Engineering. Available online at www.raeng.org.uk/change, accessed on April 24, 2015. Graham, R. (2015) Does Teaching Advance Your Academic Career? Perspectives of Promotion Procedures in UK Higher Education, Royal Academy of Engineering, London, UK, www.raeng.org.uk/teachingpromotion, accessed on April 24, 2015. Gray, P. (2009) CDIO Collaborator Survey 2008. Proceedings of 2009 International CDIO Conference, Singapore. Jungert, T. (2006) Students' Experiences of their 2nd, 3rd and 4th Years in an Engineering Program: Results Based on Questionnaires, Proceedings of 2006 International CDIO Conference, Linköping, Sweden. Malmqvist, J., Bankel, J., Enelund, M., Gustafsson, G., Knutson Wedel, M. (2010) Ten Years of CDIO – Experiences from a Long-Term Education Development Process, Proceedings of 2010 International CDIO Conference, Montreal, Canada. Martins, A., António Costa, A., Pinto Ferreira, E., Rocha, J. (2013) Assessing 6 Years of CDIO in a Computer Engineering Program, Proceedings of 2013 International CDIO Conference, Cambridge, MA, USA. Ng C. H. T. (2014) Evaluating Effects of CDIO Implementation On Diploma In Chemical Engineering, Proceedings of 2014 International CDIO Conference, Barcelona, Spain. Nguyen, G. N., Tran, N. T., Nguyen, T. T, Nguyen, D. M. (2014) The Benefits of CDIO for ABET Preparation from a Hands-on Study in Vietnam, Proceedings of 2014 International CDIO Conference, Barcelona, Spain. Yoder, B. L. (2014) Engineering by the Numbers. Washington: American Society for Engineering Education. Available online at www.asee.org/colleges, accessed on April 24, 2015. Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. BIOGRAPHICAL INFORMATION Johan Malmqvist is a Professor in Product Development and Dean of Education at Chalmers University of Technology, Gothenburg, Sweden. His current research focuses on information management in the product development process (PLM) and on curriculum development methodology. Ron Hugo is Professor of Mechanical and Manufacturing Engineering and Associate Dean (Teaching & Learning) at the University of Calgary. He is also the holder of the Engineering Education Innovation Chair in the Schulich School of Engineering. His research interests are in the areas of experimental fluid dynamics, energy systems, and engineering education. Malin Kjellberg is a Lecturer in Engineering Education Research at Chalmers University of Technology, Gothenburg, Sweden. Her research focuses on project-based learning and design-build-test projects. Corresponding author Professor Johan Malmqvist Chalmers University of Technology Department of Product and Production This work is licensed under a Creative Development Commons Attribution-NonCommercial- Gothenburg, SWEDEN, SE-41296 NoDerivs 3.0 Unported License. +46 31 772 1382 [email protected] Proceedings of the 11th International CDIO Conference, Chengdu University of Information Technology, Chengdu, Sichuan, P.R. China, June 8-11, 2015. Paper I Edström, K., & Hellström, P.-E. Improving student learning in STEM education: Promoting a deep approach to problem-solving. Manuscript in preparation. Improving student learning in STEM education: promoting a deep approach to problem-solving Kristina Edström and Per-Erik Hellström KTH Royal Institute of Technology, Stockholm, Sweden Abstract This paper addresses educational practice related to problem-solving within STEM education. A conceptual framework is shaped by conceptualising problem-solving first as an educational aim, then as a learning activity. Five principles for purposeful active learning are derived. Through this theoretical lens we investigate an active learning method called student-led exercises. In this activity students are randomly selected to present solutions to given problems, requiring them to solve the problems in advance and prepare for presenting their solutions. Drawing on the conceptual framework and informed by course results and qualitative data in the form of student interviews and teacher experiences, we analyse the teaching method. One conclusion is to challenge the learning value of activities based on teacher demonstrations of problem-solving. We suggest that student-led exercises are a cost-effective intervention, improving learning while affording more stimulating roles to both students and teachers. Keywords problem-solving, Feisel-Schmitz technical taxonomy, deep and surface procedural approaches to learning, active learning, student-led exercises Introduction Problem-solving in STEM education Lectures have long been under debate, generally based on the critique that they consist of “continuous exposition by the teacher” (Bligh 2000) and are ineffective for learning. Hence, suggestions for improving lectures using active learning include, for instance, integration of exercises (Gibbs 1992, 46-58), peer instruction (Mazur 1997, Crouch and Mazur 2001) and other flipped classroom strategies (Abeysekera and Dawson 2015). In many STEM subjects, teaching devoted to problem solving is also common. Unlike lectures, however, there is no universal notion of how problem- solving is taught. Unless it is entirely left to the students outside scheduled hours, problem-solving practice is integrated into lectures or addressed in special learning activities, often in smaller groups. In the English language sessions have names like exercise classes, recitations, tutorials, sections, or lessons, and in other languages we find the French travaux dirigés, German Übung, Norwegian øving, and Swedish Paper I övning. As an example of the extent, the authors’ technical university annually schedules about 42,000 hours of lectures and 30,000 hours exercise sessions and lessons, a considerable investment of teaching resources. The educational approaches in problem-solving sessions can vary considerably, likely with patterns of differences across subjects, countries, and institutions. The format ranges from mainly teacher-active, e.g. mini-lectures where a teacher solves problems on the board, to student-active, e.g. individual or group problem-solving practice, with teacher support available. We find cases critiquing teacher-centric formats. Redish and Steinberg (1999) identify small group recitations as one of three standard components of introductory physics courses, besides large lectures and laboratories. They say: “Recitations are often presented by teaching assistants (TAs). They may answer student questions, but the activity tends to have the TA modeling solutions to the problem on the board. Students rarely participate actively.” Similarly, in a general chemistry course, a faculty member answers questions and works example problems (Mahalingam, Schaefer, and Morlino 2008). Attendance is poor, and: “many of those who attended these recitation sessions passively copied down the solved problems and therefore did not get the benefit from going through the thought process required to set up solutions to suggested problems.” These descriptions of recitations as passive affairs for the students are familiar to us, and we aim here to consider possible advantages of active learning. This paper addresses the following questions in the context of STEM education: How can we understand problem-solving competence as a learning outcome, and as a learning activity? How can we interpret experiences of active learning methods based on problem- solving? The paper is structured as follows. First, we outline a conceptual framework for considering problem-solving as an aim of education, and as a method for learning. This is operationalized as principles for active learning. We then describe an active learning method called student-led exercises, and its implementation in two settings. Finally, these experiences are analysed through our conceptual lens. Learning problem-solving, and learning through problem- solving The quality of learning outcomes – Taxonomy To inform course design it is useful to consider the nature of quality in learning. The Feisel-Schmitz technical taxonomy (Feisel 1986) is particularly salient with regards to problem-solving. See Table 1. Judge To be able to critically evaluate multiple solutions and select an optimum solution Solve Characterize, analyse, and synthesize to model a system (make appropriate assumptions) Explain Be able to state the process/outcome/concept in their own words Compute Follow rules and procedures (substitute quantities correctly into equations and arrive at a correct result, ‘plug & chug’) Paper I Define State the definition of the concept or describe in a qualitative or quantitative manner Table 1. Feisel-Schmitz technical taxonomy (Feisel 1986). The taxonomy captures an important distinction between problem-solving with or without understanding. The lowest level, Define, basically corresponds to repeating course content. Next, Compute means following known procedures to find answers to standard problems, but limited to pattern-matching problems with solutions, this is essentially problem-solving without understanding. Therefore, Define and Compute do not, on their own, amount to good quality learning. As the taxonomy is hierarchically inclusive, any level also comprises the levels below. Thus, Explain subsumes both Define and Compute. Now, students can recall knowledge but also explain it, and solving typical problems they can also explain procedures and results. The next level, Solve, implies handling problems through some measure of own modelling. In other words, this is problem-solving with understanding. The Explain and Solve levels imply the ability to make connections within the material and with previous learning, creating a coherent and meaningful structure for combining knowledge to attack also previously unseen problems. Finally, Judge is evaluating solutions by setting relevant criteria, and proposing informed decisions. The quality of learning processes – Approaches to learning The approaches to learning originate in experiments where Marton and Säljö gave students a text to read for a test. They observed that depending on their intention, students engaged differently with the task: collecting facts in anticipation of the test (surface-level processing), and trying to grasp the meaning (deep-level processing). Learning was strongly impacted: students engaging on the surface of the text did not understand its message (Marton and Säljö 1976a). Further, Marton and Säljö (1976b) showed that the level of engagement could to some extent be influenced, as students approached the task differently depending on their expectations what the test requires. These ideas were expanded to naturalistic settings like university courses, identifying the deep and surface approaches to learning (Marton, Hounsell, and Entwistle 1984), defined by the intention: to pass the course or to understand. Seemingly, students have the capacity for both, as they can use different approaches in different courses (Gibbs 1992, 8-9), and in relation to different assessment tasks within a course (Edström et al. 2003, 10). Given the strong correlation with the quality of learning outcomes, the deep and surface approaches are often used to operationalize good and bad learning processes. This underpins many normative discussions on learning environments (Gibbs 1992, Prosser and Trigwell 1999). Notably, it is the conceptual cornerstone of constructive alignment (Biggs and Tang 2011), implying that learning activities and assessment systems should be designed in a purposive relation to the intended learning outcomes, so that, crucially, the course promotes a deep approach to learning. A surface approach is discouraged when learning activities do not invite it, and assessment demands performances of understanding less likely achieved through a surface approach. Case and Marshall (2004) make an important contribution by identifying two additional approaches specifically related to problem-solving. In both cases the study strategy is to work through problems, but intentions differ. For students adopting a Paper I procedural deep approach the intention is ultimately understanding. In contrast, students adopting a procedural surface approach: “attempted to memorize standard solution methods, in order to be able to apply these to similar problems in the test or examination. The emphasis lay on working through as many problems as possible, and remembering the solution methods given in the memoranda. Some of these students attempted to find one ideal method for solving a particular type of problem. Solving problems was seen as involving a search for the appropriate equation(s) and substituting values to get an answer.” Although this strategy often entails very hard work collecting and studying given solutions, it produces meagre results – both in terms of understanding and passing exams. Jonassen (2014) pinpoints why: “Learners too often fail to recall or reuse examples appropriately because their retrieval is based on a comparison of the surface features of the examples with the target problem, not their structural features, whereas experienced problem solvers represent problems in terms of their principles, emphasizing conceptual understanding.” Case and Marschall (2004) placed the procedural approaches as intermediate categories on a continuum between the classic surface and deep approach. They warn: “Given the strong focus in engineering and science courses on the ability to solve problems (often, as evidenced in assessment, explicitly valued more than understanding concepts), [promoting a deep procedural approach] might be a sensible strategy for helping students to succeed. However, the danger lies in that students might thereafter struggle to make the appropriate adjustment to a conceptual deep approach when required.” (p. 613). In our view this reasoning contributes to a hierarchy, in which theory comes before and above practice, and it positioning problems merely as exercises for illustrating concepts. Our stance is different, because we insist that problem-solving has a rightful place also as an aim in its own right. Especially so in STEM education, it is important not only what you understand but also what you can do with your understanding. Conceptual understanding is necessary but not sufficient for problem-solving, as students also need to develop the associated skills, strategies and habits. This includes interpreting a problem in context, shaping a plan for attack, making assumptions and finding information, working through the solution process, and appropriately evaluating and presenting the method and its results. Problem-solving as a learning activity is therefore not only a means to reach conceptual understanding; it is also about learning to solve problems. This critique leads us to suggest a reinterpretation of the deep procedural and surface procedural approaches, by placing them not between, but in parallel with the classic deep and surface approaches, see Figure 1. The implication is that the deep/surface dichotomy is essentially the same, it just appears differently in relation to what students are learning. INTENTION: Passing the test Understanding Concepts Surface approach Deep approach and theory CONTEXT: Problem Procedural surface approach Procedural deep approach solving Paper I Figure 1. Approaches to learning (our adaptation of Case and Marshall 2004). Based on Perry’s (1998) classic work, Gainsburg (2015) studied students epistemological views on mathematical methods in engineering. She identified the dualistic view, where students see their task as matching given problems with correct (given) procedures, replicating the steps of the correct solution. This involves little critical appraisal of the results, or reflection on what one learned from a problem. Instead, the culmination lies in satisfying the teacher or seeking confirmation in the answer key. In the integrating view, students try to understand concepts and procedures. They may use e.g. visualisation and dimensional analysis, though more as tactics for reaching correct answers than strategies for understanding through connection to the real physical problem. Finally, the relativistic view and the skeptical reverence are about making sense, even if it delays the route to the solution. Students increasingly connect course problems with the real problems they represent, and with the nature of problems and processes used in engineering practice. We suggest that these views could be seen as the epistemological underpinnings of Case and Marshall’s approaches. The surface procedural approach corresponds to a dualistic view, and the deep procedural approach to the relativistic view and sceptical reverence. The integrating view may be a borderline case. Although clearly associated with successful study strategies in educational settings, it does not accommodate the full aims of professional education. In summary, while the quality of the learning process is operationalized by approaches to learning, the taxonomy levels refer to the quality of learning outcomes. The relationship between process and outcome is sometimes confused by inappropriate conflation of the concepts, i.e. assuming that the surface approach gives learning outcomes on the lower taxonomy levels, while the deep approach will produce the higher levels. Since a surface approach correlates with poorly structured and short-lived knowledge, it is never appropriate – regardless of the intended level of learning outcomes. When designing teaching and learning activities, the aim is always to promote the deep approaches. Principles for active learning Seeing approaches to learning as responses to the educational context entails implications for educational practice. Learning activities involving problem-solving should influence students to engage with the problems on a higher level. Study strategies associated with a surface procedural approach, e.g. ‘collecting’ correct solutions, should be discouraged. When considering suitable learning activities, there is ample support for active learning (Freeman et al. 2014) and for stimulating retrieval of knowledge (Karpicke 2012). Active problem-solving practice should model the higher-level engagement associated with a deep procedural approach, e.g. practicing joint reasoning and trouble-shooting, and exploring variation in understanding and perspectives. There are also opportunities for practicing various competences, e.g. communication and collaboration (Edström and Kolmos 2014). While cramming towards the end of the course correlates to a surface approach to learning, regular coursework is better associated with a deep approach (Gibbs 1992, 1-11, Edström et al. 2003). In collaborative settings, inadequate study strategies are exposed, and students can be Paper I supported by the social context to try out more effective approaches. Smith et al. (2005) recommend: “To maximize students’ achievement, especially when they are studying conceptually complex and content-dense materials, instructors should not allow them to remain passive while they are learning. One way to get students more actively involved is to structure cooperative interaction into classes, getting them to teach course material to one another and to dig below superficial levels of understanding of the material being taught.” Below, we have collated a list of relevant principles, by synthesising several overlapping accounts, emphasising characteristics of purposeful active learning, rich with interaction and feedback (Gibbs 1992, 1999, Nicol and Macfarlane-Dick 2006, Boud and Molloy 2013b, a, Chickering and Gamson 1987). Good teaching will I. support students to spend sufficient time-on-task and distribute their effort appropriately (Attention) II. generate appropriate learning activity (High-level engagement) III. communicate high and clear expectations, and provide opportunities for reflection and feedback, to help students improve their work and develop their judgement (Purpose) IV. encourage interaction and dialogue with teachers and peers, modelling appropriate approaches to learning (Interaction) V. provide information to teachers that is used to develop teaching (Continuous improvement) Below we describe a format for problem-solving sessions, aiming to promote a deep procedural approach. Using these principles to frame our analysis, we will reflect on the results and experiences from two implementations. Student-led exercises Activity design Student-led exercises is a teaching method based on active problem-solving. In contrast to ‘traditional’ activity formats where a teacher solves problems on the board, students are randomly picked to present their own solutions, prepared beforehand. At course start, students are given a problem set, to prepare for each weekly session. The sessions proceed as follows: • On arrival, students tick on a list which problems they are prepared to present. • The teacher starts the session by randomly picking a student, who presents the first problem on the board. • A classroom discussion follows, considering the presented solution and any alternatives, and reflecting on problem-solving strategies. • Another student is picked for the next problem, and so on. • After the session, model solutions are posted online (optional). In principle, attendance is voluntary, however it is a formal course requirement to tick (say) two thirds of the problems, and thus students need to attend a certain number of sessions. The quality of presentations is not graded, as the activity is purely formative, but students are required to demonstrate an honest effort. Should the solution be Paper I inadequate, the teacher assists the student in leading a classroom discussion to get help in a satisfactory treatment of the problem. If the presenter is obviously unprepared, all ticks for this session are removed from his or her record. Teaching in student-led exercises The instructor must prepare the problems before course start, normally with model solutions. They are however created from scratch only once; later it is sufficient to replace some problems and dress others in new clothes. Refreshing the problems prevents old notes from compromising learning in the preparation stage, and creates opportunities for improving the learning value of the problems. Appropriate problems should be aligned to the intended learning outcomes, reflecting key aspects of the course, and addressing difficulties and misconceptions (Kember and McNaught 2007). Since the format may be novel to the students, it needs explaining – mixing enthusiasm, firmness, and empathy. At the sessions, the teacher’s role is to facilitate discussions, always encouraging students to discuss among themselves. The teacher avoids taking over, so when students miss an important point, the teacher can add it, but always last, and longer explanations go in the next lecture where all groups can benefit. Limiting group size allows most students to participate actively, and increases the chance of being picked. Ensuring a positive and safe atmosphere is crucial, making the activity a friendly way to learn together. For instance, making errors creates valuable moments that several students can learn from. This means modelling a helpful attitude and ensuring the same climate between the students. The class is invited to help: - Did anyone else get stuck here? What did you do to proceed? - Can anyone give a little hint here? The discussion should challenge students who see problem-solving as a hunt for the right answer, prodding them to approach the problems with a broader mind-set. Suitable prompts depend on what the particular problem can illuminate: - Did anyone have a different solution? Please tell us! - I notice that only eight of you ticked this problem. What were the difficulties? - What if [condition] had been [this] instead? - Why was this problem different; why can’t we use the same strategy as last week? Results from two cases Case 1: Semiconductor Devices at KTH Semiconductor Devices is a course in the third year of an engineering program. Teaching lasts seven weeks, each with two two-hour lectures and one two-hour exercise session, followed by a written exam in week eight. Students learn to derive and calculate currents inside a semiconductor device, analyse the internal state of the charge distribution, the electric field, and the current density – and relate this to how semiconductor devices are used in applications. Instructors (among them the second author) were dissatisfied with students’ ability to solve problems, also reflected in a pass rate around 60%. Correcting exams was frustrating and many students had to come back for new attempts, some repeatedly. The instructors felt that students Paper I needed better engagement in solving problems on their own, but previous interventions were largely unsuccessful. For instance, quizzes mainly rewarded strong students with bonus points, thus spending considerable teacher time without helping those who most needed support. Since the course already had teacher-led exercise classes, there were slots in the schedule where students were divided into parallel groups of about twenty. Only the interaction changed to randomly picking students for presenting their solutions. Starting the second week, there were in total six student-led exercises sessions, each with six problems. Students were allowed, even encouraged, to collaborate in preparations. Ticking 20 of the 36 problems was required. As a result, the exam results improved. The pass rate increased from previously about 60 per cent, to 78, 70, 83, 86, and 75 per cent respectively, during five years when the course design was otherwise held constant. In evaluations, students were positive about student-led exercises, typically 4,2 on a Likert scale from 1 to 5, with frequent comments that it supported effective learning. The instructors enjoyed discussing the subject on a much higher level, and students surprised them as good presenters. Trivial questions, e.g. about notation, disappeared, as did questions revealing procrastination when exam-time approached. Students also seemed more engaged in lectures. Teachers could adapt the pace of teaching, penetrating common difficulties when needed, or moving on knowing that students already ‘got it’. Increasingly, they learned to design problems addressing the most important issues, difficulties and misconceptions. The pre-course preparations increased teacher workload, but this was more than compensated when correcting the exam. Since fewer students needed re-sits, there were fewer exams to correct. Further, students seem less prone to desperate scribbling in hope for half a point here and there. With fewer poor exams to correct, the most mentally taxing work diminished. After two years of improved results, the course evaluation method was altered to investigate student feedback more deeply (Edström 2012). After the course, semi- structured interviews (Kvale 2008) were held with six students representing various grades. Interviews lasted twenty minutes and were recorded and transcribed. The qualitative data offered surprising insights into the quantity and nature of the study activity. Table 2 shows excerpts from three interviews, numbered for reference in the analysis below. Table 2. Excerpts from student interviews. Student How did you prepare? A – I tried to do as many problems as possible. I didn’t just want to do the minimum, but all of them because it is good for the exam. [Laughter] Well, for each session… I don’t know but at least six hours maybe. (A1) How was your work distributed during the course? – Quite evenly, with a little extra just before the exam, to really dive into stuff. And it was quite a lot of time [in total]. (A2) Student How did you prepare? B – We sat in a group and did the six problems, helping each other. Then the evening before I read through to get a good grip, and then I ticked them. Well, we sat maybe… how long could it have been? Five hours in the group and two hours on my own. (B1) When you study in groups, what do you really do? – We have a whiteboard [in a vacant classroom]. If we are three we do one problem each Paper I and stand together discussing it at the board. (B2) How was the exercise session, with other students up on the board? – Well, you knew the problems, so it was not extremely interesting, but at the same time you could follow extremely well. I guess it was quite good. You saw straight away if your solution was wrong. (B3) If you compare with normal [teacher-led] exercise classes? – Oh, nothing at all, I just go there. You mean normal exercise sessions where he solves problems, right? I don’t prepare for that, just copy the solution and try to follow and then use the notes when cramming for the exam. (B4) – Student exercises are better because you have worked on the problems. You should do that in teacher-led sessions too, or at least read the problems. Then you would learn more. In teacher-led exercises you mostly copy the solutions. If you are lucky you understand. But otherwise it doesn’t give much. But student exercises gave a lot. (B5) – Studying for the exam is sort of embedded. […] There were six problems every time, pretty full-scale, similar to exam tasks. You learned a lot, because you had to, in order to tick. (B6) Student How did you prepare? C – I solved the problems and read the book. I couldn’t solve the problems without reading the book. Maybe six hours per week. I would have spent less time without the student- led exercises. (C1) Anything else you want to add? – In the beginning I didn’t want the student exercises. It was a scary thought. It was a bit tough that you had to solve problems yourself. But then I realised that I would not have gotten started working like this without them. (C2) How was the recitation itself? – Well, when you are not up yourself but listening to others, it is like a normal exercise class. Even if there were errors on the board sometimes, the right solutions were on the homepage later. It was so much fun to see if anyone had solved the problems differently from yourself. (C3) Case 2: General Chemistry at the University of Oslo General Chemistry is an introductory course for students in the first semester of medical and teaching education, about 275 students in total. It was reported (Uggerud 2013 and 2014) that usually about 85 students dropped out during the semester; of the rest some 75% passed the exam. Further, the course had drifted for several years in order to support weaker students, making lecturers worried about preparation for future courses, e.g. Physical Chemistry. Instructors expressed that the most serious problem was poor attendance in the group sessions, with only about one third showing up. The instructors believed that if student engagement could be improved, exam results would benefit. Various interventions were discussed for a couple of years. Instructors were willing to experiment and hearing about the student-led exercise method (Edström 2015) they decided to redesign the group sessions accordingly. In the Oslo version, students were assigned six assignments every week, in total 78 problems over thirteen weeks, with the requirement to tick 52. Very promising results were reported (Gørbitz 2016, and personal communication). Absence was minimal, as students kept coming to the group sessions even after having fulfilled the required number of ticks. This strengthened the social context for the first-year students, supporting a sense of belonging in the class. Students also reported spending more time on the subject than they would have otherwise. The most important and somewhat unexpected result was a considerable decrease in dropouts. In the first year, about 40 students dropped out (compared to about 85 in three previous years). Of the Paper I students who wrote the exam, 85% passed. The instructors found the new way of teaching challenging, especially finding appropriate strategies for supporting students at the board, and making sure that the whole group benefited from the session, including those who had succeeded with the assignments. Five students were allowed to deliver written work due to various degrees of anxiety. Next year, the course leader carefully informed students and instructors about the aims, the setup, and the results from the previous year. Now teaching was less challenging than the first year, and only two students needed exemption from presenting. The dropout rate was still low, however the pass rate decreased to 75%. In the next version of the course, student-led exercises will continue due to these good experiences. Analysis and discussion Analysing the activity from a learning perspective In the following, we interpret the experiences using the principles for active learning derived above. Our analysis is inspired by Gibbs (1999, 43-47). While empirical data supports the validity of our reasoning, the aim here is more importantly to expose the underlying philosophy and connect it with how the teaching method is carried out. This could support transfer of the innovation, not by superficial copying of the mechanics, but through translation and transformation to other contexts. It can also inspire other teaching innovations. Principle I. Student-led recitations support students to spend sufficient time-on-task and distribute their effort appropriately (Attention) Students are encouraged to spend time-on-task. The KTH interviews show students spending six-seven hours per week preparing (Table 2: A1, B1, C1). Further, students need to attend a majority of the sessions. Particularly in Oslo they kept attending even after fulfilling the requirement. However, a word of warning is warranted. The aim is not to maximise time-on-task, just to achieve sufficient time-on-task. The key challenge is to allocate this effort to appropriate learning activity (principle II). It hardly helps the student applying a surface procedural approach to do more of the same. Further, student effort is well distributed over the whole course; they are ‘in phase’. Through continuous engagement with the subject, students create a better basis for understanding new materials and making connections. In Oslo, we attribute the lower dropout rate to the way students were helped pacing their work. Previously, they could postpone their own work until catching up became overwhelming, making it tempting to give up. The KTH students also appreciated the support in getting started and keeping going (Table 2: A2, C1). To explain why students were more active during lectures, we think they are up to speed with the materials, and have a more immediate need to understand for solving next week’s problems – not only for a distant exam. They may also feel more like active players, used to speaking up. Principle II. Student-led exercises generate appropriate learning activity (High-level engagement) The key consideration when designing teaching is whether it generates the appropriate learning activity (Gibbs 1999). In fact, unless student activity is appropriate, time-on- task equals unproductive busy-work. First of all, solving the problems constitutes Paper I good studies. Watching teachers solve problems can create a false feeling of grasping, as everything appears obvious and easy. There is therefore great advantage in having students first trying for themselves, to discover the limits of their understanding (Prince and Felder 2006). Left to their own devices, however, many students would still just try to find the right answers and have them done, i.e. adopt a surface procedural approach. Here, since they must also prepare to explain their solution, it becomes obvious that finding the answer is not sufficient, thus allowing them to challenge their habits. Thus, this activity models a deep procedural approach, inviting students to reflect more on their problem-solving strategy. Good facilitation increases the value of sessions. Since most students have prepared the problem they can follow even when presentations are not perfect (Table 2: B4, B5, C3). In contrast, we learned that teacher-led exercises fail to confront the surface procedural approach (Table 2: B3, B5). Put differently, the teacher-led exercises allow students to aim for the Feisel-Schmitz Compute level. We suggest that as students are now required to prepare how to explain their solutions, this is inherently geared to making students perform activities that correspond at least to the Explain level. To reach even higher, to the Solve or Judge levels, is contingent on the nature of the problems and the facilitation of discussions. We note that the activity is appropriate only if problems are aligned with the intended learning outcomes, and with assessment, i.e. they are in constructive alignment (Biggs and Tang 2011). This also implies that problems must reach the same taxonomy level as is required in the exam. Considerable student motivation and buy-in comes from trusting the constructive alignment, i.e. knowing that this effort will help them reach the intended learning outcomes also reflected in the assessment (Table 2: A1, B6). Principle III. Student-led exercises communicate high and clear expectations, and provide opportunities for reflection and feedback, to help students improve their work and develop their judgement (Purpose) Since students have at least tried to solve the problems, some feedback comes already from following peer presentations (Table 2: B3, B5, C3). We see discussions as valuable feedback opportunities for the whole group. This is however not inherent in the method; it needs facilitation and active prompting for e.g. alternative solutions, assumptions and interpretations. Discussions should model a deep procedural approach, making strategy explicit, e.g. use of theory and arguments, how to devise the solution, exposing the rationale and difficulties. Hence, insightful facilitation is a key task for the instructors. As students are exposed to a variation of solutions and presentations, alerting them to discern and reflect on key aspects can help them internalise the expected standards, developing their own sense of judgement. This might be a factor behind fewer poor exams at KTH: students know what adequate solutions require. Posting model solutions after the session provides minimal feedback to absent students. Principle IV. Student-led exercises encourage interaction and dialogue with teachers and peers, modelling appropriate approaches to learning (Interaction) Student-led exercises are intensely social, exposed the learning process to peers and instructors. This can be important for social motivation. In Oslo, due to the good attendance, the learning environment could fulfil its important social function better than when only a minority of the students attended. KTH interviews revealed some Paper I students preparing together in groups, essentially organising their own exercise session in advance (Table 2: B1, B2). We propose that practicing higher-level forms of engagement together in a given structure makes it easier to try out new intellectual habits. Instead of copying solutions, students are cast in more mature roles, to develop as learners and future professionals. In our experience, the demands are reasonable and within the potential for practically all students, also those initially doubting their ability (Table 2: C2, C3). In Oslo, some students balked at the thought of standing before the group, but fears subsided as the method became increasingly familiar. This underlines that while the social dimension is a useful driving force, it causes pressure. Therefore we treat problem-solving critically and students gently. Principle V. Student-led exercises provide information to teachers that is used to develop teaching (Continuous improvement) In student-led exercises teachers monitor learning while it can still be improved. As needed, we move on, or stop to penetrate the muddier issues. Over time, instructors recognise common difficulties in conquering the subject, which can inform course development. It is one thing to know the subject for oneself, and another to guide students into this foreign terrain (Shulman 1987), and we see the instructors’ increasing awareness in problem design as a significant finding at KTH. For instance, to explain two alternative solutions, a problem is constructed where students will likely choose different paths. The discussion is then facilitated to ‘discover’ and consider the merits of the alternatives. Learning effectiveness and teaching efficiency Educational development is often analysed in terms of learning only. However, to enable change, teacher workload also needs consideration. The KTH experience showed that changing teacher-led exercises into student-led is at least cost-neutral. Making a scheduled activity generate a volume of high-level engagement out of class is a way to maximise cost-effectiveness. There is an “iceberg effect”, in that for every presentation we see, up to twenty students have prepared it outside class. This not only leverages time-on-task; it has high educational value that students have worked on the problem that is being presented. When colleagues worry about inferior student presentations (‘after all we are the experts’), this objection indicates a misconception. Student presentations are the means to an end; their function is to generate the appropriate learning activity. Hence, student-led exercises should not be judged by the quality of presentations (the tip of the iceberg), but by the quality of learning generated by the whole activity, including all preparations. Student-led exercises implies no teacher abdication; it is just another, more effective way to teach. As Boud and Molloy (2013a, 2) put it, “the acts of teachers need to be judged in the light of their impact on learning”. While teacher-led exercises use the same teacher time, KTH interviews show that they generate no preparation beforehand and little high- level engagement during sessions (Table 2: B3, B5, C1). Hence, teacher presentations, even perfect ones, can yield less learning. Limitations Compared to presenting solutions from notes, student-led exercises make different demands on the teacher. Creating the safe atmosphere and sense of community requires some measure of emotional maturity, and some comfort with the subject is required, e.g. ability to think on one’s feet. If the mechanics of the method are copied Paper I without understanding the underlying philosophy, we fear that it can be implemented in ways that mainly creates busy-work. It must be assumed that students can approach the task with a surface procedural approach, and the activity should not be designed, presented, or carried out in ways that promote or condone it. Student-led exercises should not be confused with summative assessment. Students and colleagues have called it ’unfair’ that students are allowed to collaborate during preparations, and may present different numbers of problems of various difficulty. But the summative assessment, where fairness is key, is performed in the final exam. Since the exercises are strictly formative, it is learning that matters, not fairness. Would student presentations influence grading, the situation is entirely different, for which the implementation and experiences presented here do not apply. This learning activity may not be suitable for all subjects. Descriptive subjects may not be suitable for problem-solving. Also, the value diminishes if problems lack room for discussion, e.g. small tasks with one right method and one right answer. Otherwise, it can be modified for many types of subjects by varying the problem or assignment type. In other implementations, students have explained algorithms, demonstrated code, explained concepts, proposed solutions to case studies, and reflected on literature. Conclusions To conceptualise problem-solving as a learning outcome, the Feisel-Schmitz taxonomy makes a useful distinction between problem-solving with and without understanding. To conceptualise the quality problem-solving as a learning activity, we explored Case & Marshall’s surface and deep procedural approach, also highlighting their possible underlying epistemological views. While problem-solving is an appropriate learning activity for conceptual understanding, we advocated for problem- solving competences as legitimate and important learning outcomes in themselves. This implied a new positioning of the deep and surface procedural approaches in relation to the classic deep and surface approaches. We argued for designing learning activities conducive to a deep procedural approach, synthesising some principles for purposeful active learning. We described a method for teaching problem-solving, student-led exercises, and the lessons learned in two universities where the intervention led to improved student learning, in a cost-effective way. We suggested that the activity models the deep procedural approach, in a structured, collaborative setting. The first case, an engineering course at KTH, gave insights into how the activity worked from the student perspective. In particular the comparison between teacher-led and student-led exercises, so alike on the surface if anyone would peer through the door, showed vast differences. Student-led exercises generated considerable amounts of effective studies, while the teacher-led ones did not. The second case, a science course in Oslo, revealed something that the KTH case could not. This course was placed in the first semester and suffered from a high dropout rate, and standards had drifted to accommodate weaker students. The result of their implementation indicates that attrition can be addressed by creating a motivational context and structure to support learning, essentially raising expectations. Paper I While the elaborated analysis could support transfer of student-led exercises, we did not intend to be prescriptive about this particular method. Rather, we hope that this way of thinking can just as well support the design or analysis of other activities. There are many different ways, some even cost-neutral, to develop courses for qualitatively improve the experience of learning as well as the experience of teaching. Acknowledgements We thank Martin Domeij and Gunnar Malm for collegial discussions in the Semiconductor Devices course. Special thanks to Professor Carl Henrik Görbitz, University of Oslo, for providing data and reflections on their experiences. References Abeysekera, Lakmal, and Phillip Dawson. 2015. "Motivation and cognitive load in the flipped classroom: definition, rationale and a call for research." Higher Education Research & Development 34 (1):1-14. Biggs, John, and Catherine Tang. 2011. Teaching for quality learning at university. Maidenhead: McGraw-Hill. Bligh, Donald A. 2000. What's the Use of Lectures? San Francisco: Jossey Bass. Boud, David, and Elizabeth Molloy. 2013a. Feedback in higher and professional education: understanding it and doing it well. Abingdon: Routledge. Boud, David, and Elizabeth Molloy. 2013b. "Rethinking models of feedback for learning: the challenge of design." Assessment & Evaluation in Higher Education 38 (6):698-712. Case, Jennifer, and Delia Marshall. 2004. "Between deep and surface: procedural approaches to learning in engineering education contexts." Studies in Higher Education 29 (5):605-615. Chickering, Arthur W, and Zelda F Gamson. 1987. "Seven principles for good practice in undergraduate education." AAHE bulletin 3:7. Crouch, Catherine H, and Eric Mazur. 2001. "Peer instruction: Ten years of experience and results." American journal of physics 69 (9):970-977. Edström, Kristina. 2012. "Student feedback in engineering: Overview and background." In Enhancing learning and teaching through student feedback in engineering., edited by P Mertova, S Nair and A Patil. Cambridge: Woodhead Publishing. Edström, Kristina. 2015. "How to improve student learning without spending more time teaching". Keynote at the University of Oslo, 8 May 2015. Edström, Kristina, and Anette Kolmos. 2014. "PBL and CDIO: complementary models for engineering education development." European Journal of Engineering Education 39 (5):539-555. Edström, Kristina, Josefin Törnevik, Madelaine Engström, and Åsa Wiklund. 2003. "Student involvement in principled change: Understanding the student experience." In Proceedings of the 11th Improving Student Learning. Oxford: OCSLD. Feisel, Lyle D. 1986. "Teaching Students to Continue Their Education." Frontiers in Education, Arlington, Texas. Freeman, Scott, Sarah L. Eddy, Miles McDonough, Michelle K. Smith, Nnadozie Okoroafor, Hannah Jordt, and Mary Pat Wenderoth. 2014. "Active learning Paper I increases student performance in science, engineering, and mathematics." Proceedings of the National Academy of Sciences 111 (23):8410-8415. Gainsburg, Julie. 2015. "Engineering students' epistemological views on mathematical methods in engineering." Journal of Engineering Education 104 (2):139-166. Gibbs, Graham. 1992. Improving the quality of student learning. Bristol: Technical and Education Services. Gibbs, Graham. 1999. "Using assessment strategically to change the way students learn." In Assessment Matters in Higher Education, edited by S Brown and A Glasner, 41-53. Buckingham: SRHE and Open University Press. Gørbitz, Carl Henrik. 2016. Sluttrapport KJM1100 Generell kjemi H-2015 [Final report KJM1100 General Chemistry Fall 2015]. Oslo: University of Oslo. Jonassen, David H. 2014. "Engineers as Problem Solvers." In Cambridge handbook of engineering education research, edited by Aditya Johri and Barbara M Olds. New York, NY.: Cambridge University Press. Karpicke, Jeffrey D. 2012. "Retrieval-Based Learning." Current Directions in Psychological Science 21 (3):157-163. Kember, David, and Carmel McNaught. 2007. Enhancing university teaching: Lessons from research into award-winning teachers. Abingdon: Routledge. Kvale, Steinar. 2008. Doing interviews. Thousand Oaks, CA: Sage. Mahalingam, Madhu, Fred Schaefer, and Elisabeth Morlino. 2008. "Promoting student learning through group problem solving in general chemistry recitations." J. Chem. Educ 85 (11):1577. Marton, Ference, Dai Hounsell, and Noel James Entwistle. 1984. The experience of learning. Edinburgh: Scottish Academic Press. Marton, Ference, and Roger Säljö. 1976a. "On Qualitative Differences in Learning — 1: Outcome and Process." British journal of educational psychology 46 (1):4- 11. Marton, Ference, and Roger Säljö. 1976b. "On Qualitative Differences in Learning — 2: Outcome as a function of the learner's conception of the task." British Journal of Educational Psychology 46 (2):115-27. Mazur, Eric. 1997. Peer instruction: Upper Saddle River, NJ: Prentice Hall. Nicol, David J., and Debra Macfarlane-Dick. 2006. "Formative assessment and self- regulated learning: a model and seven principles of good feedback practice." Studies in Higher Education 31 (2):199-218. Perry, W.G., Jr. 1998. Forms of Intellectual and Ethical Development in the College Years: A Scheme. San Francisco, CA: Jossey-Bass. Prince, Michael J, and Richard M Felder. 2006. "Inductive teaching and learning methods: Definitions, comparisons, and research bases." Journal of engineering education 95 (2):123-138. Prosser, Michael, and Keith Trigwell. 1999. Understanding learning and teaching: The experience in higher education. Buckingham: SRHE and Open University Press. Redish, Edward F, and Richard N Steinberg. 1999. "Teaching Physics: Figuring Out What Works." Physics Today 52 (1):24-30. Shulman, Lee. 1987. "Knowledge and teaching: Foundations of the new reform." Harvard educational review 57 (1):1-23. Smith, Karl A., Sheri D. Sheppard, David W. Johnson, and Roger T. Johnson. 2005. "Pedagogies of Engagement: Classroom-Based Practices." Journal of Engineering Education 94 (1):87-101. Paper I Uggerud, Einar. 2013 and 2014. Periodisk emneevaluering av KJM1100 [Periodical evaluation of KJM1100]. Oslo: Oslo University.