October 2016

Measuring Early Learning Quality and Outcomes in Tanzania Institutional assessment for integrating early childhood measurement in the pre-primary system

Kate Anderson Rebecca Sayre

Authors

Kate Anderson is a project director and associate fellow in the Center for Universal Education at the Brookings Institution.

Rebecca Sayre is a consultant in the Center for Universal Education at the Brookings Institution.

Measuring Early Learning Quality and Outcomes in Tanzania i

Acknowledgements

This report represents the collaborative work of the Measuring Early Learning Quality and Outcomes core team, in consultation with Ministry of Education. The authors are grateful for the inputs provided by Anna Smeby of UNICEF Tanzania and stakeholders from the Ministry of Education, Tanzania Institute for Ed- ucation, National Exams Council of Tanzania, University of Dodoma, and civil society (a complete list of stakeholders consulted is provided in Annex D).

MELQO Core Team

Technical Lead: Abbie Raikes, University of Nebraska Maki Hayashikawa UNESCO Sobhi Tawil Amanda Devercelli World Bank Lucy Bassett Kate Anderson Center for Universal Education, Brookings Tamar Manuelyan Atinc Institution Fabiuola Lara Pia Britto UNICEF Ana Nieto

The Brookings Institution is a nonprofit organization devoted to independent re- search and policy solutions. Its mission is to conduct high-quality, independent research and, based on that research, to provide innovative, practical recommen- dations for policymakers and the public. The conclusions and recommendations of any Brookings publication are solely those of its author(s), and do not reflect the views of the Institution, its management, or its other scholars.

Brookings gratefully acknowledges the program support provided by the Chil- dren’s Investment Fund Foundation and Omidyar Network.

Brookings recognizes that the value it provides is in its absolute commitment to quality, independence, and impact. Activities supported by its donors reflect this commitment.

Measuring Early Learning Quality and Outcomes in Tanzania iii Acronyms

3Rs Reading, writing, and arithmetic BEST Basic Education Statistics BRNEd Big Results Now in Education DFID UK Department for International Development ECD Early Childhood Development ECE Early Childhood Education EGMA Early Grade Mathematics Assessment EGRA Early Grade Reading Assessment EMIS Education Management Information System EQUIP-T Education Quality Improvement Programme-Tanzania ETP Education and Training Policy GPE Global Partnership for Education LGA Local Government Authority M&E Monitoring and Evaluation MCDGC Ministry of Community Development, Gender, and Children MELE Measure of Early Learning Environments MELQO Measuring Early Learning Quality and Outcomes MODEL Measure of Development and Early Learning MoEVT Ministry of Education and Vocational Training MoHSW Ministry of Health and Social Welfare NACTE National Council for Technical Education NECTA National Exams Council of Tanzania PMO-RALG Prime Ministers’ Office–Regional and Local Governance RTI Research Triangle Institute SSME Snapshot of School Management Effectiveness TIE Tanzania Institute for Education

iv Measuring Early Learning Quality and Outcomes in Tanzania Contents

I. Introduction ...... 1

The measuring early learning quality and outcomes (MELQO) initiative ...... 1

The process in Tanzania ...... 2

II. Overview of the pre-primary system in Tanzania ...... 5

Pre-primary policy and planning landscape ...... 5

Current status of pre-primary programs in Tanzania ...... 6

III. Key Features of Quality Assurance and Assessment ...... 9

IV. Tanzanian perspective on measuring early learning ...... 13

V. Recommendations ...... 23

References ...... 29

Annexes ...... 31

Annex A. Functions and organization structure of MoEVT ...... 31

Annex B. Functions and organization structure of PMO-RALG ...... 32

Annex C. Protocol for national stakeholder interviews ...... 33

Annex D: List of stakeholders consulted during Phase I ...... 35

Measuring Early Learning Quality and Outcomes in Tanzania v

I. Introduction

The measuring early learning • Identifying core items across instruments to quality and outcomes (MELQO) build regional and global comparability, using initiative the most promising, technically robust, and feasible approaches for measurement, especial- he Measuring Early Learning Quality and Out- ly for low- and middle-income countries; Tcomes (MELQO) initiative aims to improve • Working closely with national stakeholders to early childhood education worldwide through map their existing assessment efforts and en- measurement of children’s development and learn- suring that the modules are informed by na- ing and the quality of learning environments. Led tional perspectives and research; by UNESCO, UNICEF, the World Bank, and the • Providing guidance to countries on how to Center for Universal Education at the Brookings move from prototype piloting to implemen- Institution, MELQO comprises an international tation at scale by assessing the institutional consortium of individuals and institutions working changes needed and the cost involved; to improve outcomes for young children by making • Designing modules to help governments take early learning assessment more accessible around measurement to scale to integrate within existing the world. MELQO’s measurement modules pro- data systems and develop policies that improve pose a core set of items with relevance across coun- early learning environments and outcomes; tries, with the goal of devising items that are glob- • Making all modules and manuals open-source ally comparable but locally adaptable. and freely available and providing technical guidance to users. While approaches to measuring early childhood development and learning have been developed The assessment modules are designed for use by across countries, less attention has been placed on national governments, teachers, and school admin- coordinating across instruments to improve the ef- istrators and can be administered by someone with ficiency of measurement, or on using the modules limited training. The emphasis is on producing data at scale across countries or regions, especially in that can be used to improve the quality of learning low- and middle-income countries. MELQO builds environments and children’s development and on the strengths of measurement initiatives that learning through links with national curricula, have already been developed through the following quality standards, and teacher/parent support and activities: training. If a sufficiently large number of countries adopt the modules and report on results, the data • Bringing together and building upon existing will also help in regional or global tracking of chil- regional early learning and quality assessments; dren’s readiness for school

Measuring Early Learning Quality and Outcomes in Tanzania 1 Two technical advisory groups have been formed information from this analysis was used to adapt through MELQO, one developing a Measure of the MODEL and MELE modules to meet the needs Development and Early Learning (MODEL) and of the Tanzanian context. another developing a prototype Measure of Ear- ly Learning Environments (MELE). The technical groups were first asked to identify constructs of in- The process in Tanzania terest and then, based on inventories of items used in existing modules, were asked to identify items that 1. Launch were viewed as being especially valuable or reliable in diverse settings. The goal for both the MODEL In July 2015, the government of Tanzania, with and MELE modules is to obtain reliable evidence leadership from the Ministry of Education and Vo- establishing validity, including predictive validity. cational Training (MoEVT),1 convened national and international experts to launch the MELQO As noted above, the purpose of MELQO is to im- initiative in Tanzania and plan for implementation prove the efficiency of measurement by providing a of piloting the modules. Participants included gov- core set of items that can be used in any country. ernment representatives from MoEVT, the Prime The items were selected as a starting point for testing Minister’s Office for Regional and Local Govern- and adaptation within countries. Because much of ment (PMO-RALG), the Ministry of Health and So- the research to date on predicting children’s school cial Welfare (MoHSW), the Ministry of Communi- achievement comes from a small number of high-in- ty Development, Gender, and Children (MCDGC), come, Western cultures, it is anticipated that some the Tanzania Institute for Education (TIE), and the of these constructs and items will not be relevant in National Exams Council of Tanzania (NECTA); all settings while others will be useful across coun- non-government representatives from academia tries. As part of the validation process, information and civil society, including the Tanzania Early on how relevant the items are across contexts will be Childhood Development Network (TECDEN), the collected and used to modify the modules over time. Tanzania Child Rights Forum (TCRF), Aga Khan University (AKU), and the University of Dodoma Table 1 summarizes the design and use of the (UDOM); and local partners from UNICEF, UNE- MODEL and MELE modules. SCO, and the World Bank. National stakeholders were joined by the MELQO Core Team from the Center for Universal Education at the Brookings In- Purpose of the institutional stitution, the World Bank, and UNICEF as well as assessment technical experts from academic institutions from Kenya, the United States, Canada, and Hong Kong. The MELQO modules are intended to be integrat- ed into existing national monitoring and evaluation systems. This institutional assessment was under- 2. Adaptation and field testing taken to understand the current policies, curricu- la, standards and monitoring and evaluation sys- After the launch, the MODEL and MELE mod- tems for pre-primary education in Tanzania. The ules were translated to Kiswahili, adapted, and

1 At the time of this assessment, the ministry was named Ministry of Education and Vocational Training, but has since been changed to Ministry of Education, Science and Technology (MoEST).

2 Measuring Early Learning Quality and Outcomes in Tanzania Table 1: MELQO MODEL and MELE Modules

Module What does it measure? How can the module be used in the long term?

Measure of Four basic areas of children’s devel- Identify areas of strength and difficulty for Development opment at the start of school: groups of children in classrooms or schools (for and Early example, whether many children already know Learning Executive function (self-regulation, their numbers, or whether working memory is (MODEL) attention, etc.) a challenge). This could include:

Social-emotional development • Informing approaches to curricula at the policy level Early mathematics skills • Setting reasonable expectations for what children will be able to accomplish by the Early literacy skills end of the year • Identifying areas where additional support is needed • Tracking changes within one country over time, or providing some degree of com- parison between countries on a core set of items, if all countries in a region agree to use the same set of items

Measure of 1. Environment and physical setting Track the overall quality of children’s learning Early Learning 2. Family and community environments and identify the areas in which Environments engagement specific actions are needed, including: (MELE) 3. Personnel 4. Interactions • Providing teachers with more training and 5. Inclusiveness support in interacting with young children 6. Program structure and • Providing professional development oppor- curriculum tunities for all teachers 7. Health, safety, and hygiene • Ensuring that classrooms have access to high-quality, age-appropriate learning mate- rials for all children

field-tested. MoEVT convened an initial adaptation the pilot experience prepared by RTI will inform workshop in August 2015. The Research Triangle the revision process and 2016 study design. Institute (RTI) conducted a five-day training with DataVision, a Tanzanian research firm, and repre- 3. Consultations with stakeholders sentatives from MoEVT. Enumerators from Data- Vision then piloted the modules and collected data In October-November 2015, one-on-one inter- on 209 pre-primary children in 20 schools in and views and group discussions were conducted around Dar es Salaam in mid-October 2015. The with national stakeholders, including the MoEVT sample of schools included higher- and lower-per- Pre-Primary Unit, the MoEVT Quality Assur- forming government schools and private schools. ance Division (formerly the School Inspectorate), The MELE module was piloted in 18 schools in late the MoEVT Department of Policy and Planning, September and mid-October. A detailed report on TIE, PMO-RALG, TCRF, UDOM, EQUIP-T, Plan

Measuring Early Learning Quality and Outcomes in Tanzania 3 International, the Tanzania Education Network 4. Planning for 2016 national study (TEN/MET), UNICEF, and others. Stakeholders were asked to share their opinions on the appe- In 2016, Tanzania plans to conduct a nationally tite for measuring early learning in Tanzania, the representative study using the MODEL and MELE needs for a national study on pre-primary learning modules. The study methodology will draw from outcomes and quality, and options on how to best the field test experiences in 2015. The specific sam- integrate the MELQO modules into the existing ed- ple size and sampling frame will be determined ucation system. Sections IV and V of the present re- through a joint process with the MELQO Core port summarize input from stakeholders on the key Team, RTI, and the MELQO Task Force. features of the existing system and the relevance, feasibility, and sustainability of measurement in the pre-primary sector in Tanzania.

4 Measuring Early Learning Quality and Outcomes in Tanzania II. Overview of the pre-primary system in Tanzania

Pre-primary policy and planning development partners, and others also play a key landscape role in the pre-primary sector. Annexes A and B display the organizational structure of the overall oEVT is responsible for policy formulation MoEVT and PMO-RALG. In addition, the Minis- Mand regulation, quality assurance and cur- try of Health and Social Welfare and the Ministry riculum, teacher training, examination and assess- of Community Development, Gender, and Children ment, standards setting, planning, monitoring and are involved in early childhood development (ECD) evaluation, and education research. in general, but not specifically in pre-primary.

The administration functions of primary and sec- In 2014 the government of Tanzania released its ondary education are decentralized to local gov- new Education and Training Policy (ETP), which ernment authorities (LGAs), overseen by PMO- calls for increasing opportunities and quality of RALG. PMO-RALG is responsible for the delivery all levels of education in the country. The policy of pre-primary (and primary and secondary) edu- shifts the official age of entry to Standard I (first cation through Councils; delivery is complemented year of primary school) from age 7 to 6, and calls by non-government provision of education. Deliv- for pre-primary education for 3-5-year-olds, with ery includes supply of teachers (payment of teach- compulsory enrollment for one year. The ETP also ers’ salaries), physical facilities, and school materi- emphasizes quality pre-primary education through als, and supervision of the Education Management adequate teaching and learning methods and mate- Information System (EMIS). rials, relevant curriculum and teacher training, and strengthened quality control and assurance. Mo- TIE, a parastatal organization under MoEVT, EVT is in the process of developing an implemen- is responsible for the design and development of tation plan, which will include a specific plan for curricula and teaching and learning materials for pre-primary, for effective application of the ETP. pre-primary, primary, secondary, and teaching edu- cation levels. TIE also conducts training programs In the short term, the government’s Short-Term Ear- to build capacity on the national curriculum for ly Childhood Education (ECE) Action Plan was de- teachers, education officers, education coordina- veloped to prioritize key issues and action areas in tors, inspectors, and other relevant stakeholders. the pre-primary subsector. The ECE Action Plan is a collaborative effort across MoEVT, MCDGC, TIE, Figure 1 displays the institutional arrangements for the Agency for Development of Educational Man- the national leadership of quality assurance and de- agement (ADEM), and partners including the World livery of pre-primary education in Tanzania. Civil Bank, UNICEF, the Global Partnership for Educa- society, the private sector, academic institutions, tion (GPE), Dubai Cares, Children in Crossfire, the

Measuring Early Learning Quality and Outcomes in Tanzania 5 Figure 1: Institutional Arrangements for Pre-Primary

MoEVT PMO-RALG Planning authority Implementing authority Basic Education Director Basic TIE Office: Policy & Planning ICT Systems Education Commissioner Division Section Coordination Curriculum development/monitoring, Oversees policy development, quality Coordinates planning, Data collection and Oversees education NATIONAL teacher training assurance, accreditation budgets, M&E Management delivery Primary & Preprimary Primary Schools Quality Primary and pre-primary Education Unit Assurance (Inspectorate) Education Section Oversees sector policies Ensure compliance to Coordinates PEDP and programs policies, regulations implementation

Zonal Quality Regional Assurance Office Education Officer Monitors and evaluates ZONAL Coordinate regional quality & implementation REGIONAL/ implementation District Quality Regional Assurance Office Education Officer Prepares and submits school Manages and monitors

DISTRICT inspection reports district education plans Ward Education Quality Ward Education Assurance Office Coordinator

WARD Oversees quality in all Ongoing school site schools in Ward supervision (4-5 schools) Primary School Head Teacher Prepares and submits school

SCHOOLS inspection reports

Aga Khan Foundation, Aga Khan University, and systems; (6) conduct a national baseline survey on EQUIP-T. Priority action areas comprise plans to: pre-primary learning outcomes and quality; and (1) develop costed pre-primary action plans to sup- (7) strengthen pre-primary sub-sector planning and port implementation of the new ETP; (2) develop coordination. and test quality program models to enable expan- sion of pre-primary access and equity, including a national satellite pre-primary model and a par- Current status of pre-primary ent education strategy; (3) update and align the programs in Tanzania pre-primary quality framework (including curric- ulum and standards) to the new ETP for children In 2014, approximately one-third of 5- and ages 3,4, and 5 with a focus on pre-3Rs (reading, 6-year-olds were enrolled in pre-primary ed- writing, and arithmetic); (4) strengthen the teach- ucation (the gross enrollment rate was 36.9, er training, development, and professionalization the net enrollment rate 33.4). According to system, including by conducting a national base- Tanzania’s Basic Education Statistics, approx- line survey on the pre-primary workforce, piloting imately 95 percent (992,356) of children are a satellite pre-primary teacher and mentor training enrolled in government pre-primary classes, program, and developing a professional develop- while 5 percent (54,013) are enrolled in regis- ment/certification plan; (5) include pre-primary tered non-government pre-primary classes. The education within school management and planning government tracks enrollment for registered

6 Measuring Early Learning Quality and Outcomes in Tanzania pre-primary schools, but the number of other primary is completion of Form IV of secondary and early learning programs provided by the private a Grade A teaching certificate (requiring completion sector or community organizations is unknown. of a two-year full-time pre-service teaching course). In 2014, less than half of teachers were “qualified,” As of 2014, 14,719 pre-primary institutions were and a majority of these had the minimum require- operating in Tanzania. These include govern- ment of the Grade A certificate. The government ment-operated pre-primary schools operating with- is considering increasing the minimum requirement in the country’s 16,648 primary schools as well as from certificate to diploma. At the same time, Mo- non-government pre-primary schools that have EVT and TIE are supporting the development of registered with MoEVT. alternative teacher training initiatives.

Pre-primary schools are overcrowded. While the There are currently no national data on school read- 2009 Basic Standards for Pre-Primary and Pri- iness outcomes. However, a national assessment in mary Education in Tanzania call for a maximum 2013 revealed that only 8 percent of Standard II pupil-to-teacher ratio of 1:25, the average ratio is students were reading with comprehension, and 1:77 (BEST 2014). The difference is stark between the majority struggled with conceptual mathematic non-government and government pre-primary tasks. Ensuring children arrive to school ready to schools, with 1:90 in government schools com- succeed is critical, as international evidence sug- pared to 1:21 in non-government schools. gests that intervening before primary school will have a significant impact on later school perfor- As displayed in Table 2, as of 2014, 13,600 teach- mance (Heckman and Masterov 2007; Reynolds et ers were teaching pre-primary (BEST 2014). The al. 2001; Lynch 2005). official requirement for teachers to work in pre-

Table 2: Pre-Primary Teaching Staff by Qualification (BEST 2014)

Grade A certificate Diploma Graduates Others Total Number Percentage Number Percentage Number Percentage Number Percentage 6,090 44.7% 511 3.7% 97 1.4% 6,902 51% 13,600

Measuring Early Learning Quality and Outcomes in Tanzania 7

III. Key features of quality assurance and assessment

re-primary education in Tanzania is incorpo- relationships). TIE has also updated a pre-primary Prated within the formal education structure, syllabus, which will provide detailed guidance for existing as a sub-sector within the primary sector teachers to implement the national curriculum. The (Standards I-VII for children ages 6-13 years). In syllabus is divided into six main areas: (1) socializa- general, no specific monitoring and evaluation tion, (2) communication, (3) health and protection, (M&E) mechanisms have existed for pre-primary. (4) caring for the environment, (5) artistic compe- Some features of the primary education system, in- tencies, and (6) identifying mathematical concepts. cluding standards and inspection procedures, have been applied to pre-primary classrooms. With the launch of the new ETP, the government is in the Inspection and service delivery process of considering how to update and improve standards its system for monitoring, evaluating, and ensuring the quality of pre-primary. In addition to TIE’s leadership in developing and monitoring the curriculum, responsibility of as- suring quality pre-primary services also spans Pre-primary curriculum across PMO-RALG and MoEVT (see Figure 1). The Pre-Primary and Primary Education Section In the past eight years, the National Pre-Primary of PMO-RALG is responsible for preparing and Curriculum (2007) has served as a guide for the disseminating operational guidelines and circulars content, skills, teaching, and learning strategies for on management and administration of pre-pri- pre-primary education. In 2015, TIE led an exten- mary classrooms. Under PMO-RALG, regional sive review and revision of the curriculum that con- education officers, district education officers, and centrates on one year of pre-primary education (as ward education coordinators periodically supervise stipulated in the new ETP).1 The updated curriculum classrooms, teachers, and overall education deliv- aligns to a new quality framework for pre-primary, ery. Under MoEVT, the Quality Assurance Division which focuses on pre-3Rs and outlines desired com- (formerly the School Inspectorate) is responsible petencies for both pre-primary teachers and pre-pri- for ensuring compliance with policies, laws, regula- mary students (including communication, recogni- tions, and other standards in the delivery of pre-pri- tion of the environment, pre-literacy, writing, and mary, primary, and secondary education.2 The

1 As of December 2015, the new curriculum and syllabus have been drafted and are under review for finalization and dissemination in early 2016. 2 In 2015, the government changed the name of the Inspectorate to Quality Assurance Division to capture the ministry’s intentions to ensure continuous improvement of quality through feedback loops and coordination with other education stakeholders.

Measuring Early Learning Quality and Outcomes in Tanzania 9 Quality Assurance Division collaborates with TIE variety of services, including home-based services, and the National Exams Council of Tanzania to community-based ECD centers, and state and non- design and develop an internal and external qual- state institutions. ity assurance framework and oversees district and ward quality assurance officers (formerly school in- spectors) to inspect schools and recommend reme- National assessment dial measures to continuously improve the quality of education. To date, the government has never assessed chil- dren in pre-primary. However, Tanzania’s recent The quality assurance and M&E system are framed experience with the assessment in early primary by several key documents. The National Basic Ed- grades presents several important lessons that can ucation Standards created in 2009 give guidance be applied to the MELQO experience. In 2013, in on design and implementation of pre-primary and response to declining trends in primary and sec- primary education programs and include require- ondary student learning outcomes, the government ments for infrastructure, environment, curriculum, launched the Big Results Now in Education (BR- teaching and learning process and materials, teach- NEd) program to fast-track the improvement of ers, school management, and capacity building. quality of basic education service delivery. While The national standards call for school inspection BRNEd does not specifically focus on pre-primary, at least once every two years, as well as ongoing it emphasizes quality learning of the 3Rs in the ear- school site supervision carried out by head teachers ly primary grades. One of the nine key initiatives and a ward education coordinator. During inspec- within the BRNEd program is 3R Assessment and tion and advisory visits, quality assurance officers 3R Teacher Training, which serve as a mechanism use the MoEVT Whole School Inspection Check- for annual monitoring of student learning for chil- list and the Handbook for School Inspectors and dren in Standard II (8-year-olds) in reading, writ- suggest recommended remedial measures as per ing, and arithmetic. government regulations. The Guidelines for School Supervision provide guidance to ward education MoEVT conducted a national 3Rs baseline assess- coordinators, heads of schools, district education ment in July 2013. With technical assistance from officers, and regional education officers on effective USAID/RTI, data were collected on a nationally rep- supervision to enhance school performance and en- resentative sample using the Early Grade Reading sure quality education. Assessment (EGRA), the Early Grade Mathematics Assessment (EGMA), and the Snapshot of School In 2014, only 22 percent of pre-primary classrooms Management Effectiveness (SSME). A total of 2,266 were inspected (BEST 2014). Data are not publicly Standard II students randomly selected from within available to indicate how many classrooms com- 200 randomly selected schools took the EGRA and plied with national standards. EGMA assessments. The baseline findings have con- tributed to the development of 3R teacher training In addition to the officially recognized national materials and the government’s commitment to im- standards for all of basic education, Operating proved student learning in these foundational areas. Guidelines and Standards for ECD Services have The National 3Rs Program includes a revised cur- been promulgated, but not formalized. In 2013, the riculum and teacher training program for Standards Inter-Ministerial ECD Steering Committee, with I and II, which is currently being rolled out with support from UNICEF, released guidelines for a Global Partnership for Education funding.

10 Measuring Early Learning Quality and Outcomes in Tanzania The National Exams Council of Tanzania is legal- and school online.3 NECTA is in the process of cre- ly responsible for all national assessment. After ating a database of all children, starting in primary the 3Rs baseline assessment, the government ap- education, which will track the performance and pointed NECTA to develop a national assessment movement of every child in Tanzania. Each child for Standard II. In 2015, NECTA worked with will have a unique identifier number, which will al- national experts to develop a 3Rs assessment for low the government to compare and track children children in Standard II and an “advanced” 3Rs as- throughout the system. sessment for children in Standard IV. The assess- ments are aligned with the national curriculum and While NECTA’s experience with primary (and are intended to show teachers, education officials, secondary) education is promising, stakeholders parents, and the public how children are faring. In identified several risks to directly applying nation- November 2015, NECTA piloted a national assess- al assessment at the pre-primary level. Primarily, ment with a sample of children in Grade 2 (3Rs as- stakeholders had concerns with ranking and/or sessment) and Grade 4. Rather than a sampling ap- excluding children in pre-primary and are hesitant proach, the government plans to assess every child to make judgments about individual children in in primary and publicly share data for every child pre-primary.

3 NECTA’s website (http://www.necta.go.tz) displays exam results publicly. Previously, Tanzania had Primary School Leaving Exams, which are being eliminated and replaced with the new assessment system, which includes 3Rs assessment.

Measuring Early Learning Quality and Outcomes in Tanzania 11

IV. Tanzanian perspective on measuring early learning

hrough consultation with national stakehold- MELQO can help gain attention to the strong Ters, the following issues related to opportuni- links of pre-3Rs and 3Rs. MoEVT has recently ties, needs, and challenges to measuring early learn- announced that the National 3Rs program will ing quality and outcomes have been identified: include pre-primary education. GPE funds are available to roll out pre-primary teacher train- 1. Priorities and expectations for early learning ing of the new curriculum. need to be aligned. Implementation of MELQO in Tanzania will be timely for the country, par- While the ETP shapes the overarching goal of ticularly with the launch of the government’s pre-primary education, specific objectives and new ETP, which has initiated new policy dia- expectations are different for various stakehold- logue and initiatives to expand access and im- ers involved in the pre-primary sector. Every prove the quality of pre-primary education. stakeholder consulted expressed that, in Tanza- MELQO is viewed as an opportunity to inform nia currently, there is an extremely varied per- the planning process for ETP operationaliza- spective on the definition of quality pre-primary tion and serve as a catalyst for further efforts education. For example, TIE, which is responsi- in advancing quality pre-primary in Tanzania. ble for development and monitoring curricula, might define quality pre-primary based on the The new policy states that the goal of pre-pri- extent to which a classroom has an adequate mary education is to prepare children for Stan- setting quality (defined by the MELQO quali- dard I. It emphasizes quality pre-primary edu- ty module constructs of language development, cation through adequate teaching and learning curricular areas, lesson planning, positive inter- methods and materials, relevant curriculum actions, etc.). On the other hand, PMO-RALG, and teacher training, and strengthened quality which is responsible for infrastructure and allo- control and assurance. The government’s pri- cation of teachers, may focus on whether class- ority on improved 3Rs achievement presents rooms have adequate structural quality (defined an opportunity to improve the quality and by the MELQO quality module constructs of measurement of early literacy and numeracy personnel qualifications, the safety and organi- in quality pre-primary school to set the foun- zation of space and materials of physical envi- dations for quality learning of 3Rs in primary ronment, etc.). Meanwhile, MoEVT, which is school. Unfortunately, the increased focus on responsible for overall supervision of the sector, quality in early education currently begins at may focus on system quality (defined by the Standard I and has mostly overlooked the role MELQO quality module constructs of engaging of pre-primary education in providing early families and communities, teacher knowledge, literacy and numeracy. Stakeholders feel that in-service training and management, etc.).

Measuring Early Learning Quality and Outcomes in Tanzania 13 Expectations for what children should be 3. Global comparability is valuable, but national learning and what and how teachers should be focus is the priority. The majority of stakehold- teaching also vary at the service delivery lev- ers expressed the view that, while the primary el. For example, teachers, many of whom have value of early learning measurement was for not received training in early childhood devel- national policy planning and implementation, opment, may not be familiar with learning and there is also value in having evidence that is teaching methodologies appropriate for 5- and globally recognized. Through its experience us- 6-year-olds. Parents often expect a very “ac- ing global learning measurement modules with ademic” setting for pre-primary where their EGRA/EGMA, MoEVT saw value in conduct- children learn English and obedience. They ing research that is internationally recognized may not consider play-based learning activities and based on global expertise. to be high quality, believing that high-quality pre-primary should look the same as quality However, there is also some hesitance to par- primary education. ticipate in research that may lead to Tanzania being ranked and compared to other countries. 2. National data on learning outcomes and qual- Often, when the concept of global comparabil- ity are needed. Stakeholders believe that, with ity arises (related to both MELQO and more recent progress in major policy decisions, there generally with other international measures), is now more than ever a need to have data to stakeholders express concern about being (1) inform operationalization of detailed sector compared to other countries. Some see the val- planning and (2) show the impact of these new ue in learning from and sharing lessons with policies. Tanzania urgently wants data on ear- other countries, especially other countries with ly learning outcomes and pre-primary quality. similar contexts (such as Kenya), while others Until now, no national-level data have existed are hesitant to be rated and compared to other on early learning or school readiness outcomes, countries and contexts. and thus having a nationally representative snapshot of the overall skill level of children at 4. The role of assessment in pre-primary education school entry will be extremely valuable. will need to be carefully defined. Stakeholders have mixed thoughts on exactly how children In addition, stakeholders would like to see the in pre-primary should be assessed in the future. links between pre-primary quality and learning With leadership from NECTA, as well as exter- outcomes for medium-term policy planning as nal technical support from donors, a compre- well as longer-term integration of measurement hensive national assessment system is emerging in the pre-primary system. As the government for primary and secondary education. While is embarking on new initiatives to improve some of this experience could be translated into the quality of pre-primary (including the 2016 pre-primary, stakeholders also recognize that launch of a new pre-primary curriculum, and children in pre-primary will need to be treated plans to improve pre-primary teacher training differently than their older counterparts. programs—including possible alternative certi- fication programs), understanding how teacher In addition, Tanzania’s experience with 3Rs background, curriculum, infrastructure, etc. assessment has taught the country that transi- predict children’s achievement will be extreme- tioning from an internationally supported re- ly valuable for decision-making around where search study assessing children’s learning out- to concentrate efforts and investments. comes and integrating learning assessment into

14 Measuring Early Learning Quality and Outcomes in Tanzania the national system can be challenging. NEC- Division at MoEVT has acknowledged the TA’s 2015 pilot assessment was generally based need to update the system to emphasize quality on the EGRA and EGMA assessments used more than quantity. Furthermore, with plans in 2013 (with technical support from USAID/ to launch the new curriculum in 2016, the gov- RTI), but has not been validated to show com- ernment will need to consider how to update parability across time or construct validity of and adapt the existing standards, which are the new measures. To ensure comparability for currently aligned to the 2007 pre-primary cur- payment-by-results funding (supported by the riculum. In addition, school inspectors, head World Bank, the UK Department for Interna- teachers, and district education officers will tional Development (DFID), and the Swedish need to be oriented on the new curriculum and Embassy), USAID commissioned RTI to sup- teacher’s guide. National studies have shown port the government in conducting a compara- that quality assurance officers/inspectors very ble 3Rs assessment in 2016. Moving forward, often provide inappropriate advice to teachers. national stakeholders and partners will need to work together to agree on a streamlined ap- Stakeholders see the MELQO study and mod- proach to reconcile the need for valid data that ules as an important opportunity to catalyze accurately portray the situation while still fol- improvement in quality assurance. In Tables 3 lowing a feasible approach that will meet na- and 4, the areas measured by MELQO MOD- tional priorities. EL (socioemotional skills, pre-academic skills such as language, pre-literacy and pre-numer- 5. Inspection and quality assurance needs a more acy, and areas that support learning across defined approach based on children’s learning multiple domains, such as executive function, needs in pre-primary, and MELQO can help persistence, self-regulation, and approaches to shape this. While a system is somewhat in learning) and quality (environment and physi- place for quality assurance, there is room for cal setting, family and community engagement, improvement. The quality standards and ex- personnel, interactions, structural support, pectations for a pre-primary classroom should inclusiveness, program structure and curricu- cater to the needs of 3-, 4-, and 5-year-olds lum, and health and hygiene) are outlined and and will be different than those standards in aligned with the existing documents in Tanza- a primary classroom. The Quality Assurance nia that define learning needs and quality.

Measuring Early Learning Quality and Outcomes in Tanzania 15 Table 3: MELQO Child Development and Learning and Linkages to Old and New Tanzanian Curriculum

B. Draft quality framework A. Old national pre-primary for pre-primary (expected to Construct measured by MELQO curriculum (2007) be launched 2016) Early math skills • Content area 5.3: Mathematics • Child competency 2.6: Foundational skills in mathematics Interpreting numbers, Counting include skills such as knowing how to distinguishing different Identifying concept name numbers, compare quantities, dimensions and shapes, of numbers, shapes of and perform basic addition, all which developing positive outlook numbers, actions of have been shown to be important on mathematics, and using numbers, various shapes, in developing later math skills. The mathematics in daily lives. times. MELQO assessment modules includes verbal counting, producing a specified set/number of objects, shape and spatial awareness, and comparison and number identification.

Early language/literacy • Content area 5.1: Language • Child competency 2.1: Foundational skills in literacy include Development of Communication relating oral language to written communication skills in Speaking, listening, language, phonological awareness, everyday life. These skills are drawing, reading photo. print knowledge, and oral language, listening, speaking, writing, as well as alphabet knowledge, and reading. Includes Kiswahili • Child competency 2.2: listening comprehension, and name and English learning activities. Recognition of the writing. environment Identifying objects found at home and school, names of things, names of living things, different melodies and sounds.

• Child competency 2.4: Reading Identifying concepts in print, sounds of letters, names of letters, syllables.

• Child competency 2.5: Writing Recognizing foundational/initial steps of writing, recognizing shapes of letters, creating letters.

16 Measuring Early Learning Quality and Outcomes in Tanzania B. Draft quality framework A. Old national pre-primary for pre-primary (expected to Construct measured by MELQO curriculum (2007) be launched 2016) Socioemotional development • Content area 5.4: Elimu ya • Child competency 2.7: This construct refers to skills that Jamii Relationships facilitate children’s competence in Knowledge, interpretation of Performance in social relationships and interactions how behavior and practices cooperation, self- with others including peers, affect people’s environments; awareness, obedience, family members, and teachers. learning different skills to identifying rights and Socioemotional skills that can be appreciate and communicate responsibilities. assessed through direct assessment through art; instill include social cognition and determination, perseverance, emotional knowledge. These skills cooperation, love, respect for may help facilitate children’s social society. functioning in school. Parent/ teacher reports of social/emotional • Content area 5.5: Values and development focus on children’s social beliefs behavior and emotional well-being. Prepare child on different principals including tolerance, responsibility, honesty, justice, and value of life. Includes character and games and religion lessons.

Memory and attention (executive functioning) Refers to skills that support learning in math, literacy, and other areas. Children’s executive function refers to the ability to focus their attention, ignore distractions, and move from one task to another. Self-regulation refers to the ability to control behavior and emotions.

General knowledge Refers to general knowledge that a child knows, including identity and the name of his/her community.

Not directly included in MELQO • Content area 5.2: Science • Child competency 2.3: instrument, but part of system Understanding of the real Creativity world in which children live, Drawing, pottery, learning concepts, skills, creating various items, and processes in conducting making different things, simple experiments with singing songs, playing objects existing in the local games, using primary environment. colors.

Measuring Early Learning Quality and Outcomes in Tanzania 17 Table 4: MELQO Quality Linkages to Tanzanian Quality Assurance System

C. Draft TIE Competency Framework Domain measured by A. MoEVT National Basic B. Guidelines for for Pre-Primary (expected MELQO Education Standards School Supervision to be launched 2016)

Environment and • Infrastructure standard • Characteristics physical setting Sufficient and appropriate of good quality Includes safety, school buildings for school organization of space effective teaching and Conducive and materials. learning; availability of teaching playgrounds and facilities and learning for sports and games; environment that sufficient and appropriate is child friendly. land according to the size of school; effective school physical facilities set up and efficient land use; adequate furniture for effective teaching and learning; a healthy and clean environment.

• School environment standard Secured and learner protecting environment, appropriate distance to school.

Family and community • Teaching and learning • Teacher competency engagement processes standard 1.5: promote Includes supporting Effective parent/ cooperation positive parenting and caregiver relationships, Engaging stakeholders child development based on mutual respect in developing early practices; engaging and understanding. education, promoting families and communication about communities in the children’s growth program. and development, developing relationships and collaborating with various stakeholders to support development of children with special needs.

18 Measuring Early Learning Quality and Outcomes in Tanzania C. Draft TIE Competency Framework Domain measured by A. MoEVT National Basic B. Guidelines for for Pre-Primary (expected MELQO Education Standards School Supervision to be launched 2016)

Personnel • Teacher standard • 5.3 Focus of • Teacher competency Includes knowledge, Well-trained and school supervision 1.1: knowledge skills, qualifications; competent teachers, Curriculum Enabling children supervision and teachers provided with management to gain knowledge training of teachers and regular professional and school based on capabilities, supervisors / directors. support, motivated and management using curriculum, committed teachers. (also relevant to identifying stages of program structure child development, • Quality assurance and curriculum). conducting research standard on quality. Effective advisory and • 5.4: Provision of professional support administrative and • Teacher competency to school, effective professional support 1.2: using teaching inspection of schools. by supervisor skills identify and support Using teaching and underperforming learning techniques teachers, facilitate based on children’s age professional and needs. (Also see mentorship of new Inclusiveness below.) teachers.

Interactions • Teaching and learning • Teacher competency Includes positive processes standard 1.4: follow the interaction, encouraging Learning that reflects the rules/procedures of reasoning, feedback/ learner’s sociocultural teaching scaffolding. context; teaching and Have good morals, learning activities use appropriate that are planned with language, acceptance consideration of learners’ in the community. diversity; competence- based approaches with a special emphasis on critical thinking; appropriate teacher workload that ensures quality teaching and learning; sufficient time devoted to learning; regular, relevant, and efficient assessment of learning; effective teacher-learner parent/ caregiver relationships, based on mutual respect and understanding.

Measuring Early Learning Quality and Outcomes in Tanzania 19 C. Draft TIE Competency Framework Domain measured by A. MoEVT National Basic B. Guidelines for for Pre-Primary (expected MELQO Education Standards School Supervision to be launched 2016)

Inclusiveness • Child rights standard • 5.5.2 Special needs • Teacher competency Includes meeting the Effective prevention education 1.2: Using teaching needs of all children. of child abuse and Ensure schools skills labor, effective support enroll children Includes catering to for most vulnerable with disabilities children with special children. and schools needs. are adequately equipped with facilities for such disabilities.

Program structure • Curriculum standard • 5.3 Focus of • Teacher competency and curriculum; Well-structured school supervision 1.6: Standard guide comprehensiveness of curriculum framework, Curriculum Maintaining daily services appropriate, relevant, management schedule, managing Includes basic program and user-friendly and school school regulations. characteristics; language curriculum. management. (Also see Finance development; curricular below.) areas (math, science, • Teaching and learning arts); Child Choice. materials and equipment standard (from national Connections with other curriculum) Time to sectors including social study and number of protection and child periods (five 20-minute protection. sessions per day, five days week).

Health and hygiene • Care and support • 5.5.4 Cross- • Teacher competency Promotion of health services standard cutting issue 1.7: Managing the and nutrition practices; Effective school feeding Issues to be health and safety of healthy environment. programs; effective addressed during the child health services. supervision Provide first aid, include gender, supervise children’s • Environmental disability, HIV/ nutrition, identify education standard AIDS, and specific ways to Effective conservation of environmental identify children in school environment. protection. various situations.

Finance • Financial audit standard • Teacher competency Income sources and Properly managed 1.6: Standard guide expenditure categories finance and materials. Managing income/ and estimates expenditures.

20 Measuring Early Learning Quality and Outcomes in Tanzania 6. In the future, several options exist to integrate satellite, privately provided, and other pre-pri- the MELQO modules into the national M&E mary programs. To ensure that all children are system. After the 2016 national study, stake- accessing quality learning opportunities, re- holders considered several different approach- gardless of the type of program, it will be help- es Tanzania could take for continued measure- ful for the government to monitor and enforce ment and monitoring of early learning quality basic quality standards. and outcomes: It would not be feasible in terms of cost or lo- a. Sampling approach for national research gistics for teachers, district education officers, or quality assurance officers to use the full set Using the validated MELQO modules for re- of MELQO modules in their current form. search studies in Tanzania will provide reliable However, the modules can be adapted for data to inform pre-primary policy planning. In practical classroom-level use. In the immedi- 2015-2016, the government is implementing ate and medium term, it may be more realistic the modules with the support of an interna- for adapted modules, such as a checklist with tional firm. Eventually, a sustainable nationally guidance, to be used primarily at the classroom led approach will be needed. and school level as a guide for teachers. In the long term, once the overall EMIS and quality One risk stakeholders have identified with re- assurance systems in Tanzania are strength- search studies is that sometimes results may ened and streamlined, data from the modules not be translated into improvements in the sys- could be integrated into a national system. tems. Some have expressed concerns that re- search studies with international support may c. Monitor every child’s learning not prioritize practical application in Tanzania. The government and partners can encourage Several stakeholders suggested that MELQO national researchers from Tanzanian universi- modules could be linked to the existing assess- ties to use MELQO modules in their own re- ment system to monitor the learning outcomes search. In addition, the capacity of members of every child. However, the majority felt that of the Policy and Planning Division of the Mo- it is not feasible to test every child given the EVT can be strengthened to ensure government length of test and training required to imple- officials are equipped with the skills to conduct ment the MODEL modules. In addition, there and analyze national research. are ethical considerations for assessing individ- ual children. It is considered ethical to screen b. Monitor the quality of every classroom for developmental delays or learning issues only if help is available to address any delays Linking the MELE module to the existing discovered. Identifying a specific score or cut- quality assurance/inspection system could help off is not possible with the MELQO modules, PMO-RALG and MoEVT monitor the quali- which are designed as population measures ty of learning environments and provide bet- to identify patterns and gaps with groups of ter support to address gaps in quality service children. provision. Various models for pre-primary education exist in Tanzania, and as the ETP The MELQO modules were designed for mea- is operationalized a large number of children surement at scale and not intended to make will be entering the system through traditional, decisions about specific classrooms, teach-

Measuring Early Learning Quality and Outcomes in Tanzania 21 ers, or children. Screening for developmental The options described above may not be mu- delays of individual children is very important, tually exclusive. In planning for future integra- but must be undertaken carefully and in close tion of early learning measurement, the gov- collaboration with health care providers and ernment will need to consider how it prioritizes educators who will provide services to treat the various uses of MELQO, from providing delays. In the short and medium term, it may data to inform planning, to serving as mod- not be feasible to plan for national assessment ules for service providers and those directly of pre-primary children until challenges in the supporting service providers (quality assur- overall M&E system (including NECTA and ance officers, district education officers, ward EMIS) are addressed. In the long term MoEVT, education coordinators), to framing a general in coordination with MCDGC and MoHSW, approach to quality early learning and teach- can consider what this might look like. In ad- ing. An evolving hybrid of the various options dition, as Tanzania gains more experience in described above may be best. In the following measuring early learning, perhaps it will be section, several key recommendations are de- feasible in the future to identify a common core scribed to plan for effective and sustainable set of items in the MODEL that can be used at measurement of early learning in Tanzania. the classroom level.

22 Measuring Early Learning Quality and Outcomes in Tanzania V. Recommendations

 Encourage continued A clear mechanism is also already in place to pro- government leadership and mote government leadership of MELQO activities. engagement with national After the launch of MELQO in Tanzania, the Mo- stakeholders EVT permanent secretary appointed a MELQO Task Force, with representatives from the Primary Regardless of which modalities MELQO is inte- Section, the Quality Assurance Division, and the grated with moving forward, it will be essential to Policy and Planning Division of MoEVT, as well as have strong government engagement. In general, representatives from PMO-RALG, TIE, and NEC- MELQO should be a nationally driven initiative, TA. The Task force is responsible for coordinating with opportunities for Tanzanian input and leader- MELQO activities on behalf of the government. ship throughout the process. Multiple stakeholders have recommended that, during the entire study process in 2015-2016, from planning and adapta-  Design research studies that tion to analysis, synthesis, and dissemination, there can directly influence system should be opportunities for representatives from improvements Tanzania to contribute their expertise and perspec- tive as well as build their capacity. Some interview- While the first national study will focus on a set ees expressed concern that in the past they have felt of specific research questions and will be valuable excluded from research efforts involving interna- to initiate strong measurement in early learning, tionally supported studies, and they expressed the stakeholders acknowledge that one study alone opinion that Tanzanian stakeholders should play cannot answer all of the country’s questions. Ad- an integral role in the national study. ditional studies in the future will have the potential to reveal other important trends in the pre-primary The government, with leadership from MoEVT, sector. With that said, besides the challenges asso- is already actively committed to improving ear- ciated with cost and logistics, stakeholders feel that ly learning. The government’s Short-Term Plan of the major risk to strictly using MELQO modules Action for Pre-Primary Education 2015/2016, ap- to conduct research is that findings will not be ef- proved by the commissioner of education in June fectively synthesized and communicated to decision 2015, outlines concrete areas to prepare for the op- makers. It will be essential to use results of future erationalization of the ETP’s goals for the pre-pri- research to identify areas in which specific actions mary sector. Implementation of MELQO is one of are needed. Table 5 presents several key research the nine key actions and complements many of the questions Tanzanians feel would be helpful to an- other priority areas. swer and indicates how answering these questions can lead to system improvements.

Measuring Early Learning Quality and Outcomes in Tanzania 23 Table 5: Examples of How Research Using MELQO Instruments Can Lead to Improvements in the Pre-Primary System in Tanzania

Research questions How this will inform policy/program improvement Do children’s scores on Evidence that shows how achievement in pre-primary leads to success the MODEL at the start in primary school will make a strong case for investment, particularly of Standard I predict for parliamentarians and high-level officials and donors. Furthermore, primary achievement understanding which specific areas have the strongest link to early primary (Standard I attendance, achievement could show the government exactly where to invest its scarce 3Rs assessment scores, resources to achieve the desired impact. etc.)? With recent national emphasis on 3Rs and 3Rs assessment by the Which items in the government and donors, there has been a major political priority placed on MODEL and quality achievement in early grades of primary. While a small clique of ECD experts measures predict and technicians in Tanzania understand the links of pre-primary success children’s achievement in to primary success, there is a need for national evidence to broaden policy school in Standard I or dialogue and action within the rest of the Tanzanian education sector. A Standard II? longitudinal study that could show the impact of pre-primary on subsequent learning on 3Rs would provide extremely valuable evidence to convince policymakers to prioritize concrete efforts in pre-primary.

What is the overall skill Evidence on the status of cohorts of children will inform teachers, TIE, and level of children entering others where children are so that they can cater teaching to their specific Standard I? learning needs. Understanding the specific gaps in, for example, pre-literacy could facilitate the development of improved training and support to teachers At what ages do children in promoting children’s early literacy. In addition, this evidence could be typically gain specific particularly useful as the government develops its implementation strategy for competencies? the ETP to best support 3-, 4-, and 5-year-olds in the system (currently the new pre-primary curriculum is for only one year, but does not specify an age).

Does higher quality lead Understanding the relationship between quality and learning can help to more gains in learning identify gaps in teachers’ pedagogies, which could lead to improved training in pre-primary? and support to ensure that teachers know how to conduct age-appropriate learning activities and have adequate teaching and learning materials.

How do children do Longitudinal evidence, which to date does not exist in Tanzania (or in over time? Do MELQO the majority of developing countries) could inform policymakers and MODEL or quality scores practitioners about the long-term consequences of the age at which children predict success in later acquire certain skills and competencies. school and life?

How do quality and Evidence on how quality and outcomes may differ by setting would have learning outcomes major implications for where the government and donors will choose differ by setting (public to invest as the new ETP is operationalized. Understanding the scope of vs. private, traditional different settings may also inform improvements to the existing standards for pre-primary classrooms early education. vs. satellite, urban vs. rural) or by teacher If it can identify particular strengths and weaknesses of different types background/experience? of settings and/or service delivery models, the government will be better equipped to provide the necessary support to the various settings.

24 Measuring Early Learning Quality and Outcomes in Tanzania  Strategically communicate and need to be carefully communicated to stakeholders widely disseminate MELQO at all levels. results

Stakeholders have expressed the need for a strong  Build capacity and communication and dissemination strategy so that understanding of measuring a wide scope of stakeholders can have access to and supporting early learning MELQO results and use the findings in their re- spective agencies/positions. Results should not only The MELQO findings alone will not improve qual- be communicated at the national level but also at ity of pre-primary education in Tanzania. But hav- the community level, so that schools understand ing evidence on how children are learning and how how teachers and children are doing. Based on quality affects how they learn will be an important past experience with the 3Rs assessment, govern- step forward in beginning to shift mindsets at mul- ment representatives see a need for targeted com- tiple levels around the needs for pre-primary qual- munication of national study results to cater to ity improvement. Sensitization and training for a specific audiences (including civil society organi- range of stakeholders to strengthen the attitudes, zations, district/ward education officials, teachers, knowledge, and skills on both how to measure and parents). One important opportunity to share and what to do with what is measured to improve MELQO findings with a large group of stakehold- learning will be necessary: ers is National Education Week, which is held in May each year and has the audience of the media, a. National ECE leaders: TIE, PMO-RALG, the general public, parents, teachers, and local, re- NECTA, MoEVT, and national leaders need gional, and national education leaders. to come together to synthesize MELQO findings and discuss options for a way for- ward. The agencies can draw from their  Strike a balance in aligning previous experience with quality assurance, while also differentiating M&E, and assessment, but given the rela- measurement in pre-primary tively nascent position of monitoring early vs. measurement in early learning in Tanzania, some technical assis- primary tance may be required.

Stakeholders feel that there is an opportunity to b. Teachers: Currently in Tanzania, approxi- align 3Rs and pre-3Rs, not only in curriculum and mately half of pre-primary teachers are not training but also in framing and communicating qualified (i.e., do not meet the official re- the new concept of measuring early learning to the quirement of a Grade A teaching certificate). public. However, there is also a need to differen- And even those who are qualified are not tiate 3Rs and MELQO. With NECTA efforts to necessarily trained on all areas of learning publish primary learning outcomes online as well for young children. The MELQO module as extensive 3Rs dissemination efforts, parents now will be valuable for the government in con- might expect to receive a score for their child who sidering strategies to improve the teacher is assessed in pre-primary. While the exact mecha- qualification and professional development nism for measuring pre-primary children’s develop- system in Tanzania. While the new pre-pri- ment still needs to be defined, it will not look ex- mary curriculum is already in the final actly the same as the 3Rs assessment, and this will stages of finalization, training on the new

Measuring Early Learning Quality and Outcomes in Tanzania 25 curriculum has not yet been developed, pre- for parents and communities that the MOD- senting a timely opportunity to incorporate EL module is not intended to test children on MELQO into a new training program that whether they “pass” pre-primary, but rather will hopefully reach all pre-primary teachers help teachers and parents understand a child’s in the near future. level of development. Existing mechanisms, such as parent-teacher associations, should Measuring the quality of early learning be leveraged to ensure that parents play an environments can also inform the govern- integral role in improving pre-primary quali- ment’s plans for a pre-primary professional ty. Different information dissemination strat- development and certification plan (part of egies, including radio and TV programming, the Short-Term Plan of Action for Pre-Pri- could be helpful in sharing MELQO findings mary Education 2015/2016). MELQO data to parents and communities. For example, will help Tanzania understand the profile using the parent report can help parents be- of “qualified” and “unqualified” providers come more aware of their own child’s devel- who are currently teaching in pre-primary opment and learn ways they can better pro- classrooms. This will help MoEVT, TIE, and mote development. TIE has suggested that a the National Council for Technical Educa- curriculum handbook for parents could be a tion (NACTE) understanding options for a potential option to help sensitize parents on pre-primary professional development plan. quality pre-primary.

c. District education officers, ward education e. Local researchers and research initiatives: coordinators, quality assurance officers, and As recommended above, MELQO will head teachers: Currently, local education be valuable in future national research in officials monitor quality of pre-primary the Tanzania. Researchers from the University same way they monitor Standards I-VII and of Dodoma, Aga Khan University, and the have limited knowledge on early childhood. University of Dar es Salaam, among others, In addition to incorporating MELQO into are already actively engaged in researching professional development for pre-primary pre-primary education. Partnering with na- teachers, training of school inspectors and tional researchers and ensuring that they are ward education officers, who currently have equipped with the skills to conduct rigorous little knowledge of what is appropriate for research using the MELQO modules will young children’s development, will be crit- promote sustainability. ical. The Quality Assurance Department at MoEVT has suggested that training materi- f. Civil society organizations, the private sec- als such as pictures, fliers, and videos could tor, NGOs, and advocacy groups: Besides be helpful modules in sensitizing ward edu- government stakeholders, non-governmen- cation coordinators and inspectors on mea- tal stakeholders, who make up a significant suring early learning quality. part of the pre-primary subsector, will also be important to sensitize on early learning d. Parents: Parents and community members measurement. They should also be sensi- will play an important role in the integra- tized on how to make use of information tion and uptake of MELQO. It will be im- gathered by modules to promote advoca- portant to sensitize parents on the purpose cy and further awareness on how to foster of the modules and set realistic expectations quality early learning.

26 Measuring Early Learning Quality and Outcomes in Tanzania  Develop complementary The pre-primary subsector in Tanzania is experienc- materials and modules that are ing an exciting time, with the new ETP, a new Ear- practical and user-friendly ly Education Curriculum, and a growing emphasis from civil society and other partners. Coordinat- In their current form, the MELQO modules would ing early learning measurement efforts with other be challenging to implement across every sin- ongoing and planned initiatives within the govern- gle pre-primary class in Tanzania. With that said, ment’s Short-Term Plan of Action for Pre-Primary synthesis of the upcoming findings from the 2016 will be beneficial. Some examples include: MELQO study as well as further review and testing of the modules in various pre-primary settings can a. Further align to new curriculum: As Tables lead to a better understanding of how to measure 3 and 4 showed, the constructs measured by and monitor how children are learning and what MELQO are already directly linked to dif- they are learning. Stakeholders have suggested sever- ferent areas within the new quality frame- al complementary modules that could be developed, work drafted by TIE. At the time of draft- including: ing this brief, the government had not yet publicly shared the new draft curriculum or a. Practical classroom level module: At a syllabus. When they are available, it will be classroom level, teachers need resources to helpful to further align the new curriculum understand how their quality affects chil- and syllabus to the MELQO instruments. dren’s learning at development. A classroom During an ECE stakeholders meeting in No- instrument could be used as a monitoring vember, TIE explained that the new curricu- module for head teachers or ward education lum will be dynamic and flexible, and it be- coordinators, or as a module for teacher lieves that the MELQO results will inform self-evaluation. further refinements to it.

b. Checklist for inspectors: The government b. Seek synergies with partners: As MELQO has recognized the need to shift the current gains momentum, there will be increased approach of inspection to focus on school demand for measuring early learning out- improvement and quality assurance. Exist- comes and quality. For example, represen- ing inspection modules are currently more tatives from the EQUIP-T School Readiness focused on quantity than quality. Quality program (a DFID-supported quality edu- assurance officers and ward education coor- cation program providing condensed alter- dinators can use elements of the MELQO native early education program to children MELE module to not only inspect but also before they enter primary) and the Fursa to give advice to teachers and heads of kwa Watoto (an initiative to test two quali- schools to improve the conditions of the ty-enhanced cost-effective government mod- pre-primary environment and pedagogical els for pre-primary education, with impact practices. evaluation to document improvements in school readiness and Standard II learning outcomes) have expressed interest in using  Link to ongoing and planned MELQO modules. initiatives in the existing pre- primary system c. Include pre-primary in planned improve- ments of the education-sector-wide M&E

Measuring Early Learning Quality and Outcomes in Tanzania 27 system: Currently, EMIS does not collect framework to improve data collection and data on learning outcomes at any level. NEC- management through an integrated EMIS. TA has a national database on standardized While concrete plans are still under devel- assessments for primary and secondary opment, both child-level and classroom-lev- school students, but this is not currently inte- el indicators from MELQO could eventually grated into EMIS. MoEVT (with donor sup- be integrated into this M&E framework. port) plans to develop a sector-wide M&E

28 Measuring Early Learning Quality and Outcomes in Tanzania References

Government of Tanzania. Basic Education Statis- Reynolds, A.J., J. A. Temple, D. L. Robertson, and tics (BEST), 2014. E. A. Mann. 2001. “Long Term Effects of an Early Childhood Intervention on Educational Heckman, J. J., and D. V. Masterov. 2007. “The Achievement and Juvenile Arrest- a 15-Year Productivity Argument for Investing in Young Follow-Up of Low-Income Children in Public Children.” Applied Economic Perspectives Schools,” Journal of the American Medical and Policy 29(3): 446–493. Association 285 (18): 2339-46.

Lynch, R.G. 2005. “Early Childhood Invest- ments Yield Big Payoff.” Policy Perspectives. San Franciso, CA:WestEd.Basic Education Statistics.

Measuring Early Learning Quality and Outcomes in Tanzania 29

Annexes

Annex A. Functions and organization structure of the Ministry of Education and Vocational Training (MoEVT) (Approved by the President on 12th February, 2015)

MINISTER NATIONAL EDUCATION ADVISORY COUNCIL PERMANENT SECRETARY POLICY AND PLANNING DIVISION DIRECTOR FINANCE AND ACCOUNTS UNIT ADMIN & HUMAN RESOURCES MANAGEMENT DIVISION CHIEF ACCOUNTANT DIRECTOR

PROCUREMENT MANAGEMENT UNIT INTERNAL AUDIT UNIT PRINCIPAL SUPPLIES OFFICER

GOVERNMENT COMMUNICATION UNIT EDUCATIONAL SECTOR PERFORMANCE PRINCIPAL OFFICER COORDINATION UNIT DIRECTOR REGIONAL AND INTERNATIONAL EDUCATION AFFAIRS COORDINATION UNIT NATIONAL COMMISSION FOR UNESCO DIRECTOR EXECUTIVE SECRETARY

LEGAL SERVICE UNIT AGENCY FOR DEVELOPMENT OF PRINCIPAL LEGAL OFFICER EDUCATION MANAGEMENT (ADEM)

BASIC EDUCATION DEVELOPMENT OFFICE COMMISSIONER FOR EDUCATION BASIC EDUCATION DEVELOPMENT PROGRAMMES AND BASIC EDUCATION TEACHER TRAINING UNIT PROJECTS COORDINATION UNIT DIRECTOR DIRECTOR

HIGHER TECHNICAL AND SCHOOL QUALITY SCHOOL PHYSICAL INFO., COMM., TECH. EDUCATION VOC. EDUC. TRAIN ASSURANCE ACCREDITATION EDUCATION AND E-LEARNING DIVISION DIVISION DIVISION DIVISION POLICY DIVISION DIVISION DIRECTOR DIRECTOR DIRECTOR DIRECTOR DIRECTOR DIRECTOR

HIGHER TECHNICAL PRIMARY SCHOOLS GOVERNMENT PRE-PRIMARY AND ICT EDUCATION EDUCATION QUALITY ASSURANCE SCHOOLS PRIMARY EDUATION SECTION DEVELOPMENT TRAINING SECTION SECTION ACCREDITATION POLICY SECTION SECTION TECHNOLOGY NATIONAL VOCATIONAL SECONDARY SCHOOLS SECONDARY ENABLED QUALIFICATIONS EDUCATION QUALITY ASSURANCE NON-GOVERNMENT EDUCATION LEARNING SECTION FRAMEWORK TRAINING SECTION SECTION SCHOOLS POLICY SECTION DEVELOPMENT ACCREDITATION LIBRARY SERVICES SECTION MANAGEMENT QUALITY ASSURANCE SECTION ADULT & NON-FORMAL AND SUPPORT MGT. SUPPORT EDUCATION INFORMATION SERVICES SECTION SECTION PHYSICAL POLICY SECTION FACILITIES ZONAL QUALITY STANDARDS SPECIAL NEEDS ASSURANCE OFFICE SECTION EDUCATION Proposed or already an Agency POLICY SECTION DISTRICT QUALITY ASSURANCE OFFICE DIVERSITY POLICY WARD QUALITY SECTION ASSURANCE OFFICE

Measuring Early Learning Quality and Outcomes in Tanzania 31 SOCIAL HEALTH HEALTH DISTRICT DIVISION WELFARE SERVICES SERVICES SERVICES SERVICES REGIONAL SERVICES NUTRITION NUTRITION HEALTH, SOC. HEALTH, WELFARE AND WELFARE LEGAL LEGAL LEGAL SECTION DIVISION SUPPORT SERVICES LITIGATION LAND URBAN SECTION DIVISION DIRECTOR URBAN DEV URBAN DEV PLANNING & MANAGEMENT MANAGEMENT DEVELOPMENT Executive Agency UNIT SERVICE MANAGEMENT IMPROVEMENT INTERNAL AUDIT INTERNAL UNIT DAR RAPID TRANSIT AGENCY DAR RAPID TRANSIT FINANCE AND ACCOUNTS UNIT POLICY AND PLANNING DIVISION POLICY UNIT TRACKING AND FINANCE INSPECTORATE SOCIAL SECTOR SECTOR SECTOR DIVISION SERVICES ECONOMIC & PRODUCTIVE COORDINATION MINISTER PERMANENT SECRETARY ICT ICT INFRA- COMM. INFOR. & DIVISION STRUCTURE TECHNOLOGY APPLICATIONS DEV DEV INFRA- INFRA- URBAN INFRA- RURAL DIVISION SECTION DIRECTOR RESEARCH PLANNING & STRUCTURE STRUCTURE STRUCTURE DEVELOPMENT MANAGEMENT DIVISION ADULT & ADULT & ADULT DIVISION DIRECTOR & PRIMARY EDUCATION EDUCATION EDUCATION EDUCATION EDUCATION SECONDARY NON-FORMAL NON-FORMAL PRE-PRIMARY PROCUREMENT MANAGEMENT UNIT ADMINISTRATION GOVERNMENT COMMUNICATION UNIT GOVERNMENT COMMUNICATION ADMINISTRATION AND HUMAN RESOURCES ADMINISTRATION & LOCAL LOCAL LOCAL LOCAL HUMAN SERVICE DIVISION DELIVERY RESOURCES GOVERNANCE GOVERNMENT MANAGEMENT GOVERNMENT GOVERNMENT DIVISION SECTION SECTION BUILDING CAPACITY BUSINESS DIRECTOR REGIONAL GOVERNMENT ADMINISTRATION Annex B. Functions and organization structure of the Prime Minister’s Office–Regional Administration and Local Government (PMO-RALG) 2015) (Approved by the President on 12th February,

32 Measuring Early Learning Quality and Outcomes in Tanzania Annex C. Protocol for national c. For parents to understand their children’s stakeholder interviews development Guiding questions for national stakeholder d. For District, Regional, and National interviews Gov’t to understand how the pre-primary sub-sector is performing 1. Tanzanian priorities for early learning: Which e. For equity and to allow targeting of chil- outcomes for ECD/early learning are most im- dren most in need portant in Tanzania (for Gov’t, communities)? a. What is the goal of pre-primary education 4. Appetite for MELQO: How much do you in the country? What does Government/ think the following stakeholder groups would your specific agency want it to accomplish? support or oppose large-scale measurement b. Which child outcomes does the Govern- of early learning quality and outcomes? Men- ment/your specific agency care about most? tion each of the stakeholder groups. Discuss c. What information is most critical for the why or why not they would support and their Government/your specific agency to track thoughts on specific risks and/or opportunities at a national level? to national adoption of MELQO. Interviewees d. How important is global comparability for can speak about past experience with 3Rs and/ the Government/your specific agency? or ideas about perceptions of MELQO. a. Policymakers 2. Perceptions of good quality: What do you b. Teachers unions think a “good” pre-primary classroom looks c. Educators like? Ask interviewee to describe what, in their d. Students opinion, a “good” quality classroom looks e. Parents like. Once they do this, ask them whether they f. Media think this description would be the same if they g. Think-tanks, NGOs, etc. had responded as a teacher or a parent. How h. Teacher training colleges would parents describe “good” quality? a. What are the children doing? (prompt 5. Linkages to primary school: How and what about structure, interactions, play-based do you need to align to the primary education activities, etc.) system? b. What is the teacher doing? a. What are the expectations of what chil- c. What does the room look like? dren should know and be able to do upon d. How are the parents engaged? Standard I entry and how does this influ- ence the content and expectations for the 3. Need for monitoring and uses of MELQO: Do pre-primary education programs? you think monitoring early learning outcomes b. What does it mean to be ready for school and quality is needed in Tanzania? How do in Tanzania? you think MELQO findings could be best used in Tanzania? Open discussion about benefits. 6. Responsibilities for long-term integration: If prompts are needed, gather opinions about Which group do you think would be best fit the following uses: to use the MELQO modules? Mention each a. Monitoring module for school heads of the groups and gauge why/why not this b. Classroom improvement as feedback mod- group would be well-suited. Note more than ule for teachers one group could use the modules. What is the

Measuring Early Learning Quality and Outcomes in Tanzania 33 capacity of this group to implement large-scale 8. Feasibility of documentation and dissemina- data collection? What capacity do each of these tion: What would be the best way to techni- groups already have and what are their weak- cally document MELQO findings as well as nesses (i.e. identify specific capacity building disseminate findings to the public?Ideas could needs)? To who is this group accountable? include: technical report (publicly or not pub- a. NECTA (national exams council) licly available), online portal, school/district/ b. Quality Assurance Unit/Inspectors region-specific reports, media briefings to dis- c. Regional Education Officers cuss results, workshops/presentations for key d. District Education Officers stakeholders on results, etc. e. Private Firm f. Head Teachers 9. Overall challenges: What are the main risks g. Teachers or challenges to integrating MELQO? If not h. Specific programs (TZ 21, Fursa Kwa Wa- already discussed, gather information on ad- toto, NGO-led programs, etc.) ditional specific challenges for measuring i. Other groups? pre-primary quality and outcomes. Can also link to risks/challenges of 3Rs: 7. Feasibility of data collection: What kind of a. What went well? What didn’t go so well? measurement is feasible? How will modules be b. How were results used by the Government, used and managed? Discuss feasibility in terms teachers, and communities? of cost, logistics, and human resources. a. Is a sampling approach best or does every 10. Considerations on study design: We have lim- child need to be tested? ited amount of flexibility in what we can do in b. Who can collect the information? Could the upcoming national study. In your opinion, teachers be the enumerators or District Ed- what are the most important questions to an- ucation Officers? Or University students? swer? What type of data would be most valu- c. How/where can the information be ag- able for Tanzania? gregated? Would this feed into EMIS? a. Difference between quality and child out- NECTA? comes in: d. Would modules need to be designed to • Public vs private schools align with existing EMIS data collection • Urban vs. rural efforts or inspection efforts? Or would a • Girls vs. boys new system be created? • High-performing vs. low-performing e. What have been challenges with assess- (PSLE pass rates) ment/EMIS/inspection in the past? (for • Whether children in different regions example-delays in administering, poor are performing better than others training/guidelines of test administrators/ b. How do Kiswahili speakers fare compared inspectors, weaknesses in assessment/in- to children whose mother tongue is not spection protocol design, errors in data Kiswahili? processing, etc.) c. Relationship between child outcomes/ quality d. Anything else?

34 Measuring Early Learning Quality and Outcomes in Tanzania Annex D: List of stakeholders consulted during Phase I

Name Title Affiliation Inputs provided via National Exams Dr. Charles Msonde CEO One-on-one interview Council of Tanzania Tanzania Institute for Dr. Meena Director One-on-one interview Education One-on-one interview and Dr. Lyabwene Mtahabwa ECD Expert University of Dodoma ECE Stakeholders Group Discussion One-on-one interview and Tanzania Child Rights Eric Guga Coordinator ECE Stakeholders Group Forum Discussion One-on-one interview and Principle Education Clarence Mwinuka MoEVT ECE Stakeholders Group Officer Discussion One-on-one interview and Hawa Selemani Education Officer MoEVT ECE Stakeholders Group Discussion One-on-one interview and Senior Curriculum Tanzania Institute for Francisca Tarimo ECE Stakeholders Group Coordinator Education Discussion One-on-one interview and Curriculum Tanzania Institute for Vida Ngowi ECE Stakeholders Group Coordinator Education Discussion Monitoring and ECE Stakeholders Group Christopher Amwine Evaluation and PLAN International Discussion Learning Specialist ECE Stakeholders Group Ian Attfield Education Advisor DFID Discussion ECE Stakeholders Group Borase Chibura ECD Coordinator PMO-RALG Discussion Regional Senior ECE Stakeholders Group Rupert Corbishley Programme Officer AKF Discussion Education & ECD Advisor, Teacher ECE Stakeholders Group Ray Harris Professional Equip-TZ Discussion Development Principal Finance ECE Stakeholders Group Cyprian Iraba Management MoEVT Discussion Officer Curriculum ECE Stakeholders Group Christina Kamamba TIE Developer Discussion ECE Stakeholders Group Alistidia Kamugisha Programme Officer TEN/MET Discussion

Measuring Early Learning Quality and Outcomes in Tanzania 35 Name Title Affiliation Inputs provided via Curriculum ECE Stakeholders Group Laurence Kunambi TIE Developer Discussion Program Officer, ECE Stakeholders Group Stephen Myamba Embassy of Sweden Education Discussion ECE Stakeholders Group Yonafika Mzungu Education Officer MoEVT Discussion Education ECE Stakeholders Group Masozi Nyirenda Secretariat Development Partners Discussion Group ECE Stakeholders Group Anna Smeby Education Specialist UNICEF Discussion Canadian Cooperation ECE Stakeholders Group Douglas Taylor Education Advisor Office Discussion National Exams ECE Stakeholders Group Wellu Mpanda Registration Officer Council of Tanzania Discussion ECE Stakeholders Group Nkahiga Kaboko Education Specialist World Bank Discussion Director Teacher ECE Stakeholders Group Frederick Shuma MoEVT Education Unit Discussion Education ECE Stakeholders Group Mwansith Kitogo Coordinator PMO-RALG Discussion Primary

36 Measuring Early Learning Quality and Outcomes in Tanzania

The Brookings Institution 1775 Massachusetts Ave., NW Washington, D.C. 20036 brookings.edu