Office Design: a Study of Environment

DOCUMENT RESUME

ED 033 529 EF 001 718 AUTTIOP Manning, Peter, Ed. TITLE Cffice Design:A Study of Environment. INSTITUTION Ccmmunity Coll. Planning Center,Stanford, Calif. Put Date May 65 Ncte 154p.; Peprinted 1966 Available from Department of Puilding Science, University cf Liverpool, Liverpool, England(Thirty Shillings)

EDPS Price EDRS Price MF- 0.75 HC Nct Availablefrom EDrS. Descriptors Acoustical Environment, Attitudes, Puilding Design, Controlled Environment, Design Needs, *Pnvircnmental Criteria, *Environmental Influences, *Environmental Research, Interior Design, InteriorSpace, *Cffices (Facilities), PhysicalDesign Needs, Physical Environment,Psychological Design Needs, Research Criteria,Social Environment, Space Utilization,Surveys, Task Performance, ThermalEnvironment, Visual Environment, *Work Environment Abstract Reporting upon a study of environment which was based on the design ofoffice buildings and office space, the study formspart of a continuing program of environmental research spcnsoredby Pilkington Brothers Limited of St. Helens, England.In this report the word 'environment' is used in thesense of the sum of the physical and emotional sensationsexperienced by people within buildings. The long-termaim of this research is the academic development cfan understanding of this "total environment". This multi-disciplinarypublication is accomplished by individualreports from the four research staff members--(1)a general survey by the architect of literature relating tc office designand environment, (2) an account by the geographer of the general characteristics, population and locationof post-war office buildings in England and Wales, (3) a report by the physicist of the results ofsurveys in offices of noise, lighting, and thermal conditionsin winter and summer, and (4) a report by the psychologistof the response of the CIS (Co-operative Insurance Society,Ltd., Manchester, England) staff to their working environmentand its implications for office design. Illustrations,photographs, charts, diagrams, appendices, andan index are provided. (KK) S

.01111N..

, 4'

4' 1 4t

U S DEPARTMENT OF HEALTH, EDUCATION S. WELFARE lL OFFICE OF EDUCATION

THIS DOCUMENT HAS BEEN REPRODUCED [WILY AS RECEIVED FROM THE

PERSON OR ORGANIZATION ORIGINATING IT POINTS OF VIEW OR OPINIONS STATED DO NOT NECESSARILY REPRESENT OFFIrlAt OFFICE OF EDUCATION POSITION OR POLICY -Pe 01111aillionia Synopsis

Office design: a study of environment

Sf B(92) UDC 725.23 1965

This l ublication reports upon a study of environment which was based on the design and performance of office buildings and office space, and people's attitudes towards the offices they use. It is not strictly a research report, in the sense that the term is normally used, nor is it a design guide. Instead it sets out to provide a picture of environment ' in the round', as it is seen from the present stage of development of the Pilkington Research Unit's studies of the total environment within buildings. The report discusses some of the major components of building environment in their physical, psychologi, and design aspects and endeavours to show the inter- relation of one component with another. It gives an account of the present state of office design and environment, summarises current knowledge, examines and compares those environmental standards that have been established, aid incorporates the Unit's own contribution to an understanding of the subjects of office design and environment. This picture of environment is presented in four stages of increasing detail. The first, and most brief, account of the principal findings and their relevance to contemporary architecture is contained in Stage One (1. 'reword, written by G. Grenfell Baines). This is followed by Stage Two (Chapter One), which summarises present trends in office design and environment, and states the lessons of the project for further environmental research. Stage Three consists of a series of chapters and is in two parts. The first part commences with an introductory chapter, follows with a description of the building type and its main characteristics (including a general picture of the office buildings which have formed the 'Survey Sample' for the Unit's investigations) and continues with detailed discussions of different environmental components in isolation. These components include the location of post war office building, the spatial, visual, thermal and aural environments, and the subjective and social consequences of building environmental design. The second part of Stage Three (Chapter 10) comprises a detailed case study of the total environment within one of Britain's outstanding modern buildings. Stage Four consists of a number of appendices which describe the research methodology and tabulate the more detailed findings of the investigations. The report is primarily addressed to other research groups engaged in systematic studies of architecture. But since it indicates an approach to office design from the basis of the environment which people working inside buildings experience, it is hoped that it will also prove of interest to office managements, their designers, and others. Published in 1965 by The Pilkington Research Unit "PERMISSION TO REPRODUCE THIS Department of Building Science COPYRIGHTED MATERIAL HAS SEENGRANTED University of Liverpool in( Pter nn in P Reprinted 1966 aefseavck U it TO ERIC AND ORGANIZATIONS OPERATING Pilkington Research Unit 1965 UNDER AGREEMENTS WITH THE U.S.OFFICE OF EDUCATION. FURTHER REPRODUCTIONOUTSIDE THE ERIC SYSTEM REQUIRES PERMISSIONQf THE COPYRIGHT OWNER,",....)4.0sc olaress Price Thirty Shillings VtreckoriSck00( oC. ArckLtEcEure ScoEick coL(calE P.OBox lone4o..t .s IAP3a..

Book designed by Alan G. Swerdlow N.D.D.

Made and printed in England by Rockliff Brothers Limited 44 Castle Street Liverpool 2 Office design a study of environment

by the Pilkington Research Unit Edited by Peter Manning

U.S. DEPARTMENT OF HEALTH,EDUCATION t WELFARE OFFICE OF EDUCATION

THIS DOCUMENT HIS PUN REPRODUCED EXACTLY AS RECEIVED FROM THE PERSON OR ORGANIZATION ORIGINATING IT.POINTS OF VIEW OR OPINIONS STATED DO NOT NECESSARILY REPRESENT OFFICIAL OFFICE OF EDUCATION POSITION OR POLICY.

Department of Building Science University of Liverpool

May 1965 Stage One: Foreword

by G. Grenfell Baines,0.B.E., F.R.I.B.A., M.T.P.I.

This report aims to provide a global picture of the en- gation into people's response to their working condi- vironment in modern office buildings, and in so doing tions, and shows some of the many ways in which the brings a new approach to the evaluation of building environments architects create may influence human be- performance. Though mainly exploratory, setting the haviour. Where other research bordering this field has pattern for further detailed studies, it consolidates exist- been content to stop at the facts as they appear, this has ing knowledge which has been gleaned from many delved into possible reasons for opinions and reactions sources, exposing some of the gaps and filling some to questions. The report demonstrates that many office others. It indicates an approach to design on the basis of buildings are not designed to take advantage of know- thinking outwards from the office worker to his work- ledge already available or are designed largely on the place. basis of assumptions made by owners and architects. It The report is compiled by the Pilkington Research Unit forecasts a trend to open planning, with deeper plan which developed out of the factory research project forms and engineered environments. Evolution in design whose outcome was the book 'Design of roofs for single- is inevitably implied. Just as the previous factory re- storey general-purpose factories', a publication which port provided a needed stimulus for design ideas based aroused considerable and widespread interest. Since this on ascertained and correlated facts, this report should achievement the Unit has continued to develop a multi- give a similar service to the design of better offices. disciplinary structure and its approach to a better under- The work has been made possible by the generosity of standing of the `total environment' within buildings. Messrs. Pilkington Brothers who will support the Unit Each element of this total environment (eg, lighting, until 1967. Members of the Building Research Station, heating, ventilation, acoustics) which hitherto has been the Institute of Directors, the Liverpool School of studied in isolation is now reviewed in the context of a Architecture and Department of Building Science have total experiencein other words, in terms of the way in voluntarily given time to the committee which advises which people encounter buildings. Studies of this kind the Unit. It is hoped that the University will be in a must be of fundamental importance to the development position to take over the financial responsibility for the of the discipline of building design. work in 1967. The Unit has evolved into a multi-disciplinary team made up of an architect, a geographer, a physicist and a psychologist. The architect, besides providing the general direction of the research, was also responsible for building design aspects of the study. The geographer was concerned with climatic and locational influences, and the physicist with measurement and description of the physical environment. The psychologist undertook studies of attitudes towards, and the social and physical influences upon, the users of buildings. While developing the long term aim of a general understanding of total environment within buildings the Unit is basing its environmental studies on a number of building types, this on offices being the first. Other projects in hand, currently, are primary schools and mental hospitals. Besides providing a more comprehensive picture of the internal environment in particular building types (which should provide valuable material for design guidance), important information on the process of inter- disciplinary working will emerge, greatly helping the building design professions to develop more fruitful methods of work. The present report provides the most comprehensive estimate in existence of post-war office construction and location. The interim statement on this particular topic, published in the Research Section of theArchitects' Journal,has already been used extensively by planners and commercial research organisations. The study also demonstrates the possibilities of psychological investi- Acknowledgements

A research project of the sort described in this bookre- ceives a vast amount of help from too many people for it to be possible to give individual thanks. But special mention must be made of the managements and staffs of the CIS and other organisations whose office buildings comprised the 'Survey Sample'. It is difficult to know why they helped us: perhaps they did not realise when they letus into their buildings for the first time that we wouldcause them endless trouble and sometimes expense,occupy their time, distract their staff, and ask all manner of questions, many of which must have seemed quite im- pertinent. Yet, having discovered what was involved, they continued to help. And without the facilities theyso generously provided, these studies would have been quite impossible.

The sources of certain illustrations are acknowledged with thanks: figure 27, British Petroleum Company Ltd; figures 26, 38, 39, 73, 74, 75, 81 and the whole page plate facing page 79, Co-operative Insurance Society Ltd; figures 80, 82, 84, Design Research Unit; figure 45, E. S. and A. Robinson Ltd; figures 2, 3, 23, West Midlands Gas Board. Figures 24 and 25 are reproduced from the Japanese magazine, Public Buildings. Figure 28 is based on a drawing by Fergus Designs Ltd, figure 36 on one by the Staff Architect to Messrs. Robinson and figure 76 on drawings provided by the CIS. The original versions of figures 7 and 19 were first published in the Architects' Journal. Quotations appear by permission of the pub- lishers or authors named in the footnote references.

With the exception of figures 24 and 25, all the line diagrams in the book were prepared by David Jeffcoate. Pilkington Research Unit

Advisory Committee G. Grenfell Baines0.8.E., F.R.I.B.A., M.T.P.I.(Chairman) Peter M. Davidson u.s c. J. B. DickM.A., B.SC., F.INST.P., M.I.H.V.E. Professor R. Gardner-MedwinB.ARCH., F.R.I.B.A., M.T.P.I. Professor A. W. HendryB.SC. (ENG.), PH.D., D.SC., M.I.C.E., M.I.STRUCT.E., F.R.S.E. Thomas A. MarkusM.A., M.ARCH(M.I.T.), A.R.I.B.A. Professor J. K. PageB.A. H. B. WrightM.B., F.R.C.S.

Staff Peter ManningA.A.DIPL., A.R.I.B.A. Pilkington Research Fellow Brian WellsM.A., PH.D., A.B.PS.S. Psychologist Miss Sheila M. TaylorB.SC. Geographer Mrs J. A. MarshB.SC. Physicist

T. Takano (Japan) United Nations Fellow (October 1963-March 1964)

J. McMillan Technician

Miss Rosemary J. Ellis (1961-1963) Mrs Diane M. Goldblatt (1963) Miss Mary Burns (1964 on) Secretaries

Consultants C. B. WilsonB.SC., PH.D. Editorial J. E. BandaB.SC.(MIN.), PH.D., A.M.I.H.V.E. Surveys of the thermal environment Contents

Stage One Supplemel it 4 Questionnaire for section Foreword by G. Grenfell Baines clerks 7 used to record preferences andattitudes towards open planning Stage Two 134 Chapter 1 Supplement 5 Letter tomanagers covering: 137 Summary and Conclusions Supplement 3 Questionnaire formanagers used 17 to record preferences and attitudestowards open planning Stage Three, Part One 138 6.1 Surveys of daylight in offices The Components of Environment 141 Chapter 2 6.2 Surveys of electric ligtv in offices 142 6.3 Scale for estimating relativepercentage of Introduction 25 Chapter 3 daylight and electric light used inexperimental study of satisfactory lighting conditionsfor Ba:kground to the environmentalstudies: clerical workers 143 the building type 27 Chapter 4 7.1 Surveys of the thermal environmentin offices during winter The external environment, andthe location of 144 7.2 Surveys of the thermal environmentin offices post-war office buildings 36 during summer Chapter 5 145 8.1 Surveys of noise levels in office buildings 146 The spatial environment 39 10.1 The first questionnaire Chapter 6 survey of environmental conditions in the CISbuilding The visual environment 48 Chapter 7 Supplement 7 Letter to all membersof CIS staff covering: 147 The thermal environment 61 Chapter 8 Supplement 8 Questionnaire for allmembers of CIS staff on environmental conditions 148 The aural environment 68 Chapter 9 Supplement 9 Simplified floor plans ofCIS building on which staffwere asked to mark the Subjective and social aspects of theenvironment 74 approximate position at which theyworked 152 10.2 The second questionnaire Stage Three, Part Two survey of The Total Environment environmental conditions in the CISbuilding: Chapter 10 Supplement 10 Letter to a sample of theCIS staff covering: The total environment:a case study of the 153 Supplement 11 Questionnaire fora sample of the CIS building in Manchester 79 CIS staff on environmental conditions 154 Stage Four Index Appendices (The number before thestop refers to 157 the chapter from which the appendixis derived) 1.1 Further studies seen to benecessary 92 2.1 Schedule of separate research tasks 93 3.1 Construction costs of officebuildings 97 4.1 Schedule of post-war officespace approved, built and under construction inthe areas of certain local planning authorities 98 5.1 The influence of office sizeon the individual, and on supervisory and managerialprocesses 104 Supplement 1 Letter to clerical workerscovering: 129 Supplement 2 Questionnaire forclerical workers used to record preferences andattitudes towards open planning and also sociometric choices 130 Supplement 3 Letter to section clerkscovering: 133 Stage Two

Summary and Conclusions Chapter 1 Summary and Conclusions

Introduction owners' and architects' ideas about people's Building in 1965 is a matter of considerable social and political im- psychological needs will forma basis for fundamental portance receiving governmental attentionon a scale policy decisions about a building's form,although there previously unknown. But current interestresides largely will be little or no evidence to substantiate in the ways and them. means of construction, especially in An example of this is the argument that,in terms of speed and quantity; whether what is beingbuilt is what spatial layouts and environmental conditions, is really needed is rarely questioned. a building Although environ- should be stimulating, for in thatway it can optimally ment is a word often heard when the design ofmodern achieve its practical ends: buildings is being discussed, it tendsto in employed ' Normal consciousness, perception and loosely and imprecisely. Sometimes it is thought can seen as a side- be maintained only ina constantly changing environ- effect of design. It is rarely treated (as it is in this study) ment; when there is no changea state of "sensory as the essential basis. Yet environment is theessence of architecture. deprivation" occurs. Experiment has shownthat a homogeneous and unvarying environmentproduces An ordered assembly of existing knowledgeof environ- boredom, restlessness, lack of concentrationand re- ment in relation to building design doesnot exist, for duction in intelligence. studies have largely consisted ofseparate and unrelated This is the psychological basis for investigations deliberately creat- for example, of lighting, heatingor ac- ing varying conditions in buildings. oustics. This has been convenient for the Office blocks in understanding which each floor has thesame layout, colour, mater- of individual topics but notvery helpful to the under- ials and climate a 'Oust asking for trouble standing of the whole, for the . . . . The sort parameters interact. The of variatior. thatwe often demand instinctively on Pilk ington Research Unit has been established with the aesthetic grounds, hasa sound physiological and express long term aim of making scientific studies of the psychological basis. A change in environmentstimu- 'total environment'. Studies of the individualenviron- lates our built-in devices to perceive mental factors and their interaction and respond with one another, rapidly to significant events and efficiencyis thereby at both physical and subjective levels,are fundamental increased. It is worth paying for variety'.' to the development of architecture. This argument, which is heard The Unit's aim being long at many conferences and term, simpler objectives are put forward by many architects whenreferring to necessary for the short-term stages of the study. This open plan offices, seems to arise froma misunderstand- exploratory investigation has been centred on the design ing of what constitutessensory deprivation. The exist- of offices because sucha method of study has enabled ence of a book with that title 2 is well knownto archi- many aspects of environment to be studied in relation to tects, but few have read it and understand thatthe effects a building type which needs some fresh thought. The described are the result of fairly members of the Unit have been working gross sensory depriva- to a common tion in the sense usually meant bypsychologists. It is purpose but they have been individually responsible for true, though, that some office environments those aspects of the study for which their are very dull qualifications and lack stimulus. Whether this hasany implication for best fit them. This report is an attemptto draw together efficiency, what the optimal limits of the work of the Unit as a whole and also environmental to present a stimulation are, and how theyare to be attained, are all unified picture of environment in thecontext of office questions yet to be answered. It is perhaps, design. therefore, premature to state that `it is worth paying for variety'. Several surveys of environment in offices havebeen con- Yet if building environments have ducted in recent years by administrators, a major influence, public health good or bad, upon people's mentalstate, as well as upon inspectors, medical officers and others. Each one of the their physical comfort and, possibly,health, then this is published reports has some interest, but bydemonstrat- not something which should be left ing that it is impossible for individuals to chance, nor to in- or groups of lay- expert investigation. There existsa clear need for re- men to possess all the necessary skills and experienceto search which will establish the facts. appraise the many aspects ofan environment, they re- One problem which has to be recognised in inforce the need for a multi-disciplinary approach. any such research is that peopleare not necessarily the best judges of what they need; they havea tendency to state The building environment a preference for the conditions with which they The ramifications of building environment are most are endless. familiar. Furthermore, it is dangerousto generalise from Conditions within a building are dependenton the surroundings, circumstances and climate outside.The form I. Noble, J: The how and why of behaviour:social psychology and relationship of buildinggroups (eg, town centres) for the architect. Architects' Journal, 1963(March 6) influences the design of individual buildings. 2. Soloman, P, (Ed): Sensory deprivation.Harvard University Building Press, Cambridge (Mass.), 1961

17 one case to another without examining the differences be- would rather work in daylight than electric light, and tween sub-samples of a population. This study has their belief in the importance of daylighting is sharedby demonstrated that office staff's response to an environ- both physicians and architects. No objectivesupport for ment are likely to differ according to the respondents' this belief is known to exist, and an assessment of what age, sex, status, and experience of working in other constitutes a subjectively acceptable visual environment buildings. Sonic recent official studies of officecondi- showed that in conditions wherepermanent artificial tions may be suspect because thismatter, so far as can lighting and an unobstructed view of windows exist?, be seen, was not taken into account. people working substantially in artificial light tendto Although their effect has to be seen in its totality, the grossly over-estimate the proportion of daylightto arti- major components of an environment have,unavoid- ficial light on the working plane. The probabilities ably, to be considered scparateiy. are that windows are really valued for their -iewto the exterior and that in the second half of the twentieth Space century this is their important function. The studysuggests Today's typical office building hasa linear plan-form that daylighting provisions might with a building width of around 35 more reasonably be to 50 ft which con- made in terms of meeting subjective needs thanof att- sists, internally, of a central corridor withshallow offices aining arbitrary physical levels. on either side. The only substantial points of difference Modern technology is influencing the function of win- with many office buildings of the immediatepre-war dows in other ways. An increase in the volume of traffic period are external cladding and fluorescentlighting in city centres has created the problem of external systems. traffic noise which is transmitted through the external fabricto The most likely explanation for the continuationof this the interior of the buildings, most of it going through type of design is habit ; a prolongation, possibly, of the the windows. Windows are usually kept shut in orderto traditional search for daylight. The feasibility of natural limit the annoyance but this affects their performanceas lighting in city centres is rarely questioned,even for first ventilators. In many modern buildings the size of wind- floor offices in densely built-up and overshadowedareas, ows has created a problem of solar radiation gains and and there is still a widespread belief that byelaws dictate the consequent likelihood of the interiors becomingtoo a minimum window-to-floor-area ratio. The virtue of hot. Windows occupy a central position ina study of large windows has been part of the philosophy of the building environment ; the problem of their design isto 'modern movement' in architecture for so long that it reconcile the conflicting requirements whichare made would be surprising if both architects and localauth- upon them. ority planning officers had not been mouldedto un- critical acceptance of the daylight 'slab'. Thermal conditions Yet another possible explanation can be attributed to Heating installations in modern office buildingsare nor- the general preference for (or at least, expectation of) mally capable of providing comfortablywarm condi- small-sized office spaces which is found in large organis- tions in the most severe winter. Nowadays officestaff ations as well as small. However,an important conse- are likely to prefer a fairly high air temperature, in the quence of departing from linear plan-forms and adopt- region of 70 to 72°F. This is often provided butventila- ing deep buildings is the probable needfor very large tionisrarely efficient and sensations of draughts, clerical areas (ie, open planning) so that people's view of whether caused by movement of airor radiation from daylight and the world outside the building shall not be the body to cold surfaces like windows,are commonly obstructed by partition walls. experienced. An examination of the attitudes towards small and large In sonic recent British owner-occupied officebuildings general office areas held by the three mainhierarchical a better control of the thermal environment has been ob- sub-groups of managers, supervisors and clerks showed tained by the installation of complete airconditioning that only managers were able to cite substantialadvant- systems. These can remove unwanted heat gains from ages for the large spaces. Other users expressed a clear solar radiation and mechanical and electricalequip- preference for smaller areas. Supervisors were the group ment, and control ventilation rates, airtemperatures most averse to large offices, their reasons mainly being and relative humidities. Airconditioning plant in terms of 'keeping track of staff' and will also creating 'esprit eliminate the ingress of atmospheric pollutionand, by de corps'. The clerks who least liked large officeswere permitting fixed windows, reduce the transmissionof not the ones who worked in such spaces but theoccu- external noise. pants of smaller, partitioned areas. This suggeststhat people may be more prepared to acceptsome (perhaps Noise unusual) environmental features if they have already Noise is perhaps the major environmentalproblem of acquired some experience of them. offices in city centres. It is commonly believed Sociometric studies have indicated that large some possible con- general clerical areas present special difficultiesin the sequences for management of the choice between large control of internal noise butsurveys have shown that and small work-spaces: small office areas containing this is not necessarily true. Although it isimpossible to single working units may help togenerate within them make unequivocal comparisons with smalleroffices or small competitive teams whose immediateloyalties are with other buildings, the sound levels registered to themselves, whereas large areas may produce in one a more building in large open offices whosearea is of the order generally collaborative whole. If this should be con- of 35,000 sq.ft rarely exceeded 55dBA. This was only firmed, the implication is that, where the performance slightly higher (and sometimeslower) than the sound of the work will be assisted by inter-grouprivalries, levels recorded in smaller office small working areas should be provided. areas in the same build- On the other ing and in offices of 20to 40 people elsewhere. Much hand, where work demands the co-operative efforts of can be done to reduce the problem withsuitable ac- many, the large areas may be found more effective. But oustic treatment of the ceiling and whatever the relation between space-size and floor surfaces. Non- group co- metallic waste bins and chairs withrubber-tipped feet hesiveness or loyalties, buildings can only be designed to will limit the amount of noise createdwithin the office, be most efficient and satisfactory inuse if the operational and it has a better chance of beingabsorbed if the space consequences of the design decisions are understood. is large, eg, upwards of 50 ft wide. A substantial reduction of window Light areas, the use of double glazing, the elimination of openinglights and the Historically, the purpose of windowswas to provide use of heavy forms of construction in preference light. It is now possible to do this to light more consistently and ones, such as curtain walling, will reduce the effects of more reliably (perhaps, therefore, less interestingly) by street noise. But the only way in which electricity. Office workers the problem will are likely to say that they be really solved will be byre-locating offices in quiet 18 zones or, in the long term, by new town planning There is need for new thought about the requirements policies. of office space, not least from buildingowners and users. Owners will need to take a more prominentpart in pre- The environmental design process paring the brief than they do now, for, after The creation of the total environment has all, they traditionally know best the use to which the building will beput. Un- been the architect's job but it is now sucha complex task fortunately, a serious problem of understandingand that, in the most enlightened offices, multi-disciplinary communication is all too common. Langdon 4 has design teams are found to be de- necessary. scribed this as happening in acute form inmechanised Decisions which affect the office environment are seldom offices: the client is insufficiently analytical inpreparing based upon research findings. This could scarcely be a detailed brief and the architect is too unfamiliar with otherwise for there have been few user-requirement the requirements to complete it himself. studies in buildings. Such topicsas lighting and ventilation and heating have been studied ingreat detail by The future office psychologists, physiologists and other specialists, but A forecast of trends in office design is difficult,for there their influence has been less than might be expected. are few indicators; any buildings which do not follow Environmental design has tended tomean the adapta- the conventional pattern are 'one-offs', individually tion of a given constructional design by the piecemeal created for individual owners and owing littleto pre- addition of lights, heaters, blinds, ventilators, andother vious or contemporary British examples. Occasionally, fittings. There has been no serious movement to design for very important buildings, client-architectteams have commercial buildings around the functions theyare studied current American practice. No othercountry meant to serve. seems to have exerted any major influence although, Environmental design is so complicated that it is not surprising that designers sometimes succeed inmanu- facturing their own problems. For obviousreasons a wholly intuitive approach cannot bevery successful, yet this is the main basis of current practice,as, for the moment, it must be. Not only is there little positive design guidance available but the building educational system has been such that there are few designers who could or would use it if there were more. The provisions of recent office legislation are rudimentary, and offerno help; it is safe to say that the standards which have been established so far will be exceeded bya comfortable margin in all new office buildings. And the history ofen- vironmental legislation (consider, for example, the Fact- ories Act) does not suggest that regulations, when they are made, will be unduly restrictive. The design of office space is unlikely, in the immediate future, to change from being a matter of arbitrary decision and personal hunch. Because of the nature of their training most architects already qualified will continue to take a fragmented view of the design team's responsibilities in this particular sector. They will regard itas their task to deal with some parts of the problem and they will leave other parts to their specialist consultants. The consultants are likely to display a lack of interest in the total environment and they may interpret theirre- sponsibilities narrowly: concerning ttomselves solely with, say, the electric lighting design, the thermal environment. Some matters will probably be taken for granted, including the place and function of windows, Figure 1 which, in all probability, will be designed for their effect Office buildings in New York. upon the external elevations. The social consequences of design of office space will not be questioned, for it is recently, the concept of bfirolandschaft', whichwas probable that none will be expected. A change of atti- developed in Germany, 3 has provokeda good deal of tude will only occur slowly, and must eventually be discussion. brought about by changes in architectural education. If the supposition is correct thata large proportion of But it is not surprising that practising architects simpli- the total demand for office space is, and will remain, for fy their design process, even if it means curtailingcon- small units of area (whether in size oftenancy or number sideration of important matters, for where environmental of private offices), then it is probable that in,say, twenty- design procedures have been evolved theyare often five years' time, the greater number of officebuildings complicated, time-consuming and suspect. built for letting will still be very much thesame as they There should be no doubt about the low efficiency of have been during the past quarter-century. There islittle present design procedures and current attitudes to office in the history of speculative office buildingto suggest environment: the results can be seen in that part of the radical experimentation. approved 120 million or so square feet of post-war office More substantial changes in office designcan be expected space already constructed. It is fair to ask how many from owner-occupiers than from propertyspeculators. office buildings really are functionally efficient, andpro- Although there are many examples of ultra-conserva- vide pleasant working conditions; howmany really in- tive buildings in the ownership of theirusers, it is this tegrated examples of commercial architectureare to be group which has been responsible for the few outstand- seen, and how many buildings still look well after only a ing departures from the conventionalpattern. The few years' use. The fault is not the architect's alone, for when he designs the buildings, the users are frequently 3. See, for example: Casson, Hugh: The executiveslum. Sunday unknown, and offices are sometimes badly mis-used.' Times colour magazine, 1964 (January 5) Moreover, there is quite inadequate feedback of inform- 4. Langdon, F. J: The design of mechanised offices. Architects ation to the architect about the performance of his Journal, 1963 (May 1 and 22) 5. Duffy, Francis: Biirolandschaft. ArchitecturalReview, 1964 buildings. (February)

19 41114.11, : _ -

Character Mo. Two divergent trends in the character of office space can 1161.01.ka. be seen. They are, on the one hand. the increasing mechanisation of clerical procedures which produce the atmosphere of clean factory areas and, on the other, the refinement and elegance of an increasing number of private offices. The trends are so different, yet so marked, that it seems possible that this is the watershed in the current development of office design. Clerical work in the past has been performed by numerous relatively well-educated but low-ranking clerks; now mechanisation makes it possible to replace clerks with machine- minders. Since the output of information from machines is so many times greater than from clerks, there has arisen a new and large strata of 'junior management' which plans and sifts the machines work. If the topex- ecutive strata is disregarded, an office hierarchynow consists of two layers. One is an enlarged officer-corps of junior executives which satisfies the need of top- management for more control information and, in so doing, creates a need for machine work. The other isan army of machine operatives. As machines increase in Figure 2 efficiency and output this second group must steadily de- The headquarters of the West Midlands Gas Boardwas trans- crease in numbers. At the moment it is still increasing, ferred from the centre of Birmingham to the suburbs: thevery due to both bureaucracy and the increasing require- large, single-storey building is at 4 o'clock from thegasometer. ments for moh data.

trends suggested by these forward-looking buildingscan Environment be discussed under the headings of location, character, Most contemporary office buildings provide onlyun- environment, amenities and flexibility. distinguished environments. In comparison with theen- vironments within pre-war offices they are likely to be Location better in some respects (visually, perhaps), and in others In the absence of specialised out-of-town sites for offices (eg, acoustically) to be worse. But a number ofowners on the pattern of industrial estates, office buildings for who have recently constructed buildings for theirown letting are likely to remain in centralor metropolitan occupation have used high quality finishes and advanced positions. Owner-occupiers are not so tied and itseems modern services to create quite impressive environments. probable that an increasing number, especially those The main features are high standards of comfort: high which have no need for frequent and personalcontact levels of illumination from electric lighting, air temp- with the public or other businesses, will take advantage eratures of 70 to 72°F and increasing use of aircondi- of the greater availability and cheapness of land, the tioning for thermal comfort, reduced noise and increased easier access and the possibilities ofa better environ- cleanliness. Environmental provision either stands still ment which are offered by the provinces and suburbs. or improves (it goes back only during historical 'dark One result of decentralisation and cheaper sites is likely ages ') so it is reasonable to assume that what is now ex- to be lower buildings and, in consequence, increased ceptional will eventually become general. efficiency in the use of space and lower building costs.

0111111h

Figure 3 Unlimited parking on a suburban site. 20 insulation are in obvious conflict;in practice it is both difficult and expensive to achieveany useful sound insulation from demountablepartitions in combination with suspended ceilings. It ispossible that such problems in the use of partitions willbe resolved by their omission.

A trend to open planning in owner-occupiedoffices In most commercial establishments0 and M departments seem to occupy a relatively lowlyposition, being more concerned with method than organisation,so an impression acquired during this studythat the building requirements of office functionshave not been appraised very critically is .probablytrue. This helps to explain the lack of any clear and positivedirections in the planning of office spaces bycommercial organisations. Taken together, the likelihood ofincreasing needs for flexibility and for inter-office communication,the use of mechanical or electronic procedures,and the findings that the lighting requirements of clerkscan, in suitable circumstances, be adequatelymet by well designed artificial lighting, suggest the suitabilityfor more widespread use of dc43, artificially lit officebuildings. These will need to contain open-plannedoffice space, with a minimum of partitions anda maximum size of window, so that staff working in the interior of thespace have an adequate view to the outside world. Figure 4 The choice between slabor block design is, however, Amenity: lavatories. one which in general can only be made byowner - occupiers: deep buildingsare unlikely to be suitable for letting, unless it can be expected thatthe unit of letting will be a whole floor. Office layoutsapproximating to this form have often been used forjunior staff employed on routine work and they are also quitecommon for highly-paid professional and technical staffin drawing offices. But they ararely used for junior and middle grades of clerical c.e management in this country, although the practice iscommon in the United States. Indeed, the most senior executivesof some firms in that 11116-__ country work in shared offices for theexpress purpose of achieving better communications andrapid decision- making. From this point of view,private rooms for any staff are less efficient thanopen offices just because they restrict opportunities for direct andinformal communication. They are alsomore expensive in both spatial and economic terms. It is possible, then, thatan important trend in the design of owner-occupied office buildings willbe towards deep, open-planned space, for this is certainlythe most flex- ible arrangement. Full airconditioningis likely to become more necessary and thereforemore common.

Figure 5 The lessons of the project for furtherenvironmental Amenity: dining room. research This office building research projecthas been a first ex- Amenities periment in the working of Side by side with improvements in a particular type of multi- the environment of disciplinary research team and, forat least half the the workplace will come improvementsand higher members of the team, standards in amenities, especially in an introduction to architecture such spaces as lava- and building problems. Thisreport must necessarily be tories, canteens and recreationspaces. The importance regarded as an interim of lavatories, in particular, ranks statement, constituting a starting so highly with women point for further studies. Besidesreviewing much work office staff that it willseem especially worthwhile for already published, the team has owners building offices on restricted budgets explored several new to give this fields. Inevitably, a great deal hasbeen left outstanding. 6 space more generous attention thanit commonly The most important points receives. of office design needing further attention are first:an examination of sizes of office occupancy, and second: studies Flexibility (or adaptability) of the nature of office work as it affects the need for office The likelihood of change in office space. work and in the use of The study has shown that theworking environments office space has become so great that itis now usual for which architects create will influence a measure of `flexibility' to be built into human behaviour a new office in many ways, and the value ofpsychological investi- building. The use of modules is convenientfor construc- gations of building design problems has tion at least as much as for adaptability been demon- in use. Increas- strated. There are a number of questionsof particular ing use of industrial methods of buildingwill inevitably psychological importance which need result in even greater use of dimensional further work to standardisation. develop the line of investigation whichhas been opened Demountable partitions are used extensivelyto meet present needs for both private offices andflexibility. Their functional requirements of lightnessand sound 6. See appendix 1.1

21 -S*

MII=IMEZ-Z

Figure 6 Windowless environment.

up, notably, perhaps, the degree to whichone environmental factor affects the subjective experience ofanother. There is much anecdotal, and little experimental,evidence in this field, yet it is likely to beat the very root of an under standing of the total environment. The use of a 'building type study'as the basis for research into the total environment has,on the whole, been successful, for it permits examination of allor most of the environmental factors. But it has thedisadvantage that it becomes difficult to separate whatunavoidably become the two objects of the work, ie, theenviron- mental aims of the long-term researchprogramme and the building-type problems thrownup by the immediate investigations. The Unit already hastwo building-type projects running simultaneously; itnow seems advisable to change one of these to an expressly environmental project. A single-parameter study wouldbe a retro- gressive step; what is wanted, perhaps, isa multi- building-type study. One possibility isa consideration of the ' windowless ' environments whichare increasingly being provided in such a variety ofbuildings as department stores, underground railways, photographic laboratories, race-track-plan hospitals, factoriesand, perhaps, even space-craft and submarines. Finally, a necessary task which remains for thefuture is the establishment of a means to evaluateand compare total environments. It is already abundantly clearthat it will be difficult to devise a method whichcan secure general acceptance: the number of thermalcomfort scales in use (for example) indicates theproblems of weighting the components ofeven a simple index.

22 Stage Three Part One

The componentsof environment

f IOUR ORGANISATION REDEVELOPMENT PREJUOICE NOISE NUISANCE DIMENSIONS HUMIDITY LIGHTING DISTRACTION SAFETY FUMES RELATIONSHIPS TACTILE TELE SAIISFAVION USE AtARIBILITY VISUAL REST ISOLATION CONCENTRATION PLEASAJITUSS RES11.IENCE. OLLABARATION SITE PRESTIGE ERGONOMICS PH' SENSATION SMELL OBJECTIVITY PROXImITY AIR mOVEMENT BRIoNNESS MODULE LOUDNESS DISTRIBUTION ATMOSPHERIC POLLUTION PRODUCTIVITY TRA CHARACTER PLAN UNIT RE PI ATION MISUSE OF BUILDING REFLECTION DESIGNER COMMUNICATION IAMBUS( AIR CONDITIONING FLOORIOEX RATIO COLOU ," LOCATION PEO PIELRAVIIIINF INSHES HAPPI .CESS STAT US FLICKER W 10 YALU AREA it 1 * u-s rtilfiitil TIST U >. BELLI 6 AN SION SEC U o RITY FUNCTION TIVITY SOL c-3 AR RADIATION WELFARE( w XPERIMENTATI FORM GLO OM ACCLIMITI CLOTHING RENT SEX FL' NCE AYE = LAWS cr., (METRO 1110 CLIMAT_ SI ?Riau? ION APATHYr coIS FO 4 RMALISATIQN v LOT FURJ ittE111YALIIA socual. ANNUITY St INGESTION L 4 (61SLATION SI IRV OWNER tlF INTERES AURAL. ?Hifi/14RM Telilltilli _I4 IlYSUCIEGVtilit ON RESENT MENT PENSION RATES EL CTRIC LIGHT E VIEW INS I ULATIIIN AC Ill AT BALM 41 E CLEANLINESS El 1 0- StIPCNIEE Cr3 VIPPEANE1115S1 ECONOMICS ANTIPATHY HE PERATURE64

PASUIVIITIUA 1110 MENITY FRIEN_ UHT I% TITU ES S UPERVI ION .1 TERFERE 4 ULATIONUL AT FRESPI REFERENCEEXTERNAL IMPACT S IP( SEA lliPPIPOUND OI WEN.6 4 L Y LTATiIO TE MAINTE NANCE RADIADIA% HEATING DI FEELI CTION 1EN AVFFUSIONIOUR ORGANI REDELOPM ENT PREJUDICE

11.111 it gRITERIANN "su FUMES REL ATIONSHIPS TA ELEPHONES SATISFACTION . VAILAIILITY VISUAL REST I ON OWN TRATION PLEAS SPATIAL

EIITli R- gli*WIN TY T WO B RAT litY EA Atlf MAP TIMA P.10111CYTMEM116115AU MINSECURITYilitturioluschfsRiA I'llut u 1011AOTY1114 TOYDONEWETAI IOETTPECROIMEVITVAtION FORM GLOOM ACCLIMIIISATION CLOTHING RENT SEX FLUORESCENCE BYELAWS METHOD EMOTIONS MICROCLIMATE SIZE GLARE ORIENTATION APATHY RUNK STS FORMALISATION INOIVIOUALITY DAYLIGHT FURNITURE MATERIALS AUDIBILITY SHAPE CONGESTION LEGISLATION SOCIOMETRY OWNERSHIP AESTHETICS! Chapter 2 Introduction

This publication reports upon a study of environment operative InsuranceSociety (the OS) in Man- which was based on the design of office buildings and chester, which houses more than 2,500 staff. office space. The study forms part of a continuing pro- (iv) Individual reports by the four research staff: gramme of environmental research which is sponsored (a) a general survey by the architect of the literature by Pilkington Brothers Limited of St Helens, and succeeds relating to office design and environment the previous investigation of factory building design.' (b) an account by the geographer of the general At the time of writing, `environment' is a word in popu- characteristics, population and location of post-war lar use, especially in connection with architecture, but it office buildings in England and Wales tends to have different meanings and implications for (c) a report by the physicist upon the results of sur- different people. In this report it is taken to be the sum veys in offices of thermal conditions in winter and of the physical and emotional sensations experienced by summer, noise, and lighting people within buildings and arising out of their use of (d) a report by the psychologist upon the response of those buildings. These sensations include bodily com- the CIS staff to their working environment and its fort, aesthetic sensibilities and social relationships; they implications for office design. Nt may originate in the external climate, the location and These four separate reports have been combined to- the social milieu, but they are influenced and modified gether to form this present `multi- disciplinary' report. by the building's design. Although an environment is experienced as a whole, The long-term aim of the research is the academic devel- with all the different stimuli acting simultaneously, this opment of an understanding of this total environment'. report has necessarily to deal with the more obviously ti A building type is a convenient base for such studies important components in isolation. The next chapter because it provides real situations in which many of the provides an overview of the main characteristics of the component environmental factors are acting together. office building type and a description of the buildings The choice of offices as such a base was made for a which formed the preliminary survey sample. In the succ- number of reasons. This building type seemed to be a eeding six chapters, present knowledge of the effects of natural progression from factories because it was an- location and the spatial, visual, thermal, aural and other workplace, with related functions and similar pur- social environments is summarised. The Unit's studies poses. The annual capital investment was unknown are drawn upon to fill some of the gaps. Chapter ten (there are no central agencies maintaining records) but consists of a case-study of the total environment in the was believed to be high. No current research into this CIS building. building type was known apart from a user-satisfaction This report represents a stage in the development of the survey by the Central Office of Informationand a Pilkington Research Unit's studies of the total environ- study of the design of space for automatic data process- ment within buildings and is therefore primarily add- ing by the Building Research Station. 3 It was considered ressed to other research groups engaged in systematic more worth-while to investigate building types where studies of architecture. But it also indicates an approach the gap between knowledge available and knowledge to design from the basis of the environment which applied to design was not very wide; contemporary people working inside buildings experience. There- practice in office design did not lag significantly behind sponse to a short course on office design and environ- existing knowledge, indicating a need for some new ment which the Unit based on this material s suggested thought. Many buildings were available for study, there seemed no necessity for the Unit to obtain staff with 1. Manning, Peter: The design of roofs for single-storey general- specialised knowledge of the building type and oppor- purpose factories. Department of Building Science, University of tunities for studying environment were plentiful. Liverpool, 1962 The form of the office project has been as follows : 4 2. Central Office of Information: Report not yet published, but (i) A series of background studies to compile a biblio- drawn upon in: Langdon, F. J. and Keighley, E. C: User research in office design. Architects' Journal, 1964 (February 5) graphy and define the characteristics, population and 3. Langdon, F. J: The design of mechanised offices. Architects' distribution of post-war office buildings. 6' Journal, 1963 (May 1 and 22) (ii) Preliminary visits to a representative sample of 4. For a detailed description of the organisation of the research buildings in Liverpool, Manchester and elsewhere. see: Manning, Peter: The organisation of a programme of research. Architects' Journal, 1964 (December 30) (iii) A series of detailed investigations of environmental 5. Taylor, Sheila: Post war office building: a survk.,. Architects conditions and other mattets.6 Physical conditions Journal, 1962 (December 19) were examined in a sample of nine buildings. Psycho- 6. See appendix 2.1 logical studies of clerical staffs' attitudes towards 7. Wells, B. W. P: Office design and the office worker. Ph.D thesis, University of Liverpool, 1964 their physical and social environment were made in 8. White, W: Liverpool University: office design and environ- the new headquarters office building of the Co- ment. Architects Journal, 1964 (October 21) 25 0.*

that to many architects and buildingowners this is a new way of thinking about buildings. It is hoped, therefore, that the report will also prove of interestto a wider audience than the present limited circle of architectural research workers.

LIVER0_OL 70 DONCASTER Estimated Estimated Population1963 60 Population 1963 60 86,910

50 50

40 4 cs

- _ 30 - 30 u.. 4=, cc 20 cic 20 co

10 rig 10 ......

0 20 40 60 80 100 300 500 700 20 40 60 80 100 300 500 700 FLOOR AREA OF BUILDINGS IN 1000 SO FT FLOOR AREA OF BUILDINGS IN 1000 SO FT

CARD!F 70 70 SOUT1MPTON Estimted ,Estimted Poputtion963 Popultion 1)63 60 60 207,220

50 50

40 40 ESJ ;13 30 g 30

tc, 20 cc 20

g 10 10

0 20 40 60 80 _ 100 300 500 700 20 40 60 80 100 300 500 700 FLOOR AREA OF BUILDINGS IN 1000 SO FT FLOOR AREA OF BUILOINGS IN 1000 SO FT Figure 7 Histograms showing size distribution of office buildings constructed post-war in fourtowns. Chapter 3 Background to the environmental studies: The building type

The term ' office ' is used very loosely. Perhaps the main distinction to be made is between the office building and the office space' within a building. Although an office building may be designed entirely for office use, it is quite likely to consist of one or more floors of office space above such other accommodation as shops, banks, public houses, etc., or be attached to a factory or warehouse. Recent office legislation= has defined office premises as `... abuilding or part of a building...the sole or principal use of which is as an office or for office purposes.' `Office purposes' include administration, clerical work, handling money and telephone and telegraph operating; `clerical work' includes writing, book-keeping, sorting p..,Jers, filing, typing, duplicating, machine calculating, 1i.1 /e1 drawing and the editorial preparation of matter for pub- I so /flat lication. This definition describes the operations which go on within the office but it does not explain its function. An explanation of this type is given in a standard handbook on office organisation and methods: '...in the 0 and M context " the office" is primarily the admin- istrative headquarters of tiltbusiness which provides management with a service of communications .... (and)... arecord of past transactions and operations.' 3 Figure 8 Most new office buildings are quite small. The building Although it has not been possible to collect statistics of Types of space within office buildings all post-war office construction, it appears that a large Not all the space within an office building is office space proportion of the total consists of very small buildings, for there are in addition the `service' spaces of lavator- ie, areas less than 10,000 sq.ft. 6 In Hertfordshire, for ex- ies, boiler rooms, canteens, lift lobbies, staircases, etc. ample, 86 percent of projects, and in Surrey 80 percent The proportion of office space to the floor area of the of projects have been below this figure. The number of whole office building may be of the order of less than 50 projects whose size is greater than 100,000 sq. ft is small up to a maximum of about 80 percent. and, outside the larger towns, individual office buildings Office space may comprise reception and waiting areas, with areas exceeding 50,000 sq. ft are rare. Most new conference rooms, general offices, private offices, typing office buildings, especially in the smaller towns, have be- pools, machine rooms (for noisy accounting machinery, tween one and five storeys, the greater number being of etc.) and filing and storage areas. Most businesses have only one or two storeys, especially when they are attach- a need for some of these spaces. Their relative import- ed to industrial premises. Much higher buildings are to ance depends on the nature and practice of each individ- be found in the larger towns, buildings over 12 storeys ual business. high being not uncommon. The 34-storey Vickers building in London is at present the British office building Size of buildings with the greatest number of floors, though it is not the Information about the floor areas of offices constructed tallest. in the post-war period has been sought from the local planning authorities. 4 The typical range of areas of office space in individual buildings was defined as: small office buildings: less than 20,000 sq. ft of office space 1. This research project has been concerned specifically with 'office space', whether it comprises the whole or part of an office medium office buildings: 20,000 to 100,000 sq. ft of building office space 2. Offices, Shops and Railway Premises Act 1963. HMSO large office buildings: over 100,000 sq.ft ofoffice space. 3. Milward, G. E. (Ed): Organisation and methods. Macmillan. In this instance, the term 'office space' refers to the total Revised edition, 1962 4. See chapter 4, appendix 4.1 and Taylor, Sheila: Post-war amount of office floor space in one building, whether office building: a survey. Architects' Journal, 1962 (December 19) this occupies the whole of it or only a part. 5. Taylor, Sheila: op cit (4) 27 Ownership local authorities to determine siting, building The ownership of post-war office buildings use, day- seems to fall lighting zones, building height, car parking, theappear- into three groups: localor national government, specu- ance of buildings and many other matters. lativedevelopers,and owner-occupier commercial firms. Only a small proportion is government owned, Building Byelaws.Although there are plans for the the remainder being approximately equally crea- divided be- tion of national building byelaws whichwill apply tween speculative and other private developers.Specu- itross the whole country, at present the Model Byelaws lative buildings have usually been commissioned by published by the Ministry of Housing and LocalGov- insurance companies, building societiesand other pro- ern,nent are u-:?,c1 by local authorities as the basis of the perty and holding companies; they oftenaccommodate byclaws adopted for and applicableto their particular a small branch office of the owners, the bulkof the areas. The Model Byelaws have no specific requirements space being let. Other privately owned office buildings of environment within office buildings (which have been built by banks, are classed newspapers, manufach. asdomestic buildings') and most of the retailers and others. Such clauses are. owners often occupy a con- intentionally, couched in generalterms eg, 'Every siderable proportion of the officespace, and let on short storey of a domestic building shall b, provided with lease that area which is superfluousto their present needs. effectual means of ventilation'. 4 Government departments are extensiveusers of office space, mostly in rented property. Comparatively few The Offices, Shops and Railway PremisesAct.Over many new Crown buildings have been built. years there have been a number of attempts to introduce legislation which would control the working Mixed uses of office buildings conditions of non-industrial employees (ie,principally clerical Buildings containing office spacevery often have other workers) but, until recently, they have been uses too and although the building's primary unsuccess- use may be ful. In 1946 a committee of enquirywas appointed by for offices, in some cases thismay represent only a small the government to enquire into (amongst other proportion of the total floor matters) area. The most common .. the statutory provisions relating to the health, wel- arrangement consists of shops on the groundand first fare and safety of employedpersons at places of em- floors and offices above. Car parkingspace is sometimes ployment other than those regulated under included within the building and might the Fact- occupy base- ories or Mines and Quarries Acts...' The committee's ment or ground floor space. report (theGowers Report ')" summarised existing legislation and made recommendations which,in re- Types of user and tenancy spect of shops and offices, were substantially that Information is needed about the the type of business in provisions of the Factories Act should applyto these which office users are engaged, for it isprobable that two building types. The report became this would have important implications a standard for for the design of those endeavouring to raise the workingconditions of office space. Possibly the most usefuldata would be a office staff, but no actionwas taken by any government frequency distribution by number of employeesand until, in 1960, a private member's bill (theMarsh Act ') SIC Orders. But at the moment thereare no figures reached the statute book. 12 This Act available which show such gave the Secretary a breakdown and, if this of State power to make `... regulations specifying the office project should be extended, it is probablethat one standards as to structure, arrangement and operation of the first tasks would be to establish to one. be applied in offices for the protection of thehealth, safety and welfare of persons employed therein...' but Size of occupancy no regulations were ever made, for the government in- Another important matter which hasnot yet been investi- troduced its own legislation and this, in theform of the gated is the range of size of office occupancies,including Offices, Shops and Railway Premises Act 1963 both owner-occupiers and tenants. This repealed information is the Offices Act and was operative from1st August needed for the planning of cities,as well as office build- 1964.73 ings, for the size of an occupancy can determine suitable The purpose of the Act is .. to make fresh provision dimensions for the building's width and depth,and in- for securing the health, safety and welfare ofpersons dicate points of access. Streetpatterns, especially in employed to work in office... premises ...' which are comprehensive redevelopment schemes in central areas, defined as . a building or ... part the sole or princi- are subsequently influenced by the size and shape of pal use of which is as an office building blocks. or for office purposes .. 14 The Act isnotapplicable to certain offices such as those in which the only staff are relatives of theem- Designers ployer or where a maximum of only 21 According to a survey of man-hours per new commission; for private week is worked. Those parts of the Actappropriate to architects,' in the years 1955 to 1957 thedesign of com- a consideration of environment are discussed in later mercial buildings (in thiscase, offices and banks) repre- chapters. sented about 30 percent of privatearchitects' work. It was also estimated that 90 percent of all commercial The people who work in the building buildings were designed by privatearchitects. It is not Numbers of office staff now possible to say that these percentages are still appli- Itis generally believed that thereare now approximately cable for, since thatsurvey, the RIBA revised the basis for its collection and analysis ofstatistics and commercial buildings areno longer quoted separately. 6. Central Statistical Office: Standard IndustrialClassification. HMSO. Second edition, 1958 7. Anon: A survey of private architecturalpractice. J Royal Inst Legislation affecting the siting and design of office Brit Archt. 1959 (April) buildings 8. Town and Country Planning Act 1962.HMSO 9. Ministry of Housing and Local Government:Model Byelaws, Series 1V: Buildings. HMSO, 1953 Office buildings are subject to three mainitems of legis- 10. Offices, Shops and Railway Premises Act1963. HMSO lation: the Town Planning Act, ° BuildingByelaws, 9 and 11. Home Office and Scottish Home Department:Health, welfare the Offices, Shops and Railway Premises Act.'0 and safety in non-industrial employment. Hoursof employment of juveniles. Report by a Committee ofEnquiry. Cmd 7664. HMSO, 1949 Town Planning Acts arethe instruments used to control 12. Offices Act 1960. HMSO development and amenities. Thcreare no provisions 13. Builder, 1964 (January 3)p. 33, reporting the Minister of specifically relating to the environment withinoffices Labour 14. See the beginning of this chapter for definitionof 'office pur- but the general application of the Actsgives power to poses ' 2$ three million clerical workers in Britain's and, if the and Keighley for an article describinguser research in proportions of men to women have not changed sub- office design. stantially since the 1951 census. about 60 percent are The findings of these surveys will be reviewed in sub- w omen. sequent chapters which deal with specific environmental Together, office and shop staff account for about 80 per- topics. Their aims and methods were as follows. cent of all employed persons other than those engaged in industrial or domestic work. The survey ol service buildings had as its main purpose the consideration of how efficiency might be in- I)/ O///e Mil creased by up-grading working conditions to the best The main types of stall' who may be housed in offices standards prevailing in commerce and industry. Itwas have been categorisedas made between April 1943 and September 1944 bya work- (I) typists ing group drawn from various government departments. Iii) secretaries The task included reviewing environmental and other (iii) machine operators standards, assessing the extent to which thesewere met iv ) clerks (including consideration of possible variations between I s) specialistseg, draughtsmen and telephonists departments) and comparing civil service conditions i) 'transitory' staffeg, salesmen, inspectors with those of 'outside industry'. The method adopted (vii) managers and supervisors was to '... get some knowledge of theory and of expert Iv iii) 'advisory' staffeg, in laboratories, studios. opinion.. "' on the elements of the problem, and to strongrooms visit contemporary industrial and commercial office (ix) executives buildings and all types of government offices. (x) service staffeg, canteen stall', drivers, building maintenance workers The office lighting survey was undertaken in February In this study emphasis has been placed upon the general 1948 by the Central Office of Information to *...obtain office areas which are staffed mainly by clerks and, to a information about the general state of natural and arti- lesser extent, typists, machine operators, and superficial lighting in offices in Great Britain with regard both visors. to the facilities available and to the workers' opinions . "s It took the form of a statistical analysis of the re- Occupational health of othce workers plies to a questionnaire completed in respect of 1,408 In a paper demonstrating the need for legislation to office workers and a correlation of those replies with the regulate the working conditions of non-industrial work- physical conditions of the 358 offices in which theywere ers, Wattleworth'" collected together a number of stat- obtained. istics bearing upon the occupational health of clerical workers. He said''' that the incidence of respiratory The Liverpool survey was commenced in 1955. Itspur- tuberculosis among clerks was 38 percent higher than pose has not been explicitly stated, but from the manner the average for the country and anaemia twice as heavy in which the findings have been published'it can be as among agricultural workers. The average mortality assumed that it was undertaken partly in the course of rate from rheumatic heart diseases, coronary heart disease, pneumonia, ulcers, diseases of the liver and neph- ritis was higher than the average for the country; 2" the 15. See, for example, a report of a talk by L. A. Hill of Shell number of spells of incapacity due to sickness in agri- International Petroleum Co. in The Guardian. 1961 (November 24). culture is 9 per 10,000, for all workers 33 per 10,000, but The results of the 1961 Census in terms of occupationwere not for clerical workers 47 per 10,000. 2I available when this report was written 16. Home Mc( and Scottish Home Department : op cit III) Wattleworth appears to assume that there may be a di- 17. Manasseh, Leonard And Cunliffe, Roger: Office buildings. rect relationship between the working conditions of Batsford, 1962 office staff and their liabilities to particular kinds of ill 18. Wattleworth, W. H: The forgotten army. Royal Society of health, but this has hardly been established. Taylor and Health Journal, 1959 (March-April) 19. Referring to a statement in the House of Commons and the Wood remarked'!" upon the lack of knowledge of the Registrar General's report for England and Wales for 1951 risk of infection to which office workers are exposed but 20. Referring to the Registrar General's report for England and drew attention to some evidence 2" that risks are less in Wales for 1951 small than in large units. McGirr remarked 24 that cleri- 21. Referring to a statement in the House of Commons by Mr Yates and to Hansard (1958, 24 January). The figures originated cal workers include a higher proportion of disabled per- in the Registrar General's report for 1951 sons than industrial workers. 22. Taylor, (Lord) and Wood, C. H: Occupational healthserv- A belief is widely held by office managements and others ices in smaller work places in Britain. Chapter 19 in: Modern that good working conditions encourage efficiency (or trends in occupational health. Ed: Schilling, R. S. F., Butter- worth. 1960 productivity') and bad conditions contribute to high 23. Stewart, A. and Hughes, J. P. W: Tuberculosis in industry: staff turnover rates. So far, no such relationships have an epidemiological study. Brit Med J, 1959 1 926 been adequately proved. 24. McGirr,P. O. M: The health aspects. Paper to a conferenceon Efficiency in the office' organised by the Industrial Welfare Society in March 1964. (Unpublished) Published surveys of offices and their environment 25. Study Group appointed by H.M. Treasury: Working condi- Since the end of the war a number of reports have been tions in the Civil Service. HMSO, 1947 published of extensive surveys of office design and work- 26. Gray, P. G. and Corlett, T: A survey of lighting in offices. ing conditions. These include: Appendix to Post-war building studies no. 30. HMSO, 1952 27. Wattleworth, W. H: Unfinished business in environmental 01 physical working conditions in the civil service" health. Royal Society of Health Journal, 1958 (May-June) (ii) the state of office lighting " 28. Wattleworth, W. H: 1959, op cit (18) (iii) environmental conditions in offices and work 29. Wattleworth, W. H: Health and welfare of the office worker places in part of the central area of Liverpool °. 2". " and others engaged in non-industrial employment. Paperto conference of public health inspectors at Scarborough, 1960. (Dupli- (iv) standards of office accommodation in Barrow-in- cated typescript) Furness "" 30. Nelson, I. D. M. and Morse, R. J: A pilot study of office (v) occupational health conditions in the City of accommodation. The Medical Officer, 1960 (September 9) London "! 31. Robinson, Alan: Working in the city: the interim report ofa (vi) a user study of eight offices which employ auto- survey. City of London Health Department, 1962 (December) 32. Langdon, F. J: The design of mechanised offices. Architects' matic data processing equipment22 Journal, 1963 (May 1 and 22) The Central Office of Information is known to have 33. Langdon, F. J. and Keighley, E. C: User research in office made a nation-wide survey of working conditions in design. Architects' Journal, 1964 (February 5) post war offices. The results have not been published, al- 34. rady Group appointed by H.M. Treasury:or. cit (25) 35. Gray, P. G. and Corlett, T: op cit (26) though the material has been drawn upon by Langdon 36. Wattleworth, W. H: (1958, 1959 and 1960)op cit (27, 18, 29)

29 of 37 businesses occupyingspace within these buildings wasisited. The sample included differentdepartments of central and localgovernment, representative professions. public corporations, insurance,shipping, industry, chain stores and the buildingindustry (table I). The areas of office space in individualoccupation ranged upwards from 200 to several hundred feet. thousand square The impressions gained from thesevisits have formed a general background picture for themore detailed investigations, and an attemptat generalising this experience is made to help in setting thescene for subsequent chapters.

Table I Analysis by Standard Industrial Classificationof business interest of 37 firms visited during A preliminary stage of project

4 _raw SIC Order No. of it AM,41'41112111 firms Figure 9 Agriculture, Forestry, Fishing Within Liverpool's business area. IV Chemicals and Allied Industries VI Engineering and Electrical Goods the normal work of a public healthdepartment, partly X Textiles to support a case for establishing protective legislation XV Paper, Printing and Publishing for non-industrial workers. Thesurvey combined in- XVII Construction formation from routine inspections andthe results of a XIX Transport and Communication 5 pilot survey specially undertaken bythe city corpora- XX Distributive Trades 6 tion, in the course of which the accommodationof 1,900 XXI Insurance, Banking and Finance 7 firms in commercial premiseswas inspected. XXII Professional and Scientific Services 2 XXIII Miscellaneous Services 4 The Barrow-in-Furnesssurvey was carried out jointly by XXIVPublic Administration and Defence 7 the town's medical officer and deputychief public health inspector, as a pilot survey of the standardsof office Total no. of businesses 37 accommodation in a mediL.n-size' 37town. Fifty offices were visited, these being chosen at random fromthose Note. Unless a firm was known to be which met the definition of theGowers' Report, "4 . a wholesaler or factor without manufacturing capacity, the classificationwas by SIC Order premises in which personsare employed in clerical work, of the original producer including book-keeping, filing, typing,multigraphing, machine calculating, and drawing' but excluding offices Description of the buildings which formedthe survey forming part of factories, workshopsor shops. The en- sample quiry was mainly observational, ie, room dimensions, All but one of the buildingswere situated in city centres. lighting and temperaturewere measured but a direct Their owners were mostly commercial question was asked about noise. The organisations survey took place (such as insurance companies) whichhad built the offices between October 1959 and June 1960. as investments and let the officespace, either directly or through agents, to a variety oftenants. A small number The City of London surveywas of the ' health and envir- were owned and occupied by central and localgovern- onment of City non-industrial workers,to compare their ment; another group was owned by commercial working conditions with those likely firms to be nroposed in which occupied the floorarea they needed and let the re- future legislation and to ascertainthe practicability and mainder on short lease. One building support by employers for an occupational health consisted of a service semi-industrial depot where the officeaccommodation for the City of London '. 39 Thewhole of the field work, was only a small proportion of the whole. which involved visits to 267 businesses, Two others including 186 were described by their usersas having originally been offices, and measurements of environmentalconditions intended for development in 1,823 rooms was carried as department stores. The out by one assistant medical majority were designed and builtas offices, but the officer of health especially appointedfor the study. The ground floors were often used for field work took place during the period other purposes. The March to Sept- total amount of officespace in a building varied upwards ember 1962. from less than 500 sq.ft; the totalarea of individual buildings in the sample varied The user-study of mechanised offices downwards from about was made by the 400,000 sq.ft. The number ofstoreys above the ground Building Research Station because it was found that floor was often 4 or 5, a smallernumber 9 or 10 storeys, many firms changing over to mechanised office proced- and one or two higher still. ures were experiencing difficulty in obtaining suitable With the exception of one building, accommodation or had noise, ventilation, which had been built or other prob- before the war, all the officeswere new since 1954, some lems. The work started about 1959 and finished in 1962, having been built very recently.The period of construct- and consisted of functional andenvironmental studies ion was typically twoyears. of eight buildings in which staff reactions to workplaces Internal finishes tended to be simpleand were commonly were correlated with physical measurements of theact- plaster for the ceilings (direct to the ual conditions. concrete floor structure) and thermo-plastic tilesfor the floors. Other floor finishes used were linoleum, woodblock and rubber The Pilkington Research Unit'spreliminary survey of offices The Unit's preliminary visitswere to 21 buildings in the 37. Population 65,000 north-west of England whichwere representative of 38. Home Office and Scottish HomeDepartment: op cit (11) 39. Robinson, Alan: op cit (31) 'large', 'medium' and 'small' officebuildings. 40 A total 40. Sec Taylor, Sheila: op cit (4) andpage 27 30 -

as long blocks or wings, witha total width of 40 to 50 feet. The officespace in these was usually served bya central corridor and the depthof the room from the windows was a maximum ofabout 25 ft and frequently less than 20 ft. Itwas rare to see such offices more than 100 ft or so long. Small privateoffices partitioned of from large open officeswere usually arranged on the building perimeter, ie, by thewindows. Areas of usable office space (ie, in thecase of tenants, rented area) varied from 61 to 182 sq.ftper person.

User requirements alafficespace Each occupier was asked whetherhis firm's work made any special requirements of the officespace. The main need was for general-purposespace and there was little demand for special provision forexcessively large or heavy or noisy equipmentor individual procedures. The office managementswere also asked what they con- AlP11111111 sidered to be important factors inthe design of office space. Most firms said they had been lookingfor good working conditions. This sometimesmeant low levels of noise but more often good lighting(by which they in- Figure 10 tended daylighting). The need for Upper floor of a department store adapted space which would for clerical use. be readily adaptable tonew layouts was mentioned;

-*to

Figure 11 General office.

sheet. Occasionallysome form of acoustic tile had been used on ceilings. Many different one firm chose their present offices becausethe stand- types of partition ards of finishes and fittings in that (standard systems and in-situ building were reason- constructions) were seen. ably good, while in alternativebuildings they were The services provided within thebuilding contract were `cheap and nasty'. heating, and electricity for lightingand office machines. Telephones were added later, usuallyby arrangement Spatial standards between the occupier and telephoneorganisations. Only a few large firms hadan established policy for de- Use of space termining their office standards.Where such standards had been fixed the minimumarea was of the order of Most office space was usedas a combination of open 60 sq.ft per person. A feeling offices together with a number of small was often expressed that rooms accomm- the heights of officespaces were inadequate nowadays: odating one or two people. Itwas noticeable, however, for example, 9 ft that the larger the organisation was sometimes held to be too low. The and its floor space, the structural design of floors often dictated more marked was the tendency for bigger and the layout of an more office because of the needto locate banks of filing cabi- open offices to be used. At the otherextreme, small org- nets over beams. anisations employing total officestaffs of six or seven might accommodate them in three,four or five separate rooms. Working hours Original planning decisionsmay exert a great deal of in- The earliest time any of thefirms which were visited fluence upon the way in whichopen office space is used. started work was 0815, and thelatest time any firm A number of buildings, for example,had been designed normally finished was 1800. Moststarted at 0900 and fin-

31 Figure 12 Figure 13 Most firms equate good working conditions withthe provision of daylight. Standardof finishes and fittings influences choice of offices.

Figure 14 'Core' area of a slab-type plan office block withintersecting wings

ished at either 1700 or 1730. Rented officespace within sort of a view to the outside. an office building was often available to the tenants only (iii) Badly shaped spaces and ill-designedlighting in- during normal working hours; without special(and stallations; people were working inpoor daylight con- possibly expensive) arrangements it mightnot be poss- ditions or with inadequate electric light,sometimes ible to enter a building before 0800or enter or leave after without a view outside. 1900 hours. It is obvious that some electric light willalways be used during office working hours in categories Lighting (ii) and (iii). Offices with the arrangements described incategory (i) Almost all managements assumed that gooddaylight normally expect to work without electric light,(especially was an essential requirement for their offices, but few during the summer months) although, it practice, it was appeared to be very critical of the widely varying standards not unusual to find that electric lights.. being used. they had. In broad terms, there seemedto be three types But this could be due toa number of factors, including of plan arrangement of general officespace in relation window design, weather condition and time to lighting: of year. Many offices in city centres, particularly thoseon lower (i) Long narrow rooms with windowson one side floors, are seriously overshadowed,so that any day- and a typical maximum depth of 3 desks arrangedat light received at points more than two right angles to the windows. or three feet from the window is indirect, being reflected frombuildings (ii) Large open areas with moderate to good electric opposite. lighting; people not on the perimeterwere working In some large organisations itseems to be tacitly accept- partly or mainly by electric light, but they hadsome ed that clerical staff should not workexcept in daylight 32 ences of attitude to the thermal environment between men and womon, and young and old may be expected, but people of tine sameage and sex, working together, may complain of completely different conditions (eg, draughts and stAly atmospheres)at the same time. Of all the environm,:ntal problems thisseems to be the one of which office nuriagersare most aware and which they find most intractable. The problemmay at times be ex- Arrilli...1116 twig trone, Toi avoid di: utes as to whether windowsshould be open or closed, one manager had instructedhis porter to go round the building opening all windowsfor five minutes in each hour, +1 This buildingreceived more fresh air than most, for, typically,the offices which were visited during the period July toNovember had almost all their windows shut tight. Practically all of the offices whichwere visited were heated by radiators placed under thewindows. Where private offices were partitioned off froman open space it was not unusual for thermal conditions in thesmall rooms and the main spaces to be quite noticeablydifferent. On L several occasions office managementsmentioned the need for heating systems to be subjectto greater and Figure 15 more local control.Ina number of buildings noticeable Indirect daylight. differences in air temperaturewere reported to occur between the upper and lower floors (theupper floors being the colder ones) and betweenrooms on the north and south sides.

Thermal conditions in sunutter Several firms reported uncomfortably hotconditions during summer, but the principal form ofdiscomfort during this time of the year seemed to be causedby glare. It was, however, difficult toassess the importance of this issue for response may be largely dependentupon recent conditions and the summer 1962 (when these preliminary visits were made)was not a noticeably hot one.

Noise Where offices are sited frontingonto busy main shopping streets or traffic routes, disturbancefrom outside noise can be substantial. The noise levelcan be reduced by closing all windows but thispresents the office workers with a choice of reducing the noise level and having to suffer a stuffy atmosphere,or of ventilating their rooms and accepting the noise. The usualcourse is to keep windows closed. Some of the officesinthesample Figure 16 which were located in shoppingareas had tocontend Internal light well. with canned musicfrom the shops beneath. Internal noises seemedto be relatively unimportant,even (ie, within about 10 to 15 ft ofa window) and it seems in large open offices, althoughone organisation ex- likely that this attitude influences the design ofthe build- plained its use of smallrooms on ings which these organisations the grounds that they rent or build. Several limited the number of people distractedby telephone post-war buildings contained internal light wells which calls. In small offices shared byan executive and his sec- were lined with white or glazed bricks which had be- retary the typewriter could not be used duringa convers- come dirty and were patterned with drain pipes. Invari- ation on a telephone or witha visitor. Most large office ably they looked depressingly dingy, yet the manage- organisations had machine sections, butthese were ments seemed certain that although their staffs would usually partitioned off from normal clerical offices. prefer an external wall to anarea wall, the area was acc- Construction work was a eptable in a way that total reliance common cause of complaints upon electric light of noise, particularly in the early days ofa building's life would not have been. when some parts were undergoing preparationsfor a On the whole, people appeared to like workingfairly new tenancy, and the noise created especially that high up in buildings (eg, 8 to 10 storeys highor higher) transmitted through the structureannoyed tenants al- because of the better opportunity for daylightand a ready established. Heavy footfalls frompublic corridors view, and of being away from street noises.Venetian and even from floors abovewere often audible. blinds, sun blinds or curtains were usually fixed after the In most private offices wherean adequate degree of priv- building contract had been completed andwhen their acy was considered necessary, the requirements necessity had become apparent. In the were upper floors of met, but in some small offices it was often possibleto be the very high buildings the view from the workingpos- aware of telephone conversations in adjoiningrooms. ition was onto very bright sky. The views from thelower floors were more likely to be of other buildings. General comments by managementsupon office design A majority of the offices used tubular fluorescentlight- Many views were expressed by different ing. managements which, while not of direct interestto this project, provided an insight into related problems of Heating and ventilation office design. The principal difficulties concerned withheating and ventilation in offices seemed to arise from thedifferent 41. This is a practice which at one timemay have been general in the civil service: see Study Group appointedby H.M. Treasury: thermal requirements of different individuals.Differ- op cit (25)

33 The cost of min officespace With such a small sample, itwas not possible to obtain any useful data on the typical costs of constructing office space." An indication of the annual costsper sq.ft incurred by tenants of rented spaceare given in table 2. The figures quoted are 'typical' rates whichwere in force in Liverpool in mid 1962. Therewere, of course, considerable variations between firms, dependingon the date of lease, type of office, thetenant, and other factors.

Table 2 Typical annual costs incurred bytenants of office space in Liverpool during 1962

:3 Cost Percent shillings/ of total sq.ft in the region of Rent 10.5 46 Service charge (to cover heating, lifts, lavatories, etc.) 2.5 11 AIL Rates 7.5 33 Electricity 0.5 2 Cleaningof rooms * 1.5 7

of windows 0.3 1 Insurance (not known) Redecoration and maintenance Figure 17 (average per year) Space needed for staff's clothing. 0.05

Toilets were often thought to betoo few in number and Total (excluding insurance)say 23.0 100 their users were critical of the standardsof the surface finish and equipment provided. Thefrequently used *Costs of room cleaning were not obtained but'. .. cleaning costs plan arrangement of lavatories for differentsexes on alt- . may normally be reckoned at 1.5 to 2.0 shillings per sq.ft of ernate floors was highly unpopular. Some building floor area per year' (Eastwick-Field, J. C:Finishes. J Royal Inst own- Brit Archt, 1957 (July)). ers were caused considerable expense by wilful damage to lifts and toilets. Complaintswere also made about the unreliable or temperamental Taking the average density in these officesas 120 sq. ft. nature of lifts and the per person, the annual costs were of the order of L140 noise they may create. A smallroom where staff can eat sandwiches and make tea was usually consideredan essential provision. Opportunityto rent a car-parking 42. But a summary of the illustratedcost analyses of office build- space within the building was often appreciated, but ings which have been published from timeto time in the Archi- cos*. at least £100 per annum and sometimesa great deal tects' Journal is given in appendix 3.1. See alsoImperial Chemical more. Industries Ltd: Standards for offices and laboratories.(The Comp- any), 1960 (April)

" 1;7

r*"

Figure IS Staff tea room. per office-worker On first moving into an office, tenants might be raced with the expense of providing partitions, floor finishes, light fittings etc. Figuresobtained from firms which had had tocarry out such work varied between 7.0 and 30.0 shillings /sq.ft.

43. These are the typical ' costs in one provincial city. Costselse- where, especially in London, could be very different 35 Chapter 4 The external environment

and the location of post-war office buildings

A large part of the environment within a building results The Unit's location enquiry developed from part of a from the wider environment outside, which is modified larger enquiry into the statistics and general character- by the building's fabric and services. The nature of the istics of this building type, for details of the office build- external environment depends in the first place upon the ings constructed in this country had not been collected building's geographical location: whether, for example, by any central agency and they were necessary for samp- it is in the north or south of England, in a city centre or ling purposes. a suburb. It is influenced also by factors of the micro- The enquiry commenced in July 1962, when every plan- location such as the micro-climate; the congestion, noise ning authority in England and Wales was circulated and and fumes of traffic; the land-use of surrounding areas asked for the total amount of new office space built since and by more abstract local characteristics. All these 1945, the number of individual projects and, for each features of the external environment of office buildings project, the floor area, number of storeys, location, own- require further investigation. ership and cost of construction. Of 145 planning authorities asked, 65 percent gave some information. 27 percent replied that they had neither the information available nor the time and staff to collect it; the remainder did not reply. Unless an authority had already maintained a separate register of applications for planning approval for office development there were formidable problems in com- plying with the request, and it is not surprising that such a large proportion of planning authorities have been unable to help. Councils which have replied have done so in varying amounts of detail: in respect of the whole area of some towns, only the central areas of others, only the major buildings elsewhere, and all for different periods of time.' Analysis of the replies has consequently been extremely difficult but the best available estimate is that in England and Wales something of the order of 120 million sq.ft of office building has received town planning approval since the end of the war. Of this figure, about 80 percent is in the south-east of Eng- land. By 1962 about 50 million sq.ft of the total approved had been built or was under construction in London itself and a further 8 million or so in the Metropolitan parts of Essex, Hertfordshire, Kent, Middlesex and Surrey and in the county boroughs of Croydon and West Ham. Other county boroughs in the South-east in which significant areas of office space had been built were Reading and Southend (0.6 million sq.ft each). In the eight new towns surrounding London 1.5 million sq.ft had been built or was under construction, Hemel Hempstead having the most with 0.4 million sq.ft. The remaining 20 percent of office space which hadre- ceived planning approval was distributed about the whole of England and Wales outside the South-east. Of sq ft the total approved in the provinces, rather more than of post-war three-quarters (18.4 million sq.ft) had already been built = 1,000,000office space or was under construction. 11.7 million sq.ft were loca- Figure 19 ted in seven of the towns and cities, ie, Birmingham (3.4 Geographical distribution of post-war office construction (built million sq.ft), Liverpool (2.2 million sq.ft), Manchester or under construction). The small circle within the large one representing London shows office construction in Croydon to the value of nearly 3 million sq. ft. (See appendix 5.1 for actual amounts). I. See appendix 4./ (1.9 million sq.ft), Bristol (2.2 million sq.ft), Cardiff (0.8 million sq.ft I, and Leeds and Derby (eachabout 0.6 million sq.ft). Most of the towns with between 0.2 and 0.5 million sq.ft of post-war officespace were located in at a the North and Midlands, but two (Plymouth and South- votaawasaaaaaramaaar'soma ..... 11111111/1Meoligaseallle MIMI flab hampton) were on the south coast. Apart from New- /1111111111110/0 castle, towns in the North-east reportedvery little office apilliMailumaimmiaailefft811111* building. _111111110Mir ,0111111.1p0.111_ Information from the county councilareas outside the South-east has been sparse. The indications fromChesh- ire and Staffordshire (which between them hada mini- 11111.7.11M1F., mum of 1.6 million sq.ft approved) were that theremay 111111L-- have been a fair amount of office building in thecounty areas of the West-midlands. This may also be true of the county areas (particularly their industrial parts) of Lanc- ismr, plum ashire, Yorkshire and Durham. Monmouthshirehad approved over 0.6 million sq.ft but the largest figure approved by any of the other five Welsh counties which replied was only 50,000 sq.ft. The outstanding fact which emerged from thesurvey was the high concentration of new office building in edlia London and the south-east of England. This hasbeen :01. II. C .4,11M k.L 11111&. due to several reasons, but firstly the needto replace buildings damaged in the war and secondly,to accommodate the increasing demand for officespace. This demand has been created by the `drift' of populationto the South-east in the last three decades and by the desire of many firms to establish the prestige symbol ofa Lon- don branch or headquarters. Figure 20 The tremendous expansion of office construction in the New office building in London suburb South-east has caused serious concern in official circles for the 'formidable transport, housing and financialpro- blems in its wake. ..', the effect of these problems upon the lives of Londoners, 2 and the depressing effect ofthe population migration upon the regions. " A policy of dispersal from London has been advocated by the Town and Country Planning Association since at least 1958 and this was eventually adopted by the Conservative government, which stated its policythus:' `The Government intend to tackle the office problemin the following way: (a) to make planning con-(by modifying the rules trol over new office build-which previously ing still (sic) more effectivepermitted re-developers to add 10 percent to the volume of the old buildings. With the aid of modern construction and planning techniques, this made it possible to add up to 40 percent to the effective floor area of rebuilding schemes) (b) to get as much govern-(over three-quarters of Figure 21 ment work done outsidethe Civil Service already Most post war offices have been built in city centres. central London as possiblework in regional and local offices and the (i) to commission an enquiry by the Economist In- possibility of moving out telligence Unit into the reasons why companies have a proportion of the their offices where they do 100,000 now remaining (ii) to accumulate and act as a clearing house for in London is to be information on possible re-locationareas outside considered) London (c) to encourage the prov-(such as the new towns (iii) to propagandise the economic and otheradvant- ision of more office centresand other centres as ages of relocation. outside the heart of Lon-distant as Norwich and The EIU study 5 has only recently been made available. don, including places rightReading) It provides a great deal of information about companies' away from London reasons for their present location, and their attitudes to- (d) to make a major effort(the principal instrument to dissuade private employ- for this will be the 2. Minister of Housing and Local Government: London: Employ- ers from opening new offi-Location of Offices ment; housing; land. Cmnd 1952. HMSO, 1963 (February) ces, or extending their pre-Bureau (LOB) which has 3. Ministry of Labour figures for 1959-60 suggestthat in that sent ones in central London ;now been set upin year the London Region gained about 40,000 insured employees from Scotland, the North of England and Wales (Townand and to persuade employerscentral London) Country Planning Association:The paper metropolis.(The Associ- already there to decentral- ation) 1962) ise work which is not clear- 4. Minister of Housing and Local Government:op cit (2) ly tied to London'. 5. Economist Intelligence Unit: A survey of factors governingthe location of offices in the London area. (Location of OfficesBur- By mid 1964, the main activities of the LOB had been: eau), 1964

37 wards decentralisation. The principalreason given for many problems as they have solved : they are exposed to location in central London was a need for contact with the noise from more than one street and, insome cities other organisations. But more than half of all the firms (perhaps especially in London) they have createdmajor who gave information had, in fact,no need to be in visual and aesthetic problems. London for that purpose. Other importantreasons given for a central location were tradition, prestige, andease of communication with the remainder of thecountry. It is difficult to assess the potential usefulnessof LOB. The annual increase in the number of jobsin central London during recent years is believed to havebeen almost entirelyinoffice employment and to have numbered at least 15,000.' So far, LOB hasbe'n able to influence the movement outwards, from the centre of London, of less than one-fifth of this number, andmost of this only as far as the London suburbs. Although the evidence of the Unit's locationenquiry is incomplete, it seems clear that the bulk ofpost-war office building has occurred in city centres, wherethe most severe environmental problems are likely to arise. Sites are expensive and therefore restricted in area; adjoining buildings overshadow; city centre trafficcreates a con- tinual noise problem; bothaccess and car-parking are likely to be difficult.- The office buildings inwhich the more detailed surveys of the Unit's environmental study have been made have, with one exception, been offices on such site, with difficulties similar to the ones described. The exception, an extensive single-storeyopen- plan office which is located in a suburb ofa large city was included because it provides an interesting alternative to the type of building necessitated by citycentre development and one which the mountingpressures to decentralise may make more common. During the preliminary visits, office managementswere asked why they had decided to takespace in their present buildings. The answers naturally ranged widely but one or two recurred frequently. One was that, at the time a move to a new building was contemplated, the building in which the offices werenow located was the only one with suitable space available. Situation was probably the most important single criterion: central location within or accessible to the commercialcentre of the city or other business foci wasa general requirement. Many businesses had been locatednear railway stations and main shopping centres because it had been thought that these features would attract staff. A few firmshad considered moving to the suburbs but had been deterred by possible difficulties in persuading office staffto follow; a small number of market surveys by individual firms had confirmed that these difficulties might be real. The economic implications ofa decision to locate an office or to rent space in a particular situationcan be substantial. LOB has calculated, for example,8that each office worker in central Londonmay cost his employers £275 p.a. more than one outside. Of this total the higher rent of office space amounts to £180, higher rates on office space £45, and the central London salary differential £50. The traditional pattern for the planning of the business areas of British cities consisted of `corridors' of buildings facing each other across the street. By itsvery nature this plan can be expected to create environmental problems. Daylight, especiallyon the lower floors of buildings, is likely to be obstructed anda canyon is created for sound to reverberate within. The London Coun- ty Council operated a `Daylight Code' 9 which essentially consisted of angular limits in plan and sectiongov- erning the bulk of the building. This supersededthe previous (and elsewhere more general) controlby cut- off angles and set-backs and was conceivedas permitting a greater variety in the form and layout of new buildings whilst ensuring that they and their neighboursob- tained better daylighting. In practice the Codedoes not 6. Town and Country Planning Association:op cit (3) appear to have been as successful as was hoped. Build- 7. This was written in 1964 ing-lined streets have sometimes been avoided, andsome 8. Town and Country Planning Association: PlanningBulletin: buildings have suffered less obstruction than they might quotingThe Guardian,1963 (October 1) otherwise have received. But it seems possible thattower 9. Ministry of Town and Country Planning:Redevelopment of Central Areas. HMSO, 1947. See also:Crompton, D. H: The buildings which have been permitted have createdas daylight code. Town Planning Review, 1955 (October) 38 Chapter 5 The spatial environment

The basic criterion ofspace in an office building is that there must be enough for incr-lsed aesthetic value and,if possible, increased the satisfactory perform- functional efficiency ance of the work. In a narrow interpretationof building economy a standard of just enough is (ii) the building fabric (eg,the structural frame) the efficiency level; should be subservient to the in practice, minimumstandards are rarely found and office space, and should there is a general expectation not detract from the functional,economic or aes- of additionalspace. Mini- thetic value of this space. mum space is regarded asmean ; more space than is functionally demanded contributesto enhanced quality and usually confers Legislation status to buildings, rooms, their The unit of space in occupants and even the work. This an office is the amount needed by aesthetic value of individual office workers,or by a particular office-pro- extra space applies throughouta building, to stairways as well as offices, to lavatories cess. Recent legislation' requires that,except in the case as to conference rooms. of rooms to which members The spatial criteria to be of the public haveaccess, applied to a building will de- floor areas should not be less pend, therefore,upon many factors, including the loca- than 40 sq.ft orroom vol- tion and purpose of the building, umes less than 400 cu.ft perperson. These requirements its place in the hier- will become effective in archy of its users and thestatus of the occupants. The 1967, ie, three years afterthe owner of a prestige office building will vesting date. There isa general requirement that while be prepared to work is going on noroom may be so overcrowded accept (and will in fact expect)a lower ratio of usable as to office space to total floor cause risk of injury to health, andconsideration must be area of building than will the given not only to the number of owner of, say, a works office. For thesame reasons, a people who may beex- pected to be working in theroom at any time, but also to managing director will be providedwith several times the space occupied by furniture,fittings, machinery, etc. more floor area than a typist. It is thereforedifficult, if There will be few not impossible, to define spatial cases of modern office buildings being standards. What can be occupied at such high densities. done is to state generalisationswhich are applicable to both office buildings and officespaces, eg: (i) whatever the quality level Previous surveys of the spatialenvironment in offices of the office (buildingor The Civil Service study 2 drew room) there must be sufficientspace for adequate attention to a minimum space per worker being dependenton standards of vent- functional performance of thework. Any space add- ilation which would avoid itional to this basic requirementmust contribute to air stagnation and body odours. To provide the 1,000to 1,200 cu.ft of air perperson per hour considerednecessary dictated 400 cu.ft of space per person, assumingthat 3 air changesper hour were obtained. Abovethis minimum,space standards were seen to be affectedby, among other things, the shape of rooms, the demandsof natural lighting, furniture and the status of botha room's occupants and their work. During wartime the generalrange of space provision in the civil service(including space occupiedby furniture) was about 50 to 100sq.ft per person. There- port recommended `comparativelylarge aggregations' of staff rather than smallgroups. Standards of 60 sq.ft per clerical worker and 40 sq.ftper typist were thought to be 'not unreasonable'where little filing and interviewing was done. The office lighting survey's analysis3 showed that for all types of employment the numberof workers perroom was most typically between two andfour. At 61 to 80 sq.ft per person for the modalgroup and a substantial

1. Offices, Shops and RailwayPremises Act 1963. HMSO 2. Study Group appointed by H.M. Treasury: Workingcondi- Figure 22 tions in the Civil Service. HMSO,1947 3. Gray, P. G. and Corlett, T: Structural column occurring within workingspace. A survey of lighting inoffices. Appendix to Post-war buildingstudies no. 30. HMSO, 1952 39 proportion (28 percent) ofover 100 sq.fz per person, Space standards standards in national and localgovernment were higher Apart from legal minima, the than in manufacturing industrywhere the comparable space standards of office workers will depend upon suchconsiderations as: space allowance was 41 to 60 sq.ft. (i) the policy of the organisation In Liverpool, gross overcrowding does not seem to have (ii) the status of the officewithin the organisation's been general. A survey made in1955 (and reported in hierarchy 1958)4 showed that thirteen rooms out of an unstated (iii) the status of the individual total in 136 buildingswere considered to be overcrowd- (iv) the type of officespace in which the work is to be ed (by Gowers Report standardsof 400 cu.ft per pers- done (eg, single private office,open-plan general on including the space occupied byfurniture). A report office) published in 1959 ofa survey made the previousyear, noted that overcrowding in (v) the furniture and equipmentto be accommodated many small rooms was due (vi) the needs of the individualor his work for quiet to the space occupied by equipment butmentioned only or privacy two rooms, out of an unstated total (used by 269 busi- (vii) the space available (whichin turn may be in- nesses in 47 buildings) where space standardsfell below the Gowers minimum. fluenced by projected expansionor change in the work). Space was measured butnot reported upon in the Bar- row-in-Furness survey. The space allotments for differentcategories of staff in In the City of London 8 only 71 Imperial Chemical Industries Limited percent of rooms fell below the Gowersspace standards and the Civil Ser- vice have been published.' Enquiriesaddressed to sev- while nearly 75 percent hadover one and a half times eral other large business organisations that minimum; overcrowding about their spat- was as likely to occur in ial standards for different grades post-war as in pre-war buildings. of employee produced two additional lists. These are comparedwith the al- Table 3 ready published lists in table 3. Space standards adopted by certain organisations (square feet per person) Civil Service Civil Service ICI Ltd. Firm X austerity long term Firm Y standard 1948 standards 1955 (may still be (now being applied in force) to all new buildings)

Permanent Secretary 500 - 550 Deputy Permanent Secretary 400 - 450 Directors Secretary Under 350 Deputy Top Secretary 250- 350Grade A 290 Secretary executives 288 Directors250 Assistant Senior Under 200- 250Secretary 200 - 250 Senior Secretary executives 210 executives 200 Assistant Two senior secretaries Secretary 150- 200Principal 150- 200 B 170 sharing 210 One senior SEO 120 -150 secretary 140 Managers 125 HEO 100 - 120 Technical Pt incipal 80- 100EO/ HCO 75- 90 Assistant staff 100 managers 85 SEO PA's HEO 60- 80 Supervisors

C Supervisor 85 Secretaries Clerks Machine 65 - 70 D operators CO 55 - 65 E}Clerical60 Typists EO / HCO 40- 60 CO 40- 60 F Typist / Secretaries 60 Typist 40- 60Typist 40 - 60 clerical 50 Clerks 50 Sources: Anon: Better Offices. Institute ofDirectors, 1960, and PRU enquiries

The author of the mechanisedoffices report 9 considered that the most surprising conclusionto emerge from his 4. Wattleworth, W. H: Unfinishedbusiness in environmental examination of space was the relativelylarge amount of health. Royal Society of Health Journal,1958 (May-June) space needed for automatic data processing 5. Home Office and ScottishHome Departm2nt: Health, welfare systems. and safety in non-industrial employment.Hours of employment The average in his eight buildingswas 85 sq.ft per mach- of juveniles. Cmnd 7664. HMSO,1949 ine position; satisfaction withthe space providedwas 6. Wattleworth, W. H: Theforgotten army. Royal Society of only 45 percent among clerical Health Journal. 14)59 (March-April) workers, up to 58 per- 7. Neuun, I. D. M. and Morse, R. J: A pilot study ofoffice cent among machine operators. Themore recent paper accommodation. The Medical Officer, from the Building Research Station 1960 (September 9) " stated that for a 8. Robinson, Alan: Working inthe city: the interim report ofa `satisfaction level' of 75 percent,"minimum space survey. City of London Health Department,1962 standards would be about 45 sq.ftfor copy typists, 55 to 9. Langdon, F. J: The designof mechanised offices. Architects 60 sq.ft for clerks and from 100 sq.ft Journal, 1963 (May 1 and 22) upwards for senior 10. Langdon, F. J. and Kelghley,E. C: User research in office management. Surveys in more than 2,500office rooms design. Architects' Journal, 1964(February 5) showed that furniture occupiesbetween 25 and 30 per- 11. That is, the point at which, innineteen cases out oftwenty, cent of the floor space. seventy -five percent of the staff would besatisfied 12. Anon: Better Offices. Institute ofDirectors, 71960 40 11111*0111111111

-ost47.71. 4.44 .4. 411.

, t p !)I r") 7

74 t crk, 'i7r1r cr"c a0 ) a P a p cl.T.P c^.2.t Cr'

C;3 Figure 23 Figure 25 Furniture: one of the factors which determinethe standards of the spatial environment. Design for Japanesegovernment offices whose dimensions are determined by units comprising workinggroups of Ike clerks. is-4 isonsomsesasits

1 Spatial considerations in the planningof office space MtI The determination ofspace allowances and room dimensions appropriate to office functions 1 has received negligible attention. One impressionresulting from visits to office buildings has been that the basisfor design seems to be quite arbitrary : office workappears to be 8ti fitted into whatever space is available,rather than the building being planned to facilitate officefunctions. A seemingly obviouscourse would be to parallel indust- A . rial processes and design aroundthe work-flow of paper. Cto f., %. However, no studies of typical officeclerical procedures 1 which could be used to determineor even help deter- : mine I % office space are known. '3 Similarly,so far ascan . II it be seen, nobody in Europe has madeany analysis of the size of working groups, a AA or of the pattern of use of office Figure 24 space. Some of these matters have been investigatedin Frequency distribution by size of workinggroups in certain Japan and, while the findingsare almost certainly not Japanese government offices Ix number of clerks in a group: applicable to British conditions, theprecedent is a use- y number of groups).

- apt .

Figure 26 Typical arrangement of desks ina regular pattern. 13. Apart from mechanised procedures.See Langdon, F. J: op cif (9) 41

) t

Figure 27 Biirolandschaft at British Petroleum Company.

1.1 Lt 1 r

Figure 28 Plan Cross aisles should be provided every 25 ft'.17 Similarly, `Aisles are usually 3 ft wide, but 2i ft suffices, except for ful one. One Japanese architect (Takano), 14for ex- main ones...3 to 31 ft is allowed from the back edge of ample, examined the sizes of working groups in certain one desk to the front edge of the one behind...when a Japanese government buildings, found that the most clerk must pass behind the chair of another to getaccess usual number was five persons and that other peaks in to an aisle . 1" There are few departures from the un- the frequency distribution occurred at 10 and 15 persons. imaginative layouts which these quotations illustrate As a result of this study his government's offices were and only one British example is known (and thata small subsequently designed to a width determined by the unit scale one, at the British Petroleum Company's London of five desks and a gangway. Similarly, to arrive at the head office) of ' biirolandschaft '. 19 This isa form of lay- size of waiting spaces for provincial employment exchanges, Yagi and Takano'" made extensive studies of 14. Kensetsu-Sho Eizen-Kyoku Kenchiku-Ka (Building Section, the manner of use of existing employment exchanges, Building and Repairs Bureau, Ministry of Construction): Akita- including the 'customers' time of arrival and departure, KenchOsha No Sekkei (The design of Akita prefectural govern- the times they spent waiting and consulting and the ment office). Kokyo-Kenchiku (Public Buildings) ?1961 (in Jap- number of people concerned. In respect of drawing anese) 15. Yagi, A. and Takano, T: Chosa to Sekkei- PESO (Survey offices, the relation of drawing board and reference and design of Personnel Security Offices). Kokyo Kenchiku tables has been systematised and the space utilisation of (Public Buildings) ?1960 (in Japanese) different arrangements shown," but apart from the 16. Langdon, F. J: Design and equipment of drawing offices. Part 2, layout and equipment. Engineering, 1959 (April 24) work on ADP systems already mentioned, the manner 17. British Institute of Management and Office Management in which individual and grouped clerical workplaces are Association: Office accommodation. (The Institute, and the Ass- organised does not appear to have been studied. ociation), 1953 Desks for clerical work are normally placed in rows: an 18. Curtis, C. R: 0 and M in the smaller office. Part 4, planning the layout. Office Magazine, (now Office Methods and Machines) office management handbook suggests that ' Main aisles 1963 (March) should not be less than 4 ft, and sub-aisles 3 ft wide. 19. Duffy, Francis: Biirolandschaft. Archt Review, 1964 (Feb.) ustrial laboratories has shown a clear correlation, in the case of that building type, between ceiling heightand cost of building. But there may still besome aesthetic justification for heights greater than thebyelaw minimum, especially in large open offices, and thiswould have to be balanced with the economicand acoustic implications of increased height. One factor which will affect sensationsof room height is the amount of light playingupon the surface of the ceiling. Where light fittingsare fixed flush with the ceiling, then the only light reaching thatsurface is reflected, .. mainly from the floor and desktops. The effect is often rather gloomy. The floor to ceilingheight of most office buildings constructed at thepresent time seems to be within the range of 9 to l I ft.

The size of individual officespaces Sizes of office spacesvary greatly, from the one-person room to Ink, completely open flooraccommodating several hundred people. This last formis very common in the U.S.A. and thereare signs of it becoming more usual in Britain. The main determinantsare the nature of the work and the policy of theorganisation.

The use of 'open-plan' offices It is possible that largeopen offices may often be more efficient workplaces than combinationsof smaller spaces because of the opportunities theyprovide for easier Figure 29 supervision and communication. Inaddition, floor space A single-person room forminga planning unit. can be used more intensively. Due to the reductionin circulation area, and the out in which desks and office equipment more efficient use of space are clustered in which open planning permits,the space requirements functional and non-regularpatterns within large open office spaces. per person in open offices are usuallysubstantially less than in small rooms. Servicingalso (eg, lighting and

MIMMIM,1111'

I I

art.Alla

"lb

46, 'Oa Figure 30 Single storey, open-plan office building(West Midlands GasBoard).

Height of office spaces ventilation) can be more efficientlydesigned. During the Unit's preliminary visits,people often ex- It is commonly expected that staff pressed a desire for greater height will choose to work in in their office space, small units of space in preferenceto large ones. But 'such as we had in our old building'(which was prob- managers, too, are likely to press for their ably Victorian). What height should staffs to be be provided does separated from those of other managers'by partitions, not appear to have received muchconsideration. The even though there may be no functional loftiness of many 19th century offices justification for derived, most like- separate rooms. A compromisecan often be arranged by ly, from lighting needs: highwindows were necessaryto obtain sufficient daylight at the far side of large rooms. 20. Hopkinson, R. G. andLongmore, J: The use ofpermanent With modern developments inelectric lighting this is supplementary artificial lighting. Architects'Journal, 1959 (Oct- no ober 8). Figure 1 in this longer a valid reason for high ceilings.A survey of ind- paper is derived from the work of the Nuffield Foundation, Division ofArchitectural Studies

43 any objective consequences for the employer or individual arising from the different sizes of working area?'''' Many studies of group and industrial psychology have demonstrated the subjective importance of groupmem- bership and the feeling of belonging. Thereare also important consequences to a person's satisfaction in his work and his efficiency in the way that he conceives his relationships with his section, department andcompany; it may be that these are affected by the size of the working area or group. Small office units hold the possibility of highly stable work groups: cohesive forces deriving from eachsep- arate group's common aims and objectives, its supervision and leadership, and the physical proximity of the members. The physical closeness of thegroup will be influenced not only by the absolute distances between working spaces, but also by the use ofa common entrance, and the usual interactions on non-business topics when staff are circulating around such focias the document stores, postal trays and telephones. But both special possibilities and special restrictions forgroup organization result from the small officearrangement. Figure 31 A group of offices separated by glazed screens. MIL providing partial screening, eg, by glazed partitions, isolated screens which define space but donot conceal, cupboards, filing cabinets and other office furniture. The output of noise from many office machines is frequently advanced as an argument againstopen offices. A common practice, and one which is often necessary, is to place machine staff and their equipment inrooms separate from those occupied by un-mechanised clerical staff. Interference from noise can be reduced inopen offices by arranging that walls and partitionsare as far apart as possible (ie, of the order of 50 to 60 ft [or more]) to reduce reflections, and by using acoustic ceilings placed as low as possible. 21

Figure 33 Members of different working groups meeting in an open-plan office. The possibilities are of a stable group witha clear concept of itself as a discrete and simple entity. Thismay be good or bad, but it effectively restricts the forms of group structure that are possible because the staff occupying small areas are kept from interaction withmem- bers of the larger working unit by the walls which divide them. By contrast, open-planning increases opportunities of inter-personal contact between members of the smaller work groupings. It is possible that there would beno change in personal relationships or group structureas a result but this seems unlikely. Open planning inevitably Figure 32 Demountable partitions. results in members of different sections coming into frequent contact with one another as a result of sharing The influence of room size on the members of working entrances and services and the necessity of passing groups through, between or around the periphery of other It is possible that offices can be built to greater depths groups. In addition, in open-plan offices, fewer workers will be seated beside partition walls and than has previously been supposed and still providea so more people satisfactory environment. However, the possibility of will have greater opportunities of contact withothers deeper buildings may be qualified by a probable im- than they would have in smaller units. portance of unobstructed views of windows, which It follows that, if the physical proximity of individuals suggests that in order to obtain the advantage of deep has an important influence on the formation of inter- personal relationships and thus of friendship buildings, it may also be necessary to acceptopen workgroups, in ing areas. situations where several groups occupyone space there Office staffs' possible preference for small officesor partitioned areas raises the whole question of theaccepta- 21. Anon: Speech communication and the shape ofrooms. bility of open plan offices. Two points arise: firstly, how Architects' Journal, 1957 (September 12) strong are the feelings involved, and secondly, are there 22. See appendix 5.1 for a full reportupon an experimental investigation of this topic, completed since this reportwas drafted 44 may occur conflicts of loyalties and identification to- respondent had earlier come, may bemore influential wards and with a particular group, becausesome mem- determinants of subjective response to office conditions bers of each section will be in very close proximity to than the physical environmental variables themselves,at members of one or more of the other sections. least within the range of physical variation found in the It might be found that, in the absenCe of partition Walls, building which formed the basis of that investigation. ls sheer physical proximity becomesa potent force in determining who a person sees as the close colleagues with Spatial considerations in the design of office buildings whom he identifies himself. If this shouldprove to be so, The shapeofOffice space then the effect could be a considerable weakening of the A particular site may impose limitationsupon a design- structural groupings which managements are often at er's freedom to shape his offices. Otherwise, theshape of great pains to promote. But, whether any different be- individual office spaces is most likely to be determined havioural or attitudinal consequences follow from closed by daylighting policy or a requirement for theuse of or open office planning, and whether these are good or private offices. bad for the company and the individual,are questions Where full daylighting from side windows is required, that can only be answered by empirical research.'2' then this consideration is the most importantdetermin- Perhaps the most celebrated attempt to apply the meth- ant. General offices (say for 20 people or more) will ods of social psychology to considerations of the effect tend to be long and narrow; the depth from window of building design on the individualwas a study of the wall will rarely exceed 15 ft. Similarly ifa large number effects of spatial arrangements on group formation in of private offices are required, then the depth ofthe pri- various types of housing. '24 This concluded that .. vate offices (which may be a maximum of 12 to 15 ft) will the relationships between ecological and sociometric tend also to be the depth of general offices. structures is so very marked that there can be little Where an office building is built for letting, the opti- doubt that in these communities passive contactsare a mum size of space for individual tenancies will deter- major determinant of friendship andgroup formation'. mine the planning unit. If tenanciesare expected to be It may be assumed that the spatial setting of the cleri- small, the tendency will be for the depth of all office cal worker has important psychologicalconsequences. space to be determined by the maximum acceptable Indeed, one of the main conclusions ofa study of depth for single rooms, say, not more than 15 ft and the government typing pool staffs was that there were '. . . width (or length) of any particular officeroom will con- significant variations between the degree of satisfaction sist of an additive series of private room widths. Iften- in different pools. Part, but not all, of the variation ancies are expected to be large, then whole floors of could be accounted for by the different sizes of the pools'. office buildings might be designed to be the plan unit in This illustration is drawn from onlyone of the many the expectation that they will be usedas open-plan studies which show the influence of group sizeon be- offices. haviour, production or satisfaction; it has been singled out because of its concern with the typing pool which is The number and sizeof floorsin an office building usually regarded as one of the clerical functions most The distribution of the total net floorarea of a building easily dealt with in a programme of rationalisation and upon several floors is most likely to depend upon one or concentration into larger working areas. more of the following : When asked to express their preference for size of office '2" (i) the site, and any restrictions (suchas floor index typists in one of the offices studied by the Unit respond- ratio) it may impose. In some locations planningre- ed in a similar way to the rest of the clerical employees, strictions may limit the height to conform with ad- ie, with a preference for the smaller spaces. Machine jacent or surrounding buildings. Elsewhere,a greater operators, on the other hand, made significantly more height may be required to be provided thana devel- choices for a large open area. Two possible explanations oper would wish suggest themselves: in the first place, the typists had only (ii) policy decisions, eg, about daylighting (a demand recently been placed in very large pools, whereas the for daylight limits building width and therefore, given machine operators were accustomed to the largerareas, a maximum length, the area per floor) or the prestige ie, they were expressing a preference for what they knew or advertising value of, say, high buildings best. An alternative explanation is in terms of actual (iii) (in respect of buildings to be tenanted)assump- differences between the two groups: itmay be that tions about the optimum size of offices for letting typists and machine operators self-select themselves for (iv) (in respect of buildings to be occupied by their their occupations on different criteria. The typistswere owners) decisions about the actual size of departments largely composed of audio-typists, with smallernum- or working groups bers of copy and stencil typists; the machine operator (v) economic considerations. For example, increasing group was mainly composed of punch operators and height results in increasing cost anda decreasing us- smaller numbers attending the tabulators and ancillary able (or lettable) proportion of the totalarea. equipment. Although the demands of the jobsmay be The office location study 29 showed that the numberof somewhat similar, typists tend nevertheless to be regard- storeys in typical post-war office buildings ranged from ed as having a proper clerical occupation, whereas three to twelve. machine operation is viewed as a quasi-industrialoccu- pation. Consequently it would not be surprising to find Theformofoffice buildings fundamental differences in the attitudes, preferences and There are two basic plan-types, which derive from the expectations of the two groups. This, indeed,was found method of providing access and services: (i) the slab- when an analysis of replies to a questionnaireon en- type block, which has an access corridor in the middle vironmental conditions showed that the machineoper- ators differed from other female grades in over 60per- cent of all cases, clearly marking themselves off as a 23. This research was continuing during the period in which this separate sub-group. 27 Typists formed the next most report was being edited and published. See appendix 5.1 24. Feininger, L., Schachter, S. and Back, K: Socialpressures homogeneous group of all the women staff, though in informal groups. Tavistock Publications, 1959 (First published differing in only one third of cases. in USA in 1950) 2nd edition 1963: Stanford, Calif: StanfordUP These considerations underline the need to specify the 25. Matey, E., Baines, A. H. J. and Stanafield, R. G:Work groups to which a particular piece of research is applic- preferences of typing pool staff in the Board of Trade. Unpub- lished paper, quoted in :Walker, Nigel: Morale in the Civil able. Research based on one sub-group has often been Service: a study of the desk worker. Edinburgh University Press treated as though its results were applicable to all clerical 1961 groups, but the investigation referred to above showed 26. See chapter 10 unequivocally that such personal variables as age, 27. See chapter 10 sex, 28. See chapter 10 status, and the type of accommodation from which the 29. See chapter 3

ammii .41

would be appropriate to a strictly functional building, whereas a ratio appropriate to a prestige building might be two-thirds "'2 or less.

Planning dimensions and modules The dimensioning of office buildings, as with many other building types, is now commonly carried out to unit dimensions, or 'modules'. These ensure a unit size for offices by providing a regular grid for the positioning of partitions. Windows and heaters, light fittings and other fittings have to be positioned in a regular pattern. The most general use of modules is along the length of the external wall and this will be related to such structural dimensions as the lateral spacing of stanchions. The ad- option of a particular module may depend upon (i) whether offices are mainly private, or (ii) whether offices are mainly open-plan. For office planning purposes modules are likely to be important only when (a) some private offices will be needed and (b) in open-plan offices when the external wall is a solid pierced by windows, with substantial areas of wall between the windows. Otherwise, modules are likely to be used mainly for their Figure 34 constructional convenience. Central corridor in a slab-type plan. IMINE111116710111Kriiilar and offices on either side, and (ii) the core-type plan, where lifts, stairs, ducts, lavatories, etc. form a central core which is surrounded by a circulation area which in turn is surrounded by office space. The horizontal and vertical circulation systems affect both the economic and functional efficiency of the building. If the building is expected to be tenanted, then the slab type plan may be inevitable, for the need will be for small units of rela- atively shallow space which can be let off as units. Be- ANL cause of its duplicated circulation, a core-type plan would tend to be economic for letting only if large tenancies (ie, whole floors, or deep offices) were antici- pated. To obtain the most efficient and economic distribution of office space, simple slab or block plan shapes are preferred. Intersecting blocks such as cross and double- cross plan-forms create overshadowing and (possibly) unacceptable areas in the `dead space' at the point of intersection.

..LLEaniii- Figure 37 Office building designed to allow sub-division of open floors into smaller spaces. Size of individual single-window rooms can be varied within limits of lines on walls between windows.

If planning is based on an expectation of private offices, Figure 35 then the minimum dimension of a single room (along Slab-type plan. the window wall, usually) will be the determining factor. This is likely to be of the order of 8 ft33' 34' 35and, because they will often be related to window units as well, modules of between 4 ft and 4 ft 6 ins are common. Where there are wide areas of wall between windows there is scope for considerable variability in the size of individual rooms. Alternative (square) modules of 3 ft 8 ins, 4 ft or 4 ft 4 ins (giving `small', 'standard' and `commodious' space alternatives) have been recommended, and examples given of individual rooms and open-plan offices plann-

30. Besides offices. this includes canteens, conference rooms, archives, and all other ancillaries which actively assist the main r work of the firm 31. Strictly, net floor area to gross floor area Figure 36 32. Manasseh, Leonard and Cunliffe, Roger: Office Buildings. Core-type plan. Batsford, 1962 , 33. Manasseh, L. and Cunliffe, R: op cit (32) Plan ratios 34. Fischer, Manfred: Von den Planungsgrundlagen biszur The usable proportion of a building's total area is most Arbeitsplatzanordnung. Das Rationalle Miro, 1962 (5) 35. Marr, A. N: (contribution to discussion): Office Buildings: dependent on its 'class'. For example, a high ratio of net report of a symposium held on 12th April 1956 at the Royal (or lettable) space 30 to gross space 31 of (say) 80 percent Institute of British Architects. (The Institute), 1956 46

1 ed on these bases. r"., A wide variety of modules has been used for office buildings, varying from about 2 or 3 ft "' up to 20 ft but there is a large measure ofaggre- gation around 4 ft 3 ins or 4 ft 6 ins up to about 7 ft.

36. Cowgill, Clinton H: Executive architecture.J Amer Inst Archt. 1959 (April) 37. Cowgill, Clinton H: Planning the law office. AmericanBar Association: Economics of law practice series, pamphletno. 4 (The Association), 1959 38. Ministry of Public Building and Works: Dimensionalco- ordination for Crown office buildings. HMSO, 1964 39. Holford, Sir William: (contributionto discussion): Office buildings: report of a symposium heldon 12th April 1956 at the Royal Institute of British Architects. (The Institute), 1956

47 Chapter 6 The visual environment

Previous surveyst$f the visual environment in offices lighted working points in each room visited in the City of Of all the components of the total environmentwithin London. 6 The illumination at both positionswas meas- offices, the one which has perhaps receivedmost atten- ured in two stagesfirst from daylight and then from tion is lighting. Even so, a studygroup, reporting in daylight plus full electric lighting. At the timemeasure- 1947 on the work it had done during thewar, concluded ments were made, 37 percent of the best and 75 percent that the most urgently needed reform in CivilService of the worst work stations received less than 30 Im./sq.ft working conditions was improved lighting.' It heldthat from daylight alone. At least 90 percent of allmeasure- a first essential in the design of an office building is the ments of illumination from electric lighting were below provision of good natural lighting, and suggestedthat the IES standard 7 of 30 Im./sq.ft which isnow current. the depth of rooms should not exceed 11 timesthe I ,347 observations of glare suggested to theone observer height (where windows occuron only one side) or four that in City of London offices thiswas not as acute a times the height (where windowsoccur on two opposite problem as bad illumination. sides). It was reported that the area of windows in im- The most satisfactory feature' of the eight mechanised mediately pre-war commercial offices usuallyexceeded offices " was the quality of daylighting, although the crit- one-third of the total external wallarea. Standards of erion of satisfaction was not, apparently, the abilityto lighting were suggested which would providevisual con- work by daylight alone. Daylight factors calculated in ditions considerably better than those whichwere curr- the worst positions varied from 20 percent downto 1.5 ent, including values of illumination on desks of 10 Im./ percent. However, in view of the practical difficulties of sq.ft (with a minimum of 8 lm. /sq.ft) for clericalwork, calculating daylight fpctors, these values must be treated and general systems of lighting to avoid the possibilityof with some reserve. Illumination from electric lighting glare from desk lamps. Light matt colourswere consid- ranged from 9 to 29 Imisq.ft but the staff everywhere ered advisable; cream ceilings and light buff wallswere were reasonably well satisfied'; the chief source of dis- recommended. More tentative suggestionswere made satisfaction appeared to be variations in intensity and for consideration of the colour of office furniture. quality in different parts of workingareas. The office lighting survey' conducted by theCentral Off- The most recent publication of the Building Research ice of Information provided a great deal of information Station's work on office environment 9 has summarised about this topic. Its more important findings were that the results of a series of studies (most of whichstill in 1948 7 percent of office staff worked inrooms with- await publication), stating that, in modern offices, less out windows (though most of these had glass parti- than a third of the working time in winter andno more tions or rooflights) and 15 percent inrooms having areas than a half in summer, can be spent without supplement- of window glazing below the then current bye-lawmini- ary lighting. In south-facing offices in London whose mum of one-tenth the floor area. 21 percent of staffre- external walls are completely glazed, 17percent of the ceived no direct daylight on their work. Theaverage staff found the daylight inadequate while,at the same value of all measurements of illumination fromdaylight time, 24 percent, working near the windows, foundcon- and electric light combined was 15.4 Im./sq.ft butthe most ditions too bright. Electric lightingwas found to be 'one frequently observed value was only 8 lm. /sq.ft. Therewas of the most satisfactory features of thepresent day apparently slightly more satisfaction with electric light than daylight, but the report remarkedon the probability that electric lighting was assessed againstlower standards of satisfaction. Criticism of electriclight tended to be of quality rather than quantity. 1. Study Group appointed by HM Treasury:Working condi- Wattleworth reported 3 that lighting conditions in tions in the Civil Service. HMSO, 1947 Liver- 2. Gray, P. G. and Corlett, T: Asurvey of lighting in offices. pool, as measured during 1955 and 1958 showedpeople Appendix to Post-war building studiesno. 30. HMSO, 1952 working in inadequate daylightor in windowless rooms. 3. Wattleworth, W. H: The forgottenarmy. Royal Society of Illumination levels from electric lightingwere below Health Journal, 1959 (March-April) 4. Wattleworth, W. H: Unfinished business inenvironmental 20 lm. /sq.ft on the majority of desks in 660out of 913 health. Royal Society of Health Journal, 1958 (May-June) suites of offices.4 In 27 percent of offices artificial light 5. Nelson, I. D. M. and Morse, R. J: A pilotstudy of office was needed during most of the day. accommodation. The Medical Officer, 1960 (September 9) In Barrow,5 illumination levels in 11 out of 50 offices 6. Robinson, Alan: Working in the city: the interimreport of a were below 10 lm. /sq.ft and in 34 percent below 20 lm./ survey. City of London Health Department, 1962 (December) 7. Illuminating Engineering Society: The IESCode: Recomm- sq.ft even when daylight and electric lightwere used endations for good interior lighting. (The Society),1961 together. In half the offices the light fittings, and inone- 8. Langdon, F. J: The design of mechanised offices.Architects' seventh the windows, were visibly dirty. Journal, 1963 (May 1 and 22) 9. Langdon, F. J. and Keighley, E. C: User research inoffice Light was measured at both the best and worst naturally design. Architects' Journal, 1964 (February 5) 48 office', even though only a minority of installations When the brightness of the task has been set, the bright- reach the levels recommended in the 1961 IES Code.'" ness of the other parts of the environment are determined. The general rule is that the task should be bright- Legislation er than its background which in turn should be brighter The requirements of the new office legislation " are ex- than the general surroundings. pressed in only general terms which few existing offices The British IES Code suggests a relationship between could not claim to meet:Effective provision shall be the brightnesses of task, background and surroundings made for securing and maintaining, in every part of the appropriate to office lighting of the order of 10:3:1 but premises. . .in which persons are working or passing, adds that becauseof the limited data available this should sufficient and suitable lighting, whether natural or arti- be regarded as general guidance only. The ratio of mini- ficial'. The Minister may prescribe standards by regu- mum to maximum illumination over the working area of lations but none have yet been made. So far as is reason- a space should not be less than 0.7.13 ably practicable windows and skylights must be kept clean and unobstructed ; whitewashing or shading to Standards of ilhonination mitigate glare or heat gains is permitted; electric lighting The 1ES Code specifies a normal minimum illuminat- apparatus has to be properly maintained. ion for the interiors of buildings of 15 Im./sq.ft. Recom- mended standards for working areas of officesare: Requirements of the visual environment General offices 30 Im./sq.ft20 The requirements of the visual environment for office Business machine operation 45 Im./sq. ft work are : Conference rooms and (i) good lighting and, possibly, executive offices 30 lm./sq.ft (ii) a view, and Entrance halls and reception (iii) good colour. areas 15 lm./sq.ft

Good lighting Recommended illumination levels, in Britain and in Good lighting in offices has two aspects: (a) functional other countries, are ultimately based on judgement and ie, the service of the work and (b) aestheticthe creation are compromises between higher levels of illumination of a pleasing and appropriate character. Both are quali- required for higher standards of visual accuracy and the tative, but whereas the first is now largely definable in economic consequences of very high illumination levels. quantitative terms, the second still remains the province Standards differ between countries (table 4). of emotion and artistry (or design judgement). The two aspects are more than complementary for it is possible Table 4 that 'function' may constitute a significant proportion Comparisons of recommended illumination values for of the aesthetic content, and vice versa. general offices in different countries. lm./sq.ft (approx) In Britain the authoritative guide to lighting is the 1ES Code. 12 This publication states the three chief aims of France 1961 minimum 30 good lighting as the promotion of work (or other act- recommended 60 ivities), the promotion of safety and the creation, in con- Germany (draft 1962) 12 to 25 junction with the structure and decoration, of a pleasing Britain 1961 30 environment conducive to interest and a sense of well- Sweden 1962 30 to 100 being. The realisation of these aims involves: U.S.A. 1959 100 to 150 `(a) careful planning of the brightness and colour U.S.S.R. 1959 fluorescent 20 to 30 patterns within both the working area and the surr- incandescent 7.5 to 15* oundings so that attention is drawn naturally to the important areas, detail is seen quickly and accurately Source: Commission Internationale de L'Eclairage: Report of and the room is free from any sense of gloom or mon- Committee S-4.2: Lighting codes, regulations and legislation. otony; Proceedings, Vienna, 1963. Volume C. Paris (The Commission), 1964. (b) using directional lighting where appropriate to *Why the U.S.S.R. should recommend different values for differ- assist perception of task detail and to give good mod- ent light sources is not clear, it is presumably due either to greater elling; practical difficulties of achieving higher standards with incandes- (c) controlling direct and reflected glare from light cent lamps or to the higher running costs of such an installation. sources to eliminate visual discomfort; (d) in artificial lighting installations, minimising flick- While legal minimum standards of illumination will er from certain types of lamp and paying attention to tend to be exceeded by a majority of lightingusers 21 the the colour rendering properties of the light; 10. Mr. J. A. Lynes has pointed out that the only obvious con- (e) correlating lighting throughout the building to clusion to emerge from this summary of previous surveys is that prevent excessive differences between adjacent areas where lighting has been measured against a physical criterion, and so to reduce the risk of accidents; such as the IES Code recommended illumination values, it has (f) installing emergency lighting systems where nec- invariably been found inadequate. Where it has been assessed subjectively, it has always been found fairly satisfactory essary'.13 11. Offices, Shops and Railway Premises Act, 1963. HMSO Since this present report is primarily concerned with the 12. Illuminating Engineering Society: op cit (7) design of the workplace environment (and little con- 13. Illuminating Engineering Society: op cit (7) cerned with the functioning of the building as a whole) 14. ie, the angular size of the smallest detail which has to be seen, meas.:red in minutes of arc subtended at the eye items (e) and (f) above will not be considered further. 15. In the IES Code this is taken to be within a range of between (approximately) 90 and 100 percent ' of the maximum per- Brightness patterns forma Ice of any visual task which ideal conditions could yield' The pattern of brightness in a workplace has three main 16. By reference to the task's reflection characteristics 17. Illuminating Engineering Society (of the USA): IES lighting parts - the visual task, the immediate background to handbook. (The Society) 3rd edition. New York, 1959 that task and the general surroundings. The size of crit- 18. Illuminating Engineering Society: op cit (7) ical detail" determines an appropriate value of bright- 19. Illuminating Engineering Society: op cit (7) ness of the task for an acceptable high level of visual 20. Compare with the ' average minimum value of daylight illumination tolerable for clerical work' of about 5 lm. /sq.ft reported performance 15 and this is expressed in terms of the re- in Illumination Research Committee: Daylight illumination nec- quired illumination 16 on the task. These are the methods essary for clerical work. Illumination research technical paper developed in Britain. In the U.S.A. methods based on no. 19. HMSO, 1937 analyses of the process of vision have formed the basis 21. No regulations determining standards of illumination for office premises have been made yet (November 1964). In fact- of the very different recommendations of the American ories the generally applicable legal minimum level of illumination IES.17 (6 lm./sq.ft) is almost always exceeded in practice 49 recommended illumination values of the IES Codeare at present in advance of general practice. This is apparent from the standards of the Ministry of Public Build- ing and Works 22 for government offices (table 5) and most of the offices visited during this project. In only one of the six offices in which levels of illumination from electric lighting installations were measuredwas the present IES standard exceeded.The other buildings are probably representative of average or better-than- il average conditions, yet they allfell below the IES standard. Several of the buildings were substantially overshadowed; for this and other reasons it was rarely possible to take measurements of daylight factor. 2" r P Table 5 , -we Standards of illumination for government offices. Im./sq.ft

Location or type of Standards Standards task introduced in applicable 1960 (not yet before 1960 universal) (may still exist in buildings Figure 39 o:mpleted Light needed on vertical plane. before 1960) Control of glare General office work 23:15* 10-6* Glare is defined as 'a conditionofvision in which there Book-keeping, typing is discomfort or a reduction in the ability tosee signifi- and work involving cant objects, or both, due to an unsuitable distribution sustained reading 23-15 15-10 or range of brightness or to extreme contrasts, simul- taneous or successive, in the field of view'. "a Popular Source: Society of Civil Servants: Office accommodation and usage of the term may not imply the same experience as allied subjects. 1963. the one defined. Glare is not always immediatelyor First figure represents approximate average illumination when lamps, fittings and decorations are new and clean. readily apparent and people may complain of glare only when the condition is extreme. It may be direct,as from

Figure 38 Most clerical tasks are on horizontal surfaces. Directional characteristics of lighting alight fitting or window, or reflected,as from the polish- In office buildings illuminated largely by daylight ed surface of a desk. through side windows, lighting will tend to be mainly directional and have pronounced modelling character- Factors influencing direct glare include the brightness istics. Electric light fittings will be mounted overhead, and area of the light sources, their position and the at average brightness of the room background. Tolerance frequent intervals; from suchan installation the illumination received at a point on the working plane will limits, based on good current practice', of Glare Index tend to be omni-directional and shadowless. Measure- applicable to the office (not the task)are recommended ments of illumination are commonly made on the horizontal plane but many office functions (eg, filing) require 22. See: Society of Civil Servants: Office accommodation and the working illumination to be on vertical surfaces and allied subjects. (The Society), 1963 (January) 23. This point is discussed later in this chapter this may be more difficult to provide by electric lighting. 24. Illuminating Engineering Society: op cit (7) 50 in the IES Code; no comparable standards exist in effect upon an environment. This effect is bothphysical other countries although it is believed that the matter is (eg, colours with a high Munsell value "° willreflect under active consideration in several Europeancount- more light and therefore contributetothe general ries. illumination), and subjective (colours contributingto Reflected glare has to be controlled by the positioningof the character and 'mood' of a space). light fittings and use of a matt finish for potentialre- The colour rendering characteristics ofan environment flecting surfaces. depend upon the type of light and the colourand reflection characteristics of the surfaces. Daylight Control of flicker con- , ining direct sunlight will givea rendering of a colour Stroboscopic and flicker effects from fluorescent fittings different from sun-less north light; varioustypes of can be largely eliminated by suitable installation and electric lamp have varying colour rendering circuitry. properties.al The IES Code suggests that, where colourdiscrimin- ation is critical, illumination levels should be highand not less than 45 Im. /sq.ft.:$t In the offices in which illuminationsurveys were made, the usefulness of high reflective colours for theceilings had been recognised, but values for wallswere found to be much lower and for floors, variable. : °'In theone building where the office staff's opinion abouttheir colour schemes was sought the general attitudewas critical." This was surprising for of all the officeswhich have been visited this has the most highly developed pattern of decorations. It is subdued yet sophisticated and reasonably well integrated through all theroom surfaces, furnishings and fittings, yet the staff chosethe words 'cold', 'mediocre' and 'dull' to describe it.It seems that the criticisms were, in fact, of the very sophis- tication of the 'colour design,' and its tendencyto cool colours."5 Langdon foundsimilar resultsin his studies ofmechanised offices: 'almost all the colourtreatments were rated too cold'. ""

Aesthetic effects of lighting It isnot possible at this stage to say anything about qualitative aspects of office lightingover and above those affecting its functional performance. Whilstmost people believe that such higher aesthetic values exist,it has yet to be provedthatthey do. Certainly, no visual environments experienced during thecourse of this studyoffer any suggestion that lighting in officescan be better than utilitarian. Figure 40 A 'visual rest centre' and a view of the outside world. Source of illumination Dueto two factors, (i) the congestion and consequent A view overshadowing of central areas and (ii) the high levels of Most typesof office work are said to create a need fora illumination now considered necessary, daylight in offi- `visual rest centre', where the eyes havean opportunity ces is increasingly coming to be regarded as, essentially, to refocus upon a more distant scene. Accordingto the an amenity, while electric lighting is used to provide the Building Research Station," 'Not much is known of working illumination. This, at least, is the directionof the factors which determine what isa good visual rest informed opinion in Britain and the 1ES Code" has centre, except that it should under no circumstances be been expressly framed in these terms. Experience during bright in relation to the work'. Yet the most popular this study suggests that for most officesin city centres form of visual rest centre is undoubtedly a window with this approach has to be acceptedas the only realistic a view. Nearly 90 percent of the 2,500 respondents to one. The Building Research Station has developed the the first CIS questionnaire" considered that itwas concept of PSAI I in which electric lighting is used important to be able to see out of the office, and though to brighten up apparently darker areas and providethe the psychological need for a link with the outside world working illumination in positions distant from windows (with a realisationof the time of day and state of weather) `without destroying the essential character anddirection is generally appreciated, the actual nature of the needis little understood. A point which may not be generally 25. Collins, J. B. and Hopkinson, R. G: Flicker discomfortin recognised is that provided a window is locatedat eye relation to the lighting of buildings. Trans Blum EngSoc.(Lon- level and is large enough, a view throughcan be ob- don). 195419, (5) 135-158 tained from deep (ie, at least 60 to 70 ft) withina room. 26. Illuminating Engineering Society:op cit (7) In the allocation of window space practice varies be- 27. Building Research Station: Some general principlesof the lighting of buildings. BRS digest no. 70. HMSO, 1954(October) tween different organisations. It is reported that insome 28. Discussed in chapter 10 firms in the U.S.A. the higher executivesare accommo- 29. Milward, G. E. (Ed): Organisation and methods.Macmillan. dated in air-conditioned inner rooms without windows Revised edition, 1962 so that space with natural daylight is available for staff 30. Value is one of the ordinates in the Munsellsystem and is a measure of the lightness or darkness of a colour. It ranges from performing clerical work or machine operations. 29 it is 0 (black) through the darker and then lighter coloursto 10 (white) probably much more general, though, for executivesto 31. See table 1 in: Gloag, H. L: Colouring in factories.Factory be located on the perimeter, and lower-rank staffto be building studies no. 8. HMSO, 1961 seated in the interior. 32. Illuminating Engineering Society: op cit (7) 33. For report upon these surveys, see later in this chapter and appendix 6.2 Colour 34. See chapter 10 The colours employed within an office in the form of 35. Manning, Peter and Wells, Brian: CIS: re-appraisalof an decorations applied to wall and ceiling surfaces and the environment. Interior Design, 1964 (May/June) 36. Langdon, F. J: op cit (8) finish of floors, furniture, fittings, etc, havea complex 37. ie, the latest (1961) edition

51 source of daylight, they: (i) provide views out of and into the building (ii) create surfaces of very low reflectivity at night (iii) are usually the principal means of ventilation and will therefore admit any atmospheric pollution which tit- rtfti.f may be present in the external environment (iv) provide a weak link in the external fabric ofthe building allowing heat to escape from the interior and heat gains from solar radiation to be admitted (v) allow the passage of noise fromone side of the wall to the other more readily than most 'opaque' wall constructions. If windows are providing the working illumination, then views of bright areas of sky with attendant discomfort from glare may be inevitable. Where working illumination is provided by electricity, then the windowarea can be reduced and the view selected. Ina high building, for example, a downward view may be preferable toan upwards or horizontal one. In deep office buildingsthe need will be for large areas of windows to provide the best views for people working in the interior of the building. This may make necessarysome form of external sun-shading device. Figure k i The real problem of window design is the needto recon- Working illumination provided by electric light in deep office space. cile conflicting requirements. As a general rule, the

of the dominant daylighting'. "8 Its authors describe itas suitable for rooms (with windowson one side) of 20 to 40ft depth; it does not seem to be appropriate forthe deep spaces which are a current tendency in office design. "9 So far, there are few office lighting schemes MEM in existence which have been designed to theBRS PSALI method and it is not clear whether in ultimate effect the results differ significantly from well-designed straightforward electric lighting systems. It can be expected that office workers will prefera day- lit environment to one illuminated by electric lighting, usually on the score that it is better for the workers'eyes to work in daylight. This point has not been proved ''' but nevertheless it is heldfor example, by over two- thirds of the staff of the CIS, 41 and, judging by the designs of some recent Crown buildings, this is probably an important consideration in the design of new office buildings for the Civil Service. 42 An interesting finding of this present study 43 has been that people yium;;;Avo may not be 41, able to distinguish between the differenttypes of light: staff working up to about 75 feet froma window in a deep office building which is continuously illuminated by electric light greatly over-estimated the proportion of daylight in their total working illumination.It is possible that the general preference for daylight isbased upon an expectation and experience of crude lighting installations. Bickerdike, referring to the user-opinions reported in the COI survey of office lighting"suggest- Figure 42 ed that reactions to the use of supplementaryelectric Window forming the basis for a modular planningunit. lighting were prejudiced by known conditions which were not comparable with the kind of environments possible today. 45 38. Hopkinson, R. G. and Longmore, J: Theuse of permanent supplementaryartificiallighting.Architects' Journal,1959 Daylight (October 8) 39. A more detailed discussion of this topicoccurs later in this The recommendations of the IES Codeare based on chapter the assumption that for most buildings,especially those 40. Weston, H. C., speaking ata conference organised by the in built-up city centres, it will benecessary to supple- British Lighting Council, said that therewas no medical evidence of ocular ill-effects from the use of fluorescentlighting ment daylight by electric light used constantly duringthe 41. See chapter 10 day. Values of daylight factor recommended are 'mini- 42. The Society of Civil Servants bookletwhich sets out the mum requirements ... for amenity and function . various agreed rules governing accommodationcommences the For offices, this value is 1 percent." Highervalues are section on Lighting by drawing attentionto the Civil Service's recommended in the appropriate Code ofPractice 47 own Study Group's declaration that the first essentialof all clerical accommodation should be adequatenatural lighting but these can rarely be obtained withouttop lighting 43. Reported later in this chapter and are consequently inapplicableto the great majority 44. Girl y, P. G. and Corlett, T:op cit (2) of city office buildings. 45. Bickerdike, John: Lighting. Paper in:Office buildings: report of a symposium held on 12 April 1956 at the RoyalInstitute Windows of British Architects. (The Institute), 1956 46. Illuminating Engineering Society:op cit (7) Since most office buildingsare multi-storey, windows 47. Council for Codes of Practice: BritishStandards Institution: can only be provided in the perimeter walls. These have Code of basic data for the design of buildings.Chapter 1: Light- ing; Part 1: Daylighting. British Standards several effects upon the environment. Besidesbeing the Code of Practice C.P.3: Chapter 1: Part 1 (1964) (The Institution),1964 52 a Figure 44 Ohmruction of windows h) nearby buildings ings opening lights arc almost certain to requireweather stripping and it is likely that theycan be opened only during the occasional day when the air is still. In buildings which have mechanical ventilation,windows will be both cheaper and more efficient if they arc Figure 43 Windows: surfaces of very low reflectivity fixed. 'Psychological' reasons for openable windows at night time. may be advanced 51 but it is probable that staff's initial

Is 0

MM.

Figrsro 45 A tall building on a low site; practically all stair obtainan interesting view from their desks. greatest possible amount of glass is needed for light and concern is reduced with time. Openable windowsmay view while the least amount restricts heat gains and losses be preferred for ventilation in the events of fireor break- and the admission of street noise. Many office buildings down of mechanical ventilating equipmentand for are planned on the basis of an overriding concern with access to cleaning cradles." Opening part even the maximum admission of daylight, (eg, 'the highestposs- ible proportion of the outside walls ofa building should 48. Imperial Chemical Industries Limited: Standards consist of windows.. for offices .')when, even now, with present and laboratories. (The Company), 1960 illumination needs and the constantly rising standards 49. See. 4or example, Markus, Mamas A: Daylightwith insula- which can be expected, reliance has increasinglyto be tion. Pilicington Brothers Limited, 1960 placed on electric lighting. 50. See, for example. Imperial Chemical IndustriesLimited: op cif (48) Heat loss through windows, and condensationon win- 51. See, for example, Joedicke, Jurgen: Officebuildings. Crosby dow glass will be reduced by insulating (eg, double) Lockwood, 1962 glazing. '° However it may not be economic touse double 52. Bailey, George R., Gilbert, F. L. andWilds, Horace W: glazing only to reduce heat losses" for heatmay also Commercial office buildings. Paper in: Performanceof buildings. Publication no. 879 of Building Research Institute,Washington escape by leakage around windows. In very high build- D.C. (The Institute), 1961 53 selection of screening do ices suchas curtains. brise- soleil and venetian blinds is included in the Arttiteetc Amnia/ design guides.

Curtain walling A post-war development in window design is theso- called curtain wall. which consists of proprietary dry- fixed light frameworks supported from thebuilding's main structural frame, into which windows and wall panels are inserted as required. Although the daylighting requirements may be met thereare likely to be related environmental problems. For example, prevention of weather penetration may dependon mastics whose life is unlikely to exceed 10 to 15years. 55 In a limited survey of curtain walls, the typical nuisances were found to be wind penetration and rattle, andsummer heat gains...,, Because of the many materials comprised in curtain wall constructions, the prediction of their heat transmission characteristics is complicated. Thermal discomfort (and extremes of hot and cold conditions) is quite common. The total heat loss througha whole acct curtain wail construction is probably high and thereare strong possibilities of interstitial and surface condensation.' Such light constructions increase the opportunities for noise penetration.5'

Orientation and aspect of office Inahlings It is usually recommended that office buildingsshould, Figure 46 wherever possible, be orientatedso that the two long The problem in window design is to reconcile conflictingrequire- sides are exposed to the east andwest, ensuring that all ments. In this building. windows are large to obtain the maximum offices may have some direct sunlight. When the amount of light and take advantage of the view. Venetian blinds sun is are provided to minimise sky glare and solar heat gains. low this arrangement may cause discomfort dueto both glare and heat gains from solar radiation. Ifopen-plan offices are used, then a north-south top of windows for ventilationmay create draughts exposure limits pen- and cause discomfort andpapers may be blown off desks. Pivotted opening lightsare liable to foul blinds and curtains. In an approximately southexposure, double glazing consisting of an external skin of1 inch heat-absorbing glass and an internal skin of 32oz glass may create temperature conditions in an office during periods of direct sunlight 5 to 7 deg F lower than in similar officesentire- ly double glazed with 32 oz sheet glass. r." Aguide to the

Figure 4S Opening light fouling venetian blinds.

53. Private communication 54. Markus, Thomas A: Element design guide:Screens and louvres: General. Architects' Journal, 1963(January 2) 55. Markus, Thomas A: The glass curtain wall.Architects' Journal, 1958 (January 23) 56. Anon: Working details revisited: External Figure 47 no,-loadbearing walls. Architects' Journal, 1962 (November 7) In this aireonditioned building the windows arc kept locked 57. See: Ramon, Michael: Light. cladding,3: condensation. shut, but they may be reversed for cleaning. Avenetian blind Architects' Journal, 1960 (March 10) works within the double glazing. Telephone and electric points 58. Committee on the Problem of Noise: Noise:final report. are located below the window. Cmnd 2056. HMSO, 1963 titration of sun in summer, permits deeper penetration tings. Considerations affecting the choice between tung- in w inter (when direct sunlight may he very acceptable) sten filament and fluorescent lamps are listed in table 6. and minimises the need for and size of sun-shading de- ices. Table 6 Comparison of characteristics of tungsten filament and biclighting in 11w offsee.s in the Survey Sample hot cathode fluorescent lamps for general lighting in Most of the offices in the Survey Sample are compara- offices. tively shallow. ie. they were obviously designed to be lit mainly by daylight. The only building in which the Characteristics Tungsten Fluorescent electric lighting was itiretided to be used during most of the day was the CIS building. Here, behind the totally Efficiency 8-14 Int watt 27-50 Im 'watt glazed exterior, the electric lighting is used freely at all depending on times of the day, winter and summer. Another building colour rendering in the Survey Sample was an extensive single storey properties and open-planned office accommodating about 600 people, size of tube, the which differs from all the other buildings examined in lamps with the lower efficiency thatit obtains its daylight illumination from plastic rooflights. The electric light in this building is switched usually providing on and off in three stages by an automatic photoelectric the higher quality control. Although the other offices had been more con- light ventionally designed to be lit mainly by daylight, there Life Normally taken asNormally taken as was a tendency in many of them for the electric lights to 1,000 hours 5.000 hours be switched on as soon as the staff arrived in the morn- Suitability for Unsuitable for Range of lamps ing and to be left on for the rest of the day. colour critical colour with different col- Many of the difficulties and inconsistencies experienced discriminationdiscrimination our rendering in measuring daylight in single-storey factories''" have characteristics also been found in surveys of daylight in office build- available ings. Some of the buildings are at a considerable dist- First cost of Relatively low Rclatively high ance from the Department of Building Science, and so, lamps and even if an overcast sky existed in Liverpool, the likeli- fittings Relatively low hood of finding the same conditions thirty or more miles Running costsRelatively high away was small. Even when the cloud cover was eight- Heat emission Relatively high Relatively low eighths, most attempts to measure daylight factor were from lighting unsuccessful because of rapid variations in sky lumin- installation ance. In addition, in air-conditioned offices with sealed Lamp especiallyHighlighting points General and perm- windows, it was impossible to fix an external reference suitable for: of interest anent lighting cell. Some measurements have been made in spite of these Light fittings give either direct lighting (ie, downwards difficulties, on seventy-one desks of a representative area only), indirect (ie, upwards only) or, most usually, a of a top-lit single-storey office building and on fourteen combination of the two ie, illumination downwards desks of an office on the 15th floor of a side-lit building." onto the working plane, with a smaller proportion up- The average of one set of readings of daylight factor in the top-lit building was 4.6 percent, with a range of 2.4 wards to light the ceiling. Direct lighting is the most efficient in terms of illumination on the horizontal work- to 8.2. Two sets of measurements in the side-lit building ing plane, but it creates strong contrasts of bright light were taken on consecutive days, and during both the sky sources against dark ceilings and, possibly, gloom. in- was completely overcast. Yet the second set of measurements differed from the first by between 7 and 30 per- direct lighting is the least efficient system possible, it cent, in spite of generally similar sky conditions. The gives shadowless lighting and eliminates modelling and present state of development of field techniques for textured effects, besides making the ceiling the brightest measuring daylight factor is quite unsatisfactory and the element in the room. National Illumination Committee Daylight Steering To avoid glare, the average brightness of the surface of Committee and Advisory Group has set up a panel to a fitting should not be greater than 1,000 ft-lamberts review present knowledge on the subject and, if possible, (with a maximum value at any point not exceeding 1,500 ft-lamberts), unless it is screened so that it is not visible to propose better methods. within an angle of 45° to the horizontal."' A cut-off of the light source can be achieved by (a) the shape and Electric light General lighting, ie, a substantially even pattern of light- position of the surface of the fitting, (b) by concentric- ing over the whole working plane, is now the normal in- ring, egg-crate or other diffusers, or (c) by coffering or stallation in offices. It is provided by either (i) a regular otherwise shaping or screening the ceiling surface. pattern of light sources, (ii) lights in continuous strips, or (iii) luminous ceilings. 59. No description of daylighting design methods are given in In Britain, the standard basis for the design of electric this report. Some sources of information are: lighting installations is (i) calculation of illumination by Building Research Station :Estimating daylight in buildings. the lumen method" together with (ii) calculation of BRS Digests (second series) nos. 41 and 42. HMSO, 1963 (December) and 1964 (January) glare index. "" This involves decisions (or information) Hopkinson, R. G: Architectural physics: Lighting. HMSO,1963 about : Walsh, John W. T: The science of daylight. Macdonald, 1961 (i) the geometry and dimensions of the room Lynes, J. A: Developments in daylight prediction. Light and (ii) the reflectivities of the surfaces of the room Lighting, 1963 (May) 60. Phillips, R. 0: Natural lighting investigations. Appendix (iii) the choice of lamp 2.6-1 to: Manning, P:The design of roofs for single-storey (iv) the choice of fitting general-purpose factories. University of Liverpool, Department (v) the positioning of the fittings, ie, their spacing, of Building Science, 1962 height ant orientation 61. See appendix 6.1 for details of daylight surveys (vi) the use and maintenance of the room. 62. See, for example, British Lighting Council: Interior lighting design. (The Council), 1962 For the design of an installation which will provide gen- 63. Illuminating Engineering Sockty: op cit (7) eral lighting conforming to the criteria previously dis- 64. Building Research Station: Artificial lighting of building cussed, thcre are two matters of choice which affect the interiors: lamps and fittings. BRS digest no. 81. HMSO, 1955 character )f the environment the lamps and the fit- (September) 55 Figure 49 Bare batten fittings cut of from view by design ofceiling construction.

All the systems examined were installedduring the period 1945 to 1962. The mean illumination levelsin the four offices with fluorescent installationsvaried between 19 and 100 Im./sq.ft; the twotungsten installations gave values of 9 and 18 Imisq.ft. In onlyone of these offices was the illumination level higher than the IESrecom- mendation of 30 Im./sq.ft for generaloffices; all the others were lower, some very muchso. (But the value of illumination recommended by theIES during the period 1955 to 1961 and applicableto general offices was 20 Im./sq.ft). The three schemes whichwere designed to provide a known value of illuminationfailed to do so by between 8 and 27 percent. A perfectly even illumination isdifficult to achieve and, if monotony is to be avoided,is possibly undesirable anyway. But an important requirement in officedesign is that internal planning anduse should be as little restricted as possible, anda substantially even illumination is therefore essential. The IESrecommendation of a minimum value of the ratio of Emin to E max (ie, 0.7) has been achieved in onlyone building. This has a fluorescent installation with a mean illumination of28 Im/ sq.ft. The building with thevery high illumination levels has the next best ratio of 0.6;46the remaining four schemes have values of either 0.4or 0.5. The calculation of glare index isa relatively new practice, requiring photometric data whichis not yet readily Figure SO available; it has not been possibleto compare the per- An indirect fitting in a private office has beenchanged by the formance of these lighting installationson that basis. user for a direct one in order to obtain a higher working The practice of group replacement illumination. of fluorescent lamps has been advocated for severalyears. 67 Where it is To maintain the efficiency of the lamps, thecondition of adopted the lamps are changedwhen their 'residual the fitting and wiring insulation, andto reduce the value' has become so low that itis cheaper to replace amount of dirt deposited on both lamp and fitting,it is them all at once rather than individuallyas they fail. often recommended that ventilation bepermitted up- Half the offices where lightinginstallations were surveyed use the group replacement wards, through the fitting. This usuallyallows some method, but they tend to light to be spilled onto the ceiling. be the larger ones, for thepractice is comparatively rare in small offices. In two of the officeslamps are Electric light in the offices in the Survey Sample replaced after the expiration ofa definite period of time. During this study, surveys of electric lightingwere made This has two advantages: the workcan be planned in in six offices. The number ofsurveys was too few to draw adv2nce and it can also be timedto coincide with clean- any general conclusions from the results, but they have been useful in indicating the magnitude ofthe differ- 65. See appendix 6.2 for furths,r detailsof these surveys ences which might exist between the standards thatman- 66. At its ' high' illumination stage. The valuefor the 'low' illu- agements or designers have intended to provide and mination stage is 0.2 67. Robinson, W. and Strange, J. W: Themaintenance of light- those actually achieved." ing installations. Trans ilium '':ng Soc. (London) 1955, 20(S). 56 ine, thus reducing the cost. people did feel this to be important. Only twelveper- In a third office, a complicated procedure for bulk recent considered it unimportant. 7' placement has been laid dow n.It ..icludes failure re- Hopkinson and Longmore have referred " to what they placerent at four w eck I.), inter als and wholesale replace- believe to be the great psychological need that people ment (except of lamps preciously replaced individually) have for daylight. They instanced, asan extreme example at the end of the lamps' economic life. In the fourth of the effects of daylight deprivation, thecase of the fluorescent installation the lamps are replaced individ- Scandinavian neurosis 'Lapp sickness', andsay that ually w hen they fail. 'This kind of evidence is by no means conclusive, but if In the two offices which are lit by tungsten filament one wants to believe that daylightisnecessary for lamps the lamps are replaced individually. But group human well-being, it reinforces that belief'. In framing replacement is not usually adopted for this type of lamp, subsequent recommendations on lighting and building possibly because lamps can be replaced by unskilled form they then take it as axiomatic that a physical pre- people. dominance of daylight illumination is a necessarycon- In all the offices the light fittings are cleaned on a syst- dition for a suitable environment for clerical workers. ematic basis, mostly at six-monthly or longer intervals This is probably an accurate reflection of the assumption by contractors working outside normal office hours. made in most current office building design, andso far Fittings are normally washed in warm water containing it has not been seriously questioned. soap or detergent; in two offices they are treated with an It was apparent from the replies to another question that antistatic solution. Ventilated fittings have been in- people feel there are clear qualitative differences between stalled in only one building. Yet tests performed in the daylight and electric lighting. Asked whether they felt it U.S.A. on flourescent fittings have shown that after a to be as good for their eyes to work by electric light as period of twelve months, the percentage light output of by daylight, only 13 percent replied 'yes '.18 percent the ventilated fittings was 96 percent of the original, but gave a neutral answer and 69 percent replied that electric of the unventilated fitting only 72 percent." light was not as good as daylight. It was stated earlier that the desirable pattern of bright- On the face of things, it would therefore seem that de- ness in a room is one where the visual task is the bright- signers would be well advised to ensure that daylight is est part and where there is a gradation downwards from the dominant feature in the lighting of working spaces, this to the more distant surfaces. The brightness of a for it appears to be a necessary condition for producing surface depends upon the amount of light allowed to fall a subjectively satisfactory environment. If people feel upon it, and its intrinsic qualities of reflectivity. they have a need for something, they will feel discon- Light colours were used for the walls and ceilings of all tented or deprived if they are aware that itis not six offices in which the lighting surveys were made, present, but their cleanliness varied considerably. The reflect- However, what if people are unable to distinguish ivity of all the ceilings was estimated" to be within whether they have or have not what they think they the range 64 to 75 percent; the reflectivity of the walls need? In a wholly glass-clad but deep office building the was measured and found to be between 43 and 63 per- respective contributions to the total illumination of day- cent. Excessive brightness contrasts are liable to be light and electric light are not readily apparent, yet they created by light-coloured tasks seen against dark desk must be established before the relationship between act- tops; in general, brightly polished desk tops should not ual physical conditions and the beliefs and attitudes be used because they may cause reflected glare. In most connected with the presence of windows, and with nat- of the offices desk tops were matt, with reflectivities of ural and artificial lighting, can be examined. between 15 and 48 percent. There was a small group of As the questionnaire responses showed that daylighting desks in one building which had dark green, glossy sur- was regarded as being so much better for the eyes than faces, where the reflectivity was as low as 5 percent, but artificial lighting, and that being able to see out of the they were only in temporary use. building was rated as being very important, it seemed The most common types of decoration were semi-gloss that distance from a daylight source should be regarded or emulsion paint. The external walls of one building as one of the most prominent factors making for the were entirely glazed and only the ceiling and partitions satisfactOriPess of the individual's working environment. had to be decorated. The ceiling was metal with a Accordingly, an objective appraisal of the individual's stoved on finish and the partitions were covered with ability to ,discriminate between daylight and artificial plastic sheeting, so both were easily washable. In only light hwebeen made." If the individual is unableto one building was wallpaper used in the general clerical make satisfactory discrimination then, clearly, this will areas and the management said that 'it is treated with influence the prima facie case for high levels of daylight more respect and therefore keeps in a good condition illumination which currently dictates the need for de- longer than paint'. signing narrow buildings, or narrow spaces arranged So far as could be judged by relatively inexperienced ob- round open light wells or courts. servers, none of the installations were uncomfortably The setting of the investigation was the new head office glaring. In most of the offices the colour of the electric of the CIS. This is a very deep office building withcon- lighting was pleasant but in one an irregular combina- tinuous floor-to-ceilingglass curtain walling. The electric tion of warm white and daylight was used which gave a lighting installation consists of fluorescent fittings inset rather untidy appearance. In another, because the fitt- in the ceiling with a plastic 'egg crate' flush with the ings were flush with the ceiling, very little light 'spilled' onto that surface and it looked dark and perhaps dirty. 68. Test referred to by Robinson, W. and Strange, J. W:or) cit (67) Clerical workers needs of a visual environment 69. See appendix 6.2 In the system of permanent supplementary artificial light 70. Hopkinson, R. G. and Longmore, J: The permanent supple- (PSALI) which the Building Research Station has re mentary artificial lighting of interiors. Trans Ilium Eng Soc. cently recommended for use, 7 ° 71. 72 it is assumed that (London). 1959, 24 (3) 121-148 71. Hopkinson, R. G. and Longmore, J: The use ofpermanent daylighting is, in itself, an important condition for a supplementary artificial lighting. Architects' Journal, 1959 (Oct- good working environment. The replies to a question- ober 8) naire about the working environment in their new build- 72. Building Research Station: The permanent supplementary artificial lighting of interiors (PSALI). BRS digestno.135. ing addressed to the 2,500 head-office staff of the Co- HMSO, 1960 (June) operative Insurance Society show that such an assump- 73. See chapter 10 tion is in line with common belief. The response to a 74. Hopkinson, R. G. and Longmore, J: The use ofpermanent question which asked whether it was felt important supplementary artificial lighting. op cit (71) 75. Wells, B. W. P: Subjective responses to the lighting installa- to be able to see out of the office even if there was plenty tion in a modern office building and their design implications. of artificial light to work by was overwhelmingly that Building Science, 1965, 1 (1)

57 ceiling surface. The light fittings are in continuous rows of the same thing. One value was then subtracted from at 5 ft 2 ins intervals, and they are in constant use. The the other, thus gis Mg the amount and direction of sub- sample of indk iduals whose discriminationwas being jective error. The data required for the analysis was appraised was composed of general clerical workers and completed by measuring the distance of each of the did not include any such specialised groups as typistsor desks from the nearest daylight source. machine operators. All the workers were working Analysis of the data has shown that where electric light open-plan office space and had an uninterrupted view is used permanently, then the further an individual works of the window walls. away from the nearest daylight source the greater will be The experiment consisted essentially of relating stated his tendency to over-estimate the proportion of daylight beliefs and subjective experience of the subjects' physical to electric light (figure 52). It was also found, from a conditions, so that both questionnaire and physical comparison of staff's desk positions and their replies to a measurement techniques were necessary. The staff were question on the relative goodness of daylight and arti- asked to complete a questionnaire during the course of ficial light for the eyes, that distance from the nearest a morning. The experimenter made physical measure- window has no influence on the belief that artificial ments of the lighting and obtained the appraisals during lighting is not as good for the eyes as daylight. the same morning. The questionnaire contained quest- Similar cross tabulation of desk positions with replies to ions dealing with a wide variety of environmental topics a question of whether it is important to be able to see but was specifically the vehicle of three questions dealing out of the office (even if there is plenty of artificial light with beliefs about the physiological value of daylight, to work by), has shown that the strength of a person's the importance of a view from the building, and the comfort of the combined electric and daylight illumination. The relative proportions of electric light and day- II

30 ($TIMAti 70

ACTUAL

I-1 10-17 13-15 11-11 70-21 30-31 LO. 51 -IS la.) lag 171 1.51 I.341 1.351 1.141 00) 1.171 0:STANCI IAN WNW 001.1611t MO reel

Figure 52 Study of clerical workers' ability to discriminate between daylight and electric light : plot of actual percentages of daylight and subjective estimates of percentage at increasing distances from nearest daylight source.

subjective estimate of the importance of being able to Figure 51 see out of the office is not related to his distance from The photometer and lighting judgement scale. the nearest window. Further, it was found that there was no difference between the comfort of the illumination light were estimated by each worker in a face-to-face over a very great range of physical differences in both situation during which the total level of illumination total illumination level and the amount of direct day- from both sources was measured by the experimenter. lighting. The photometer had, for convenience and in order to There was some temptation to write into the question- provide a standard stimulus background to the judge- naire some such enquiry as, Do you feel that you re- ment forms, been mounted on a rectangular board and ceive a sufficient proportion of daylight where you painted pale grey so that its reflectivity was similar to work?' On face value, this would have answered a most that of the desks on which the subjects worked. On the important practical question, ie, that amount of direct board was a pile of typed quarto forms, and beside daylight required in order that an office worker shall find these, a pile of completed forms face-down, which gave his lighting environment satisfactory. Nevertheless such a 10 inch x 16 inch area of white paper. The typed in- a question was not included because it was felt that, as structions on the left-hand pile had to be attended to, so there were obvious differences in the proportion of day- the task simulated a normal clerical procedure quite light received at different distances from the windows, it closely. would be likely to provoke responses that were based on The document for the subject's attention asked him to the apparent fact of having more or less daylight than try to estimate the relative amounts of daylight and other people, rather than upon the personal satisfactor- electric light illuminating the form, and to do this by iness of the visual environment. using a scale ranging from zero to 100 percent (no day- It might have been supposed, from the great proportion light to total daylight). In each case, the experimenter of questionnaire replies dealing with the importance of made sure that, before the estimates were made, the re- daylight, that the respondents would be very sensitive to spondent had correctly interpreted the instructions and its presecne or absence. On the other hand, factorswere understood that the level of illumination in the building at work making an accurate subjective estimate of it depended on both sources of light. The experiment took exceedingly difficult. The subjects were asked to estimate place during August at a time when the sky was dull the relative amounts of daylight and electric light illum- and almost uniformly overcast. inating the sheet of paper on which the scale was drawn. The daylight illumination available at the time of the In the absence of visual cues like shadows, and where experiment at each working position was found by sub- there are no gross differences in the spectra of the two tracting measurements made at night time of electric light sources, such a judgement is, of course, nearly im- lighting only from the combined measurements of day- possible. In fact, without a single exception, the subjects light and electric light. From this it was possible to com- looked up and around in the direction in which they pare the actual percentage of the illumination repre- were facing (ie, towards the windows) before making a sented by daylight with the worker's subjective estimate judgement. The real judgement therefore seems to have been made in the terms of apparent brightness. would make it possible to havee.en deeper buildings There appear to be two significant factors influencing than those envisaged by BRS. The Station's the way in which the judgements digest says, were made. In the first 'This is a system in Which daylight suppliesa dominant place there were floor to ceiling windows whichprovided source of lighting, and carefully designed artificial large bright areas presumably suggesting light- important ing adds supplementary illuminationto the less well-lit sources of illumination. Secondly, the artificial lighting parts of the building. It is important tostress the dom- system produced an unbroken stretch ofeven illumin- inance of the daylighting both in quantity ation which provided, except in the and direction case of those people the artificial lighting issupplementary.'The cond- very near the windows, very much more light than the itions examined for the Unit's experimentincluded the windows themselves. Physiological adaptationwould responses of people seated from between fifty and therefore take place in terms of the electric lighting seventy-five feet away from thenearest window, the day- rather than the daylighting. This could be predicted to light component at these distances tendingto be in the have two consequences: firstly, that the largewindows region of I Im./sq.ft or less. Therefore, by BRS'sown and the light reflected on vertical surfaces (whichwould, definition, it follows that the PSALI solution in fact, be a mixture of electric light and daylight) is ap- would propriate only to shallow buildings whosesite utilisa- be sufficient to give those people whowere comparative- tion will be of a lower order than that of the ly deprived of direct daylight the impression particular that they building considered here. Indeed, PSALIhas been were not so deprived; secondly, the brightness of the described as being applicable to 'Single-sidelit rooms of fluorescent lighting, compared with the windowsthem- 20 to 40 ft depth or two-side litrooms up to 80 ft in selves, would give the same people the subjectiveim- depth'. 7`Itis possible, of course, that PSALI would pression of being in a brightly illuminatedarea. The fact be suitable at greater depths, and that it is merelyits that people's ideas about the physiological advantages premises which need revision. However if the experi- of daylight, and of the need fora view to the exterior, mental findings of this present research thatthe satis- are independent of a physical context reveals attitudes factoriness of the lighting environmentwas not signifi- or prejudices rather than a genuine subjective response cantly influenced by distance from thenearest window io the environment. It is likely that these attitudesare are accepted, then it may be assumed that building and based upon some common factor, perhaps custom, or lighting designs such as theones adopted for the CIS experience of older and less successful lighting solutions, buildingareatleast assatisfactoryasPSALI such as small windows or poorly designedor manu- could be expected to be. The CIS also has factured electric light fittings. Whatever this a compara- common tively low ceiling (10 ft) and thereforeattains the re- factor might be, it appears to be of littleconsequence in duced building and heating costs (besides the highsite environments such as the one examined, wherethe utilisation) which are claimedas typifying PSALI individual is not able to distinguish the existenceof con- installations. ditions which in fact infringe these sensibilities. In a discussion of the The findings of this investigation would mechanics of PSALI, the seem to raise following statement is made: 'Thenecessary levels of some interesting questions connected with the Building supplementary light have to be determined by the Research Station's PSALI recommendations and re- the quirement to raise the brightness of theseremoter parts assumptions on which they are based. of the room until there isno longer any sensation of lack The BRS recommendations have been proposedto of light'.In practice, in the case studied at the CIS provide a lighting environment in which daylight isthe building, as in most other large office blocks visited by principal source of illumination and in which, assuming the Unit, clerical workers are seated facing either inthe that a satisfactory minimum illumination exists,the em- direction of the nearest window, phasis is placed on brightness distribution rather or with it to their left than hand side. The appearance of gloominess ofthose parts illumination levels. So long as a sufficient level of illum- of the room distant from the windows dependsupon ination is available for doinga particular task, then the viewpoint. The two illustrations (figures 53 and54) show causes of discomfort in the visual environment will very clearly the subjective differences created by the largely turn on such considerations as glare, either from direction in which workers are facing. intense light sources like nearby windows,or from contrasts in the brightness of different parts of theroom or working surface. The result of such conditionswill be unfavourable physiological adaptation of theeyes. That is to say, the influence of comparatively brightareas in the visual environment is to push up the level of adaptation so that objects and areas whichare in fact well illuminated, give the subjective appearance of darkness and gloominess. The principal PSALI method of reducingcontrast within a room is by grading the level of artificiallighting J from the inside of the room towards the windows,the highest levels of artificial light being furthest fromthe windows. To keep the brightness contrast fairly low it may be necessary to use '.. smaller windows than would otherwise have been installed, and itmay also mean the provision of built-in glare shields'." With high levels of artificial illumination inside thebuilding, there is less need for a large daylight component and apart from the windows being made smaller, it ispossible to reduce the height from which the daylight pene- Figure 53 trates the building - in other words, to have lower ceil- Experimental area viewed from inside towards windows. ings. The economic advantages of lower ceilings,reduced running costs and higher site utilisation havebeen given as cost savings which may be balanced againstthe capital and running costs of PSALI. " 76. Hopkinson, R. G. and Longmore, J: Thepermanent supple- The conception of PSALI is a boldone, for it goes mentary artificial lighting of interiors: op cit (70) beyond merely making recommendations for lightingto 77. For example in: Building Research Station:op, cit (72) an intimate concern with the design of the building it- 78. Hopkinson, R. G. and Longmore, .1: Theuse of permanent supplementary artificial lighting. op cit (71) self. However, the Unit's investigation, if confirmed, 79. Building Research Station: op cit (72)

59 several modern windowless factories, at leastone of which has been constructed foran industry which does not have a technical need for windowless conditions. Windowless design for factories for economic rather than technical reasons is becomingcommon in several East European countries; it has, ofcourse, been common in the U.S.A. for many years and experimental windowless schools are now inuse there. "' A recent Russian paperstated that 'windowless buildings must be made subject to more stringent requirements with respect to lighting and conditions of health than buildings with natural daylighting'. Possibly for thesame reason the IES Code "" suggests that a computed value of illumination may be increased bya factor of 1.5 to take account of 'special practical circumstances'when the interior has no natural lighting.

Figure 54 Experimental area viewed from windows inwards.

Everybody feels that daylighting isnecessary to their well-being, but the factual basis for this has notyet been demonstrated. Though it is possible to sympathise with the belief, it should not be allowed topass unquestioned to the status of an axiom of building design. In framing their recommendations Hopkinson and Longmore have treated it in this way and, in specifyinga dominance of daylight, have imposed severe limitationson the design and structural form of office buildings. It isvery likely that buildings can be built muchmore deeply than has yet been regarded as desirable and still produce satisfactory conditions.

Windowless environments A logical next step from permanent supplementary electric lighting may be the creation of permanently windowless environments. In the absence of knowledge of people's need for daylight, the defence of windows rests largely on emotional grounds. In 1960, when the British Lighting Council organiseda one-day conference on the theme 'Is daylighting really necessary?' " the main speakers were able to presenta stronger case for design which disregarded daylighting considerations than was the unorganised 'opposition' to refute it. No windowless offices in Britain are known, but thereare

80. The conference was reported briefly in theArchitects' Journal, 1960 (May 3) pp 677-8 81. Larson, C. Theodore: School environmentsresearch and the evaluation of windowless classrooms. Paper in: School Building Research. Publication no. 1008, Building ResearchInstitute, Washington DC. (The Institute), 1963 82. Zoz, N. I: K voprosu ob iskusstvennomosveshchenii prom- yshlennykh zdanii bez estestvennogo sveta (The proolemof the artificial lighting of factory buildings withno natural daylighting). Svetotekhnika 1960, 6 (11) (Building ResearchStation lib- Figure SS rary communication no. 1100. 1961, September) Windowless office building in New York. 83. Illuminating Engineering Society :op cit (7) 60 Chapter 7 The thermal environment

People have to maintainaheat balance with their en- were inadequate' means af maintaining temperatures at vironment: they lose heat from their bodies by radiation, 65-F, and 180 complaints of draughty conditions in convection and (via perspiration and respiration) evap- 'colder' weather." Sixteen cases of 'inadequate'means oration, the rate at which this happens dependingon the of ventilation were reported.' 120 underground offices environment and their activity. The problem of the and workrooms (excluding shops) were reported and thermal environment in offices is a matter which varies their use was .deplored unless they are exceptionally with :he season: buildings are heated in winter (and commodious and are provided with a really effective people may wear more clothes) so that the rate of loss airconditioning system... they are often damp, cold, of heat which would occur in a natural environment is dusty, badly ventilated and depressing...9. to reduced. In order to provide comfortable conditions in Eleven out of fifty offices in Barrow" showed temp- summer, the rate of heat loss may have to be increased eratures below 65°F, two being below 62°F. Old and artificially, for example, by introducing currents of air, dirty electric bar radiator fires were found to be used or by cooling the air. frequently; enquiries revealed that women wanted higher working temperatures than men and auxiliary equip- Previous surveys of the thermal environment in offices ment was often provided for this purpose, though it It is only since the war, with themore general use for might be used only during the early part of the week office buildings of light systems of cladding andan ex- before the central or other general heating gave suffici- tensive employment of glass, that thesummer thermal ent temperature.' problem has become conspicuous. Most reportedsur- In the City of London, '... every conceivable type of veys of the thermal environment within offices have dealt heating apparatus was encountered... except ... the only with conditions during winter. coal fire or stove'.'" There wa:, no heating inany of the The study panel for the civil service report' found that offices between June and September. A sample ofmeas- in the nine modern commercial offices they visited the urements taken in 290 offices during March and April temperatures maintained varied between 64 and 70' F. 1962 showed that 63 tnrcent were in Black's comfort Temperatures in buildings where the Ministry of Works range of 64 to 72°F, '" but there were nearly 9 times as was responsible for heating were commonly 60°F (which many measurements above 72°F as therewerebelow was a governmental wartime standard). In the commer- 64 °F. No attempt was made to seek opinions from office cial offices natural ventilation via windowswas usually supplemented by forms of mechanical ventilation to . Study Group appointed by H.111. Treasury: Working condi- some parts of the buildings. Government offices were tions in the Civil Service. HMSO, 1947 usually ventilated entirely by naturalmeans. The report 2. The term airconditioning is sometimes used incorrectly when recommended that the most generally comfortable temp- mechanical ventilation is intended. Airconditioning is Thepro- erature in offices was 65°F, and that airconditioning" cess of treating air so as to control simultaneously its temperature, humidity, cleanliness and distribution to meet the requirements was seldom necessary. of the conditioned space'. (Reprinted by permission of: American A study was reported in 1954 of a correlation oftemp- Society of Heating, Refrigerating and Air-Conditioning Engineers: erature measurements in 'about 55 rooms' in two build- Heating, ventilating and air conditioning guide 1960. (The Society) ings with the occupants' assessment of the conditions." New York, 1960 Bi-annual publication). Air-conditioning involves heating, refrigeration, washing, de-humidification and filtration This indicated that the most generally acceptablecom- 3. Black, Flora W: Desirable temperatures in offices:a study of fort zone is likely to be between 64 and 72°F, that occupant reaction to the heating provided. J Inst Heat Vent Engrs. women appear to be more critical than men of temp- 1954 (November) erature extremes, but prefer slightly higher temperatures, 4. Bruce, W: Man and his thermal environment. Technicalpaper no. 84 of the Division of Building Research, (NRC 5514). National and that they are increasingly likely to complainas the Research Council (of Canada). (The Council), Ottawa. 1960 temperature falls below 64 or 65°F. (February) Although they are better adapted than men to meet 5. Yaglou, C. P. and Messer, A : The importance of clothing in air- changing environmental conditions, conditioning. J Amer Med Assoc. (October 11)1941 1171261 -1262 women are more 6. The original is not precise, and it is not clear whether it is insensitive to .. those changing conditions that tend to dividual rooms or suites of rooms which is intended produce a cooler environment'.4 Part of thisgreater 7. Wattleworth, W. H: Unfinished business in environmental sensitivity might be explained by physiological differ- health. Royal Society of Health Journal, 1958 (May-June) 8. Criteria not stated ences, but variations in the .. quality, amount and 9. Criteria not stated design of clothing is a very significant factor.' Brucere- 10. Wattleworth, W. H: op cit (7) fers to work by Yaglou and Messer° which demon- I I. Nelson, I. D. M. and Morse, R. J: A pilot study of office strated that men and women wearing similar clothing accommodation. The Medical Officer, 1960 (September 9) were comfortable in similar environments. 12. Robinson, Alan: Working in the City: the interim report ofa survey. City of London Health Department, 1962 (December) In 160° out of 913 suites of offices in Liverpoolthere 13. Black, Flora W: op cit (3)

61 workers of the temperatures they desiredbut the experi- higher rates of air movement menter considered that if the temperaturesfound had than those mentioned are proved uncomfortable the office staff hadonly to open desirable. their windows (which they didnot do) to secure a re- (iii) The air movement Bruce has observed that this duction. In sixty percent of the offices allwindows were should be variable rather shut. is an assumption which is than uniform and mono-usually made, but that actual The thermal measurements taken by Robinsonduring tonous, for the body is physiological observations the period June to September provideda somewhat stimulated by ceaseless have not been made to sub- similar distribution to those taken inwinter. It is not change in the environment.stantiate it.'" possible to infer the actual comfortconditions which (iv) The relative humidityPage, in a review of human existed. of the air should not thermal comfort, has Summer temperatures and ventilationat all seasons exceed 70 percent, and suggested that relative were found to be the least satisfactory features of mech- should be preferably muchhumidities between 30 and 70 anised offices. '+ Machine rooms warmedup more rapid- below that figure. percent are' normally ly than clerical spaces, the temperatures inwhich varied satisfactory'. '" more widely, often from uncomfortably low levelsto (v) The average temp- In practice it is rarely possible peaks above 85"F. Temperatures were 'close to the point erature of the walls and to ensure that the mean of optimum comfort' during thegreater part of winter other solid surroundings days. radiant temperature is higher should not be appreciablythan air temperature. Writing in 1964, and drawing upon a number of BRS lower than that of the air,Langdon and Keighley 20 say surveys of offices, Langdon and Keighley 13 reporteda and should preferably bethat, with noise,' stuffiness' general finding that few officesare not heated sufficient- higher. The combinationwas the worst feature of the ly, whereas unsatisfactory ventilation anduncomfort- of cold walls and warm offices covered by all the able summer conditions, already reported inrespect of air often causes feelings BRS user surveys. They mechanised offices, were more generally experienced. of stuffiness. remark that although the Over 40 percent of the occupants of modernoffices sur- subjective experience of veyed in London found their offices too hot in summer. stuffiness is well known, it is In offices with particularly bad combinationsof en- not yet possible to define it in vironmental conditions this proportion ofstaff experi- terms of thermal conditions. encing discomfort rose, for example,to 66 percent of the (vi) The air at head levelBruce 21 refers to a 'generally total staff when their offices were sited on main streets should not be distinctly accepted practice of allowing with south-east to south-west aspect and hadmore than warmer than that near thenot more than a 5 deg F air dry nine-tenths of the external wall glazed. floor, and the heads of bulb temperature difference the occupants should notbetween head and feet levels.' Legislation be exposed to excessive Elsewhere in his same paper Standards for the thermal environment prescribedin radiant heat. Bedford mentions his own office legislation 1" are either so low that they would be conviction that the difference exceeded in almost any modern buildingor couched in should not exceed this figure very general terms. Thus, ' Effective provision shall be and, preferably, should be made for securing and maintaininga reasonable temp- less. Chrenko 22 has shown erature .. .'. Where the work does not involvesevere that sources of high level physical effort a temperature of less than 16 °C(60.8 F) radiation should not elevate is not considered reasonable after thefirst hour. The the mean radiant temperature Minister has power to make regulations varyingthis above 65°F by more than 4 standard. No method of heatingmay be used which deg F. It is generally allows any injurious or offensive fumesto escape into agreed that variations in the room and methods of heating whichare considered temperature over the whole to be injurious may be prohibited by regulation. A ther- area of a working space mometer has to be provided in a conspicuous position should be small. on each floor. 'Effective and suitable provision shall be In addition, the air shouldThis requirement, more made for securing and maintaining, by the circulationof be free from unpleasant specifically the avoidance of adequate supplies of fresh or artificially purified air,the odours'. body odours, is the normal ventilation of every room... in which persons are em- basis for recommended rates ployed to work.' Standards of ventilationmay be (but of ventilation for winter have not yet been) prescribed by regulation. conditions. 23 Thermal comfort, it is seen, dependson a complex of Requirements for a pleasant thermal environment factors, mainly air temperature and vertical The make-up of a thermal, environment, which temperature besides gradient, air movement, rates of airchange, relative satisfying human physiological needs, isalso pleasant to humidity, radiant temperature and ventilation. experience, can be expressed only in generalterms. Per- haps the best statement is by Bedford whosummarised the requirements in the followingway. " '(i) A room should be as cool as is compatible with comfort. 14. Langdon, F. J: The design of mechanisedoffices. Architects (ii) There should be Journal, 1963 (May 1) adequate air movement, 15. Langdon, F. J. and Keighley, E. C: Userresearch in office but there should be no local design. Architects' Journal, 1964 (February 5) draughts. At the room 16. Offices, Shops and Railway PremisesAct, 1963. HMSO 17. Bedford, T: Comfort in relationto warmth and ventilation. temperature customarily J Inst Heat Vent Engrs. 1954 22 85-114 maintained in winter in 18. Bruce, W: op cit (4) Great Britain the velocity 19. Page, J. K: Human thermal comfort. ArchitectsJournal, should be about 30 ft/min; 1963 (June 19) 20. Langdon, F. J. and Keighley, E. C:op cit (15) velocities below 20 ft/min tend 21. Bruce, W: op cit (4) to cause feelings of 22. Chrenko, F. A : Heated ceilings andcomfort. J Inst Heat Vent stuffiness. In summer Engrs. 1953, 20 375-396 months, or in hot factories, 23. Institution of Heating and VentilatingEngineers: Guide to current practice. London (The Institution), 1959 62 Thermal indices lation...' for today's conditions. Bedford'swork was Since the thermal environment,as experienced by people, primarily with industrial workers and ifchanges in dress consists of the inter-related effects of several variables, habits during the past 30or so years are taken into con- it is not sufficient to specify only one, although in pract- sideration then his values of 58to 66"F EqT are on the ice thisis the procedure commonly adopted. Many low side for today's office workers. attempts have been made to devise an index which will Draughts are a verycommon cause of complaint in offi- represent in one figure the composite effect atany one ces. Bruce "" quoting Houghten et al time of the different variables and there described three are at least types of draught which are 'any localsense of cooling thirteen different indices in use in various parts of the ... caused either by an excessive movement of air of world and for different purposes."' In Britain the best normal temperature, by air havinga normal velocity but known warmth indices are corrected effectivetempera- a lower temperature, by excessive radiation from ture (CET) and equivalent temperature (EqT). the Equiv- body to a cold surface, or byany combination of these alent temperature only incorporates three of the four three effects . . Bruce doubts whether people basic variables: air temperature, are mean radiant tempera- capable of distinguishing between the last ture and rate of air movement, but it is commonly two causes. It ac- is likely, too, that few office workersor their manage- cepted in this country as being suitable forthe specifica- ments realise that sitting next to cold surfaces like filing tion of comfort conditions in winter, ie, whenbody heat cabinets or windows can create these sensations. loss is not being controlled by evaporation. Corrected The management of one office in the Survey Samplewere effective temperature takes account of airmovement, concerned that relative humidities in their relative humidity and (by the building were use of a globe thermo- low. Their staff was in good health but theyfeared the meter) radiation and air temperature. possibility of a virus infection which might,they thought, occur if people's throats tended to be too dry. Measured Conventional standards for the thermal environment RH's in this building were of the order of 30 The specification of a thermal environment percent, then, is a ie, at the lower end of Page's acceptablerange of 30 to 70 complicated matter. Normally legislative, departmental percent. Unfortunately, in very cold dry weather, there and recommended standards do no more than suggest is no simple method of control of relativehumidity in air temperatures and, possibly, ventilation rates. Clear- buildings which depend upon naturalmeans of ventila- ly, this is no more than part of the thermalproblem al- tion. though, in Britain's temperate climate, it isperhaps the most important one. A more general (and withsome Summer thermal comfort types of construction, serious) objection is that only Summer thermal problems are experienced inmany minimum temperatures applicable to winterconditions modern factories" and occur in other buildingtypes in- are specified; summer thermal conditions and their cluding offices" and schools. Theoptimum tempera- possible consequences are eithernot foreseen or ignored. tures for comfort for factory workers in In giving evidence to the Gowers summer have Committee, some been found to be of the order of 2to 4 deg F higher than representatives of clerical workers advocated minimum those for winter. 4" People become acclimatisedto the (winter) air temperatures of 65°F "5 butthe Report as published recommended only the standardsof the Fact- ories Act 2" (60°F after the first hour).A value close to this (18 C or 60.8 °F) has been incorporatedin the new 24. Thirteen indices are listed in Webb, C. G:Thermal comfort Act." The Code of Practice for Heating and discomfort: a review. Indian ConstructionalJournal 1961 and Thermal (December) and 1962 (January) (A BRSpaper). See also: Insulation "4 recommends 65 to 67°F forgeneral and Givoni, B: The nature and application ofthermal indices. 'In private offices. The Code of Practice forVentilation '!" the field of building' bulletinno. 73-74. Technion - Israel Inst- recommends minimum rates of freshair supply of be- itute of Technology. Haifa, 1960 tween 420 and 1,000 cu.ft per person 25. Home Office and Scottish Home Department:Health, welfare per hour depend- and safety in non-industrial employment. Hoursof employment ing on the volume of officespace available to each of Juveniles. Cmnd 7664. HMSO, 1949 (March) person. The IHVE Guide "" suggests 65 °F for general 26. Factories Act 1961. HMSO and private offices with two air changes 27. Offices, Shops and Railway Premises Act1963. HMSO per hour (Ito 28. Council for Codes of Practice for Buildings: British 2 for private office). A writer in theInstitute of Directors Standard Code of Practice CP 3 Chapter VIII (1949),Heating and thermal handbook on office design "1 states that, whileair temp- insulation. (The British Standards Institution),1949 eratures in the range 65 to 68°F have been thecustomary 29. Council for Codes of Practice for Buildings:British Standard design values, the sedentarynature of office work and Code of Practice CP 3 Chapter I (C) (1950)Ventilation. (The British Standards Institution), 1950 the present-day tendency towear light clothing suggests 30. Institution of Heating and VentilatingEngineers: op cit (23) a more suitable range of 65 to 70°F (where theupper end 31. Institute of Directors: Better Offices(The Institute), 71960 of the range will be particularly suitablefor women and 32. Imperial Chemical Industries Limited:Standards for offices people over 40 years of age);temperatures at the lower and laboratories. (The Company), 1960 end of this range will be more acceptable 33. Society of Civil Servants: Officeaccommodation and allied when systems subjects (The Society), 1963 (January) of radiant heating are used. In the CivilService there is 34. Black, Flora W: op cit (3) an understanding that heating should be providedto 35. Bedford, T: Basic principles of ventilationand heating. H. maintain room temperatures ofup to 65°F with 60°F as K. Lewis. 2nd edition, 1964 36. Heating and Ventilation (Reconstruction) a minimum. One large industrial organisation"2 has Committee of the decided that an acceptable temperature Building Research Board of the Departmentof Scientific and Ind- for sedentary ustrial Research: Heating and ventilation ofdwellings. Post-war workers is between 65 and 68°F, thoughconsideration building studies nu. 19. HMSO, 1945 has to be given to humidity and air flowand the differ- 37. Billington, reviewing work of Munroand Chrenko. See ent requirements of men and women. Billington, N. S: Comfort at work. J Inst HeatVent Engrs. 1953, 21 (215) 141-4; Munro, A. F. and Chrenko, Criteria for determining the beginning and F. A: Effect of radiation end of the from surroundings on subjective impressionsof freshness. heating season have been established for theCivil Ser- J Hygiene, 1949, 47 (3) 288-96 vice. 3'I 38. Page, J. K: 1963.'op cit (19) 39. Bruce, W: op cit (4) 40. Houghten, F. C., Gutberlet, C. and Winter thermal comfort Witkowski, E: Draft temperatures and velocities in relation to skintemperature and In the introduction to herpaper, Black 34 reviewed a feeling of warmth. Trans ASHVE 1938, 289 number of comfort zones which have beenadvanced for 41. Manning, Peter: The design of roofs forsingle-storey general- the thermal environment, including equivalent purpose factories. University of Liverpool, Departmentof Build- temp- ing Science, 1962 eratures of 58 to 66°F, 35 62 to 66°F 36 and 66 to 72'F. 37 42. See, for example: Knight, J. C.and Knight, J. L: The air- It has been suggested 38 that the range of airtemperatures conditioning of multi-room buildings. J InstHeat Vent Engrs. put forward by Bedford (60 to 68°F)may be a bit 1962 (April) low, especially if air temperature is 43. Hickish, D. E: Thermal sensations ofworkers in light ind- considered in iso- ustry in summer. J Hygiene, 1955 (March)

63 warmer conditions which exist in summer, and wear lighter clothing but, due to either higher external temperatures or a build-up of heat within buildings. comfort zones are sometimes exceeded. The situation in office; is made worse by rising environmental standards and increasing business mechanisation, for the output of heat from lighting installations and office machines is often an inconvenient addition to the summer thermal

environment. The point is illustrated in a table publish lulAt ed by Page. " This shows the make-up of the cooling load for a particular room situated in Lat 52 N which has an electrical load of 6 watts/sq.ft (say, 90 to 100 Inv./ sq.ft of illumination, and no business machines). The output of heat from the electric installation represents 44 percent of the total cooling load. Thermal conditions within buildings in summer may require consideration of all the four main thermal parameters. Because relative humidity is an important consideration if people perspire, equivalent temperature (which takes no account of this parameter) is ustudiy considered unsuitable for assessing summer thermal environments. As yet no single comfort index is entirely satisfactory for defining the summer thermal environment. Conditions are sometimes specified in terms of Figure 56 The .nost common form of heating is by hot water radiators separate values for the different parameters, but a num- ,nounted beneath windows. ber of recommendations have been made in terms of corrected effective temperature (CET). Page, 45 following c,,h- mum, say, of 75 to 77 F) should be sought. A more ers, has discussed some of the defects of CET, the mo: usual method of radiant heating in offices is to mount important of which are an exaggeration of the effectsif the radiating surface within the ceiling and it must then humidity at temperatures below 75°F and the under- be ensured that the surface temperatures and mounting rating of the effects of air movement at higher effective heights are such that they do not cause uncomfortably temperatures with high humidities. warm conditions at the head. Chrenko discussed some A standard for optimum conditions of CET for light studies of subjective reaction to heated ceilings andpro- industrial workers in summer in England has been given vided a table of maximum desirable temperatures inre- by Hickish as 64.4 °F with an upper limit of 71.0°F. 4 " lation to size of panel and height of ceiling. Heating These recommended temperatures are low and during systems need to be zoned to allow for variations in hot weather could hardly be obtained in the typical office orientation, height within the building and localcon- which is not airconditioned. ditions for example, differences in density of people The currently acclpted basis for design for continuously and heat-producing equipment. occupied airconditioned buildings in the United King- The design of the thermal environment has to cope with dom " is an internal dry-bulb temperature of 70°F and relative humidity not exceeding 50 to 55 percent, subject to the difference between the inside and outside temperatures not exceeding 10 deg F. This condition is necessary because a person entering a cool airconditioned space from a hot outdoor environment with sweat on his skin and in his clothing can experience a sudden chilling of his body. Direct solar radiation may increase the CET within a room very considerably. For example, some work reported by the Knights" of measurements made during April 1961 in an unheated London office building in which no lights were burning has shown some substantial differences between the temperatures inside the south-west and north-east corners. Although the air temperatures were the same (between 71 and 74°F on four different days) in both corners, when the blinds were not used the globe temperatures were 23 deg F higher than the air temperature in the south-west but only 2 deg F higher in the north-east. Similarly, CET was 14 deg F higher than air temperature in the south -Hest but only 2 deg F higher in the north-east. Use venetian blinds reduced these differences: when the slats a fully closed the globe temperature was only 5 deg F higher and CET only 4 deg F higher than the air temperature in the worst case of the south-west corner. Figure 57 Radiant panel heating marking ceiling. Engineering design considerations The method of heating used in an office building is likely to be a choice between mainly convective systems (eg, hot water radiators, forced-air convectors) or mainly 44. Page, J. K: Lighting and environment. Architects' Journal. 1962 (June 20) radiant systems. Under-floor heating produces the most 45. Page, J. K: 1963 op cit (19) favourable vertical temperature gradients, but the pract- 46. Hiekish, D. E: op cit (43). Recommendation attributedto ical problem is to provide sufficient heat within the room Bedford. The standard used was that at least 80 percent of people without creating uncomfortably hot floor surfaces. The should be thermally comfortable 47. Institution of Heating and Ventilating Engineers:op cit (23) extreme upper acceptable limit of floor surface tempera- 48. Knight, J. C. and Knight, J. L: op cit (42) ture is considered 49 to be 80°F and it is usually re- 49. See, for example: Bedford, T: 1964, op cit (35) commended that temperatures lower than there (a maxi- 50. Chrenko, F. A. op cit (22) 64 the conflicting requirements of differentseasons. For example, the object of winter ventilationis to admit sufficient air to maintaina fresh atmosphere without causing draughts at working level.In summer it is nec- 84 essary to provide air movement at workinglevel and high rat tes of air change. A 12 Natural ventilation is dependenton weather conditions $0 and is therefore highly variable.A design method which 18 'pax 1160. takes into account either stackor wind effects has been 75 aim developed by the Building ResearchStation "2 but it 74 assumes that the inlets and outletsare measurable and, 12 mean glebe in any case, is likely to besuitable only for shallow lemperatere 10 *a.. Oa buildings. The core ofa deep building will be meeding- *M.O. ws woo...... min glebe 6$ ly difficult,if not impossible, to ventilateby only temperature ' natural' means. ss 64 If windows are to provideventilation, then there isa conflict between the ways in whichthey fulfil their role 62 in winter and summer. For winterventilation high level 60 openings and small aperturesare to be preferred, '" ss while in summer low level openings,cross ventilation, 56 and high rates of air movement may be necessary to 54 maintain comfortable conditions. in high buildings there 52 are substantial problems of windpressure and air infilt- 50 ration through the cracks aroundopening lights. A mechanical ventilatingsystem has many obvious advantages but, because ofnumerous secondary effects on 46 other parts of the building, it is difficultto ascertain its true extra cost. Including extraspace for equipment and ducting, it might be of the order of 7to 13 percent of the total building cost. A4 Although 4esternal its use is still exceptional 38 jdry bulb there is a growing tendency for airconditioning to be 36 temperature used, especially in tall buildings,because of wind prob- 34 blems, noise, air pollution, solarradiation heat gains, and the increasing output of heatfrom office equipment 32 and electric lighting. The interiorsof many deep-planned 30 office blocks are likely to requireall-year-round cooling. 28 Figures provided by the Knights 55suggest that the cap- 26 ital costs of a complete airconditioning installation used 24 the whole year round might be 20 percent of total build- 22 ing costs (including furnishing)or 3 times as much as a conventional heating system. 20 sunshine ours Measurements of thermal conditions in offices 0900 During this investigationmeasurements of the thermal 1100 1300 1500 1100 environment have been made withinoffice buildings with the object of learning howclosely typical types of TINE Hours 1817 office building design conformto the recommended Figure 59 standards. The following isa summary of the main Thermal conditions on 7 February1963 in an office building findings. °" which is heated by ceiling panels andnaturally ventilated. Time (GMT) 1110 Measurements of winter thermal conditionswere made in Mean air temperature (°F) 1510 six buildings, mostly during 72 72 the period Januaryto Mean relative humidity (per cent) 34 35 March 1963 when external dry-bulb Mean air movement (ft./min.) temperatures varied 25 51 Average EqT (°F) 70.6 70.7

around freezing point. In spiteof the extremely cold conditions, the interiors of all theoffices were warm and, in general, comfortable. Therewere, however, a number of complaints of draughts. Airtemperatures in all six 111111111111.117 - buildings never fell below 66°F,and the temperature of the surrounding surfaces andfurniture" did not differ very much from air temperatures. Nearlyall measurements of equivalent temperature fellwithin the comfort range of 66 Ftnd 72°F,the mean value being 70.2°F. Air temperatures in all theoffices tended to increase from floor to ceiling but inno case was the difference between the floor and a level 6 ft abovegreater than 5 deg F.

51. See: Page, J. K: Fundamentalsof ventilation design. Archi- tects' Journal, 1963 (July 3) 52. Building Research Station:The principles of natural ventilation of buildings. BRS digestno. 34. HMSO, 1951 (September) 53. Although the one buildingin which thermal measurements were made during winter which had thispattern of windows did not appear to be very satisfactory in thisrespect 54. Imperial Chemical IndustriesLimited: op cit (32) 55. Knight, J. C. and Knight, J.L: op cit (42) Figure SS 56. See appendices 7.1 and 7.2 fordescription of survey methods Simple measurements of the thermal environment. 57. As indicated by globetemperature measurements 58. Ellington, N. S:op cit (37)

1 11 A MR OT OR 11 U RADIATORSORDER 10 OfficeNo. 2/1 WINDOWS 22nd. Janu ry1963 10 122Q 1340 Office No. L211 9 .155 1400 9 210. January 1963 8 1015

7 7 6 6 .1 5 5

12 74 76 72 66 70 72 74 76 AIR TEMPERATURE F AIR TEMPERATIRE F

CEO:186 NEAINS 11 0rear0 11 10 10 ricoo.14t , 9 lit 'aNrc1963 9- e- 4 7 7 6 6 1 5 5--

4 3 3 2 2- 1

68 70 72 74 76 68 70 72 74 76 AIR TEMPERATURE F Figure 60 AIR TEMPERATURE F Vertical temperature variations in fouroffices with different types of heatinginstallation. In graph A, the group ofcurves at the left hand side shows temperature times. The right hand curve shows variations at a position 1 ft. fromthe window wall at different temperature variations at a position 20 ft. distantfrom the nearest windows. In graph C the left handcurve shows temperature variations 6 inches from curve is attributable to down-draught from and/or the external wall. The distinct changeof shape in the All other curves are representative of air infiltration through the windows. conditions within the occupiedareas. The largest gradients occurredin a building which has Measurements of summer thermalconditions were made a ceiling panel radiator system. Horizontal'gradients' when it was forecast that the weather would bewarm were generally level, though, aswas to be expected, the and sunny, with shade air interiors of deep buildings tended temperatures reaching over to be warmer than the 70°F. Unfortunately there perimeter. Because of the cold were only two such periods in weather, windows were the summer of 1963, the rarely opened during the period of first at the beginning ofJune, the surveys and, pro- the second at the end of July. bably as a consequence, air speeds During these times itwas were low, being most- possible to survey only three ly between 15 and 30 ft/min. buildings, two naturally The experimenter felt that ventilated and one airconditioned. some of the offices were stuffy and inone of them body In the airconditioned odours were evident. Airconditioned building the majority ofCET's offices maintained were between 68 and 70°F (ie, within optimum levels of relative humiditywhereas all the offices the comfort zones which relied upon natural mentioned earlier); themaximum being 73.5°F. Inone ventilation had mean values of the naturally ventilated between approximately 30 and40 percent; a few in- buildings, at positionsnot in direct sunlight, the CETwas mainly between 71 °F dividual values dropped to 27percent. and 75°F with a minimum Systematic studies of staff of 68°F so that, for muchof response to the thermal en- the working day, the comfort vironment were made inone building, over a period of zone was exceeded. In the other naturally ventilatedbuilding the CET time, by means of interviews andquestionnaires. In the was also high, being mainly between70°F and 72°F. early stages of the investigationthe building had only One of the importantfactors governing the recently been occupied, theairconditioning system had summer thermal environment is theamount of direct solar radia- not been adjusted to its working trimand varying temp- tion penetrating the building.The amount transmitted eratures and draughts were causingwidespread dis- by bare unshielded glassdepends on its thickness comfort. When the installationhad been brought under and colour and on the angleat which the radiation strikes control, the temperaturewas generally regarded as comfortably warm (sometimes,too warm and slightly stuffy); experience of draughts diminished. 59 59. See chapter 10 66 T 96 1 94 ,4 92 90 es ;pax ills %/ tni umpire ere

.., . .. . . 112 . ...,...... ean 119be emperatwe so . 76 ...------reiniplirgit:re 76 -...... /iNitelt 72 :teLpeiratvre 70 68 66 64 62 60 04 0.50.61.01.01.0 0.1 0.1 penekIne pours 0900 1100 1300 1500 1700

TIME Hours 1ST 16MT1Hour] Figure 61 Thermal conditions on 31 July1963 in a naturally ventilated office The peaks on the curve for maximum globe temperaturewere due to sun shining directlyon some of the globes. Time (BST) 1110 Mean air temperature (°F) 1600 76 79 Mean relative humidity (percent) 45 Mean air movement (ft./min.) 35 Average CET (°F) 25 72.5

it; the transmittance ofsolar heat energy throughtypi- cal clear glass at normalangle of incidence is about80 per cent-le, substantial. The effect ofvenetian blinds has already been referred to;" 'shade factors' for various other types of shadingequipment are given in the ASHRAE Guide. " In theairconditioned building the vertical white ' venetian 'blinds, when used, seemed to be effective in preventing highradiant temperatures. The management of anotheroffice was experimenting with several types of blindand trying to decide which was the most effective. The third buildinghas thick grey curtains but it was not possibleto take sufficient measurements to assess their effectivenessas a form of solar heating control. The staff in the airconditionedoffice had been instructed to draw the blinds on the south-easternside of the building at the end of the workingday so that heat gains in the early morning should beminimised. Apart from this one instance, office staff in other buildingsin the Survey Sample did not anticipatesolar heat gains and blinds were never drawn until thesun had been shining forsome hours. Although the surveyswere made on some of the hottest days of the year, conditionsnever became so severe that people were not ableto carry on with their normal work, and they were well belowthe limiting (ie, extreme) CET's of 90.1 to 92.2°F quotedfrom F. E. Smith 62 by Page 63 which are appropriateto 'sedentary work'.

60. Knight, J. C. and Knight, J.L: op cit (42) and (48) 61. American Society of Heating,Refrigerating and Air-conditioning Engineers: op cit (2) 62. Smith, F. E: Indices ofheat stress. Medical Research Coun- cil, memorandum no. 29. HMSO,1955 63. Page, J. K: 1963 (June 19)op cit (19) 67

r. Chapter 8 The aural environment

With the increase of traffic in towns and cities, the use of aspects of the mechanical offices. 'In machinerooms lightweight construction and larger windows in office noise levels at working positions usually exceeded94 buildings, and the growing use of office machinery,noise phons (83 dB)'.There was some relation between the in offices is becoming an ever increasingproblem. Noise type of work performed and the level of noise found is commonly believed to affect health,although this has tolerable: machine operators accepted higher not yet been confirmed by experimental research. levels of The noise than punch operators who in turn acceptedhigher Wilson Report' says that if the World HealthOrganisa- levels than clerical workers. tion's definition of health ('.. a state of complete physi- The Building Research Station's cal, mental and social well-being, and survey of post-war not merely an London offices" is reported to haveshown that large absence of disease and infirmity') is adopted,then, since numbers of the occupants are disturbed by people's well-being is diminished by noise, street noise noise affects 45 percent overall, and 65 percent in buildingson main health. But, as the report later shows, in themore narrow streets. Internal noise is less of a general problem and (and usual) meaning of the word, there is no evidence mainly occurs in large clericalareas and typing and that the noise levels customarilymet with domestically machine rooms. Within the offices the noiseclimate, and socially' (say, within offices) produceany direct from all causes, including street noises, and measureable physiological' effect movement of upon the average chairs, closing of filing cabinets,etc, is often in the range person. No reliable studies of the effects of noiseon nor- 60 to 77 dBA, rising to 80 to 95 dBA. mal mental health were known to the Committee.Noise, It has been stated in a BRS Digest '"that, apart from whether continuous or intermittent, above'about 90 dB' noise at firs. and second floor levels being has been shown to cause a significant slightly less increase in the than at ground level, traffic noise isnot greatly influ- number of errors made; it has not been possibleto iso- enced by height above ground. The Wilson late any harmful effect of noise alone below Report" this level.' mentioned some studies by the LondonCounty Council Despite the absence of proof, most office staff, especially which seem to confirm this statement andreferred to the those doing non-routine work requiringconcentration, BRS /COI survey of post war offices which are convinced that a noisy environment results in their may be the source of the statement. It is believedthat the effect work being less efficiently performed thanit might occurs not only in street `canyons': it can also be found otherwise be. Part of the reduced efficiencymay result with comparatively isolated tall blocks from the repetition of effort involved in catching where the sides up with and rear of the upper storeysare not screened, as they an interrupted train of thought. are nearer ground level, from noise generated in distant streets. Previous surveys of the aural environment in offices There have been few studies of acoustic conditionsin Legislation offices in Britain. The recent Act 12 has provisions enablingspecial regula- The civil service study group 4 reported that(twenty or tions to be made to safeguard the health so years ago) complaints about noise were frequent. It and welfare of found that six of the seven commercialbuildings in noisy locations which it visited were equipped,in noisy parts of the buildings, with either double windowsand mech- I. Committee on the Problem of Noise: Noise:final report. Cmnd anical ventilation or mechanical ventilation 2056. H MSO, 1963 (July) alone (so 2. Committee on the Problem of Noise:op cit (1) that windows could be kept shut). Althougha majority 3. Bewail, H: Noise. Paper in: Office buildings: report of a of the government offices which thegroup inspected had symposium held an 12th April 1956at the Royal Institute of at least one noisy side no special precautionsagainst British Architects. (The Institute), 1956 noise were seen except a few 4. Study Group appointed by HMTreasury: Working condi- cases of double windows in tions in the Civil Service. HMSO, 1947 very important rooms. 5. Wattleworth, W. H: The forgottenarmy. Royal Society of The Liverpool r'ports madeno mention of noise except Health Journal, 1959 (March-April) to remark 8 that, in the centre of urbanareas, windows 6. Nelson, I. D. M. and Morse, R. J:A pilot study of office cannot remain open because of street noises and accommodation. The Medical Officer, 1960(September 9) fumes. 7. Robinson, Alan: Working in the city: In nineteen of the 50 offices in the the interim report of a Barrow survey" noise survey. City of London Health Department, 1962(December) was said to be disturbing. The investigators thought it 8. Langdon, F. it The design of mechanisedoffices. Architects' `... true to say that the amount of natural ventilation Journal, 1963 (May 1) 9. Reported, so far, only in: Langdon, was governed by the amount of noise.' Externalnoise F. J. and Keighley, E. C: User research in office design. Architects'Journal, 1964 (Feb- interfered with conversation in 19percent of the offices ruary 5) in the City of Londonsurvey 7 and internal noise did the 10. Building ResearchStation:Noise and buildings. BRS same in another 12 percent. 3 percent of offices had digest no. 38 (second series). 1963 (September) 11. Committee on the Problem of Noise: double windows. Noise wasone of the least satisfactory op cit ( I) 12. Offices, Shops and Railway PremisesAct, 1963. HMSO 68 employees from noise or %ibrations. No such regulations have been made yet. '

The nature of the acoustical problem There are two main acoustical problems in offices: (i) of reducing noise which may irritateor distract staff or interfere with direct or telephone conversations (ii) of securing privacy where this may be needed. These problems require consideration of bothsound absorption (ie, reduction of reflection and hencereverber- ation time) and sound insulation. Noise experienced within an office originates bothout- side and inside the building. The relativeimportance of these two origins differs from officeto office but, especially in buildings located in busy citycentres, street noise is often the major problem. There is little that can be done to restrictexternal noise in existing buildings and, givena noisy central location, not much more for new buildings, especially those with light external cladding, although specially detailedfixed windows and mechanical ventilationmay help a little. A suggestion has been made 14 that where externalnoise

Figure 63 Noise source: telephones.

Figure 64 Noise source: desk calculators.

Figure 62 ieve the required sound insulating characteristics. These A machine section separated from general clericalareas. characteristics are virtually impossible to obtain if de- is likely to be a critical factor in the acceptability of mountable, lightweight partitions are used. A different the approach has been attempted in the United States where environment of a proposed new office building,a noise it is reported that built-in masking noises have been level survey should be conducted on the site before de- used in 'several important office buildings' sign is commenced and the actual range of to ensure exposure est- adequate speech privacy between offices separated by ablished. This information could then be usedto influ- such partitioning. 16 ence decisions about the total area of windows and their detailed design. Noise criteria for offices Noises within office buildings are likely to originate Because the sound pressure level ' (SPL)given by from several sources, including: a sound level meter does not give an adequate (i) office machinery (eg, typewriters and calculators) measure of the subjective effects of sounds, quantities which indicate (ii) telephone conversations and other voiced sounds this have been developed. They include: (iii) footsteps and slamming of doors originating Loudness level (LL) in phons within offices, outside in corridors, and overhead Speech interference level (SIL) in decibels. (iv) ventilation plant and lift machinery (v) internal constructional and office-fitting work and will be of two types, air-borne and structure-borne. The problem of securing privacy in offices is fundamental to the design of the building for, without interfering with ambient noise levels, the only effective 13. September 1964 way is 14. Miller, Laymon N: Case histories of noise control inoffice through massive construction. In other words, it ispre- buildings and homes. Chapter 22 in: Noise reduction (Ed: ferable that the office spaces where privacy is important Beranek, L. L.) New York, McGraw-Hill. 1960 should be selected in the early stages of design, and their 15. Watters, B. G: Speech-privacy considerations. AppendixC to Chapter 20 in: Noise reduction (Ed: Beranek, L. L.),op cit location and/or construction decided expresslyto ach- (14)

69 The loudness level of a sound is defined as the sound Recommended standards pressure level of a 1,000 c/s tone that sounds equal in The following standards have been advocated by differ- loudness to the sound being rated. It can be calculated ent authorities: from the octave-band levels by a method dueto Noise level within an officean upper limit of 55 Stevens. dBA 27 based on SIL The speech interference level is an approximate guideto Partitions: insulation the interfering effect of noise on speech. It consists of Where at least one room has the arithmetic mean of the octave-bandpressure levels a need for quiet or privacya minimum of 45 dB 2' in three bands: 600 to 1,200; 1,200 to 2,400 and 2,400to Clerical offices where noise 4,800 c/s and ignores the contribution of the lower and is not a major nuisance aminimum of 20 to 30dB 23 higher frequencies. Floors: insulation A SIL of 45 to 60 dB is considered 'slightly difficult' Insulation (airborne) for for telephone use;' a SIL of 50 to 55 dB is considered floors plus, preferably, a acceptable for secretarial, clerical or typing rooms.'" resilient finish a minimum of 45 dB "" An extension of these units has been made by Bera- Reduction of impact sound nek 19. 2" who found from a series of studies of noise in on floors by provision of offices that the subjective ratings of theoccupants cor- resilient finish a minimum of 10 phons :" related well with both the speech interference levels and Reverberation time (empty rooms) loudness levels when there was a difference of 22 units Small private-office 0.75 seconds 32 between the LL (in phons) and the SIL (in decibels). He derived a series of noise criterion (NC)curves from which the permissible sound pressure level in eightoct- ave bands can be read. The loudness level for each curve is 22 units greater than the speech interference level, the SIL in decibels being given by the NC number of the curve. NC valuescon- sidered suitable for certain office situations in Britain, based on a table of recommended noise criteria for offices by Beranek, are given in table 7. The valuesare applicable to offices in normal operation but withno- body speaking at the particular desk or conference table which forms the focus of the situation.

Table 7 Recommended noise criteria for offices

Recommended maximum PP/6- MOM Situation NC value Board rooms, conference rooms, private offices for use of one person 30 to 35 Offices occupied by between two and ten people 35 to 40 Figure 65 General offices, occupied by more In this building typing pools are small and the rooms have large than ten people 40 to 50 areas of sound absorbent materials, including carpeted floors. Offices in which business machines, including typewriters, are the main source of noise 50 to 55

16. Stevens, S. S: Calculation of the loudness of a complex Based on Beranek's recommended noise criteria for offices. See: noise. J Acoust Soc Am. 1956, 2$ 807-832 Beranek, L. L: Revised criteria for noise in buildings. Noise 17. Council for Codes of Practice; British Standards Institution: Control, 1957 3 19-27. British Standard Code of Practice CP 3 Chapter III (1960). Sound insulation and noise reduction. (The Institution), 1960 The A-weighting network on sound level meters is often 18. Anon: Noise. Chapter 2 in: Better offices. Institute of Direct- Jn ors, ? 1960 used to specify particular noise levels. This is a single 19. Beranek, L. L: Criteria for office quieting based on question- value approximating to the sound heard by the human naire rating studies. J Acoust Soc Am. 1956, 2$ (5) 833-852 ear, which is used without reference to the sound press- 20. Beranek, L. L: Revised criteria for noise in buildings. Noise ure levels in particular frequencies. Control, 1957, 319 -.27 21. Beranek, L. L: Criteria for noise and vibration in buildings Beranek has pointed out 21 that the same A-scale reading and vehicles. Chapter 20 in: Noise reduction (Ed: Beranek, L. L.), may be obtained for a wide variety of shapes of noise op cit (14) spectra. Moreover, the values of the eight octave bands 22. Young, Robert W: Single-number criteria forroom noise. are necessary for the engineering design of noise control J Acoust Soc Am. 1964 (February) measures, so a single number cannot be adequate. How- 23. Committee on the Problem of Noise: op cit (1) 24. See, for example: Purkis, H. J: Transport noise andtown ever, it has been shown recently 22 that noisiness as rated planning. Journal of Sound and Vibration. 1964, 3 323-34 by 'executive office personnel' can be correlated with 25. Stevens, S. S : op cit (16) 'A' sound levels. The Wilson Committee's recommenda- 26. Parkin, P. H. and Humphreys, H. R: Acoustics, noise and tions 23 were given in dBA, and the BRS's recent buildings. Faber (second edition), 1963 27. Committee on the Problem of Noise: op cit (1) London Noise Survey 24 has also used these units. A 28. Cowell for Codes of Practice; British Standards Institution: simple method of converting readings of dBAon the op cit (17) sound-level meter into 'Stevens phons' 23 has been given 29. Council for Codes of Practice; British Standards Institution: op cit (17) by Parkin and Humphreys:" the dBA reading is multi- 30. Council for Codes of Practice; British Standards Institution: plied by 1.05 and 10 is added. op cit (17) There are a number of other noise ratings in use but the 31. Acoustics Committee of the Building Research Board of the ones used in this report are dBA and Noise Criteria, for Department of Scientific and Industrial Research: Sound insula- these are the two most commonly used in connection tion and acoustics. Post-war building studies no. 14. HMSO, 1955 32. Council for Codes of Practice; British Standards Institution: with buildings. op cit (17) 70 General office 1.0 second (may be partition (which would be an expensive component) is Office-machine room increased to 1.25 seconds only about 35 dB.'" In rooms which have suspended Typing pool J -for 'very large offices') ceilings the detailing of the connection between partition and ceiling is critical for the ceiling void provides a The Code of Practice "4 contains a recommendation that, simple route for the passage of sound from one side to in the interests of noise reduction, typing pools should the other. This space is very difficult to soundproof be restricted to about twelve machines to a room. effectively, especially if it contains piped services or One industrial firm has adopted standards " based upon other irregular obstructions. the distance over which a conversation should be poss- The choice of floor finish offers an opportunity for re- ible when normal speaking voices are used: ducing one of the causes of internally created noise ie, Private offices: audible 10 ft (equivalent "6 to footstepsat its source. A resilient finish to absorb the distance of SIL of 45 dB) energy from footsteps is especially needed in open-plan General office: audible 5 ft (equivalent to SIL offices. The ceiling can be used to reduce, by absorption, distance of of 51 dB) the general level of noise. Machine rooms: audible 3 ft (equivalent to SIL distance of of 55 dB) Surveys of noise within offices During this study measurements of noise have been Constructional considerations made in a number of offices within a total of nine build- External walls which are of heavy construction tend to ings, the object being to make comparisons of typical minimise the transmission of sound (eg, of street noises) conditions with recommended criteria. Measurements but windows constitute a weak link, for they allow the were made during working days but not at rush hours-1" passage of noise more readily than most wall construct- so the results should be reasonably indicative of general ions. Compare, for example, the insulating value of a 9 conditions. They show that the noise levels in most cleri- inch brick wall (50 dB) and a closed openable' single cal offices were well up to, and usually exceeded, the window (18 to 20 dB). 37 In order to maintain a tolerable Wilson Report's upper limit of toleration of 55 dBA in level of noise within rooms in heavily trafficked areas it buildings in which 'speech is of great importance'." In may be necessary to use fixed double windows and mech- business machine rooms, noises were Jsually of the order anical ventilation. of 70 to 75 dBA. The recommended NC values were ex- ti

Figure 66 The space above a suspended ceiling is difficult to soundproof. 33. Council for Codes of Practice; British Standards Institution: op cit (17) The minimum effective spacing of the two panes is about 34. Council for Codes of Practice; British Standards Institution: 4 inches but greater widths are preferable, especially for op cit (17) 35. Imperial Chemical Industries Limited: Standards for offices low and middle frequency sounds3s (eg, city centre and laboratories. (The Company), 1960 (April) traffic). Sound reduction is increased by providing an 36. On basis of table 6: maximum speech interference levels, in : absorbent lining to the cavity and it is advisable to sep- Council for Codes of Practice; British Standards Institution: arate the frames of the two panes. Curtain walling, be- op cit (17) 37. Council for Codes of Practice; British Standards Institution: cause it consists of lightweight members which are dry- op cit (17) jointed, provides greater opportunities than traditional 38. Council for Codes of Practice; British Standards Institution: systems of cladding for external noise to penetrate to the op cit (17) interior. 39' 4° 39. Committee on the Problem of Noise: op cit (I) 40. Markus, T. A: The glass curtain wall: Fire resistance and The choice of materials for internal walls and partitions sound transmission. Architects' Journal, 1957 (December 12) includes solid walling (eg, in-situ concrete or brick or 41. Council for Codes of Practice; British Standards Institution: block), framed walls constructed in-situ (such as timber op cit (17) framing with glazed or boarded panels) and proprietary 42. Burgess, Roger A: Insulation of partitioning as a design factor. Architects' Journal, 1962 (January 31) demountable systems of partitioning. The sound re- 43. See appendix 8.1for details of methods used in noise duction value of a 4 inch brick wall plastered both sides is surveys about 45 dB; 4' a very high standard for a demountable 44. Committee on the Problem of Noise : op cit (1) 71 NC:50 h,,. , 70 70 ..,,,ii11111 81111111'11ml .. ii3111110 ii1111111111111

1111111111N.. 60 60 CID

50 50 UJ Nt.:55 win 60 Cop

LAJ 30 30

r:'20 20

63 125 250 500 1000 2000 4000 8000 63 125 250 500 1000 2000 6000 8000

FREQUENCY CYCLES PER SECOND FREQUENCY CYCLESPER SECOND Figure 67 Figure 69 Noise spectrum in centre of large general office (about 500 staff Noise spectrum in small machine room (4 operators). Plastic in one undivided space). Cork floor,acoustictileceiling tile floor. hard plaster walls and ceiling. (75-80 dBA) (59-61 dBA).

NC =50 70 70 0000"1" mimmollhhhhillinWhiH110111111141 111111 11111111mmhiii '''1111 60 inallob- 60 n 50 w Ii

lig40

20 -1

I 1 i

63 125 250 500 1000 2000 6000 8000 63 125 250 500 1000 2000 4000 8000

FREQUENCY CYCLES PER SECOND FREQUENCY CYCLES PER SECOND Figure 70 Figure 63 Noise spectrum in centre of large machine room (more than Noise spectrum in large general office (55 staff). Plastic tile 100 operators). Floor, ceiling and three out of four walls surfaced floor, hard plaster walls and ceiling, metal furniture (58-67 dBA). with sound absorbent material (70-73 dBA).

ceeded in some frequency bands in almost all of the offices visited. However, there is little point in achieving 70 low NC levels in private offices if they are not private,

and in some of the offices in which the recommended hhi, 60 criteria had been achieved, noise and conversation in 40 adjoining rooms could be heard. The spectra measured !!!!!!!:::1111114111111411114iNki

in general offices followed the NC curves better than 1.61 50.646,,k , h.,. the spectra measured in machine rooms, where the noise .. 1 levels decreased little, if at all, at the high frequency end ^. L&I 1 of the spectra. 40 mm , . NI11Mh h , '1111111111 Traffic was a major source of noise in the private offices .1"1 ^a h235' and in many of the clerical offices too. The effect of Q. opening windows was to raise NC values by approxi- ea- 30 .1.i,,I,Iii, 4 II mately 5 to 10 units. '. The staff in one large, fully airconditioned and open- .n 20 4,1 planned building with fixed glazing bordering upon a busy main road were asked about the acoustic conditions. Two-thirds stated that their building was not distractingly noisy; what noise they heard was believed to 63 125 250 500 1000 2000 4000 8000 originate inside the building from normal conversation, movement and office machinery. Yet the noise levels FREQUENCY CYCLES PER SECOND were, in fact, in the region of 50 to 55 dBA in clerical Figure 71 Noise spectra in empty private office. Spectrum shown in broken spaces, ie, at the Wilson maximum, and 70 to 75 dBA line represents noise levels with window closed (Window closed in machine rooms. The fact that the acoustic conditions 45-47 dBA: open 52-56 dBA). 72 in this building were notworse must, in large measure, be due to the quality of the detaileddesign: acoustic ceiling, rubber floor, rubber tippedchair legs and fixed windows. In another building, inone very large and completely open office space accommodating about 500people, the experimenter felt that, although the measurednoise levels were high (about 55 to 65 dBA), the general impression was one of surprising quiet; the occupants didnot seem to have any difficulty in talking or using thetelephone.

73 Chapter 9 Subjective and social aspects of the environment

In an article intended to introduce the subject of social may affect both the brief and the feedback. Failureto psychology to architects, Noblewrote .. . as architects get a proper return of information about theeffect of his we help to shape people's future behaviour bythe en- building upon the users may vironment we create'. This belief is mean that the architect's very widely held accumulated experience of practicecounts for less than amongst architects and, if it is true, it is ofconcern not might be expected: hissuccess in providing he best only to architects but also to behaviouralscientists and possible environment may not be the general public. much greater at the end of his career than itwas at the outset. Clearly, this While an architectmay be curious about the way this sort of practice is non-scientific, for science shaping of behaviour happens, his is cumula- primary concern is tive and depends upon previous experience,each prob- with the building. His questions,therefore, can be ex- lem being tackled as one of pected to be framed in such a class about which a great terms as,In what ways deal is known in at least a generalsense. should a desire for people's wellbeinginfluence the de- The notion of influencing and sign of the building?' Any architect shaping behaviour seeking the answers through the building environmentcan be interpreted in to questions such as this aboutan office building will be many ways, but it is convenient to discuss it under optimistic. For although, over the last the few years, the headings of health, efficiency, amenity,and personal Building Research Station has conducteda number of and social effects. investigations involving psychologists andother behavioural scientists, the published literatureon systematic Health and quantitative studies of qualitativeaspects of archi- Health studies in commercial tecture is negligible. 2 or industrial buildings tend to concentrate upon thosefeatures of the environ- At present, design decisions affectingthe social environment which present a hazard to the physicalwell-being ment of office buildings are made almost entirelyon the of the workers, insofar basis of expectation or personal prejudice, as they have effects that are rather than measurable ;n terms of affected health.The manner in knowledge. For example,an architect may feel that lay- which behaviour may be affected byenvironmental out must be based on smallrooms, because people are conditions in this instance is dictated more happy in small groups 'with which they by the resultant can readily health conditions. For example,a number of behavioural identify themselves' 3 than theyare in large ones. Yet consequences follow total or partial deafness caused there seems to be no properly established by bases for such inadequate sound reduction,or respiratory problems a belief, nor for many other unquestioned attitudes caused by bad thermal conditions. which are the starting point of These, however, are an architect's design pro- rather gross examples of the influenceof the environ- cess. ment, and ones which are not likely In the practice of his profession, especially to bulk large in a perhaps in study of office environmentswhere the statutory re- the design of workplaces, the architect suffersfrom two quirements are met. disadvantages. Firstly, his knowledge ofuser-requirements is often not first-hand. Hemay, for instance, have Efficiency to deal with a client who has onlya superficial know- The type of environmental study ledge of his rank and file clerical workers which has been most and their re- extensively pursued is into thenature of the optimum quirements. This situation is likelyto be particularly physical conditions at which true where the client is a developer, top-level a given group of people will executive, operate most efficiently while performinga given range or, in the case of the civil service, from anotherministry. of tasks. The architect's second disadvantage is that he is rarely Strother has drawn attentionto the evolution that has in a position to assess thesuccess or failure of his build- taken place in research into the ing, except through the working environment: comments of his professional 'The studies of the World War Iera and the twenties cen- colleagues and the specific complaintsof his clients. tred primarily around an input-output Page,analysing some of the other communication approach to the human problems of industry. The humanbeing was re- problems of the building industry, usedthe description of 'a system of zero feedback', and thisseems to be an appropriate summary of present practice in this con- 1. Noble, J: The how and why ofbehaviour: social psychology text too, for it often happens that the architect'sonly for the architect. Architects' Journal,1963 (March 6) connection with a building,once it has been occupied, is 2. Though publication of both thefull report on the study of with particular decorativeor constructional problems. mechanised offices (? completed 1960)and the CO I study of The role of the architect, post-war offices which was undertaken in 1961must be imminent as understood by the client, 3. See, for example, Salmon, G: Settingthe scene. Chapter in: does not appear to carry responsibilityfor the happiness Better Offices. Institute of Directors,London (The Institute), and efficiency of the occupants (especiallywhere the ? 1960 4. Page, J. K: The building industry:a system of zero feedback. purpose of the building is not clearly 'social') and this The Builder, 1962 (August 31) 74 garded largely as an isolated phenomenon affected by a essential to analyse what ismeant by the universally held variety of environmental conditionswhich could in- belief, ' I know what I film'. crease or decrease his output'.' He also referredto the In any study of satisfaction there historical importance of Elton Mayo's is the additional prob- extensive study lem that people arc probablynot able to be objective at the Hawthorne Works of the WesternElectric Com- about individual features in their pany." Previous to the Hawthorne study environment, but re- it had freq- spond to the 'halo ' effect of uently been observed that higher the situation. That is, their amenity levels of such judgements depend a great dealon the total satisfaction environmental factors as illumination,ventilation and they get from a situation, and heating resulted in increased this will largely turn on output and this was gene- social relationships. This, of rally believed to be due course, implies *hat, strictly, to more favourable physio- studies of amenityare not separable from studies of the logical stimuli resulting inmore favourable physiological total psychological situation. responses. This was thought to be true of thoseresults of the Hawthorne experiment where higher output follow- Personal and social effects ed improved visual conditions,but it was also found The study of the individualand social effects of anen- that reduction of the illuminationlevels, in some cases vironment is necessarily to such a degree that the workers a task for specialists, ie, ' human were hardly atilt to see scientists', but the interests ofthe two main groups well enough to work, stillresulted in increased pro- sociologists and psychologists duction. The interpretation drawn, - are different. For ex- and one which has ample, recent advances in been confirmed by much subsequent electronic data processing, experimentation, with Us many implications for clericalworkers of auto- was that workers responded more to theinterest being mation, have created shown in them as a good deal of interest. '9 Sociolo- persons, as marked by the r'search, gists may ask 'what than to the prevailing physical conditions. are the social consequences when automatic methods are introduced Mayo's research is the classic into an industry?', example demonstrating while the question for psychologistsis, ' in what way are the inadequacy of environmentalresearch which deals only with overt stimuli and individuals affected when conditionsin the industry responses. Strother summ- change?' Such a change mightresult from the intro- arised its influence by writing'If the 1920's could be call- duction of EDP systems ed the era of the human engineering or (the subject of this re- approach, the 1930's search project) from the physical marked the beginning of the human setting for work. The relations approach'. transference of people from small This may be theoretically and heterogeneous groups academically true, but in composed of clerks, typists, fact the human engineering (or comptometer operators and `ergonomic') approach junior management, into formally to environmental studies is still determined homo- a vigorous reality, poss- geneous groups like large typing pools, machine ibly because much applied rooms research is financed by in- and document stores isone example of a change of dustrial companies whichare more interested in the facts environment which may affectindividuals significantly. of increased efficiency than inthe.r causes. The Haw- In studies of people'sresponse to environment, a build- thorne experiment showed thatcause and effect do not ing can be treated as though it necessarily need to be understood is a catalyst, a relatively for the production of inert agent, but one vitalto a particular process, in this favourable effects. The exactway in which behaviour is case the work of the office. However, thebuilding differs influenced in such circumstanceshas still to be fully in a very important described; it is not yet in respect from the catalytic agents of a form likely to be of any help chemistry. These facilitate only to the architect in planning his building. one kind of reaction between substances, whereas thebuilding may facilitate Amenity many. For example, most people, botharchitects and laymen, would hold that colours, Apart from studies of the behavioural intensities of illumina- consequences of tion, and textures are capable ofinfluencing mood and environmental conditions, there hasmore recently been feelings. some formal research on subjective reaction to design Another way in whicha building interior may act as features in the home and at work.Inherent in this type a of research is the notion that catalyst to behaviour is throughits arrangement of if the conditions that spaces. For example, common entrances people say they find most congenial to different de- can be provided, partments mean that there aremany more opportunities then the influence of theenvironment can be nothing for inter-departmental but beneficial. This is the 'market contacts than if there weresep- research' type of app- arate ones. The results of these roach which aims at establishing interactions need to be the most satisfying and studied from both the social and pleasant working conditions, and the company's points regards these aims as of view for the individual'ssocial world is potentially being sufficient in themselves. widened, and it may well be that Everyday experience, however,provides a warning. In inter-departmental working is facilitated, by such informalcontacts. making judgements aboutconditions that he has not Yet another way in which thebuilding may influence be- experienced, or which he hasexperienced for only a short time, the individual haviour is by the size of the workroom. What happens, cannot be in a position to for example, whena clerk moves from a small office set- make a choice basedon knowledge and reason. His ting to a large open one? Inwhat way do his sentiments judgement is not objective andtakes as its reference change towards both his immediateworking group and point a limited subjectiveexperience. The problem is made more complex when prejudiceand attitudes, not specifically connected with the judgements being made, 5. Strother, George B: Changingconcepts of teaching and re- are superimposed. Both Jahodaand Raven9 have search in business. Introductionto: Social science approaches to drawn attention to the prejudicethat leads people to business behaviour (Ed: Strother,George B). Homewood, III: choose the conditions that they Richard D. Irwin; London: TavistockPublications, 1962 are used to rather than 6. Rothlisberger, F. J. and Dickson, the ones that they are not. W. J: Management and the worker. An account ofa research programme conducted by The prejudice of preferringwhat has been experienced the Western Electric Company,Hawthorne Works, Chicago. (rather than what has not)may have a good basis in Harvard University Press, Cambridge(Mass), 1939 7. Strother, George B: common sense: the individual can, in thisway, make op cit (5) S. Jahoda, Marie: (ina paper to the annual conference of the sure that he does not suffer too muchfrom other British Psychological Society in people's personal likes 1963). See: Some social psycho- or lack of judgement. Neverthe- logical consequences of architecturaldecisions. Abstract in Proc less, the response is conservative,and impedes innov- Brit Psych Soc Conf. Reading.Abstr Bull B P S. 1963 16(51) ation in the human environmentin a way quite out of 9. Raven, J: (in a paperto the annual conference of the British keeping with an age of experimentation Psychological Society in 1963). See:Human needs and the de- and technical sign of dwellings. Abstract inProc Brit Psych Soc Conf. Read- discovery. In discussing buildingdesign it is therefore ing. Abstr Bull B P S. 1963 16(51) necessary to examine in some detail the attitudesand 10. See, for example, Scott, W.H: Office automation and the prejudices that shape subjectiveresponse; it becomes non-manual worker. Organisationfor Economic Co-operation and Development. Paris (The Organisation),? 1962

1111111=111111111M0111111111111MMIII1111MINIMMINI his department or company as a whole? What group sizes tend to have cohesive and which disintegrative effects on social and working groups? Are they affected at all? How does the individual's concept of himself change; does the worker regard himself as reduced in stature by being given less obvious prominence in a larger group? Does he feel more or less ambitious in the new setting? Are there more far-reaching social consequences? Many personalrelationshipswillundergo radical changes when new policies for the grouping of staff are introduced. The middle grade clerical worker who previously had a regular typist, or even a secretary, may now work with a pool typist, or more frequently, a dict- ating machine. The typists will work in a typing pool and no longer have a boss or a personally responsible place in the life of a department. The changed relationships between people can be expected to have important counterparts within individuals. Formalisation, and the concentration of working groups, offer many opportunities for the introduction of more autocratic measures in the supervision and management of staff; this is another way in which the individual and social status of the office worker may be affected by building design. It has been traditional that clerical workers, more than any other working group, have had the best chances for upward occupational and social mobility. Because of this, clerical work has drawn a much higher proportion of well educated people than could otherwise be expected. Managers, executives and even directors have been drawn from the 'rank and file'. With the changing structure of clerical occupations, and other social changes like universal literacy and increasing opportunities of higher education for the more talen- ted, it is difficult to recruit the same level of individual. Instead his place is being taken by the female clerk and machine operator who are quite adequate for the mechanical clerical and accounting functions and other remaining clerical procedures which have been de-skilled by work study. The design of the work place must take into account not only the changing nature of clerical work, but also the internal changes that are taking place in clerical occupations themselves. "

I I. For an examination of the evolution of office work from the sociological point of view, see Lockwood, D: The black-coated worker. Allen and Unwin, 1958 76 Stage Three Part Two

The total environment cl°

wow' l'w oorEdi" 10 ...0.-.,, lo ' 0110- :;1. .- . Igi i .4ifs OF grit P

lig;i..4 W ;III 1 ili: ,,, Z141:i.zZ 5,1,44' 111 sa '

...aril% Z. ::4 1". ,.. 4.4.11 fig;' ":. .fr-.,' ;fill;till:.!r , i:1- driVit '1;"::,- fie;-a ttii.. ;I ! EiE, Magi El':

a a;

1111111"f! Merl'

emir ,131 1151

Oa. Chapter 10 The total environment

A case study of the CIS building in Manchester

The Pilkington Research Unit's long-term aim is the development of an environmental discipline in which the components are studied as a whole. This chapter consists of a case study of the total environment in Britain's tallest office building: the headquarters of the Co-operative Insurance Society Limited in Manchester, which was completed in 1962.1. 2 The building consists of an extensive podium which is about 175 ft by 250 ft and contains the ground to fourth floors. A tower containing twenty-one further floors rises above. The staff number about 2,500. The Unit's research interest in the building developed from a letter received from the General Manager of the CIS 3 who said that he had read of the office-building project in The Guardian4 and would be glad to offer facilities for study. These were readily accepted, for not only was the building probably the most advanced of its kind in the country, but the physical and human problems of populating it with a large staff which had previously worked in a variety of conventional and much smaller office buildings presented an unusual opportunity to study the human aspects of a modern environment. 5 Figure 73 An environment results from the satisfaction of a func- Interior of one of the offices used by the CIS before its move tional need for space. At its simplest, the requirement into the new building. may be entirely utilitarian; in a more complex form, there may be substantial overtones of a subjective type, made which involve the expression of subjective prefer- eg, the projection of the company image.' But with all ences.Translated into a building these decisions designeven of utility buildings - decisions have to be have a permanent effect upon people's behaviour and comfort. In this chapter an attempt will be made to describe how the environment of the CIS was created, how the decisions were made, what sort of environment .71 Pi has resulted, and how the occupants have reacted to it.

The reasons for the new office building With the post-war expansion in commerce the CIS grew from being a comparatively small insurance company to one of the largest in the country; this resulted in an expansion of staff and radically increased space requirements. Post-war building, however, did not keep pace with this expansion and the CIS was obliged to

1. For an account of the early studies (particularly the psychological investigations) in this building see:Wells, Brian:Psycho- logy in the office.The Guardian,1963 (October 23) 2. For a short re-appraisalc'fthe CIS environment after its first 18 months in use see:Manning, Peter and Wells, Brian: CIS: re-appraisal of an environment. Interior Design. 1964 (July/ August) 3. Dinnage, R:(Co- operative Insurance Society Ltd.): letter to P. Manning, 1962 (July 13) 4. Anon:Research unit's inquiry into office design.The Guardian, 1963(July 121 5. The building has been described and illustrated inmany tech- Figure 72 nical journals. See, for example,Scott, Keith N: Manchester's The former head office of the CIS. skyscraper. Archit. Build. News, 1963 (January 16) 79 rent eight other offices in the vicinity of its old head most senior executives: the General Managerand the office. These premises, being oldand scattered, created Investments Secretary. No departmentwas solely con- muck unnecessary movement anddelay in the process- cerned with the project, and ing of documents, and no senior staff were occu- some inconvenience to the staff. pied full-time upon it (untila premises controller was In 1953 the Board decided in principleupon the building of a new office block. appointed five months before constructionstarted). The O and M section had been establishedtoo late to take much part in the design of the buildingand so heads of The basis on which the sitewas chosen departments were given sole resonsibility forthe layout The first major decision whichhad to be made was of their working areas. where the new building should besited. The natural Other company staff closely involved, location appeared to be as executives a site close to the existing main rather than policy makers,were an assistant actuary building. It was, however, consideredwise to explore and the premises controller. The assistantactuary was the possibility of movingto the suburbs because the responsible for the mechanics of themove from the old building was to be an administrativehead office and not premises to the new, and for providingmanagement- generally used by the public. Itwas believed too that, staff communication on the subject away from the centre of the city, both site and building through special `newsletters' and articles in the housemagazine. The would be cheaper. Itwas also expected that staff would appointment of premises controller be easier to recruit. was a new one, created specifically for thenew building. His responsi- The suburban site which appearedto offer the greatest bility was to oversee the construction advantages in terms of pleasant location, of the building good trans- and thus acquire an intimate knowledgeof it, organise port, reasonable cost and a large potentiallabour force the occupation of the building, control was at Wythenshawe, a Manchester overflow the rt nning of area. The such services as canteens, recruit technicalstaff for the

siiro MIN A' .1111810

Aiiiduakt

Figure 74 The location of the CIS building in relationto the city of Manchester. (The cathedral in the foreground). labour position was felt to be particularly attractive as operation of the building, and provideplanned maint- the Society was mainly concerned withrecruiting enance and cleaning programmes. women, and it was known that large numbers of office workers commute daily from there to central Man- The appointment of the designteam chester. In common with similar employers,the CIS At an early stage in the Company's has a rather high turnover of consideration of its women staff, and expected new building, Mr G. S. Hay, Chief Architectto the that continuing losses in its head officeforce would be CWS in Manchester, had been replaced by local workers. appointed architect for the development; subsequently MrGordon Tait of Lon- However, the results of a pilot survey of the availability don was invited to act as Associate of female labour, conducted by the CWS with Mr Hay. The Market Re- joint architects then advisedon the appointments of search Unit, indicated that girlswere not interested in consulting structural, electrical andmechanical engin- working in Wythenshawe, but wantedto travel into eers. According to the General Manager, it hadalways Manchester for shopping, for dancing duringthe lunch been considered possible that the building hour, and for many other city activities. The would be fully report air-conditioned, and the consultant mechanicalengineer was so positive that the proposal to move toa suburb was therefore asked, soon after his appointment, was abandoned and it was instead decided to build in the to investigate the financial implications ofsuch a decision. city centre, on a site adjacent to the old headoffice build- A party consisting of ing. a CIS director, the Investments Secretary, and two architects and the mechanicalengineer made a visit to North America to study thedesign of The company staff responsible for planning tall office buildings, and the problems of The ultimate responsibility for planning the project heating, venti- was lating and airconditioning, and verticalmovement of in the hands of a steering committee whichconsisted of staff, mail and documents. three (later four) directors and two of the company's This visit appears to have hada major influence upon 80 the building's design. On the deputation'sreturn the architects were asked to preparea scheme for a single . large block of between 20 and 25storeys, to proceed on the assumption that curtain walling wouldbe used, and, providing the cost was not prohibitive,to employ full airconditioning. A previous intention of adoptingopen- planned offices was confirmed. The architectsand probably the other members of theparty, too, had obviously been greatly influenced by thework of the American architects Skidmore, Owingsand Merrill, especially perhaps, by their administrativebuilding for the Inland Steel Company in Chicago. Later, a third firm of architects and interiordesign consultants (Design Research Unit)was appointed to take responsibility for the interior decor andfurniture. The largest part of their interior design workis to be seen in the entrance hall, diningrooms, and in the executive suite, but they were also responsible forthe partitions, furniture and colour schemes in the generaloffice areas.

Figure 75 Aerial view of the CIS building whichshows the three main components of its form: the office towel, the servicetower, and the podium.

The reasons and policies underlyingcertain important features of the design The building has two features whichmake it somewhat remarkable. In the first place it isunusually tall and secondly, the interior was conceivedas an example of open planning'. The principles for design proposedby the project steering committee were threefold: (i) the building should addto the prestige of the Soc- iety and the Co-operativemovement (ii) it should improve theappearance of the City of Manchester (iii) it should provide first-classaccommodation for staff, with a view to attractingnew and retaining existing staff. 6 The General Manager has said,7 'If I may use a rather hackneyed phrase,we wanted to impress an " image "on the people of Manchester ardelsewhere of the strength and size of the CIS throughthe medium of a modern office building. Very fewpeople see our balance sheetor read our chairman'sreport but this new building will be a constant reminder that the CIS isa large and first-

6. Manage, R: Design policy for corporate buying. Paper"`1, Figure 76 the General Manager of theCIS to the International Desia.i Section through the CIS. Congress, 1961. (Duplicatedtypescript) 7. Dinnage, R: op cit (6)

- sters and nervous people would feel when walking across one of the podium floors, each of which contains approximately 35,000 sq.ft of office space exclusive of service areas. He had felt that the open spaces and great numbers of people might result in unpleasant self- consciousness, and therefore decided to include a certain amount of partitioning to break-up these very large areas. Thereafter, partitioning had just crept into the plans of the tower floors. 9 It is possible that the original conception of open plann- _ _ _ . o 0. ing was not strongly held, or that the custom of physic- 1 ally separating departments was too strong. It seems E 7 1_1 more likely, however, that many departmental managers 117%07 had been vociferous and successful in their de- Figure 77 mands for partitioning around their departments. Some A typical open-plan floor in the tower. small areas within departments are enclosed by partitions. An explanation for this was given in terms of st ,o01111..c-1-1 ,f supervisor needs: a supervisor could do his job more !--L1 1J easily if he could clearly see where his responsibilities 1171 1 r IL'I ! ended. Asked whether supervisors would be able to 31-71 I supervise more efficiently if their sections were screened L__11 .J_J 01 1 off and separated from other staff, a respondent replied cD:= that he could not think of any reason why this should be Uo the case, but that`supervisors preferred to be screened onn a I L__1IL b off from one another." 01 J L.3 L r- Eira-U.3 CA L7 0 L1301 _I

tmot-: F.,1] HE nn E 0 ;71 !---1 norm :71

rr_rx1 t ) a L7 r....3 CI L7 L7 I7 Li 0 L7 L7 t_1 C-2M Mr= C3 01=1=1

LI =MI= 0 i L=I 0001=1 0 t=1 ELI CD C_-IA u no [DM C7 11 0 000 0 CD0 0 0 0 0 0 0 E-J1 L-3 10 0 0 0 C7 0 00 0 L=7O1:::) 000 0 000 ogoo0 o cm C1=3 1=1 0 0 0000 00 00 000 no 0 Gr---r-i 4 0000 0 0 0 0 000 mg=, 0000ogonn Figure 78 Plan of one of the podium floors. Figure 79 Chief clerks office. class insurance office. We have already made use of the model in our advertising material and hope to extend The practice in the old office buildings, by which chief this when the building is completed. We believe that clerks had raised platforms approximately 10 ft square these new premises will have an impact upon our staff, in size which were screened but not separated-offas including the senior staff, by giving them a feeling of offices has been carried over to the new building, where pride.' they have an area 2 x 2 modules in size (approximately One of the reasons for the original decision to use open 100 sq.ft) rimed above the floor to a height of approxi- planning was said to be that there is a work-flow in in- mately 10 inches and surrounded on three sides by glass surance just like that of industry. Examples were given screens. The reasons given for the platform are better of the processing of a new policy and the settling of a provision for the supervision of staff, a degree of privacy claim, where many departments would be concerned. behind which clerks can be interviewed and a status The need for flexibility in both work-flow and depart- symbol and prop to the authority of the chief clerks. In mental size, because of changes of work method and fact, as chief clerks are placed on the perimeter of the increasing mechanisation, was underlined. In fact, in building so that they can face inwards towards the staff, many cases the policy does not seem to have been followed consistently and a substantial amount of partitioning has been used. The reasons for enclosing areas and 8. See footnote 9 9. On seeing the draft of this chapter, the management stated separating departments have been sought, for this action that partitions have only been provided in the following cases: was in direct contradiction to the stated policy of open i. private offices for officials, which were always intended planning. In the course of the project it seemed virtually ii. divisions between quite separate departments impossible to determine a particular reason why this was iii. cloakrooms iv. storerooms done or who was mainly responsible: The management, v. interview rooms however, do not necessarily endorse those impressions vi. where isolation was justified on grounds of noise of their partitioning policy formed by the Unit and re- The management also states that no partitioning has been intro- ported in this and the two succeeding paragraphs. ducedto break up large areas ; no partitioning has 'just crept A number of explanations has been given. One senior into the plans' 10. The management emphasises that all supervisors donot have executive said that he had tried to imagine how young- separately screened sections 82 some of them have difficulty in supervisingbecause of the reflections There were certain eatingdistinctions in the uldprem- on the glass of their partitions from the ises; a staff dining windows behind them. Theycan be easily seen but can- room for the rank and file employees, not easily see! a dining room for chief clerks andjunior officials, anda separate dining room for senior Totally enclosed private officeswere not envisaged at all ,fficials. The newar- in the early stages of rangement consists of threerestaurants on the twenty- planning but the strong objections fourth floor and put forward by the senior employees a cafeteria in the basement. Of therest- brought about a aurants, one is for visitors,one for all grades of official change in policy. " Thesystem of private officeswas simplified by having only and another for chief clerks.Junior officials and chief two sizes, their allocation de- clerks have now moved pending upon seniority, andavailable only to employees apart; as one chief clerk des- with the rank of `official' cribed it, into officers' andsergeants' messes. Rank-and- (ie, executive). Theyare con- file members of the staffeat in the basement cafeteria. structed of floor-to-ceilingpartitioning which follows It the line of the modules forming seems unfortunate that there shouldbe a physicalsep- the planning grid for the aration of twenty-odd floors office space. A clear criterionof the size of private office between the two sets of derives from the company's dining services, for not onlymust this be an unecon- rating of seniority. The rank omical arrangement, but it of 'senior official' entitlesthe holder to a 3x 3 module introduces the possibility of the creation of a substantialpsychological gulf between (approximately 240 sq.ft) office; therank of `official' en- management and workers. titles its holder toa 3 x 2 module (approximately 160 sq.ft) office. The criterion isa clearly understood one, and has the rationale that The physical environment senior officials aremore freq- Spatial uently liable to hold conferencesand receive visitors in The use of a plan-module forthe office space has already

Figure *0 Officials dining roomon the 24th floor. Figure 82 The modular basis of the design.

been mentioned; its dimensionis 5ft 2in. in both direct- _ ions, determined,so far as can be seen, quite arbitrarily. It consists of asquare of twenty-five 1 ftsquare floor tiles bounded bya 2 inch wide strip. The ceiling hasthe same overall pattern of 5ft 2in.squares, but it has a marked directional character,for the finish consists of ribbed metal pans witha recessed 5 ft by 1 ft light fitting in the centre. The total area of officespace (:e, excluding lavatories, lifts, stairs, etc.)on one of the podium floors isabout 35,000 sq.ft; on one of thetower floors about 7,500 sq.ft. Gross areas per clerk (ie,including circulation,coat hanging space, etc.)vary from about 77 sq.ft in themore densely occupied podiumfloors to 83 sq.ft andupwards in the tower block. Netareas per clerk (ie, the actual desk and chair space and immediate circulation)range upwards from about 45 sq.ft.

Visual The outer walls of theCIS are clad withan elegantly detailed system of curtainwalling. Glazing extends the top of the perimeter from _ heater outlet (which isonly 9 Figure $1 inches high) to the ceiling,and there are Basement cafeteria. no opening lights. Although the`window' area isso extensive, it could not be sufficientto provide an adequateamount of daylight for working in the their offices. The occupantsof the 3 x 2 moduleoffice inner areas of either podi- were given the choice of three completedecors but were not able to take part of 11. The management has one and part of another. Senior explained that theoccupants of the officials had a greaterrange of choice. private offices had some influenceupon the decision to change from partial screening to full-heightscreening 83 necessary and filtered by re-circulation plants in theser- vice tower, the air being extracted and suppliedthrough dual-purpose light fittings. Because of the lightweight construction, the buildinghas a low thermal capacity and therefore an elaboratecon- trol system is used. The secondarywater and the air distribution systems in the main towerare arranged in several zones according to height andaspect. The temperature of the water supplied to the units is adjusted for each individual zone to suit the outdoor conditions. The podium also has perimeter inductionunits, but because of the very large floor areas additional air supply is necessary in the central workingareas. This is provided through dual purpose light fittings anda high velocity distribution ductwork. The air isre-circulated in these areas through the light fittings, alternatefittings being used for extraction. All boilers, refrigerators, pumps and otherair conditioning equipment are controlled froma central panel in the boiler house. Surveys of the thermal conditions in thisbuilding were made during both winter andsummer. '4 Figure 83 The glass curtain wall.

um or tower. This was clearly recognisedat the design stage by both architects and owners and the electric lighting installation is used constantly andwithout re- straint. The windows in the tower providemagnificent views and adjustable vertical white blinds,which are both attractive and efficient, are usedto provide protection from sun or sky glare. Levels of daylight in this building havenot been measured, for they 'would be largely meaningless.Most, if not all, of the 19 tower office floors receive daylightfrom a virtually unobstructed sky, but much of thepodium, in which rather more than half the total numberof staff work, is obstructed by nearby buildings. The electric lighting installation consists ofone 5ft 65 watt warm daylight' fluorescent tube per 5ft 2in.square module mounted within a fitting whose plastic louvred under-surface is flush with the ceiling. Measurements of the illumination at night 12 have shown the typicalrange of illumination on the working surface of desksto be between 24 and 30 lm. /sq.ft with amean of 28 lm. /sq.ft. The ratio of minimum to maximum illuminationis 0.8, so that both the value of the illumination and its consist- Figure 84 The central control in the basement for thethermal environment ency are high - though the working illumination is not and mechanical services. so high as the value of 36 Im./sq ft which was intended. It has only been possible to estimate the glareindex; it There were two surveys ofwinter thermal conditions.One seems, however, that the value achieved is lower (ie, was in January 1963, during a period of continuous better) than the recommended 19. frost. On this occasion,a number of measurements were The reflectivities of the ceiling and desktops are fairly taken on the eighth floor, which is completelyopen and high (75 and 32 percent respectively) but thefloor is very without any private offices,over a period of 2 days. The dark (8 percent) and becauseso little light is reflected number of occupants at the time of thesurvey was about back to the ceiling, this surfaceappears dark too and, seventy-five. The arrangement of the deskswas such that especially in some areas of the podium, givesa rather no person was seated nearer than 4 ftto the perimeter gloomy impression of the interior. heating and airconditioning units.The globe thermometers which marked the measuring stationswere distri- Thermal buted over the occupiedarea, none being nearer to the The design criteria for the heating, ventilationand air- windows than 4 ft. conditioning of the office areas were: i" During the two mornings of thesurvey external con- Winter 70°F air temperature ditions were very similar, as indeed theywere inside the Summer 75°F air temperature building. There was a slow rise in themean globe temp- Relative humidity 50 percent at all seasons erature on both days and, as might be expectedfor peri- Fresh air changes per hour 2.0 to 3.9 (a minimumof ods without direct sunshine, the maximumglobe temp- 3.1 in the podium) erature was recorded at a stationnear the centre of the The airconditi' ing for a typicaltower block floor room, the minimum on the northern side of the building. works in the 1., owing way. Primary air,discharged During the afternoon, and due tosome sunshine, globe through nozzles within induction units situatedon the temperatures at two positions on the south-westernface perimeter of the floor, inducesa flow of room air of the building increased to above75°F. In comparing through a re-circulation grille at the front ofthe unit. The re-circulated air passes througha coarse filter and over a coil (which is either heated or cooled) before mix- 12. See chapter 6 and appendix 6.2 ing with the primary air and being dischargedinto the 13. Anon: Air conditioning for theCo-operative Insurance buil- room through the top grille. ding, Manchester. Heating, 1962 (September) The air from the central part of each floor is cooled if 14. See chapter 7 and appendices 7.1 and7.2 for a brief account of these and other surveys of thermal conditionsin offices 84 the results obtained in this office with the comfort ranges Similar surveys of summer thermal conditionswere made, quoted in chapter 7, it was found that : again on the eighth floor, duringthe period 11th to 13th (1) the relative humidities were all well within the June 1963. A maximum shade airtemperature of 80°F range 30 to 70 percent was recorded at the nearby meteorological stationon (ii) the rate of air movement seldom rose above what June 11th; maximumCET inside the building at posit- is considered to be the upper limit for the comfortof ionsnot in direct sunshine was ashigh as 72°F. Globe people doing sedentary work. Values of 40 ft/min. or temperatures of over 80°F were recorded at onlytwo above were recorded for only 6 percent of the total positions. These occurred between 1300 and 1420hours readings, and none were above 43 ft/min. Theaverage at a position near the southern corner and, after 1600 values of air movement wereabout 23 ft/min hours, at a position on the westerncorner where, because (iii) the mean globe temperatures varied throughout the blinds had not been drawn, thesun had shone dir- the two days by about 2 deg F; the variations of the globe ectly onto the globe thermometers. temperatures at individual positions were similar to The next day was not quiteso hot nor was the sunshine the variations of the means except when thesun shone so prolonged, but external air temperatureswere above brightly. Horizontal temperature gradientswere small 70°F for the greater part of the working day.The mean except during periods of bright sunshine globe temperatures varied between 74.0 and 75.2°F; (iv) in contrast with the results ofsome other sur- the highest ding was 80°F. veys,air temperatures in this office (mean values of On the 131June there was an abrupt change for the between 73 and 74°F) were high worse in the weather, but globe temperatureswere only (v) equivalenttemperatures were generally within the a few degrees lower than on the pre ious days. Air range reported as being suitable for normally clothed temperatures varied between 73°F on the llth to 70'F sedentary adults, of 66 to 72°F. " Theupper limits of on the 13th. Relative humidity was stable at 50 to 60

88 96 86

92 82 90 eo el 78 es 76 max globe temperature 84 82 72 mean globe max globe temperature so 70 .... temperature min globe 78 se "Mean globe temperature 76 _ _ temperature 66 gxtgrnpl dry 64 72 ouw temp 62 79 min e 60 -Umlauting 6d se ss 56

62 52 60 50 1.0 0404 0.5 0 93 '0.70.9 sunshine hours 46 0900 1100 1300 1500 1700

TIME Hours BST (614T.1 dour] Figure 86 3 Globe temperatures on the eighth floor during 12June 1963. external bulb 'temperature percent. The rate of air movement varied between 12 and 40 feet per minute with an 32 ---V average of 23 ft/min. 30 Aural Za Measurements of noise level were madeon the ground, 26 second, fourth, eighth, and twenty-third floors,and included four large general offices, two machinerooms, a conference room, a private office and the staffcafeteria. 20 The noise in the occupied general officesoriginated 0 0 0 0 0 0 0 0 sunshine mainly within the rooms. In the large generaloffices on hours the second floor, ie, the largest andmost open space in 0900 1100 1300 1500 1700 the whole building, the noise levels compliedwith Beranek's criteria' and the highest readingwas only T:ME Hours 6MT 2 dBA above the Wilson Committee'srecommendation'" of 55 dBA. Although the noise levels inthe machine Figure 85 Globe temperatures on the eighth floor during 15 January 1963. the comfort ranges were exceeded for short periods 15. See, for example, Black, F. W: Desirabletemperatures in offices: a study of occupant reaction to the ho...togprovided. only when there was direct sunshine, and this could J Inst Heat Vent Engrs. I554 (November) have been avoided if the venetian blinds had been 16. Winston, N. S: Comfort at work. J inst HeatVent Engrs. drawn more often 1953, 21 (215) 141-4 (vi) the vertical temperature gradientswere small, 17. See chapter 8 18. Committee on. the Problem of Noise: Noise:final report. none exceeding 3 deg F between floor and ceiling. Cmnd 2056. HMSO, 1963 (July)

35 . ..

I 70 - 111114, 10 1111," Mlillii!!!!!IPI.IiiIiIIIIIIIIII ____4

60 60

50 50 Nela

40 40

4.11111%,",NC=35 30 30

20 20

63 125 250 500 1000 2000 4000 8000 63 125 250 500 1000 2000 4000 5000 FREQUENCY CYCLES PER SECOND FREQUENCY Figure 87 CYCLES PER SECOND Figure 89 Noise spectrum in unoccupiedconference room on twenty-third floor (43-44 dBA). Noise spectrum in card punchingroom where twenty-eight machines were working (70-74dBA).

10 TOriii, "1"11111,

60 1,11 II 60 111,. 1 I 50 --NtersO 50

30 30

20 20

63 125 250 500 1000 2000 4000 0000 63 125 250 500 1000 2000 4000 5000 FREQUENCY CYCLES PER SECOND FREQUENCY CYCLES PER SECOND Figure 88 Figure 90 Noise spectrum in open-plangeneral office on the second of the podium (54-57 dBA). floor Noise spectrum in staff cafeteriaat 1300 hrs. (72-73 dRA). have points of rooms exceeded Beranek's criteria,and were of the order common reference and bemore at ease in of 75 dBA, they familiar company. Theage range was from fifteenyears were not unreasonably high bycomp- old (female juniors) arison with similarmeasurements in machinerooms in to elderly (male sectionclerks). other buildings. The experimenter playedthe dual role ofinstigator and In private offices, conversation chairman. A number of and movement couldbe general questions aboutthe new heard from the adjoining environment were put to eachgroup but the direction of working areas. the discussion was allowed to follow the interestsof the Preliminary interviews with thestaff participants andwas not re-directed unlessthe dis- An important requirement cussion became irrelevantor a mere catalogue of of a study of officeenviron- plaints. comment is a specification ofthose features ina building which are influential in What was being soughtfrom the meetings determining a person'ssubject- was not an ive reactions to his workplace. appraisal of the buildingbut an estimate ofthe features In order to obtainthis that were subjectively of representatives of the rank-and-fileclerical workerswere greatest importance, forthe interviewed, and their views interviews were to actas a guide to the questions on any aspect of the environ- be asked in questionnaires. later to ment that they felt to beimportant to themwere sought. Obvious features suchas The interviewswere completely free-ranging, heating, lighting andventilation were raised, though with more general questions together the experimenter made about liking thenew build- sure that a number ofspec- ing. The following is ific topics were introduced. a summary of the sort ofattitudes Eight discussionswere held, on each occasion commonly expressed andthe quotations, with have been selected where used, groups of six plus the interviewer.The composition of because they were typical. the differentgroups varied considerably in Because the interviewerhad to start the respect of a number of specific questions discussion, age, sex, occupation and status, butthe composition of was asked about the a particular discussion group airconditioning inductionunits which, to the staff, was usually fairly homo- resent the heating installation. rep- geneous. The participantswere always drawn from the The first transcription same department so that the members comes from a single interviewwith a group of typists. of a group should was typical of the It 86 response to the system in theearly

.11 Figure 91 Interview with a group of clerical workers.

days of occupation when teething troubleswere being senior management, however, has disclaimedany know- sorted out, and it also suggests that position in theroom ledge of such a directive and pointed out that only was an important variable, a view supported by other negligible damage to the office floors has occurred,al- testimony. though stilletto heels are now freely worn. (Interviewer): 'What about the airconditioner: doyou Ceiling height was treated more seriously. It emerged find it to your liking?' many times that those who found it too low were those ' No, we don't.' (loud agreement) who either sat in the middle of the roomor who sat fac- I've got cramp up my arm and up my back'. ing inwards. Many used the epithets' oppressiveor de- ' It isn't constant: you perhaps go in in the morning and pressing' to describe the effect of a low ceiling. it's really warm, so off comes your cardigan and you're The colour scheme was a topic which provokeda good really warm. You've been in aboutan hour or so and deal of discussion and disagreement. Itwas not possible then it goes cold and you start getting a terrible draught to form any general impression as the range of agree- on your feet, and this week we've noticed it as though ment and disagreement was very great and did notap- it's going up to the ceiling and then hitting youon the pear to be associated with age, sex, place where the back of the neck'. respondent worked, or indeed anything else. (Interviewer, to girl who had not agreed with others about Lighting, both daylight and electric, is one of themost draughts): important components of the internal environment. Yet, `How about you, do you get draughts?' surprisingly, it was not a topic which created muchcom- I don't know, and yet the girls on the next row of desks ment or feeling. The following was typical of many say they do. We don't notice it much where I am'. discussions: Also germane to the discussion of the airconditioning (Interviewer): 'What about the lighting, is this adequate?' system were questions of stuffiness and odours. Odours 'Oh, the lighting doesn't bother me personally'. did not seem to present nuch of a problem, but the Ì like the way it's covered up- you don't get the glare number of reports ot stuffiness raised the question from it'. whether this might be due to the individual's working (Interviewer): 'Oh, I see, you're talking about the fluores- position on a floor, or the density of occupation of that cent lights; I was thinking about the illumination ingen- floor. eral'. The cleanliness of the new building was a question which Òh, I think it's quite light enough'. was also relevant as the building is effectively sealed and Ì've got no complaints about the lighting'. the entire intake of air is filtered before being circulated. (Interviewer): What about glare from the windows?' A girl commented : 'Here you keep nice and clean, where- No, no we don't get any of that'. as before you were never clean. Your underclothes were We haven't had anything like that'. filthy, and your body was filthy with the dirtyou were 'The only thing is on the other side of the building, ifit's kicking up, but here the atmosphere is very, very clean'. sunny, you get the sun. But then, of course, you've got Irritation with the noise made by the ventilating system the venetian blinds and you can just adjust themto keep was sometimes mentioned, as was the noise of the mech- the light off your face without making it dark'. anical document conveyor, but most people commented However, a view through the windowswas felt to be of on the comparative quiet of the new building. This was great importance; not necessarily a pleasant view but perhaps not surprising as the former head office stood merely the opportunity to see out. The followingarose beside a main line railway station. when the colour scheme (not the fenestration)was under Floor treatment was a topic in which almost noone was discussion : interested (possibly because it is a good surface) and, in The windows here are marvellous because, before,we itself, provoked no comment. There was a number of were caged in by bars and windows that you couldn't complaints from women that they were being made to look out of at all. So the windows, lighting andpleasant wear flat heels so that they did not damage it. The surroundings here are a great boon.' 87

AIMMOMMINIIIIIIIIMIII11011111111111 Well, in the roomwe were in before it was just four walls with windows large offices, which itwas felt, foster a passion forsym- up in the roof.' metrical layouts and obsessional tidiness. You didn't know what itwas doing outside unless it Section clerks, used to more autonomy and privacy,commented on the was brilliant sunshine beating downon the back of your question whether they would rather neck (laughter). Butapart from that you didn't know have partitioning what the weather between sections: was like; whether it was rainyor 'Oh, yes, partitioning is important. foggy, or quite nice. Reallyyou had no idea.' It's very distracting having other sections aroundyou, maybe having dis- The feeling that it was importantto be able to see what cussions or something'. was going on outside wasvery widely held. The opposite view was never strenuously Say something has gonewrong on your section, you put in discussion, althougha can't tell your people off in private small proportion of people feltthat it was a question of without every other very little consequence. section knowing about it'. Of the specific features (Interviewer): 'Do you find it easieror more difficult to discussed, the one whichocc- supervise what's going on?' upied the greatest amount oftime was the lavatory Well, I don't say it's facilities. Men were much more awkward but, if something less interested in the toilet goes wrong, they've all got to know'. accommodation than women, butnevertheless they Typical comments from rank-and-fileclerical workers rated them highon their list of priorities of office de- were: sign. A number of specific criticisms were made of the You feel guilty when you get up'. new lavatories (even though theyare of a standard very and much higher than is usual), mainly in terms of over- You feel self-conscious ifyou have to walk across the crowding, overheating, washbasin design, and the lack of space for standing handbags. room to the front of one of the cabinets.You walk back The men's concernwas across the room again and you feelI don't suppose more in terms of privacy, cleanliness,and number of you are fittings. By contrast, the emphasis but you feel that you're beingwatched'. of the womens' in- Perhaps most commonly of all though: terest was on the attractiveness of thetoilets. One girl You feel they've gotyou where they can keep an eyeon commented about the lavatoriesin the previous office: you'. It was a horrible place andmade you feel contaminated At a more general level, by going in there, and itwas very gloomy and depress- many people felt that they ing'. would prefer to work ina smaller group: You get a more friendly atmospherein a small place'. Other typical commentswere: There was a great deal of indirect `Nice toilets seem to addto your personal comfort criticism of the somehow'. new building in people saying that they hadbeen much and happier in the previous office, despitethe poorer physical conditions. 20 It plays a big part inyour working day.' Toilets are also places which These comments, ofcourse, are no more than impress- women connect with their ionistic, and it must be emphasised, personal attractiveness, placesto appraise themselves in were obtained soon after the move into thenew building and at a time when a mirror and to make-up. They stressed theneed for nice many of the staff may have been feelinguprooted from colours and elegant fittingsand felt that thesespaces a way of life to which they had should be at least as wellappointed as the rest of the grown accustomed. They building. la do, however, providean overall view of the attitudes of the staff towards theirmanagement and their new build- Apart from the discussion of thephysical characteristics of the building, ing, at that time. The questionswere not always the a good deal of time was devotedto in- same, and the emphasis of interest varied tangibles, like the effects of thenew building on such with every matters as personal relationships, group. Their value was not intrinsic,but consisted work attitudes, and rather in providing an exploratory morale. Such discussions largely and descriptive back- arose from the pro- ground to a more comprehensivestudy. The impress- bing of attitudes to thebuilding in general, to tower ions gained from them block design, and to theopen planning of the interior. were subsequently used in the A large proportion of those planning of questionnaires andmore systematic surveys, people interviewed were experiments, and programmes of proud of working in this talland very impressive build- interviewing. ing. One clerk expressed it thisway: The questionnairesurveys of environmental conditions I feel very different whenanybody says to me, " Where do you work ?" and I The first questionnaire say "at the CIS" and theysay Having learnt, from interviews, "Oh, that big place". You feel something of the stairs different'. response to their building, the next I don't know. You don'tfeel as if you're step was to examine someone but these attitudes in amore specific way and on a larger you see they seem interested. I don'tknow, you just feel different with working somewheredifferent... I don't know what it is, but I thinkmost of us feel that way'. Although no one confessed to being personally afraid of 19. It is interesting to note thatthe standard of care taken by the height, many peoplecommented that therewere employees in the toilets has improvedout of all recognition. A plenty of others (alwayssupposed to be female), who 5 pm inspection of all male andfemale toilets showed themto be were. Several people said that they wouldbe very un- immaculate: this was certainlynot the experience in the old happy if the horizontal members buildings. It seems, therefore,that the provision of higher amenity of the windows were levels has resulted in muchbetter behaviour and treatment of not there. property the The discussion of thenew building often led to consider- 20. These comments stimulatedan interesting remark from the able talk about staff/management management, which is reproduced: ' This is relations. Many and, in my experience, a strange reaction people felt that these had sufferedwith the move into the true but possibly transitional. I havecon- cluded that the more perfectelle makes the environmentalcon- new building. There was a feeling thatthe rank and file ditions the more theoccupants will complain about deviations were being regimented. from the ideal. An example ofthis occurs during hot weather `It's more like thearmy now'. conditions when the occupantsreadily complain if thetemp- and erature inside the building reaches74°F when a telephone callto neighbouring non-airconditionedbuildings shows that temp- `There are too many regulations;it was very easygoing eratures of 80°F are being suffered without before'. of spoiled children complaint. The simile comes to mind and if the criterion isa happy This appeared to be due, in atmosphere and not an efficientand healthy one then the C.I.S. part, to the management and building may not be the right supervisors' zeal to keep everythingcrisp and tidy; to answer. For this reason social amenities become very important.Nothing has been said of ban any personal effects, andto insist on desks being social amenities in the our tidied every night. The report, and it may be that theyare an new, more autocratic form of essential integral part inmodern design to counteractthe supervision was regarded " regimentation," the " regulations" and the " coldness " referred as being a consequence of the to in the report' 88 -

scale. The method adopted was to circulatea quest- conception of the building is basically an open plan. ionnaire to all the 2,500 employees. the replies to which The fact that nearly half the staff said they preferreda were then analysed statistically. smaller partitioned area, must constitutean implicit Perhaps the first thing a building owner and architect criticism of the type of accommodation provided. The would want to know from such an enquiry is whethera question of optimum size for workingspaces has been building had been produced which was satisfactory for the subject of further study. the greatest number of people. The best indication of Summarising the findings of a broad analysis of the first this is provided by the response to one of the groups of questionnaire, it was found that the buildingwas gen- questions: more than 80 percent found their work no erally regarded with favour by the majority ofoccu- less enjoyable in the new building than it had been in the pants, in spite of strong criticism of a number of feat- old and three-quarters of these actually stated that their ures, notably the lift service and the airconditioning work had becomemoreenjoyable. 73 percent of the staff (both of which have been improved since) and the colour said that their personal relationships with their coll- schemes. eagues were either unaffected or even more friendly than Without going into the detailed analysis of theanswers previously (the remaining 27 percent found theirper- to the questions it is possible to make some general sonal relationships less friendly). Only a small number comments about the pattern of the results. The position disliked working in a tall office block ; 75 percentpre- or aspect of the building at which the individual works ferred the new building to the old. Taking the building did not seem to have a very great influence upon theway as a whole, there is little doubt, then, that it is well-liked he responded to his environment. It may be, ofcourse, by its occupants. that there were marked differences in particular cases, The teething troubles with the airconditioning system but it did not seem that this variable had much general have already been mentioned ; the first questionnaire influence. There was a significant difference in theans- showed that the uncomfortable thermal conditions wers of men and women to nearly all the questions, and which were experienced in the early days of the build- this is a finding of considerable importance. ing's occupation weighed very heavily with the staff. The influence of floor was very marked indeed. It was Lifts, too, were initially troublesome, and a majority of not surprising to find that the ground and first to fourth the users were critical of their functioning. Nearly 40 floors in the podium produced consistently significant percent of the staff considered that the toilet accommod- results. This might have been predicted from the ob- ation was inadequate in some degree. vious physical differences between them and the tower In spite of the impressions given by the interviews, the floors, although it would be hard to specify apriori(or ceiling height was generally approved, and over 80per- from the physical measurements which have been made) cent found the electric lighting satisfactory. Very few the causes of such marked differences. Significant differ- people were aware of any glare from the windows. ences between most of the floors were found to exist but 65 percent never found their office distractingly noisy. most such differences are explicable in terms of their The colour scheme, however, came in for a great deal of manner of use and type of staff. For example, some criticism; it was the characteristic of the environment floors are predominantly typing pools or machine rooms, which the greatest number of occupants felt might be others house such specialists as solicitors in private improved. offices. Asked to check a list of words they considered descript- The younger workers have a different view of their ive of their building, the majority of the staff expressed surroundings from their elders, and this is not surpris- themselves in terms which were, no doubt, the guiding ing, particularly as the building is so much a product of aims of both owners and architects: modern, light,com- contemporary thought. It would be interesting to trace fortable, pleasant, friendly and efficient. further the direction of their opinion andsee whether There was no significant change, for better orworse, in they are the most satisfied or the hardest critics of the the number of people suffe. ing from headaches or eye- design. This is something which is not easy toguess for strain. A question was asked about attitudes to the new it depends upon their expectations and these,presum- staggered working hours which were made necessary in ably, are less influenced by experience than those of the order to relieve the load of a concentration of staff on older groups. both the lift service and public transport system. The Of the male grades only the officials stand out. The psychological interest of the question lay in whether the reasons may be in terms of their general level of satis- move into the new building had created any substantial faction with the amenities they and their departments measure of discontent and, if it had, whether it was in- enjoy, or it may reflect vested interest:many officials fluencing people's general satisfaction with thenew were personally involved with some stages of the plann- building. In fact, less than one-fifth expressedany sort of ing and layout and they were all given private offices disapproval. with a choice of carpet and colour schemes. Their physi- The remaining questions differed from the others in that cal conditions are, therefore, completely different from they did not call for a direct assessment of the actual en- those of all other staff. Finally, the very fact of being ina vironment, but instead were concerned with the individ- small office implies a different experience of the air- ual's assessment of the ideal environment. Asked wheth- conditioning unit which was the source of much er they felt it important to be able to see out of the office, general criticism. 24 even if there was plenty of electric light to see by, the response was overwhelmingly that it was important. The second questionnaire Another question asked whether it was as good for the The first questionnaire enquiring into staff reactionsto eyes to work by electric light as by daylight: consistent- the physical environment was circulated in October ly, the great majority felt it was better for theireyes to 1962, shortly after the move into the new building. A have daylight to work by. It must, however, be noted follow-up questionnaire to a sample of the total staff that though these questions are complementary in some was circulated in August 1963, with the object of seeing respects, the underlying attitudes and beliefs upon which what changes in attitude had occurred in the interven- they depend may be very different. For this reason, they ing ten months." were subsequently made the subject of a specific experimental study which has already been reported in chapter 6 (the visual environment). 21. In October 1962 22. See appendix 10.1 for the first questionnaire A final question asked in what sort of office the respond- 23. This study was in progress while this report was beingpre- ent would prefer to work, and offered three alternatives. pared. See appendix 5.1 28 percent opted for a large open one, 45 percent fora 24. For a more detailed analysis of the response to this question- smaller partitioned area, and the remainder stated that naire see:Wells, B. W. P.:Office design and the office worker. PhD thesis, University of Liverpool, 1964 they had no preference. The result is interesting, for the 25. See appendix 10.2 ;..r the second questionnaire 89 MI.

A comparison of the answers showed that, in most ways, the building had decreased as a subject of interest. The one significant change was in the staff's attitude to the thermal environment. Temperatures were more steady and fewer people reported unpleasant draughts, but there were more complaints of uncomfortably high temperatures and the air was thought to be less fresh.'" Some change of feeling could be expected for, at the time the building was occupied (and the first questionnaire circulated) it was generally agreed that the airconditioning system was working at well below its operational standard. However, it is clear that the changes in opinion were by no means entirely favourable.'

Overall assessment of the environment A judgement of the success or failure of the total environment within a building necessarily takes account of all the contributors: owners, users, design team, con- structors, maintenance staffand of all the contributory factors: economic, physical, social, and psychological. At present there is no way of attaching a meaningful weighting to any of these and appraisals of the total environment can only be made on an individual basis and by comparison with other buildings. No criteria exist. This is undoubtedly the finest office environment in which the Pilkington Research Unit has worked. It is the most thoughtful, and of all the buildings in the Sur- vey Sample it is the best designed,test detailed, and best built. It is among the best kept, but certainly the most expensive. The CIS building is a major advance on the `routine' office block and, without doubt, among the two or three most beautiful office buildings in the country. As a functional working tool it is almost certainly satisfactory, but it has an additional business justification, for it is the outward and visible sign of an active and determined company: forward looking, adventurous and successful. The building provides an excellent physical environment for the staff, incomparably better than the conditions which existed in their former building. Whether it provides a better social environment is more open to doubt, but only because this is not a fixed thing, and there are as yet no means of measuring it with the degree of accuracy which can be applied to space, light, heat and sound. The Unit's general impression is that, on the whole, and now that the management's initial zeal for order and discipline has relaxed, the building is a considerable success. It has attracted many more applicants for employment with the Society, and is something of which the majority of the staff are proud.

26. The management comments that it has agreed with the workers' union (the GuildofInsurance Officials) that the average temperature maintained within the building should be 72°F, and that this caters for the majority of the staff, who are lightly clad females. As a result, many males mildly complain' of the heat 27. For a more detailed analysis of the response to this second questionnaire see: Wells, B. W. P: PhD thesis, op cit (24) Stage Four

Appendices Appendix 1.1 Further studiesseen to be necessary

Many aspects of office design have been surveyed during 15. Inter-action of environmental factors: technical and this investigation and new studies have been initiated in economic Consideration of effects uponone environ- certain fields where there was a lack of information. mental factor of variations in design of other factors,eg, There is, however, much work still remaining.The technical and economic consequences fora heating in- following are some of the many research topics which stallation of variations in daylight factor,use of double deserve further study. glazing, varying illumination from electric light,etc. I. Sizes of office occupancies Determination of thefreq- 16. Inter-action of environmental factors: psychological uency distribution by number of staff and/orarea of Examination of the degree to which one environmental office space occupied by the typical officeoccupier', in- factor affects the subjective experience of another. cluding both owner-occupiers and tenants. 2. Types of office occupancy Examination of thetypes of business which use office space (perhaps by Ordersof the Standard Industrial Classification') to ascertain the principal uses to which office space is put. 3. Functional requirements of officespace Study of typical clerical office functions to determine themost common patterns of use of office space. This would include consideration of the work-flow ofpaper, the size of working groups, the functional relationship of different clerical activities, number and frequency of visitors,etc. 4. Planning and constructional modules Considerationof the factors influencing the choice of planning andcon- structional modules to see whether it is possibleto determine a rationally-based system of dimensioningfor office space. 5. The external environment Further consideration of the effects of the external environmentupon the environment within office buildings. 6. Building components and finishes Technical andecon- omic codification of alternative components and finishes available, including external wall construction, windows, floor and ceiling finishes and partitions. 7. Electric lighting Technical, economic and subjective studies of alternative lighting installations providing different standards of quality and illumination. 8. Heating Technical and economic studies ofalter- native forms of space heating, to codify the alternatives available. 9. Building economics Collection of additionaldata re- specting the capital and running costs of officespace. Comparison of annual costs of major design alternatives such as `deep - blocks' and `slabs'. 10. Experimental aesthetics Experimental investigations of the aesthetic qualities of office environments. 11. Behavioural consequences of size of officespace In- vestigation of differences in behaviour and workingeffi- ciency of staff working in small and large officespaces. 12. Noise A comparative study of the distraction caused by noise and other people in large and small offices. 13. Stable environments An examination of stabilised environments to see whether they reduce stimulation and create boredom or restlessness. 14. Windowless environments Technical, economic and psychological study of windowless environments ingen- eral. 1. Central Statistical Office: Standard Industrial Classification. HMSO. Second edition, 1958 92 Appendix 2.1 Schedule ofseparate research tasks

Subsequent to the preparatoryenquiries and planning for this project, a list ofseparate research investigations was drawn up. This appendix schedulesthese investigations, explains theirpurpose, states what was actually achieved and describessome of the problems encountered. Priority A (in the second columnof the schedule)was intended to be applicableto the stage of the projectrep- resented by this report. PriorityB was assigned to investigations which could beundertaken if the duration of the project was extended.

Ref. Priority Type of Description Purpose of Intended no. investigation Staff Work done Problems investigation method of study consultants encountered investigation involved Preliminary

1 A Postal Definition of To provide a By postal enquiryGeographer Postal enquiry numbers and basis for samp- to all local 40 percent of characteristics ling buildings to all county andplanning planning county borough of office space for study authorities authorities constructed in planning auth- were unable England and orities asking to provide the Wales since the for total amountinformation war of office space built since 1945, number of individual projects and certain other matters 2 A Field Preliminary (i) to enlarge Visits to build- Architect None visits to offices Survey Sample ing owners, Investigation in Birmingham and other not undertaken (ii) to look for architects and (forerunner of members of because of regional differ- tenants in the Unit visits to NE ences (if they Birmingham shortage of and SW of exist) in design, time England) and attitude of office staff

Planning and Structure and BuildingEconomics 3 A Theoretical,Cost analysis To provide a possibly (i) ExaminationArchitect Examination of (of construction basis for gen- of published Method of postal costs) Geographercost analyses of eralisations cost analyses presentation about relative offices which of cost analy- (ii) Enquiries to have been pub- and absolute architects of ses and basis costs of different lished inThe of analysis buildings in Architects' constructional Survey Sample tends to differ elements, parti- Journal from building cularly those to building. affecting the en- Too few build- vironment of ings in Survey office buildings Sample to provide sufficient cost data for statistical analysis

93 Ref.Priority Type of Description Purposeof Intended Staff Work done no. Problems investigation investigation method of study consultants encountered investigation involved

4 A Theoretical,Studies of To provide a Elaboration andArchitect None supported costs-in-use of check upon checking of in- Geographer by postal office buildings assumptions of formation about enquiry cost which costs of rent, might be used rates, heating, in economic cleaning, electri- comparisons of city, etc which design alterna- was collected tives during the preliminary visits. Collection of data from building owners about costs of heating installations

5 B TheoreticalComparative To codify the (i) Review of Architect None technical and alternatives literature Physicist economic available to the (ii) Collection Quantity studies of build- designer and comparisonsurveyor ing components of trade and finishes: literature (i) external wall (iii) Comparison construction of costs (ii) windows, (iii) curtain walls (iv) floor finishes (v) ceiling finishes (vi) partitions

6 A TheoreticalEconomics of To provide a Review of Architect, None structure basis for gen- literature, con- possibly eralisations sidering especi- Structural about the effect ally: engineer of structural (i) high v low and Quantity design upon the buildings surveyor design and econ-(ii) wide v omics of office narrow buildings space (iii) alternative bay dimensions

7 A TheoreticalOffice buildings To provide a Review of Architect Review of and fire background of literature literature with information particular ref- about effects of erence to effect fire considera- of fire considerations upon tions on office design of office environment space (omitted)

8 B Theoretical,Maintenance To provide in- (i) Review of Architect Review of possibly and cleaning formation aboutliterature literature field the maintenance(ii) Elaboration and cleaning and checking of liabilities of cost data office buildings collected during for use in theo- preliminary retical analyses visits of design alternatives

Spatial Environment

9 A Field Comparative To provide Measurement of Architect Review of study of use of information offices in use, Geographerliterature space in clerical about space and comparison departments requirements of space utilisa- Study of space and utilisation tion in (say) utilisation, desk and extent to typing pools, plans, etc, in 7 which this is accounts offices, modern office affected by etc, between buildings (by building design different firms T. Takano, and different United Nations types of business Fellow attached (eg, chain store to the Unit) administration and the civil service). Con- sideration of structural bay dimensions, window modules, type and dimensions of desks, etc

94 Ref. Priority Type of Description Purpose of Intended Staff Work done no. investigation investigation Problems method of studyconsultants encountered investigation involved Visual Environment

10 A Field Daylight and To obtain in- Observation of Physicist (i) MeasurementDifficulties of electric light in formation whichuse of day and Psychologistof daylight Use may be used : daylight meas- electric-lighting.Architect factor in two urement al- (a) in the form- Measurement of buildings, one ulation of a ready reported daylight factors, side-lit, the in factory rational policy glass trans- other top-lit for illumination study. missions, reflect- (ii) Measure- Difficulties of design for 'vides, glare ment of illumina-siting external offices, and conditions, etc tion from reference (b) as a check found in use electric light in upon current photo cells in and comparison six buildings large air- BRS,IES with predicted (iii) Examina- conditioned attitudes and values; measure- tion of lighting advocacy of buildings with ment of daylight maintenance fixed glazing. PSALI. factors from unob- practice adoptedQuestionable Study of the structed wind- in six buildings feelings of value of day- ows, wholly (iv) Experi- light illumina- office workers overshadowed mental study of on the import- tion measure- windows and total caviron- ments in office ance of windowsinternal light ment. including buildings in as a source of wells. Measure- visual environ- congested city daylight and/or ment of illumina- ment, in one centres as a means of tion levels from building contact with theelectric lighting (v) Experi- outside world installations and mental studies of comparison with attitudes of office predicted values. workers to their Consideration of visual environ- condition of light ment fittings and measurement of improvement in illumination values resulting from cleaning the fittings. Comp- arison ofcondi- dons in wide and narrow office buildings. Inter- views with office workers in buildings having good and bad daylighting and (if possible) buildings without daylight I I B Theoretical Economics of To provide an (i) Review of Architect None alternative indication of theliterature Lighting lighting designs economic effects(ii) Design of engineer of design alternative Quantity decisions affect- lighting installa-surveyor ing the quality tions for given of the electric situations and lighting calculation of their Standard Costs Thermal Environment

12 B TheoreticalTechnical and To codify the (i) Review of Architect None economic study alternatives literature Heating and of alternative available to the (ii) Design of ventilating forms of space designer alternative heat-engineer heating ing installationsQuantity for given situa- surveyor tions and calculation of their Standard Costs 13 B TheoreticalConsideration of To assess magni-(1) Review of Physicist None technical and tude of such literature Architect economic effectseffects and, if (ii) Design of Heating and upon heating worthwhile, to installations for ventilating installations of provide data or particular situa-engineer variations in to indicate tions and calcu- Quantity other design trends for con- lations of surveyor factors, eg, sideration duringStandard Costs (i) daylight design process factor (ii) double glazing (iii) solar radiation (iv) illumination levels of electric lighting

ti Ref.Priority Type of Description Purpose of Intended Staff Work done Problems no. investigation investigation method of study consultants encountered investigation involved

14 A Field Thermal condi- To provide in- (i) MeasurementPhysicist tit Measurement tions in winter formation about :of thermal con- of therms! ,:on- (i) performanceditions in office ditions during of heating buildings (hori- winter within the installations zontal and verti- offices of six (ii) thermal re- cal temperature organisations quirements of gradients, and (five buildings) office workers calculation of (ii) Experi- comfort indices) mental study of and comparison the total en- with office work- vironment, in- ers' reactions cluding the (ii) Comparison thermal environ- of thermal con- ment in one ditions in large building private and open offices

15 B Field Draughts, in- To seek inform-litInterviews Physicist Experimental cluding body ation about one(ii) Physical study of total heat loss by of the most measurement environment, radiation to coldcommon causes including surfaces and of complaint sensation of windows about environ- draughts, in one mental condi- building tions in offices

16 A Field Thermal To assess therm-(i) Summer Physicist Measurements of conditions in al conditions in thermal surveys,Psychologistthermal condi- summer offices exposed calculation of tions during hot to sunshine, andthermal comfort days in three to compare withindices and :Atildings external climatecorrelation with meteorological data (ii) Comparison with office workers' reactions

Acoustic Environment

17 A Field Noise w ithin To provide in- (i) Measurement Physicist (i) Measurements offices formation aboutof external of noise levels in noise levels noises, and nine buildings within offices reduction pro- (ii) Experi- vided by building mental studies of construction total environ- (ii) Measurement ment in one of noise created building, in- within offices, cluding staff (typing, tele- response to their phoning, etc) aural environment

Social Environment

18 A Lab/field Comparative To establish Structured inter-Psychologist Comparative Insufficient study of the degree of con- views, question- study of beliefs time within the attitudes of sonance naires, paired and attitudes ofproject to managements, comparisons of office managers,await suitable architects and different office supervisors, andopportunities clerical staff to layouts, attitude clerical staff to of studying environmental scales the optimum architects' priorities and size of clerical attitudes standards, working areas (owner-occupied buildings)

19 A Lab/field Social relation- To provide dataSociometry Psychologist Study of socio- ships within of social, psycho- preferential links different sizes oflogical and within and be- office space management tween sections interest on the sharing very influence of large areas and spatial relation- those in separate ships ones

41.11111.1111".""'- Appendix 3.1 Constructioncosts of office buildings

Although the economics ofa particular building project must eventually 'oe basedon the annual costs, a matter of immediate cor,-:ern will bethe possible level ofcon- struction costs. The Architects'Journal has published a number of cost analyses ofoffice buildings, anda simplificotion of these is givenin table 8. Table 8 Cost analysis of 13 officebuildings, abstracted from cost analyses published inThe Architects' Journal.

Grouped Elements Cost in shillings/sq.ft Percent of total cost (normalised to November1963)

Mean Min. Max. Mean Min. Max. Preliminaries, insurance, contingencies 6.5 1.4 11.7 7.3 1.6 11.9 Work below ground floorlevel, drainage 6.7 2.8 11.8 7.1 2.5 9.9 Structural frame, structuralupper floors, roof, staircases 21.2 13.4 38.2 22.5 14.3 34.9 External walls, windows, doors 18.5 10.3 31.1 19.7 14.3 27.0 Internal doors,screens and partitions, internal finishes 19.7 13.3 33.8 20.9 13.0 29.7 Heating and ventilating installation 9.9 4.7 21.2 10.6 4.2 20.9 Electrical installation for light and power 5.2 1.8 8.8 5.3 2.7 8.1 Other services, lifts, etc. 6.3 1.9 16.5 6.5 2.1 13.3

No. of buildings in sample= 13 Appendix 4.1 Schedule ofpost-war officespace approved, built and under construction in theareas of certain local planning authorities (Information available at October 1964)

London and Metropolitan Counties Floor areas in sq.ft to nearest 100 sq.ft.

County and Built Under Approved Total Qualifying Location construction not yet approved remarks started (minimum estimate) Essex Not Not Not 791,600 1955 to 1960 (metropolitan parts) available available available Floor space approved in new buildings, rebuildings and extensions (excluding all outline applications and ancillary offices less than 5,000 sq.ft)

Hertfordshire Not Not Not 507,700 1958 to 1960 (central, west, available available available All new buildings east and north) approved

Kent (west of and Not Not Not 995,600 1955 to 1960 including the Medway available available available Permission for towns, MailingRD, offices (including MaidstoneBand major changes of RD, Tonbridge UD use) local offices, andTunbridge Wells) sub-regional offices and offices over 5,000 sq.ft for industry

London central 35,549,300 11,295,300 7,862,300 54,706,900 1948 to 1961 New buildings, rebuildings and extensions Londonrest of 2,085,300 1,751,100 816,400 4,652,800 1955 to 1961 county New buildings, rebuildings and extensions. Middlesex 6,363,700 2,633,500 5,423,700 14,420,900 1955 to 1963 All approved schemes over 5,000 sq.ft Surrey Not Not Not 3,021,100 Mid 1957 to Mid 1962 available available available Approvals Larger County Boroughs (population greater than200,000 in 1961)

County Borough Built Under Approved Total Qualifying construction not yet approved remarks started (minimum estimate) Birmingham 2,250,000 1,152,300 350,000 3,752,300 1951 to 1962, central area, major projects Liverpool 1,584,000 587,500 Not 2,171,500 1951 to 1962, central available area, all projects Manchester 1,876,800 Not Not 1,876,800 Post-war, central available available area, major projects Sheffield 265,300 159,900 Not 418,200 1956 to Sept 1962, available central area, major projects Leeds 600,000 Not Not 600,000 1948 to 1962, central available available area, major projects Bristol 1,879,500 280,000 1,003,200 3,162,700 Post-war to end 1963, central area, all projects Nottingham 170,300 177,900 313,500 661,700 Post-war, whole area, all projects Kingston- upon -Hull 130,600 Not Not 130,600 1948 to 1960, central available available area, all projects Bradford 281,000 Not Not 281,000 Post-war, whole available available area, all projects Newcastle-upon-Tyne est. 250,000 In 'built' Not 250,000 1949 to 1961, central available area, all projects Leicester 216,600 in 'built' 106,000 322,600 1950 to 1962, whole area, major projects. Stoke-on-Trent 52,400 44,200 70,000 166,600 Post-war, central area, major projects Coventry 173,000 124,000 Not 297,000 Post-war, whole available area, major projects Croydon 1,238,900 1,647,200 3,772,900 6,659,000 Post-war to end 1963, whole area, major projects Cardiff 807,500 Not Not 807,500 1956 to 1962, whole available available area, all projects Portsmouth Not 67,000 200,000 267,000 Post-war, central available area, major projects Plymouth 378,800 Not 30,300 409,100 Post-war, central available area, all projects Southampton 247,400 Not Not 247,400 1957 to 1962, whole available available area, all projects

99 Smaller County Boroughs (population less than 200,000in 1961)

County Borough Built Under ApprovedTotal construction not yet a ppro%ed started (minimum Qualifying estimate) remarks

Barnsley 64,500 Not Not 64,5(K) 1946 to 1962, whole available available area, all projects

Barrow-in-Furness Not 105,000 Not 105,000 Two office blocks available available only

Bath 77,400 Not Not 77,400 1946 to 1962, five available available office blocks only

Birkenhead Two office blocks only, no details

Blackburn 3,100 83,600 51,000 137,700 1948 to 1962, whole area, all projects

Blackpool Very little new office building

Bolton Very little purpose built office space except in connection with industrial development

Bootle 64,000 Nil 155,000 219,000 Post-war, four projects

Brighton 239,400 50,600 187,700 477,700 1946 to 1963, major projects

Burnley No purpose-built office space except in connection with industrial development

Burton-on-Trent 60,800 Not Not 60,800 1946 to 1962, whole available available area, all projects

Bury Not Not Not 141,800 1947 to 1962, available available available permissions Derby 538,900 Not Not 538,900 Post-war, three major available available office buildings

Dewsbury 34,700 78,900 12,400 126,000 1947 to 1962, whole area, all projects Doncaster 158,700 10,400 280,500 449,600 1949 to 1962, whole area, all projects

Eastbourne 50,000 26,500 Not 76.500 Two office blocks available only, very little new office building

East Ham Very little office building Gateshead 14,400 Nil 24,200 38,600 Post-war, central area, major projects Halifax 65,900 Not Not 65,900 Post-war, major available available projects Ipswich 143,100 Not Nil 143,100 1948 to 1962, central available area, major projects Middlesbrough 129,200 Not Not 129,200 1955to 1962, major available available projects 100 .

Smaller County Boroughs (continued,

County Borough Built Under Approved Total Qualifying construction not yet approved remarks started (minimum estimate) Northampton 77,700 25,100 Not 102,800 1951 to 1962, major available protects Norwich 106,000 8,000 19,200 133,200 1952 1°1962, some major projects Oldham 64,100 49,400 16,000 129,500 1945 to 1962, central area, major projects Oxford est. 200,000 Not 80,000 280,000 Post-war, not more available than this figure Reading 595,800 In 'built' 369,100 964,900 Post-war, whole area, major projects Rochdale One office block only, no details Rotherham 202,400 32,800 1,600 236,800 1946 to 1962, whole area, major projects Salford Nil 61,800 Nil 61,800 One office block only in post-war period Smethwick 126,400 31,200 137,000 294,600 1948 to 1960, whole area, all projects Southend 376,500 202,400 232,000 810,900 1955 to 1962, whole area, major projects. Southport 23,100 Nil 28,000 51,100 Three office blocks only, central area South Shields 43,500 Nil 5,200 48,700 1954 to 1962, whole area, major projects Stockport 32,200 48,000 Not 80,200 1948 to 1962, central available area, major projects Tynemouth 7,000 Not 30,200 37,200 Two office blocks available only Wakefield 155,100 5,400 Nil 160,500 1945 to 1962 Wallasey No new office space built. Walsall 139,500 34,000 133,000 306,500 1955 to 1962, central area, major projects Warrington 129,000 Not Not 129,000 1948 to 1962, whole available available area, all projects West Bromwich 84,700 Nil Nil 84,700 Post-war, three projects West Ham Not Not Not 406,700 1955 to 1961, available available available approvals greater than 5,000 sq.ft West Hartlepool 17,600 Nil 30,000 47,600 Post-war, central area Wolverhampton est. 200,000 60,000 Not 260,000 Post-war, central available area Yarmouth Nil Nil 25,000 25,000 No office blocks erected, one block approved York 154,000 In 'built' 47,000 201,000 Post-war Newport Not Not 204,000 204,000 1949 to 1962, no available available details of offices built 101 Other County Councils

County Council Built Under Approved Total Qualifying construction not yet approved remarks started (minimum estimate)

Bedfordshire 160,600 85,400 38,900 284,900 1949 to 1962, details for Luton and Bedford only

Berkshire Not Not Not 526,900 1955 to 1961, planning available available available consents in ten local authority areas

Buckinghamshire Not Not Not 1,117,200 1955 to 1960, permiss- available available available ions (including major change of use) greater than 5,000 sq.ft in central and southern parts of county Cheshire Not Not Not 948,900 1948 to 1962 available available available Seven major projects, Cornwall no details

Isle of Ely 5,200 Nil 3,500 8,700 Three office blocks only

Gloucestershire Very little new office space except in connection with industrial development

Kesteven Very little office building Somerset Three projects under construction, 32 approved, not started yet. No details Staffordshire 828,900 44,700 257,600 1,131,200 1947 to 1962, incomplete information from ten local authority areas Suffolk East Very little office building Suffolk West Not Not Not 143,000 Five office projects available available available approved since 1961, plus office space in connection with industrial development Sussex West 68,900 158,000 31,600 258,500 Incomplete list of major post-war office blocks in four local authority areas Westmorland 46,200 3,500 Nil 49,700 1957 to 1962, six office blocks only in town centre of Kendal Yorkshire E.R. Not Not Not 100,200 1948 to 1962, permiss- available available available ions in Haltenprice and areas adjacent to Kingston-upon- Hull 102 Other County Councils (continued)

County Council Built Under Approved Total Qualifying construction not yet approved remarks started (minimum estimate)

Yorkshire N.R. One large building, other development over new shops

Yorkshire W.R. Very little new office space except in connection with industrial development

Anglesey 40,400 3,800 5,700 49,900 1947 to 1962

Cardiganshire 20,700 Nil 35,500 56,200 Three office blocks built and another approved

Flintshire 40,900 700 12,600 54,200 Post-war, in four district councils

Merionethshire 11,300 18,200 Nil 29,500 Three office blocks built, one under construction

Monmouthshire Not Not Not 572,500 1945 to 1962, available available available approvals

Montgomeryshire 21,000 Nil Nil 21,000 Post-war, four office projects only

Pembrokeshire Six projects only, no details

Radnorshire Very little office building

New Towns

Basildon 38,400 38,800 Not 77,200 To December 1962 available Bracknell 254,300 None Not 254,300 To December 1962 available

Crawley 197,900 None Not 197,900 To December 1962 available Harlow 160,600 58,300 Not 218,900 To December 1962 available Hatfield 3,000 15,000 Not 18,000 To December 1962 available Hemel Hempstead 184,400 180,000 Not 364,400 To December 1962 available

Stevenage 139,300 None Not 139,300 To December 1962 available Welwyn Garden City 121,800 None Not 121,800 To December 1962 available Corby 40,100 1,700 Not 41,800 To December 1962 available Cwmbran 26,200 None Not 26,200 To December 1962 available

Newton Aycliffe 12,500 None Not 12,500 To December 1962 available Peterlee 8,400 15,000 Not 23,400 To December 1962 available

103 Appendix 5.1 The influence of office sizeon the individual andon supervisory and managerialprocesses

Summary response to lighting' indicated that it may well be possi- On Tuesday 26th May 1964, the experimenter conducted ble to build environmentally satisfactory offices at much meetings with four consecutive groups of clerks which greater depths than has previously been supposed. How- totalled nearly three hundred individuals, andrepre- ever, the possibility of deeper buildings was qualified by sented the whole population of one of the podium floors the probable importance of there being an unobstructed in the new CIS building. Usinga recorded talk and view of windows, which suggests that in order to get the instructions, together with slides and questionnaire-type possible advantage of deep buildings, it may also be reply forms, attitudes towards the open plan officewere necessary to accept open working areas. However, size probed and sociometric data gathered. of office area is potentially a factor determining the In the afternoon of the same day, a group of supervisors formation and interaction of groups, and is therefore a and managers was seen collectively and data concerning question of considerable importance to both employees the attitudes of the individual members of thegroup to- and managers. The question of whether given office wards, and professional feeling about, open planning sizes constitute pleasant and acceptable work places will were obtained with the aid of paired-comparison slides also influence the design policy. Therefore, both object- and a questionnaire. ive and subjective response patterns must be studied in A number of hypotheses concerninguser attitudes to relation to any investigation of the optimum size of open planning, and its social consequences were form- office spaces. ally tested. These findings, together with the organisa- Question 19 of the first CIS questionnaire asked the tional requirements stated by supervisors andmanagers, staff of the CIS to state their preference for the openor were drawn together to demonstrate the human factors the smaller office. Differences in the response pattern of involved in optimizing the internal layout of office build- the sub-groups were found but the results indicate a very ings. The conclusions drawn from the threesources of widely held preference for smaller offices or partitioned clerks,supervisors, and managers are summarized areas. However, the question of office size, impinging as separately at the end of this appendix. it does on management, supervisory and building economic considerations, as well as the individual and social Statement of purpose psychological ones of the users, is too important to be The purpose of this appendix is to establish the influ- settled by what amounts to a referendum on the pref.erence of room size on the personal satisfaction of the ences of the clerks. Certainly, the needs and attitur les users, and to appraise the relative benefit of open plan expressed by the choices are important, but their in1- versus small office arrangements for supervisors and portance will be relative to the other considu:ationsin- managers. The degree to which social interactions are volved in office layout. determined by spatial relationships is examined, and A multiplicity of small office spaces is still, as it has bee' 7 their effects upon working efficiency and job satisfaction in the past, normal practice. However, in the last fey `,/ discussed. years, an increasing number of arguments has been ad- % vanced in favour of the open plan office. The most per- Form of the appendix suasive of these are in terms of building economics and This appendix is composed of three relatively discrete flexibility of use; the least persuasive refer to the needs, parts, each of which is presented as a separate research satisfactions, attitudes and working efficiency of the report. However, the summary of background theory users. (The latter type of argument lacks the empirical and research has been dealt with and presented at the evidence usually produced in support of the former). beginning as each of the parts relates to thecommon If the arguments for deep block construction and open type of attitudes towards, and the influence of, open office space were dependent for their success upon argu- office areas. ments based on organization and costs, then there is Part One describes an investigation conducted with gen- little doubt that they would rapidly gain acceptance. But eral clerical workers and the results obtained. Parts Two it is generally the arguments for small working areas, and Three describe investigations conducted withsuper- based on human requirements and management con- visors and managers respectively, and the results ob- siderations, which prevail. These tend to be polemical tained from them. Part Three is followed by 'Con- rather than rational; their success being due toconser- clusions', in which the main findings of each partare vatism in the form of preferring the familiar to theun- drawn together and discussed, and the generalcon- familiar, and because architects and managers are prone clusions stated.

Background of theory and research 1. See chapter 6 The findings of the study of clerical workers subjective 2. See chapter 10 and appendix 10.1, Supplement 8 104 to taking as their frame of reference how they them- environmental conditions. Where such selves would react to certain conditions. studies have been made, they have been variouslydescribed as studies of The literature on theopen plan design is replete with 'group geography ', ' comment which, in the absence of good supportive psychological geography', 6mic- evid- roccology ',and 'spatial ecology'.This latter term ence, can only be taken as an attempt at identification seems particularly apt in the context of studies in with the clerical worker. Thevery terms used in stating which the influence of distance and physicalbarriers between the problem are heavily weighted withthe attitudes of individuals and groups is the ecological their authors. In an address to the Association aspect emphas- of Ind- ised. It will therefore be used withthese connotations ustrial Medical Officers, McGirr, 3 referringto open plan throughout the rest of the appendix. offices, spoke of ' herding clerical staff into them' and Perhaps the most celebrated attemptto apply the meth- described them as 'soulless subtopiaof impersonaliza- ods of social psychology to the questionof the effect of tion '. He rejected the counter-evidenceof their accept- building design on the individual ance in America on the grounds that was that of Festinger open planning is and others. " They studied theeffects of spatial ecology congruent with the American national aim of ' together- on group formation in various types of housing,and ness ', whereas it is a 'principle fundamentallyat vari- concluded that: ance with our national concepts of individualityand 'the relationship between ecological personal privacy, and with and sociometric a healthy yearning for non- structures is so very marked that therecan be little conformity'. doubt that in these communities passive McGirr's position has clearly been contacts are derived from refer- a major determinant of friendship andgroup forma- ence to his own standards. His beliefs in the need for tion'. individuality and personal privacy are middle class senti- Similarly, Gullahorn 1" showed inan office setting that ments which are not necessarily relevantto the majority frequency of social interaction andsocio-preferential of workers in the large clericalorganizations. As Lock- choice were closely related to spatial woodhas pointed out, recent social and relationships. This economic raises the question of whether, inan organization like forces have created paperwork industriesmanned not by the CIS where departments the non-professional members of the are broken into smaller middle classes, but sections working side by side, informalgroups arise mainly by members of lower socio-economicgroups. more easily between individuals assigned The nature of the work is also to different different: it has been formal groupings. This present studywill contrast some mechanized and de-skilled and isnow largely in the of the friendship patterns found in hands of young female workers. It open working areas has currently more in with those of some relatively smalland enclosed areas. common with light industrial assembly work than with In view of the research findings the clerical work of a generation on spatial ecology re- ago. It may very well ferred to above, one mightexpect there to be a number be, then, that the sentiments expressedby McGirr arc of important managerial not those of the modern office worker. but consequences arising from the only research decision to adopt either theopen or the closed office will settle the issue. plan. One might expect, for example, Unfortunately, the research is not done before that the single the con- section occupying a small office wouldoffer the greatest clusions are presented. Even suchrespected figures as opportunity for the formation of McGirr (who was a member of the institute a stable work group. of Directors' In such surroundings, cohesive forceswould derive from main committee in the publication of' Better Offices'), the fact that each member of the reach far beyond the evidence when primary (or face-to- discussing open face) group shares thecommon working objectives of office planning. In thepaper referred to above, he went that group. The small office on to say : area produces a closeness of the group determined not only by the by neglecting the social studies already physical distances existing in between working spaces, but alsoby functional dist- small group performances, thesegentlemen (ie, meth- ances involved. That is, people are broughtcloser to- ods engineers and business efficiencyexperts), may be gether by the passive interactions taking sowing the seeds of future frustration place as a re- and breakdown. sult of using a common entranceand common circula- I believe that as doctors incommerce and industry tion space, and also by the non-businessinteractions at we have a duty to protect those entrustedto our care such foci as the filing cabinets, postal against the potential evils ofan entirely mechanistic trays, and telephones. Unlike the open office, all theseinteractions will approach by work study'. take place on an intra-group basis. Thus This, of course, constitutesa clarion call to industrial the small office medical officers to oppose innovation arrangement produces the best possibility forthe form- on exceedingly ation of a group witha clear identity and concept of slender evidence and grounds. Almostnothing is known itself as a discrete and simple entity. of the consequences of open plan organisation and it is The open plan office allows formore possibilities of premature to support that it may be thecause of future inter-personal contact and frustration and breakdown. However, group formation. Common although the entrances and circulation spaces will increase chain of evidence is nota strong one, and is probably the number of inter-group contacts. Withoutpartitions, both the coloured by personal feelings,one certainly cannot dis- physical and functional distancebetween groups is re- regard McGirr's apprehensions; indeedthey are shared duced and onr might expect, fromthe research findings by many planners, managements, andtheir staffs. referred to above, that the McGirr concluded his paper by saying: number of inter-group `We must make ourselves familiarwith studies in group dynamics and breakdown; if satisfiedthat the 3. McGirr, P. O. M: Environmentaltargets in offices and com- human element is being neglected fora wholly mech- mercial premises. Trans Assoc IndustMed Offrs. 1959, 9 (3) anical approach to work,we ought to be fearless in 4. Lockwood, D: The black-coatedworker. Allen and Unwin, our denunciations'. 1958 5. Sommer, R: Leadership andgroup geography. Sociometry, This is, of course, a sentiment withwhich few people 1961,24 99-110 would disagree, but the links connecting 6. Moreno, J. L: Who shall survive?A new approach to the human happi- problem of human inter-relations. Nervous ness and psychic balance with space utilization andstaff and Mental Disease deployment have yet to be shown. The Monograph Series no. 58. Washington,1934 whole question is 7. Sommer, R: Studies in personalspace. Sociometry, 1959, 22 so important from not only a personal point ofview, but 247-260 also from those of management anddesigners, that re- 8. Festinger, L., Schachter, S., andBack, K: Social pressures search into the subject is clearlynecessary. in informal groups. A study ofhuman factors in housing. The question of the influence of Tavistock Publications, 1963. FirstPublished in USA in 1950. space on the individual Second edition 1963: Stanford,Calif:Stanford UP and on his social and occupationalrelationships is one 9. Festinger, L., Schachter, S., andBack, K: op cit (8) which has received surprisingly littleattention in view of 10. Gullahorn, J. T: Distance andfriendship as factors in the the great attention devoted to almost all gross interaction matrix. (In the SociometryReader, ed : Moreno, other aspects of J. L. Glencoe: Free Press, 1960)

105 friendships would increase.The individual may then expressed will be greater in theopen areas than in the find himself a memberof two groupsthe formal group of his section, and smaller semi-partitionedones. an informal friendship groupcom- posed of members of different livothesis 5 There will bea higher proportion of re- sections. ciprocal choices in the smaller The existence of largenumbers of intra-group work areas than in the friend- larger ones, thougha greater number of isolates. ships (ie, within the formalwork unit), is a factor which has been shown to be One further hypothesis will beproposed for which the related to group effectivenessand morale. ". 12 results concerningspace preferences expressed in the Unlike the small office, theopen plan area three parts of this appendixwill be compared and allows fol. the easy formationof inter-group friendships. con- If these are made trasted. One would notexpect that the attitudes andre- at the expense of the intra-groupones, then one might quirements of clerks, supervisors and expect on theoretical groundsthat the managers would morale and effectiveness of necessarily coincide, andso the following hypothesis is the work unit might suffer. proposed : The proliferation offriendship groups betweenwork units is also a potential Hypothesis 6 There will bea marked difference between force working againstmanager- ial control. The the attitudes of clerks,supervisors and managersto open status roles, lines of communication, office planning. and group loyalties intended to operate within a work- This last hypothesis ing group may well beundermined by the competing cannot, of course, be tested within claims of the informal the context of thispart of the appendix only. Itwill group. The influences exerted by therefore be examined, in the the immediate primarygroup can be very strong in- context of the findings of deed, influencing the all three parts, in theconclusions at the end of the attitudes, expectations: andbe- haviour of not only the appendix. pliable individual, but alsothe average and strong personality. 13.","0" An appraisal Population and samples used of these effects is thereforea prerequisite to the management decision of whether The sample usedwas dependent primarily upon loca- to adopt the open plan: whether the advantages tion. The nature of theinvestigation demanded that are outweighed by the dis- there should bevery large working areas, with advantages. An approachto this question will be made some in this appendix. smaller areas screened-off.This inevitably meant that the population fromwhich the sample could be As the preceding discussionhas shown, the questionof drawn the acceptability of the must be on one of the podiumfloors. As, for experi- open plan office dependsupon a mental reasons, the sample multiplicity of considerationsand viewpoints, had to be extended toan some of entire floor, onewas sought where there was consider- which are amenableto psychological analysis.First of all there are the feelings able homogeneityamongst the workgroups. This and attitudes of theusers; secondly there are the social would exclude floors havingtyping pools or machine and organizationalconse- rooms on them. quences of the different spatialarrangements. Finally, there are the requirements Only one floor adequatelymet the experimental require- of supervisors andmanagers. ments and it was therefore Each of these aspects has selected. Three smallareas been investigated and therewere then picked out for special sults described in thisappendix. attention as being fairly well defined fromthe surroundingareas. They serve as the 'small areas' to be contrasted PART ONE: GENERALCLERICAL WORKERS with the remainder, Hypotheses to be tested or 'open areas', in the rest of thisappendix. Each of the small areas was enclosed The investigation of officespace reported in this app- on three sides by a window endix is predominantly wall and two internal walls:each working section faced heuristic, and to a lesserextent the window wall and had theopening to the larger office theory-derived. Consequentlythe number of hypo- behind it. theses to be tested is kept to a minimum. However,a number of specific propositions Each of the three smallareas houses one single section, are suggested by the and a previous results of the research survey showed that the nature of their so far, and by existing theoryon spatial ecology. work and their workinginteractions with other sections, The hypotheses were in no way different fromany other section compos- may be categorized as being oftwo basic types. The first concerns staff ing the three departmentson this floor. In fact, though attitudes towards large the three departmentshave partition working areas; the secondconcerns the social and there- screens between fore the organizational them, and have separatemanagers, they are nevertheless consequences of providing either all parts of a larger large or small workingareas. department and there is therefore considerable similarity betweenthe work and method of The findings of the first CISquestionnaire study 17re- vealed a good deal of general organisation in each. support for the smaller All of the people working office area and further analysisof the results suggested on the floor chosen were invited to participate inthe study and, with the that attitude to office sizemight be related to the size exception of office in which the of those away sickor on holiday, all did. They composed respondent was working. It isnow 295 general clerks in the proposed to test rigorouslythe hypotheses that they grades below section clerk(ie, suggest. supervisor). The samplewas therefore fairly homo- The hypotheses relating geneous as it did not includeany other such categories to staff attitudes to large work- as machine operators, typists, ing areas may be statedformally as follows: managers or supervisors. Hypothesis / The smaller workingarea will generally be preferred to the larger. Hypothesis 2 People working 11. Goodacre, D. M: Group in large open areas will be characteristics of good andpoor relatively better disposed performing combat units. Sociometry,1953, 16168 -178 towards large areas than those 12. Horwitz, M. and who work in smallerareas. Cartwright, D: A projectivemethod for diagnosis of group properties.Human Relations, The theoretically expectedconsequences of open office 397-410 1953, (6) planning have been referred 13. Sherif, M: A study to at length in the dis- of some social factors inperception. cussion of background theoryand research. The influ- Arch Psycho!. 1935, 27 187 14. Sherif, M: The psychologyof social norms. Harper, ence of proximity on friendship formationhas been dis- York, 1936 New cussed generally and the hypotheses constitute threeex- 15. Asch, S. E: Effects ofgroup pressure upon the modification pected derivations of the and distortion of judgements. effect. They may be stated (In : Readings in SocialPsychology, formally as follows: Ed. Maccaby, E. E.,Newcomb, T. M. and Hartley,E. L. Hypothesis 3 The number of Methuen, 1959) friendship choices made 16. Roethlisberger, F. J. and and received between people Dickson, W. J: Managementand will be progressively fewer the worker: an account ofa research programme conducted with increased distance. by the Western Electric Company,Hawthorne Works, Chicago. Hypothesis 4 The number of Harvard University Press, Cambridge(Mass), 1939 inter-section preferences 17. See chapter 10 106

1 Of the 295 respondents, 214 workedin the large open was to be dictated by the exigencies of work ratherthan areas, and 81 in the smaller enclosed areas referredto any other factors. above. The age and sex composition of thegroups work- As large numbers of staff ing in the two types of were involved, a fairly large area was as follows: the youngest room was needed to accommodate them. To this end age group of 15 to 19 year olds composed 43 the percent of recreation (coffee) roomwas kindly put at the experi- the population, and older workers, the age groups over menter's disposal. As the subjects arrivedfor the sess- 40, accounted for 21 percent. Theage groups between ions they were mixed up as much 20 and 39 were fairly evenly distributed as possible in the seat- minorities. The ing arrangement by assigning themalternately to the ratio of females to maleswas of the order 2 to 1. These two sides of the room. On each seatwas placed a ball proportions are closely similar to thedistributions for point pen, a stapled set of questionnaire the staff as a whole. " reply forms and an explanatory letter (Supplement 1). In order that the programme should Method of investigation run to the strict time limits set, and so that the research materialfor each It was decided at the outset thatthe best way of collect- of the four sessions should be presented ing the sort of highly personal in exactly the data asked for by this same form, a tape recording and a magazine of slides section of the investigation would be a group admin- was used. A slide was being projectedon the screen for istered questionnaire. In thisway, though people would the whole of each session, the first carrying be drawn together in large numbers, a title, the they could be last thanking people for their participation,and in this drawn not as groups butas aggregates, giving essentially way a focus of attention was maintained duringthe isolated conditions for the individual,but having the entire session. advantage of processing large numbers at one time. The After allowing a minute or two for peopleto settle down experimenter could think ofno alternative that would and read the introductory letter, the recorder have been as satisfactory in preventing was start- discussion be- ed and the following introductorymessage given : tween the individuals, and keeping the replies within a `You are probably very surprisedto be addressed by very restricted period of time. a tape recorder rather than directly from the platform Any such group investigation needsthe complete sup- but there arc very good reasons for this. port and co-operation of management. Thefirst app- The first reason is connected with the experimental roach was to the general manager who accepted in procedure, that is, being sure thateveryone is pre- principle the need for the method, and therefore for the sented with exactly the same situation, and givenpre- partial disruption of departments. Hewas, moreover, cisely the same amount of time and explanation. kind enough to advocate the This matter with the depart- would be virtually impossible to achieveon each of mental managers concerned. The threemanagers inmany occasions that this has to be done. volved were most understanding, and together with their The second reason arises because, in orderto under- chief clerks planned the time and thearrangements with take such a very large survey involving the experimenter. The date decided so many upon was Tuesday people, one has to work tovery strict timings to en- 26th May 1964 during the morning. Inorder that the sure that individuals and departments are not put toa sessions should be completed beforelunchtime, when it great deal of inconvenience. was feared that confidences and notes might beex- I hope that the initial shock is changed, the first session now over and that you was set at nine o'clock in the will feel inclined to sit back, enjoya cigarette, and morning half an hour after the commencementof relax. work and the half hour between 10 and10.30 a.m. was You will all have had time to read left open for coffee. Each session my covering letter was scheduled to take with the questionnaire, but I should just liketo under- half an hour and there were thereforeto be four sessions line the fact that the information being gatheredis not each of which would involve one quarter ofthe personn- on behalf of either the Management or the Guild, el in each department. The chief clerks and supervisors though both these bodies support the study. took the responsibility for making up the groups, which At several points in the questionnaire ithas been

A- 6.

ism at

Figure 92 1E. Wells, B. W. P: Office design andthe office worker. Ph D Group administered questionnaire. thesis, University of Liverpool, 1964

107 stressed that any information given will betreated following discussion. confidentially. This questionnaire, unlike previous Each item is presented under the three headings of ones, asks for names and a good deal more personal Rationale and construction of the instrument ',Form data. Some of this might appear to bean intrusion of analysis of data ', andResults and discussion of into your personal affairs and benone of our business. results'. The conclusions to be drawn from the results This is a perfectly fair point of view, but insupport of of each item, and their implications for the hypotheses our request may I make two brief points: the first is being tested, have been deferred to the separate 'Con- that the data can be generalised and is badlyneeded clusions' section at the end of the appendix. to guide the designers of office buildings, andone hopes that the lessons learned would be usedto the Item One advantage of members of your profession in the Item One requires no special explanation, dealingas it future. The second is to underline the fact thatper- does with personal details and matters of fact.Thein- sonal privacy will be absolutely maintained. formation collected serves the threefold function ofpro- Whether you are prepared to helpus by completing viding the demographic data about the sample, speci- the questionnaire is, of course, entirely foryou to fying the age, sex, grade and working section of there- decide, but we hope that you will be kind enoughto spondent so that they can be related to the form of the do so. However, if you do haveany reservations per- replies, and finally to indentifying the respondent for haps you would be kind enough to complete thedocu- the analysis stage of the sociometric study. ment just the same and let me know about them before you leave. If, after explanation,your reservations Item Two Attitude Scale still persist, then you could feel free to destroy the Rationale and construction of the instrument document before leaving. This item consists of a scale for the measurement of I am quite sure that the practical purpose of this attitudes towards the open plan office. A form ofmeas- questionnaire, like the others you have been kind urement was sought which would allow numericalcomp- enough to reply to, must seem very unclear. AndI arisons to be made between individuals and between must apologise for not being more lucid on thisocc- groups.Themethod chosen to construct the attitude asion, but it is very nearly impossible to be specific scale was the method of paired comparisons forcom- about one's own special beliefs and interests without plete data which is described in detail by Edwards. 21 influencing other people's. However, thisprogramme The use of paired comparisons to constructan attitude is now virtually at an end and it will then be possible scale is analogous to the method used for ranking physi- to explain the objectives in detail, perhaps in the cal objects intheabsence of a physical scale. In such a form of an article in The Review.'" I hope then that situation the most accurate method is tocompare all you too will feel that you have participated ina possible pairs of the objects to be ranked and thento thoroughly worthwhile project. make the single judgement of which of thetwo is of the Thank you for your attention so far,now to turn to greater magnitude on a given criterion.Theaccuracy of the questionnaire itself. the method is further increased by usinga large number Where detailed instructions are given I will read them of judges and obtaining average judgements. Attitudes through to make sure that no one falls behind and may also be ranked and assigned a position on a scale that the whole group moves at thesame speed. by a similar procedure. Part One The first step in preparing such a scale isto collect a As you see, the first part of the questionnaire is de- number of statements representing attitudes either posi- voted to your personal details. tive or negative towards a given psychological object. In Please write your initials and surname in the blank the present instance some 30 statementswere collected space provided, then put a tick in the appropriate box together, all of which represented attitudes conveying for your sex, grade and age group. some degree of favourableness or unfavourableness to- The final piece of information required isyour de- wardstheopen plan office. These came either from the partment and the name of your section clerk'. records of interviews with members of the CIS,or The instructions, written at the head of each reply were a distillation of attitudes expressed at other times. section, were then read for each reply section. There- Thenext step was to weed out those statements which cording, allowing times for the respondents toanswer were too similar in content, those which were notcom- the questions, took exactly 20 minutes. The final item of pletely relevant, those which were factualor likely to be the recording was: interpreted as factual, and those which containedmore `That then is the end of the questionnaire. Willyou be than one single thought. kind enough to hand it to me personallyas you make Application of these criteria reduced the number of suit- your way out. Once more, very many thanks for hav- able statements down to twelve. Theywere then paired ing taken part'. in all combinations and orders and presentedto coll- The respondents then left the hall, passingas they did so aborators who were asked to choose from eachcom- the experimenter who was waiting at the door withan bination the one which they felt to be most favourable assistant to collect the completed questionnaire forms towards the larger office area. At this stage fourstate- and answer any queries. ments were dropped from the scale as not being clearly enoughdiscriminated from other statements. Items forming the study: their rationale, construction The remaining eight statements were then re-assembled and results in paired-comparison form and presentedto a new set Although the data for this study was collected on a of judges. The judges at this stage and subsequentlywere questionnaire-type reply form, it is not a true question- university personnel comprising teachers, students and naire study at all. Rather it is a portmanteau collection research students who knew nothing of this project,and of an attitude scale, a sociometric record form anda were not in any way concerned with office work orthe record form for independently presented pictorialmat- planning of office space. erial. A spot check of the results was made after twenty-five Because of the diversity of the items composing the study, judges had made their judgements. From this itwas evi- the most convenient and coherentway of describing dent that the eight statements were being sufficiently well them and their results is to deal with them individually discriminated to continue. In all, 64 subjects actedas in the order in which they appear in the original reply judges. form.2° In the form itself each item is referredto as a numbered 'Part'. However, as this term has also been 19. The CIS staff magazine used to refer to the three separate studies comprising 20. See Supplement 2 21. Edwards, A. L: Techniques of attitude scaleconstruction. this appendix, the term `Item' will be substituted in the AppletonCenturyCrofts. New York, 1957 108 An internal consistency check revealed that the judge- Form ofanalysis of ments made by the 64 judges were indeed quite consist- Three methods of analysiswere applied to the data. The ent, and that the scale values were therefore derived first was to test the homogeneity of thedata, the second from reliable data. A test of significance and unidimen- to rank the statements in the order of frequency with sionality showed that assumptions of additivity and uni- which they were chosen, and the thirdto derive scale dimensionality along a psychological continuum were values for the attitude of the respondents towardsopen justifiable in the present case.'2'2 planning. The final order of the statements,as ranked by the The cross-tabulation and chi-squaredanalysis of data judges, is given in table 9 together with the calculated from the CIS questionnaireshad demonstrated the scale values. Their order of presentation in the quest- great degree of variability that may exist within such ionnaire was randomised, and this order is shown in the data. Age, sex, grade, and previous workplacewere all last column of the table. shown to be determinants of the mode ofresponse to a Table 9 variety of environmental conditions. Previous workplace Rank order of the statements forming the attitude scale would, of course, no longer be relevant aftertwo years in the new building and was therefore omitted from ScalePos JudgedStatements the Order in present analysis. Also omitted was the respondent's valueor rank question- Neg grade as the sample was entirely composed of junior naire general clerical grades and excluded typists, machine 0 Neg 1 There is an uncomfort- 6 operators, supervisors and managers. Because of the able feeling of being promotion system for junior grades, age and gradeare watched all the time in virtually equivalent. a large office However, two additional variables were included for 0.43 Neg 2 The larger offices make 2 cross-tabulation. The first was the department in which one feel relatively the respondent worked. This was included in orderto unimportant take account of any unsuspected physical influences,or 0.80 Neg 3 The large office creates 5 those arising from differences in morale, supervision,or less feeling of perman- group composition. The second variable was the size of ence and security than the area within which the respondentwas located. This a small one was included in order to allow for a comparison ofre- 1.13 Neu- 4 The size of the office 4 sults deriving from areas which were relativelyopen and tral itself is not really those which were smaller and walledon three sides. important Three such areas were located and, taken together,are 1.29 Pos 5 Any disadvantages of 8 subsequently referred to simply as the `smallareas '. the larger office are off- It had been intended to take the statement with which the set by greater advantages respondent was most strongly in agreement (ie, theone 1.45 Pos 6 The social life of a large 7 double ticked) and weight its value in relationto the office is likely to be others. This proved to be impossible as, ina great num- better than that of a ber of cases, either more thanone statement had been small one double ticked or it had not been done at all.

1.84 Pos 7 A good working atmos- 1 phere is more likely to Results and discussion of results be found in a large office (1) Test of the homogeneity of theresponse pattern than a small one As with the first CIS questionnaire, the datawas 2.30 Pos 8 A large office is definitely 3 cross-tabulated and analysed by the chi-squaredtest for better to work in than a the independent samples. small one Table 10 Analysis of response to statements byage group Statement no 1 2 3 4 5 6 7 8 Total z Number of respondents Agegroup 15 19 36 93 25 63 68 99 90 49 523 2.31 127 20 24 16 49 12 23 35 42 35 25 237 2.25 64 2529 3 12 2 9 11 10 9 30 34 3 59 3.64 16 4 8 3 2 4 4 9 6 40 5.40 12 3539 3 11 2 4 7 6 7 5 45 1.70 40 and over 13 20 37 13 30 20 29 30 24 203 7.27 63 82 210 57 131 145 190 180 112 1107 22.57 295 NS Table 11 Analysis of response to statements bysex

Statement no 1 2 3 4 5 6 7 8 Total z 2 Number of respondents Sex Male 27 72 20 41 43 51 60 42 356 3.51 100 55 Female 138 37 90 102 139 120 70 751 1.65 195 82 210 57 131 145 190 180 112 1107 5.16 295 Df =7,NS

22. Additive quality of the scale is essentialwhere it is intended to make comparisons between the strength of attitude heldby different groups. It will usually hold only incases where the psychological continuum is add itiNe 23. See chapter 10 109

...... 1001M0411.144.1411.0..4001...... ,11.4.0WV.AOW* Table12 Analysis of response to statements bydepartment in which employed

Statement no 1 2 3 4 5 6 7 8 Total Z Numberof respondents Department Department A 27 60 14 40 45 63 58 32 339 1.94 85 DepartmentB 25 67 21 43 39 50 58 43 346 5.08 94 Department C 30 83 22 48 61 77 64 37 422 2.20 116 82 210 57 131 145 190 180 112 1107 9.22 295 Df=14, NS Table 13 Analysis of response tostatements by comparison of large openareas and smaller semi-enclosed areas

Statement no 1 2 3 4 5 6 7 8 Total y Number of respondents Open areas 72 144 46 101 102 139 142 86 832 4.89 214 Small areas 10 66 11 30 43 51 38 26 275 12.65 81 82 210 57 131 145 190 180 112 1107 17.54 295 Df=7, P < 0.02

Table14 Of the four analyses performed, only in the one shown Final rank order of the statementsforming the attitude in table 13 was a significant resultfound. The three scale statements in which there was the greatest discrepancy Rank Statement between the observed and expected frequencieswere Frequency Percent- order those numbered2, 1 and 5. To statement 2, itwas the chosen age of respondents in the small office areas whoreported more respond- frequently than the expected value derivedfrom the ents table that the larger office makesone feel relatively un- 1 The larger offices makeone important. Similarly, in thecase of statement5, it was the respondents in the small areas who feel relatively unimportant 210 71.2 more often than 2 the expected frequency reported that There is an uncomfortable the large office feeling of being watched all creates less feeling of permanence and securitythan a smallone. The distribution of replies to statement 1 the time in a large office 190 64.4 3 The social life of a large was such that it was the people in the largeareas who, more frequently than could be expected, supported the office is likely to be better than that of a small one 180 61.0 proposition that a good working atmosphereis more 4 likely to be found in a large office than The large office creates less a small one. With feeling of permanence and only two rows in the table, the exactconverse will also hold in each case, ie, a response of less security than a small one 145 49.2 than the expected 5 The size of the office itself is value in one row will be mirrored bya frequency greater not really important 131 44.4 than the expected value for the other. 6 To summarize the results, the chi-squared Any disadvantages of the analysis larger office are offset by showed that the respondentswere homogeneous in their responses to the attitude scale on the variables of greater advantages 112 38.0 age, 7 A good working atmosphere se:. and department, but significant differenceswere fou. . to exist between the occupants of is more likely to be found large and small in a large office thana areas. The difference was due to people in the largeareas being better disposed towards small one 82 27.8 open offices than the 8 A large office is definitely people working in the smallerones. better to work in than a small one (2)Ranked distribution of scores 57 19.3 A comparison ofthe rank order derived from theexperimental data and the rankings madeby the judges by summing the scores fromboth types of area. The when the scale was established yieldsan interesting re- results were as follows: sult. The order of the statements listed above,when ranked Table 15 by the judges, was 2, 1, 6, 3,4, 5, 7, 8,where 1 is the state- Mean values of individual medianscores grouped by ment least favourable towards theopen plan office, and type of working area 8 the most favourable. A rank order correlation of these Type of Number ofSum of two rankings, using Kendall's coefficient Mean T, 21yields a office space respondentsindividual coefficient of0.714. The critical value of values Tat the 0.05 median level of significanceis 0.64. Thus significant correlation scores exists between the two rankings, implyinga trend in the results such that the less favourablea statement is to the Large area 214 207.39 0.969 large office, the more frequently will it tendto be chosen. Small areas 81 60.04 0.741 Large and (3) Deriving the scale value small In view of the fact that table 13 hadrevealed significant combined 295 267.43 0.907 differences between the pattern of resultsobtained from people working in the largeopen areas and those working in the smaller ones,separate scale values were de- 24. T S where S is a score based on the rived for each. An overall scale value in(n-1) comparison of was also derived each item with every other item 110 It has been shownthat the mut' al point of this atti- Ten paired comparisonswere required in the present tude scale. te. the pointat NCI) neith..r a post's,:nor a instance. Lach picturewas designated by one of the negative feeling towards the open plan office ise \pres-ed. letters L M N 0 P. These is 1.125. A score of less than 1.125 were chosen in preference to represents a negat is e A 13 E as being less likely attitude to the large office to connote a progressise area. and a score abuse this series. As an added precaution, represents a positise attitude. the letters were random- lied when they were assigned Table 15 shows that the scale to a layout so as to pre- values derived for resent the respondent from forming spondents working in both largeand small areas are a response set in terms of. say. the plan being progressivelymore open negatise and so. ofcourse, is the combined result, Flow- from I to P. The order in whichthe objects were paired eser. the more 'legalise position on the scaleis occupied abm c or below one another), by the respondents from the and the order of pres- small areas. In other words. entation were also randomited. the people who work in small officeareas are more opp- An 'example slide' accompaniedby another slide show- osed to large offices than thosewho actually work in ing how the reply sheet them. was to be marked was shown to the respondents before theexperimental slidesproper. The technique requires that Item ThreeLayout Preferences a judgement shall be recorded in knerycase, and this was asked for in the Rananak and cam /radian of theinstrument instructions. This item consisted of use photographs(which showed an office floor partitioned in carious ways) which were Form of analysis of 'dons to he ranked in order of preference froma paired com- Paired comparison tableswere constructed from the re- parison presentation. ply form in which, for The photographswere of a scale model which repre- any pair, the letter at the top represents the preference choice and theletter in the left hand row represents thenon-preferred alternative in that particular comparison. Itwould perhaps have been desirable to calculate the coefficientof consistence for ti each judge at the time of groupingthe results but this was not possible in the presentcase as, for values of n tig\144" 4"1"111 less than 6, significancecannot be proved. 4 Having prepared the pairedcomparisons table, the % 0 ranked order of preferenceis quite simply derived by see 0sie summing the columns and attachingthe highest rank to the highest score. - The homogeneity of theresponse pattern was tested by preparing separate tables foreach of the variables of age, sex and department and thencomparing them. Separate tables were also preparedfor the respondents P working in the largeareas and those in the small ones, 71/ and these results compared.

Results and discussion of results Comparison of results byage group

Table 16 07101;i\- 15 to 19 years

L M N 0 P Totals L 10 93 90 12 205 115 108 114 2$.11- N 110 447 32 17 34 26 109 0 35 11 91 25 162 113 15 99 100 327 L Totals 295 53 391 338 173 1250 Ranked Figure 93 order of Sample paired comparison of office layouts. preference 3 5 1 2 4 n-125 sented, in simplified form, the mainelements of a CIS Table 17 podium floor. Desks and chairs fortwo hundred work- 20 to 24 years ers were shown. In one picture the modelwas quite open and without partitioning, except for thecentral service L MN 0 P Totals area. The others showed it divided by panel partitions L into two, four, six and eight smallerareas. Figure 93 12 44 43 12 111 M shows a typical paired comparison. 52 49 53 52 206 N The method of paired comparisons isusually employed 20 15 24 16 75 0 21 where a relatively large number ofobjects has to be 11 40 17 89 P 52 ranked, and where discrimination isanticipated as being 12 48 47 159 difficult. In this instance, the numberof objects to be Totals 145 ranked was not large, nor was it difficultto discriminate 50 181 167 97 640 the magnitude of the difference betweenthe alternatives. Ranked The method was used because itwas hoped that it might order of reduce the chance of respondentsadopting the single preference 3 5 1 2 4 criterion of prefering the large or the small area (or vice n=64 versa), and thus not allowing theproper weight for any actually preferred intermediate alternative. 25. Wells, B. W. P: op cit (181(Statisticalappendix 5)

111 Table IS Comparison of results by se \ 25 to 29 )ears Table 22 NI N U P Totals Male replies 9 I II 3 24 L M N 0 P Totals NI 14 13 15 15 57 N 4 2 I I 8 I. 18 74" 76 20 188 6 0 14 3 23 NI 82 81 84 83 330 12 0 14 12 38 N 26 19 37 21 103 0 24 16 63 18 121 Totals 36 3 52 37 22 150 P 80 17 79 82 258

Ranked Totals 212 70 297 279 142 1000 order of preference3 5 I 2 4 Ranked n 15 order of preference a 5 I 2 4 n100

Table 19 30 to 34 years Table 23 Female replies L M N 0 P Totals L M N 0 P Totals L 2 9 10 2 23

M 10 9 II I1 41 27 133 127 26 313 N 3 3 5 I 12 M 165 157 171 166 659 0 2 I 7 0 10 N 59 35 61 43 198 P 10 I II 12 34 0 65 21 131 50 267 166 26 149 142 483 Totals 15 7 36 38 14 120 Totals 455 109 570 501 285 1920 Ranked order of Ranked preference 3 5 2 I 4 order of n 12 preference 3 5 1 2 4 n 192

Table 20 Table 24 35 to 39 years Male and female replies combined

L M N 0 P Totals L M N 0 P Totals

L I 8 10 2 21 L 45 207 203 46 501 M 12 10 12 II 45 M 247 238 255 249 989 N 5 3 4 2 14 N 85 54 98 64 301 0 3 I 9 4 17 0 89 37 194 68 388 P 11 2 11 9 33 P 246 43 228 224 741

Totals 31 7 38 35 19 130 Totals 667 179 867 780 427 2920

Ranked Ranked order of order of preference 3 5 1 2 4 preference 3 5 1 2 4 n-13 n=292 Comparison of results by departments

Table 21 Table 25 40 or more years Department A

L M N 0 P Totals L M N 0 P Totals

L 16 43 42 15 116 L 9 63 64 8 144 M 46 49 52 49 196 M 76 71 77 76 300 N 19 13 27 16 75 N 22 14 30 16 82 0 20 10 35 19 84 0 21 8 55 14 98 P 47 13 46 43 149 P 77 9 69 71 226

Totals 132 52 173 164 99 620 Totals 196 40 258 242 114 850

Ranked Ranked order of order of preference 3 2 4 preference 3 5 1 2 4 n=62 n=85 112 011100.

Table 26 With only one exception, the rank Department B order of preferences for the objects comparedwasexactly the same ie, N 0 L P NI. This can be represented L M N in the following 0 P Totals way :

L 16 64 60 20 160 M 77 76 79 72 304 N 0 L P M N 29 17 29 24 99 0 33 14 64 29 140 P 73 21 69 64 227 1 Totals 212 68 273 232 145 930 The exception was table 19 in whichthe first and second Ranked preferences were reversed bya very small margin. order of Apart from this, the rankingsrepresent a consistent preference for smaller work preference 3 5 1 4 areas. There was no difference between the pattern of replies for it -93 men and women, and no differences between different agegroups, or different departments, or between those workingin large open Table 27 spaces and those in smaller areas. Department C Item FourSocial Considerations I M N 0 P Totals Rationale and construction of theinstrument This item called for the respondentsto make socio- L 20 80 79 18 197 preferential choices amongst theircolleagues. M 94 91 99 101 385 From the point of view of construction,devising a socio- N 34 23 39 24 120 metric reply sheet isa perfectlycommon-sense procedure. 0 35 IS 75 25 150 Moreno, 2" who was principally responsiblefor the in- P 96 13 90 89 288 ception of the sociometric method, givesa general background to the preparation of sucha document in his Totals 259 71 336 306 168 1140 book 'Who shall survive?'. Themost fundamental prerequisite in using such a technique is,of course, having Ranked the confidence of the respondents.In the present inst- order of ance, the request for sociometric datacame as the cul- preference 3 5 I 2 4 mination of a two year research projectwithin the company. The experimenter was therefore n very well known to the respondents and on good personalterms with many of them, having spenta great deal of time working on the floor where this investigationwas made and mixing socially in the coffee lounges and Comparison of results by size of workingarea recreation room. It might also be expected thatthe study had derived a Table 28 certain amount of acceptance fromthe fact that it was Small areas supported by the trade union,to which every employee belongs. At all stages in the projectgreat care had been L M N 0 P Totals taken to secure the goodwill of the unionso that it would be able to recommend theinvestigations to its I I 56 60 11 138 members. M 69 66 70 65 270 As Moreno, and many others since,have pointed out, it N 24 14 17 16 71 is essential to be specific aboutthe conditions under 0 20 10 63 15 108 which the preference choice isto be made ie, whether P 69 IS 64 65 213 the choice is a work choice,a dining choice, a choice of sports companion, etc. The population fromwhich the Totals 182 50 249 212 107 800 choices may be drawn must also bespecified if the pattern of interactions is to be interpretable. In thiscase, the Ranked criterion of the socio-preferentialchoice was 'Which order of people would the respondent liketo work beside?' The preference 3 5 1 2 4 universe of choice was limitedto other people working on the same floor. n=80 The restriction of possible choicesto a particular floor allowed for most of the possibleface-to-face social inter- Table 29 actions which might take place duringworking hours Open areas since most between-floor communicationhas to happen either by telephoneor via the document conveyor. L MN 0 P Totals Between-floor preferenceswere allowed for under the heading of Additional information', inwhich each re- L 34 151 143 35 363 spondent was asked to give the numberof his particular- M 178 172 185 184 719 ly good friends or acquaintanceson other floors. N 61 40 81 48 230 Selltiz and others 27 have drawnattention to the desir- 0 69 27 131 53 280 ability of stating, where feasible,that the experimenter P 177 28 164 159 528 will arrange for the individual'spreferences to be met if possible. This was not possible inthe present case, and Totals 485 129 618 568 320 2120 instead the experimenter took thealternative course re- Ranked order of preference 3 5 1 2 4 26. Moreno, J. L: op cit (6) 27. SeUtiz, C., Jabal', M., Deutsch,M. and Cook, S. W: Res- n=212 earch methods in social relations. Methuen, 1959

113 commended by Bjerstedt: It may therefore often be their internal sociometric structures. Howl the pres- advantageous to make the situation cognitively experi- ent study is specifically concerned with the inter-group mental, to stress that nothing will happen as a result of relationships and so it must be possible for the respond- the findings. If there is no need for report-distortion the ent to express his choices within his actual universe of subject is likely to give a true answer; that is, after all, choice, rather than an arbitrarily limited one. Therefore most natural and least laborious'. the results for the entire floor were treated together, with As well as having the general goodwill and support of a corresponding gain in the validity of the data and a the subjects,itis also absolutely vital that they are loss in the rigour of the analytic techniques available. assured of their anonymity and that their reply forms The main objectives of the analysis were threefold. First- will be treated in an absolutely confidential manner. The ly to establish the number of choices directed outside of written and spoken instructions made this abundantly the section in which they were made. Secondly, to est- clear, and also that the data was to be cod :d and trans- ablish the number of reciprocated choices made within ferred to punched cards and the questionnaires imm- (i) the open areas and (ii) the partitioned areas X, Y and ediately destroyed. It is, though, the experimenter's be- Z, and to compare them. Thirdly, to compare the num- lief that the greatest reassurance came not from promises bers of isolates found in the open and the partitioned made at the time, but from people's experience of the areas...2 way in which other confidences, derived from two years Established technique.; exist for calculating the number of contact, had been maintained. of isolates that might be expected to occur on a chance No limit was set to the number of choices that a re- basis within a given group, and probability distributions spondent could make. Preferences were to be given in have been derived. "..'1."5 However, these distributions ranked order, and spaces were allowed for up to ten depend upon the subjects making a known number of names on the reply form. As an additional motivator to choices each, and the choices themselves being made the hesitant, the tape-recorded instruction Please try to within a single group. The method is not therefore appli- list at least three people' was given half a minute after cable to the present case, where the problem is to com- the first set of instructions. A total of two minutes was pare two sub-groups of the same universe. This is not, allowed to list the preferential choices, and one further however, to say that any special difficulty exists in deter- minute to complete the additional information about mining the number of isolates in the subgroups, merely friendships with people on other floors. that it is not possible to determine the number that may be expected to occur in each of them on a chance basis. Form of analysis of data Moreno "'' advocated the use of a specified number of The sociometric method is used principally as a tool for choices, ie, Name three', 'Name five', etc, because of the examination of the social structure of groups of less the advantages of being able to calculate the probable than about forty people, and more usually still, for number of reciprocal choices and isolates to be found in smaller ones. The techniques of analysis are therefore a given group. He argued that allowing more choices most developed for the small group. The original meth- would not alter the results; a higher choice rate would od described by Moreno" and still the most popular, merely increase the frequency with which the more pop- entails drawing a diagram in which preferences or re- ular were chosen. This he termed thesociodynamic jections are represented by lines between the individuals effect '. However, Bjerstedt "7 produced evidence to show forming the group,after which the diagram is rearranged that such an assumption is incorrect as stated. He dem- to make it perceptually more simple. It then remains onstrated that the effect of allowing more choice is not to look for any trends, to establish who are the isolates the maintenance of the status quo, though most of the in a group, and who are the 'stars'. The procedures additional choices do indeed go to the more popular involved are not systemmatic and the pattern tends to members of the group. The important difference is that be exceedinglydifficultto comprehend withlarge small numbers also go to those people who would be numbers. described asisolates' under conditions of restricted To simplify the analysis, methods were developed in choice. Restricted choice is therefore more useful for which the diagram was replaced by the matrix. Forsyth identifying the popular members of a group than for and Katz"') developed a method in which the individuals determining the number of true isolates: what is gained of the group were listed in the same order down the rows in analytical and computational advantages is lost in the and along the columns. The matrix was completed by consequent attenuation of the results. Thus, as one of entering the choices made by individuals in the row bear- the main hypotheses of the present study concerned the ing his name and in the column of the person receiving number of isolates in the sub-samples, the number of that choice. This gave a picture almost as complicated choices made was left unrestricted when analysing the as the diagram, but it could be simplified by rearranging data. the order of the columns and rows so that the entries in the matrix clustered as closely as possible about the main diagonal. The diagonal is, of course, composed of empty cells as it is formed by the point at which the individuals name appears in both the column and the row. 28. Bjerstedt, A: The methodology of preferential sociometry. Sociometry Monographs, no. 37, 1956 The arrangement shows, in the clusters, those people 29. Moreno, J. L: op cit (6) who have chosen one another relatively frequently. The 30. Forsyth, E. and Katz, L: A matrix approach to the analysis well separated entries represent the people who have not of sociometric data: preliminary report. Sociometry, 1946, 9 chosen one another. 340-347 31. Festinger, L: The analysis of sociograms using matrix However, the method of matrix analysis so far described algebra (In the Sociometry Reader, Ed: Moreno, J. L. Glencoe: is, as Festinger pointed out, 31 most useful for establish- Free Press, 1960) ing changes within a group from one time to another 32. A reciprocated choice is one where two individuals mutually and is of little value in demonstrating the existence of choose one another. An isolate is defined as a person unchosen, on the criterion of the test, by any of the other members of his subgroups, or for comparing two groups. Festinger's group own special contribution was to suggest a technique to 33. Katz, L: The distribution of isolates in a socialgroup. (In allow for the determination of cliques within a group. the Sociometry Reader, Ed. Moreno, J. L. Glencoe: Free Press, The above represent the three main types of analysis 1960) applied to sociometric data. There are almost innumer- 34. Moreno, J. L. and Jennings, H. H: Statistics of socialcon- figurations. Sociometry, 1938, 1 342-374 able variations of them, but none allow fora comparison 35. Bronfenbrenner, U: The measurement of sociometric status, of the social organisation of physically separated groups structure and development. Sociometry, 1943, 6 363-397 both of which form part of the same universe of choice. 36. Moreno, J. L: Who shall survive? Foundations of sociometry, group psychotherapy and sociodrama. Beacon House, The,,usual situation involves taking two groups, each of New York, 1953 which represents a separate universe, and comparing 37. Bjerstedt, A: op cit (28) 114 -- Results and discussion of results 15 to 19 85 percent The analysis of results was undertaken intwo stages: 20 to 24 80 percent the first to determine the effecton socio-preferential 25 to 29 65 percent choice of personal variables: the second to determine the 30 to 34 79 percent effect of the spatial variables. A summary of much of the 35 to 39 85 percent data for both is also given in the first stage. 40 or more 71 percent As with the case of sex, it is clear thatage group also is a Stage I ( The effect of personal variables) Very potent factor in determining the preference choices. The first task in the analysiswas the manual cross- Again, the finding is neither novelnor unexpected, but tabulation of the total number of sociometric prefer- a necessary step in evaluating the factors responsible for ences made by each sex and age group. socio-preferential choice. Table 30 summarises the number and proportion of socio- (iii) Friendship choices to other floors: preferential choices made and received by eachage and Column 14 shows that, as might be expected, thenum- sex group. The data serves to show the influence of age ber of friendship choices to other floors isgreater with and sex as determinants of choice, and also therange of the older than with the youngergroups. This is presum- friendship choices to other parts of thebuilding."' ably because they will have hadmore opportunities to Choices to other parts of the building will be dealt with make contacts over the years and will, perhaps, have mainly in stage two, but the topic is also includedat this worked in more departments. stage as the data has been cross-tabulated with theage The 'average' row of column 14 also shows that itis the and sex of the people making them. It is thereforeposs- men who make the greatest number of choices to other ible to consider the results of table 30 under threecon- parts of the building. With junior grades there isno venient headings: reason to suppose that the men have any greater opp- (i) sex as a determinant of socio-preferential choice ortunity to interact with members of other floors than (ii) age as a determinant of socio-preferential choice have the women. This is perhaps dueto the existence of (iii) friendship choices to other floors. such secondary social networksas the sports club and (i) Sex as a determinant of socio-preferential choice: the union, both of which are almost entirely in the hands Column 3 shows that the average number of choices of men. The inforn-al social networks of thewomen, by made by men and women is approximately thesame; contrast, tend to depend to a greater extenton meetings columns 4 and 5 show that men are more likelyto refuse during working hours. The largest such network de- to record their preferences than are women. rives from the lunchtime dances in the recreation hall, It was found that the men, who represent 31percent of where girls who work and eat together also danceto- the sample, in fact received 36 percent of all the stated gether. Lunchtime dancing is now, by custom,an en- preferences; whereas the women, who represent 69per- tirely female affair. cent of the sample received 64 percent of the total of It might be expected that, in cases where the workre- choices made. However, as columns 6 and 7 show, the quires co-operation, the resulting inter-personalcontact great majority of choices made by a given sex were dir- would be most important in determining socio-prefer- ected towards members of the same sex. This held for ential choice. However, there is comparatively littleco- the entire range of age groups, and it thereforeappears operative work on the floor where the studywas made: that sex is an important determinant of choicepattern. with only few exceptions, clerks workingon this floor This is not, of course, a novel or unexpected finding; have fairly discrete tasks. Some clerks transfer data from indeed, any other result would have been surprising. The letters to proposal or application forms, from docu- value of the findings are not therefore intrinsic but they ments to cards, or from cards to documents. And in this establish the relative weight to be given tosex when the respect there are no differences between sectionsoccu- influence of other determinants is considered. pying separate, semi-enclosedareas and those in the (ii) Age as a determinant of socio-preferential choice: open ones. Being a member of one work section rather The youngest age group, ie, the 15 to 19 year olds,com- than another amounts, therefore, to a mainly administ- posed the numerically largest group and made the great- rative allegiance coupled, perhaps, witha special con- est absolute number of choices. In all, they made 48 per- cern for a given business area. Thus, as the frequency of cent of the total preference choices recorded, butre- working interactions was fairly low, andwas comparable ceived only 35 percent of the choices themselves. Con- for the two types of area studied, it has not been given versely, with the exception of the over 40 agegroup, all any special place in this discussion. other groups received more choices than they made. It appears therefore that older working companions are Stage 2 (The effect of spatial variables) generally more acceptable than younger ones. The effect of distance as a variable affecting the forma- It is interesting to note from columns 6 and 7 that there tion of friendships may be regardedas operating on two is a tendency for the proportion of men being chosen by levels. The first level is concerned withgross distances of women to rise with age group. On the other hand, the the sort existing between different parts ofa building. proportion of women chosen by men declines with the The second is concerned with very much smallerdist- older age groups. The reasons for this are not apparent ances separating individuals in the same part of a build- from the results and so one must postulatesome other ing. Both will depend to a large extentupon functional factor: differences, for example, in the relativeattract- relationships. iveness of the two sexes for one another at differentages. Columns 14 to 18 of table 30 summarize the number of Inspection of the cells in the centre of the table where friendship relationships claimed by workerson the floor the figures are in bold type shows that each age group surveyed to exist between themselves and workerson tends to choose predominantly from the members of its other floors. They show that it is the higher podium own age group. This is most marked for the numerically floors which attract most choices; the higher floors in larger age groups: members of the smaller ones,pre- the tower next; the lower podium floors third. Thelower sumably, are more cut-off from members of theirown tower floors attract fewest choices and the explanation age group. With the single exception of the 30 to 34 year for this is probably that these floors haveno functional age group it is members of the next younger or older age relationships with the sub-departmentson the floor group who are next most frequently chosen, ie, the age surveyed whereas the other groups of floors do. It might groups to the immediate left or right of the groups in the bold type diagonal. In fact, the proportion of choices 38. A substantial number of the respondents completing therest of made to members of the same age group and those imm- the questionnaire left this particular item blank and theywere ediately adjacent are: therefore not included in the analysis, though their number is shown in column 4 of table 30. Their names were also excluded from the preferences made by other people and so they effectively ceased to exist as members of the population studied 115 SummaryTable 30 table of sociometric preferences cross-tabulated by age group and sex sexandgroupsAge ondentsofNumber(1) resp- named Numberof Average Number Percent- Choices to (2) of(3)number of resp- age of ondents(4) respond- (5) males(6) femalesChoices(7) to groupChoices(8)15 to 19to year 20 to 24 year 25 to 29 groupChoices(9) to Choices(10)group to year 30 to 34 year 35 to 39 year 40 - age groupChoices(11) to groupChoices(12) to Choices(13) to otherchoicesFriendship(14) floors to Choices(15)lower tofloors otherChoices(16) floors to nearestChoices(17) six to remainingChoices(18) to formpletingcom- givenencesprefer- givenencesnamedprefer- not re-plying centage4 as(columnents per-replying not ber Num- Per- cent Num-ber Per- cent Num-ber Per- cent berNum- Per- cent berNum- Per- cent berNum- Per- cent berNum- Per- cent berNum-group Per- cent berNum- Aver- Num- Per- age berof podium cent berNum-in podium Per- cent berNum-tonerfloors Per- in cent berNum-tonerfloors inPer- cent years:15 to 19 of umncol- 1) (ofcolt) (ofco12) (ofcolt) (ofco12) (ofcolt) (ofco12) (ofco12) co12)(of (ofcol 1) (ofcol 14) col(of 14) col(of 14) col(of 14) TotalsFemaleMale 116 9521 469378 91 4.04.3 11 92 8.7 117 5364 24.914.070.3 325352 27 86.029.775.1 227275 48 60.152.758.6 103122 19 27.226.020.9 2314 9 15.4 4.92.4 10 82 2.12.22.1 1510 5 2.63.25.5 2421 3 5.15.63.3 209293 84 2.24.02.5 49311 8 21.416.714.8 140105 35 47.850.241.7 45341 I 15.416.313.1 593920 20.118.723.8 years:20 to 24 1 TotalsFemaleMale 573621 261158103 4.64.44.9 67 10.914.3 4.5 229876 37.573.813.9 163136 27 62.586.126.2 755223 28.732.922.3 124 7450 47.546.848.5 2217 5 16.5 8.43.2 13 94 5.75.03.9 12 75 4.64.44.9 1115 4 7.05.73.9 144 6480 2.53.81.8 2014 6 13.917.59.4 503416 53.134.720.0 2620 6 25.018.1 9.4 481830 33.328.137.5 years:25 Maleto 29 3 18 6.0 1 25.0 14 77.8 2 11.1 3 30 Totalsto Female34 12 9 4563 5.35.0 43 25.0 25 I1 39.724.4 3834 4 60.375.622.2 57 11.1 12 9 20.019.016.7 14 95 22.220.027.8 15 96 23.820.033.3 06 13.3 9.50 972 14.315.611.1 503218 4.26.03.6 II 92 22.028.111.1 1410 4 28.031.322.2 'I 4.03.15.6 2312II 46.037.561.1 years:Male 6 27 4.5 2 1 25.0 23 85.2 4 14.8 0 .0 1 3.7 6 22.2 I 17 3 2 -i 35Totals Femaleto 39 93 4821 5.47.0 3 25.025.0 29 6 60.428.6 1519 39.671.4 0 .0 2 1 4.24.8 I 1 5 23.822.9 24IS 9 42.955.550.0 23 6.29.53.7 48 16.719.014.8 3215 3.65.02.8 85 25.033.317.6 1810 8 66.747.156.3 0' 11.8 0.06.2 40 23.512.50 years:Male 5 21 4.2 13 37.5 17 81.0 4 19.0 0 2 1 6 7 TotalsFemale 94 3413 3.33.8 4 20.030.8 17 0 50.0 0 1713 100.0 50.0 1I 2.97.7 .0 3I 8.87.79.5 0 I 2.904.8 93 23.126.528.6 826 23.515.428.6 12 6 46.235.328.6 401030 4.42.56.0 10 3 25.030.023.4 8I7 20.023.310.0 04 10.013.3 0 1812 6 45.060.040.0 over40 or : 19 1 2 1 1 TotalsFemaleMale 3819 145 7669 3.84.03.6 2513 39.740.638.7 852758 58.635.584.1 6049 41.464.515.9 231 2.92.11.3 13 67 9.09.28.7 43I 2.83.91 . 4 22 14 8 15.218.41 1 . 6 213514 24.130.418.4 683731 44.946.948.7 148 7672 3.84.03.9 3524I I 23.633.314.5 473017 31.823.639.4 20I 1 9 13.512.514.5 462224 31.130.631.6 combinedMalegroupsAll age : 75 329 4.4 21 21.9 252 77 31 totalsGrandFemale 166 241 1020 691 4.2 5433 18.316.6 371119 36.476.617.2 649572 63.682.823.4 361286 75 41.435.422.8 276195 27.028.224.6 753144 13.4 7.44.5 935241 12.5 9.27.5 794138 11.6 7.75.9 136 8650 13.312.515.2 402707305 2.92.44.1 133 5578 18.819.418.0 277177100 39.244.032.8 995049 14.012.416.1 198101 97 28.024.133.1 -"millainmemaiallmamm"0" .0.... ma111...s he expected that, other things being equal, the higher Tables 31 and 32 are concerned with thesheer effect of tover floors would attract more choices than the lower physical distance on the differentage and sex groups. floors because they contain manymore workers (ie, 736 They were prepared by measuring,on a scale floor plan, compared with 441). Thus there exists the possibility of the distance between the person making thepreference greater numbers of random contacts between individ- choice and the person receiving it. Thename of every uals entering or leasing the building, or in the canteens, individual participating had previously beenwritten smoke or recreation rooms. against their desk position on the plan in orderto facili- The possibility that between-floor friendships wouldbe tate such cross-referencing. most numerous where floors shared thesame shift was The totals row of table 31 showsa very clear pattern: investigated and found untenable: not because ofany the number of socio-preferential choices decreases stead- contrary evidence, but because of the form in which the ily with distance. The only exceptionto the trend is the results were collected. Floorswere grouped together in case of distances of more than 36 ft, where the number an arbitrary way on the reply form, whereas the shifts is slightly above that for the 25to 36 ft group, and is pre- themselves are evenly distributed throughout thebuild- sumably the result of grouping together several possible ing. Therefore some departments in each of thegroup- distance categories. ings given on the reply form share thesame shift as the One interesting effect apparent from the results in col- experimental floor. But ifany relationship between umn 2 is that the younger age groups choose the highest inter-floor friendships and shift did exist, theeffect proportion of people from those withinone desk dist- would be diluted: the shiftsare staggered by only five ance, and the general trend is for this proportion to fall minute intervals and so there is a great deal of overlap with age. Conversely, inspection of column 6 shows anyway. that (here is a general trend for older groups to selecta higher Undoubtedly though, the most interesting findings aris- proportion of their workmate choices from theover ing from the data on between-floor friendshiplinks is 36 ft distance. This is an effect that might be expected their very great number. An average of 2.9 friendship from the tendency, already demonstrated, for choicesto links with people on other floors of the buildingcon- be made from members of thesame age group. As the veys the impression of an organization in which there is older age groups are in a minority, the individualscom- a great deal more interaction between individual mem- posing them must be relatively more isolated fromone bers of the company than might be supposed fromthe another, and the number of choices to greater distances way in which work is organized. For, not only are inter- re atively higher. personal contacts between floors minimised bythe Table 32 compares the distances involved in the socio- mechanical document conveyor but there is alsoan preferential choices made by men andwomen. The freq- active management policy of stoppingany between. uency distribution of choices at various distances is floor traffic. broadly the same, but there is a slight tendency formen The explanation no doubtowes much to the social prov- to draw a higher proportion of their choices from isions within the building though, for thereasons dis- further away. As with the older age groupsmen, being a cussed above, no causal link can be demonstrated. minority, are relatively more isolated fromone another. The social consequences of spatial relationships within Having been shown to be an important determinant of a restricted part of a building is, however, a matter more socio-preferential choice, sex would therefore beexpect- closely concerned with the hypotheses being tested. The ed to result in choices being made over somewhatgreat- data necessary to test the hypotheses is contained in er distances. tables 31, 32 and 33. Socio-preferential choice seems, therefore, to bea result Table 31 Distance of respondents from sociometric choice cross-tabulatedwith age group

Distance of respondent from choice

(I) (2) (3) (4) Age group Number of (5) (6) choices 12ft or less 13 to 18ft 19 to 24ft 25 to 36ft made by More than 36ft group Number PercentageNumber PercentageNumber PercentageNumber Percentage (of col I) Number Percentage (of col I) (of col I ) (of col 1) (of col 1)

15 19 469 187 39.9 129 27.5 64 13.6 41 8.7 48 10.2 20 - 24 261 III 42.5 78 29.9 34 13.0 22 16 6.1 2529 63 20 31.8 12 19.0 12 19.0 17.5 8 12.7 3034 48 18 37.5 10 20.8 12 25.0 4 8.3 4 8.3 3539 34 12 35.3 6 17.6 7 20.6 4 11.8 5 14.7 40 or more 145 46 31.8 28 19.1 -- - 19 13.1 19 13.1 33 22.8 Totals 1020 394 38.6 263 25.8 148 14.5 101 9.9 114 11.2

Table 32 Distance of respondents from sociometric choicecross tabulated with sex

Distance of respondent from choice

(1) (2) (3) (4) (5) Sex Number of (6) choices 12ft or less 13 to 18ft 19 to 24ft 25 to 36ft made by More than 36ft group Number PercentageNumber PercentageNumber Percentage Number Percentage Number (of col I) Percentage (of col I) (of col I) (of col I) (of col 1)

Male 329 114 34.7 75 22.8 63 19.1 40 12.2 37 11.2 Female 691 280 40.5 188 27.2 85 12.3 61 8.8 77 11.1

Totals 1020 394 38.6 263 25.8 148 14.5 101 9.9 114 11.2

117 'fable33 Summarywill,: of sociometric choices made and reciprocated within sections and departments. including the number of isolates ineach

(1 (21 (3 (4 (5) (6 (7 (8) (9 (10) Ipe ofNumber Number Aerage NumberPercent- Choices to Choices to ReciprocatedReciprocatedUnchosen °thee of rest,-of named number of resp-age of members of members of choices withinchoices withinindividuals space ondents prefer- of prof-ondentsrespon- respondents respondents single single (isolates Coln- ClICCS erencesnot dents on section own depart- section department pleting given given replyingnot ment form replying (Column 4 as per- Num- Per- Num-Per- Num- Per- Num- Per- Num- Fer- centage ber cent ofber cent ofber cent of bcr cent of her cent of of cols col 2 col 2 col 2* col 2* colt 1 plus 4

Open areas 174 779 4.5 40 18.7 498 63.9 710 91.1 147 37.7 207 53.1 8 4.6 Small aleas 67 241 3.6 14 17.3 195 80.9 229 95.0 80 66.4 84 69.7 6 9.0

241 'Totals 1020 4.2 54 18.3 693 67.9 939 92.1 227 44.5 291 57.1 14 5.8

The percentage is calculatedfromtwice the actual number of reciprocated choices. This was donebecause each reciprocal choice insolestwo choices taken from the total in column 2

of the simultaneously acting vectors "" deriving from age, composition of the sub-samples, differences in socio- sex and inter-personal distance. metric pattern must be attributable to differences in the Table 33 summarizes the sociometric choices made and size of area in which they worked. reciprocated within sections and departments, and the It was found that the social organization of the sections number of isolates found in each. The data from the working in small areas was internally more cohesive, small areas is compared with that from the remainder of though the proportion of isolates was higher, and the the floor. Column 1 shows the number completing and number of wider links with other members of thesame column 4 the number refusing the sociometric question. department much smaller. Column 5 shows that the percentage of workers in the The higher average number of socio-preferential choices large and small areas who did not state their preferences made by members of the open areas, coupled with the was almost identical. ''' lower proportion of reciprocations, shows that the Column 3 shows that there exists a very substantial socio-occupational network existing in the two types of difference between the average number of choices made area are fundamentally different. In the small areas there by workers in the open areas and those in the small. exists a fairly tight social group, whereas the social links People working in the larger office spaces produced the connecting people in the open areas are much less tightly greater outgoing choice volume: in sociometric terms knit. they exhibited a higher level of expansivity ' than workers in the small areas. Item FiveChoice of Working Position Columns 6 and 7 show that the socio-preferential Rationale and construction of theinstrument choices of workers in the small areas were very much This item represents a direct and empirical approach to more frequently made from among members of their the question of preferred office size. own section, and somewhat more frequently from mem- The stimulus was a projected photograph of the model bers of their own department, than were those of the used for the earlier paired-comparison items. The photo- open area personnel. Columns 8 and 9 show that this graph had been taken square-on, and from a higher greater proportion of choices made within a single angle, to present what amounted to a three-dimensional section or department was also paralleled by a much floor plan, though it differed from a floor plan in that greater proportion of choices reciprocated within them. it was more representational and less schematic. The fact that the proportion of intra-departmental re- Four working areas of different sizes were partitioned- ciprocated choices was much higher for members of the off from each other. The smallest area contained only small areas than members of the large is, though, a con- ten desks, the next largest contained thirty, the next sequence of the great number of intra-section recipro- sixty, and the largest one hundred. Except for the small- cated choices. In fact, the number of reciprocated est area, divisions of approximately the same size as each choices with members of the same department, other of the others in the model are to be foundon the floor of than members of the same section, is very much lower. the CIS which was chosen for the experiment andare The results therefore show that there exists a much therefore familiar to the respondents. The smallestarea greater degree of internal cohesion" amongst the mem- represents perhaps the idealisation of the small working bers of sections working within the smaller areas than area but there is only one example of it in the whole of amongst those working in the open ones. However, the the building, a wages section on the seventeenth floor. number of isolates is greater. Column 10 shows that, In response to the projected picture the respondentwas though their numbers are small, there are proportionally required to mark on a half-tone reproduction in there- almost twice as many in the small areas. If spatial relationships are important factors in determining the socio-preferential choice, then one might anticipate such 39. Vector: ' A directed magnitude in the life space;a force a result as a higher proportion of workers in a small area causing psychological locomotion '. (English, H. B. and English, must perforce be relatively physically isolated by work- A. C: A comprehensive dictionary of psychological and psycho- ing in a corner or beside a wall. analytical terms. London. Longmans, Green. New York. David McKay Company Inc. 1958) To summarize the results, it was found that both age 40. Many of the refusals were of the form no particular prefer- and sex were important determinants ofsocio-preferential ence', or' I get along quite well with most leople '. They are not. choice but, allowing for this, they tend to operate within therefore, necessarily a refusal to co-operate but may reflecta the framework of spatial relationships. As the sections genuine lack of strongly held preferences 41. Cohesion: 'The total field of forces inside and outside the working in both the open and the small areas were group, which tend to keep it intact '. (English, H. B. and English, essentially similar in both the nature of the work and A. C: op cit (39) 118 ply form, the desk at which he would choose to work. attaining the uniformity of conditions, equipment,de- The slide had a brilliance and tonal range much greater sign and occupation was out of the question. Even ifthis than the print and thus appeared very much more de- had proved possible, it would not have been possibleto tailed and realistic. get views uncomplicated by extraneous considerations Photographs of real offices would have been preferred such as the view from the windows,or distorted pro- for both this item and the paired comparisons one, but portions due to the photographic perspective. Itwas decided that photographs of real buildings would introduce too many uncontrolled variables, and thata model would be more likely to producea choice due solely to the respondent's preferred size of office. A completely open question was includedas to the basis for the respondent's choice, anda one minute pause allowed in the programme for a reply.

Form of analysis of data The total floor area from which the respondent could make his choice was divided into four smallerareas and numbered from I to 4, as in figure 94. Theareas themselves were then divided into zones and numbered from I to 4. Zones I to 3 each represent therow number of the desk from the window. Zone 4 contains the remaining desks in the centre of the floor. Area I has only 3zones as it is composed of only 3 rows of desks. The frequency with which eacharea and zone was chosen was then cross-tabulated with theage and sex of the respondent, and also the type of officespace in which he or she was working, ic, in one of the smallareas or in the remaining open areas (tables 34, 35 and 36).

Results and discussion of results The totals columns of tables 34 to 36 show thevery great Figure 94 The photograph of the model with the areas andzones superimposed.

Table 34 Choice of working position : summary of results cross-tabulatedwith age group

Age group Arca I Area 2 Area 3 Area 4 Totals Zone Zone Zone Zone

1 2 3 1 2 3 4 1 2 34 1 2 34

15 to 19 3422 35 5 1 4 18 6 3 I 12 2 1 0 126 20 to 24 24 1 0 17 2 I 0 9 1 2 1 5 0 1 0 64 25 to 29 5 00 1 4 00 3 1 1 0 0000 15 30 to 34 1 1 1 3 0 00 20 0 0 2 1 1 0 12 35 to 39 400 3 1 20 1 0 1 0 0 1 00 13 40 242 2 19 0 1 0 1 1 8 0 2 1 0 1 62

Totals 92 6 5 8110 5 3 41 98 2 21 5 3 1 292

Table 35 Choice of working position : summary of results cross-tabulated withsex

Sex Area 1 Area 2 Area 3 Area 4 Totals Zone Zone Zone Zone

1 2 3 1 2 3 4 1 2 3 4 1 2 3 4

Male 34 1 3 25 6 1 2 12 3 2 1 7 2 1 0 100 Female 58 5 2 1 56 4 4 296 6 1 14 3 2 1 192

Totals 926 5 81 10 5 3 41 9 8 2 21 5 3 1 292

Table 36 Choice of working position : summary of results cross-tabulated withtype of office space occupied by respondent

Type of Area 1 Area 2 Area 3 Area 4 Totals office space Zone Zone Zone Zone

1 2 3 1 2 3 4 1 2 3 4 1 2 3 4

Open areas 55 5 4 666 3 2 31 6 72 18 42 1 212. Small areas 37 1 1 154 1 2 10 3 1 0 3 1 1 0 ; 80

Totals 926 5 3 81 105 41 98 2 21 53 1 292

119 frequency with which a desk immediately adjoininga the sexes, though the type of office 'in which the respond- window was chosen. In all 81 percent of choiceswere ent was working did result in a significant difference. within Zone I, with the remaining frequencies dropping The main source of this difference was due to respond- steadily from Zones 2 to 3 to 4, ie, 10, 7 and 2percent ents in the open areas choosing Area 2 beyond the freq- respectively. uency expected from the total table, whereas respond- Tables 37, 38 and 39 were prepared by groupingthe ents in the small areas chose Area 1 beyond the expected zones together and summarising the results by choice of frequency. area only. This was done in order to allow for a chi- The final order in which the four areas may be ranked squared test of the homogeneity of the sub-samples,and on the basis of frequency of choice is 1,2,3,4, with very would have been impossible with the large number of little difference in the frequencies with whichareas 1 and empty or low value cells in tables 33 to 36. The method 2 were chosen. Both areas may be regardedas falling of k independent samples was employed. 4.2 within the definition of a small office; the results there- The chi-squared tests showed that the pattern of results fore show a clear preference for the smaller working was not significantly different between the age groups or area. Table 37 Choice of working position : frequency with which eacharea was chosen, cross-tabulated with age group Age group Area 1 Area 2 Area 3 Area 4 Totals 15 to 19 38 45 28 15 126 1.56 20 to 24 25 20 13 6 64 0.39 25 to 29 5 5 5 0 30 to 34 3 3 2 4 12 0.70 35 to 39 4 6 2 1 13 40 28 20 10 4 62 3.22 Totals 103 99 60 30 292 5.87 Df=9, NS Table 38 Choice of working position : frequency with which eacharea was chosen, cross-tabulated with sex Sex Area 1 Area 2 Area 3 Area 4 Totals Z. 2 Male 38 34 18 10 100 0.53 Female 65 65 42 20 192 0.27 Totals 103 99 60 30 292 0.80 Df= 9. NS Table 39 Choice of working position : frequency with which eacharea was chosen, cross-tabulated with type of office space occupied by respondent

Type of Area 1 Area 2 Area 3 Area 4 Totals office space X" Open areas 64 77 46 25 212 2.54 Small areas 39 22 14 5 80 6.73 Totals 103 99 60 30 292 9.27 Df =3, P< 0.05 Table 40 Summary of reasons given for choices* of working position CategoryReason Frequency Number Percent of respondents 1 Natural daylight 94 32.2 2 Near window (not further specified) 76 26.0nearness to 3 More or brighter light 44 15.1 windows 4 View from window 12 4.1 5 Not too near windows 7 2.4 6 Preference for office size (either large, mediumor small) 73 25.0 7 Privacy (usually back or side of room) 20 6.8inter-personal 8 To be surrounded by people 14 4.8 9 considerations Near airconditioning outlet (perimeter of floor) 12 4.1 10 Miscellaneous unspecified ** 10 3.4 11 No reason given 9 3.1 Total number of reasons given 371 n =292

* Some respondents gave more than one reason ** Idiosyncratic reasons andthose which were givenless than 3 times (le, by less than one aseparate category percent of people) were not assigned

42. Asin the analysis of the first CIS questionnaire.See: Wells, B. W. P: op cit (18) Table 40 summarizes thereasons given by respondents for making their choice such as the general level of officenoise, ventilation and as they did. It had been supp- heatingthat they might feel osed that the replies wouldhinge mainly on preferred to be involved, and to size of office as the mental avoid discussing thecontent of the document until after seta' for thinking about office it had been returned. size should already have been created and, except inre- Question 1 summarised the six spect of size, the areas from whichthe choice could be principal ways of physi- made were essentially similar. cally separating clericalworking units, and asked for However, the reasons reactions to them on given for making the choicewere more frequently in a five point scale ranging fromvery favourable to very unfavourable.Direct rankingswere terms of a positive desire to workbeside a window. not asked for as it was not (Combining categories I to 4 revealsthat 77 percent of known whether the alternatives could be rankedon a unidimensional scale, ie, respondents offered thisreason). Reasons involving x or y could both be office size or other inter-personalconsiderations cover very acceptable'. Indeed, any only 37 percent of respondents number of categories couldbe equally acceptable,or (combining categories equally unacceptable, fora variety of different reasons. 6 to 8). It would, though, bemisleading to suppose that Five point scaleswere also used for questions 2 to 6 for office size was nota major determinant of theway in which the choice responses to a number of questions aboutthe personal was made. The distribution of choices and supervisory to different areas and the consequences of different office sizes. way in which these have been The investigation, being shown to be influenced by therespondents' own work- exploratory rather than theory- orientated, posed questionsconnected with a range of ing experience make it clearthat it is. The resultsserve to show that such demonstrably topics from the way in whichclerks are treated in large important psychological offices, to the ease with which determinants of preferencebehaviour are not always supervisory duties may be apparent from a verbal report. carried on in large and small officeareas. Such items are empirical and thus have faceand not scale validity. On the other hand. question 7, which PART TWO : SECTIONCLERKS is the attitude scale de- Hypothesis to be tested scribed and used in Part One,does have scale validity. Many free-ranging conversations The penultimate item in thequestionnaire was acom- with section clerks (ie, pletely open question asking for supervisors) had conveyedto the experimenter the im- either an elaboration of the replies given to specific questionsin the questionn- pression that theywere the group most firmly opposed aire, or comments to the open plan offices. The on any other matter concerning comments recorded in supervision which was not touchedon in direct quest- chapter 10, derived from theformal group interviews ioning. held shortly after themove to the new building, typified their views. Greater familiarity The final item of the questionnairewas the paired com- with the open planre- parisons of alternative office sulted in no betteran opinion of it; supervisors still felt floor layout, the construct- that they and their sections would ion and presentation of whichhas been described fully be happier and more in Part One. effective in small office units.Accordingly, the following hypothesis may be formallystated: Hypothesis: Results Supervisory staffs will chooseto have their The section clerks ratings of own groups separated off from thoseof other super- the alternative methods of visors. de-limiting an officearea are given in table 41. Apart from testing this formal In view of the fact thatno one registered a reply in the hypothesis, the study also Indifferent' column, the data serves to establish what the floor supervisorfeels to be may be given simple the relative advantages anddisadvantages of the open plan office.

Population and samples used floor to ceiling The sample used in thestudy was every section clerk panel partitions working on the experimental floor.A total of 20 individuals were involved, all male. Frhigh. panel partitions Method of investigation A questionnaire (Supplement4), was addressedperson- ally to each section clerkand prefaced by a letter from Iwide gangways the experimenter (Supplement3). Both questionnaire and letter were themselves sentout under another covering letter from the General Manager,which requested the recipient to co-operate fully in clear glatitions ss completing the question- par naire and to attend agroup session which had beenarr- anged. The questionnairewas circulated on Tuesday, 19th May, with the instructions that it was to becom- lincabinets pleted by the afternoon ofTuesday 26th May, and tables brought along to a group sessionin the recreationroom where the final item could becompleted. I The recreation room sessionwas held at 3.0 p.m. The no ph managers who were involved in the study(whose re- I sults are presented separatelyin Part Three) also att- 1 1 1 1 1 1 1 1 1 1 1 ended. The first itemon the agenda was to project the 1 1 1 1 302520 paired comparison slides, whichcomprised the last item 15 10 5 0 5 10 15202530 on the questionnaire. Following thisthe experimenter WEIGHTED SCALE VALUE gave a short talk on his research interests and, after tea, FAVOURABLEIUNFAVOURABLE led a discussionon the psychological implications of as- Figure 95 pects of office design, with particularreference to office size. Ranked order and degree offavourablenessunfavourableness to the alternative methods of dividingoffice space. Construction of the questionnaire The instructions givenon the questionnaire requested the respondent to reply 43. Mental set: 'a to the questions strictly interms preparatory adjustment or readinessfor a of supervisory considerations, particular kind ofction or experience'. (English, H. to ignore any others English, A. C: op eft (39) B. and

121 Table 41 Section clerks' ratings of alternative methodsof dividing a single officespace for 4 to 6 sections, comprising 100 to 150 people.

Type of space division Very Moderately Indifferent Moderately Very Totals favourable favourable unfavourable unfavourable (i) Sections not physically separated from one another I 2 0 5 I2 (ii) Five foot high panel 20 partitioning between sections II 2 0 3 4 (iii) The use of filing 20 cabinets, telephone and postal tables, etc as boundaries between sections 0 6 0 5 9 (iv) Floor to ceiling panel 20 partitioning between sections 15 2 0 1 2 (v) Wide gangways 20 separating sections 2 4 0 4 10 20 (vi) Clear glass partitioning between sections I 5 0 3 11 20 n=20 graphic expression on a bi-polar frequency histogram in the histogram. (figure 95). In drawing this, the Very favourable' and The table of results and histogram together showa very `Very unfavourable' replies have been givenan arbitrary strong preference for panel (ie, opaque) partitioning, r weighting, and the 'Moderately' held replies givenno and for partitions extending from floorto ceiling in pref- added weight. Each alternative has been ranked in the erence to partitioning only 5 ft high. Widegangways, histogram downwards from the most favourablealter- clear glass partitioning, and theuse of filing cabinets, native to the least favourable. Even if no differential etc., as methods of separating sectionswere ranked in weighting was given to theVery favourable' replies, slightly ascending order but occupied much the ranked order would not change, the same except that alter- unfavourable position in the ratings. Theabsence of natives (v), (vi) and (iii) would occupya tied rank rather any physical separation between sections was regarded than being separated by marginal differencesas they are very unfavourably and ranked last for acceptability. Table 42 Section clerks replies to questions no 2 to 6 QuestionQuestion Replies number Definitely Moderately Undecided Moderately DefinitelyTotal Yes Yes No No 2 Are the members of a section more aware of themselves as a member of a particular section if they are physically separated from other sections? 14 3 0 2 1 20 3 Is there any tendency for people to be treated less as individuals and more as functionaries in the larger offices? 11 6 0 2 1 20 4 Do you feel that the actual size of clerical areasthat is, the number of people working in any one spacehas any real importance from the supervisory point of view? 18 0 0 1 6 If it was a question of 1 20 choice, would you choose to have your section located by themselves in a screened off area ? 15 2 0 1 2 20 Large Small Definitely Moderately No Moderately Definitely Total easier easier difference easier easier 5 In your experience is it easier or less easy to carry out the specifically supervisory duties of a section which is situated within a large clerical area or a small one? 0 3 1 4 12 20 122 1 he results make it clear that the section clerks' prefer Table 44 ences arc for as much pm acyas possible for their Se,:tion clerks: results of paired comparison study of sections. office layout preferences Questions 2 to 6 sought section clerks'lox s on other as- pe:ts of space di% ision. The frequency with which the I. M N 0 P Totals respondents selected particular categories of answerto explain their attitude is given in table 42. 3 15 16 4 38 Ihe results are quite clear and demand no statistical 17 16 17 16 66 treatments. Theyrepresentconsistently favourable N .5 4 4 5 l8 beliefs about the personal and super% isoryconsequences 0 4 3 lb 5 28 of working in small and separate units. !) 16 4 15 IS 50 Section clerks selected the statements composing the attitude scale to office size with the frequencies given in Totals 42 14 62 52 30 200 table 43 and they may he rankedas shown. Preferred rank order 3 5 I 2 4 'table 43 n 20 Rank order of section clerks response to attitude scale A simple expression of the preferred order, reading from left to right, is for the area divided in the followingways: Rank Statement Frequency order chosen NOLPM r r r 1 The larger offices make one feel 16 relatively unimportant The large office creates less feeling 12 J l__ J J L of permanence and security thana 777, small one The replies thus show a consistent preference for the 3 'there is an uncomfortable feeling 10 smaller partitioned area over any larger grouping. of being watched all the time ina large office Summary of results 4 The social life of a large office is 9 To summarize the results for this part of the investiga- likely to he better than that of a tion, it was found that a strongly unfavourable attitude small one existed amongst the supervisors towards the open plan Any disadvantages of the larger 4 office. The replies to every one of the questionssupp- office are offset by greater advantages orted this view. A discussion of why this should beso, 6 A good working atmosphere is more 3 how the results relate to those of the clerks andmana- likely to be found in a large office gers, and the implications for the hypothesis tested, will than a small one be postponed to the conclusions at the end of theapp- I The size of the office itself is not endix. really important 7 A large office is definitely better to 2 PART THREE: MANAGERS work in than a small one Hypothesis to be tested Like the questionnaire addressed to section clerks, the one addressed to managers was intended to serve the dual purposes of assessing the degree of approval felt Taken in the aggregate, the pattern ofresponse of each towards open office planning and of establishing what individual displays a generally held and stronglyun- were felt to be the relative advantages and disadvantages favourable attitude towards the open plan. of open plan design. The informationwas also required Question 8 was an entirely open question which sought to make comparisons between the attitudes and require- an elaboration of the replies made to the previous quest- ments of managers, supervisors and clef ks, with a view ions, and also for any other comment the respondent to establishing the various factors involved in drafting wished to make. user requirement reports. The predominant comment made was that it is,or would The general hypothesis, relating to the findings of each he. easier to create an 'esprit de corps', 'good morale' of the three parts of this appendix, is that differences will or sense of belonging' in the smaller office area. Com- exist, and that these differences will dependupon the re- ments of this sort occurred in 60 percent ofcases. The spondent's functions, expectations, and experience. In difficulty of keeping track of the whereabouts of staff in the case of managers, the general impression gained the large area was mentioned only slightly less frequent- from them in the course of interviews and conversations ly, ie, in 55 percent of cases. was that, because of the organizational flexibility allow- Most of the comments made are summarized inthis ed by the open plan, it had considerable advantages single reply: over a number of smaller offices. There were, however, some 'II would) strongly advocate theuse of full partition- reservations about the advantages to be gained from ing between sections as it eliminates extraneous noise, having a completely open plan as itwas felt that too ensuring greater concentration on one's work. Small- large an area could be overwhelming for theperson er units have the effect of people understanding each working on it. other as individuals and fosters a better team spirit. The single hypothesis to be testedmay therefore be In a large unit the tendency is for people to wander off formally stated as follows: to chat with friends on another section whereas ina Hypothesis: Managers will tend to prefera small num- small unit, it demands a little morecourage and is a ber of open office areas to a larger number of smaller more positive action to walk out of the unit for other ones. than recognised reasons'. Table 44 summarizes the results for the pairedcom- Population and sample used parison study of office layout preferences, and shows The personnel selected to take part in this studywere the frequency with which the office layout designated at senior executives of the company, all of whomhad the column head was preferred to the layout designated management responsibilities. They were selectedon a by the row.- random basis from the list of suitably qualified individ-

123 uals.I ifteen such people were involved in the study. therefore mean that the rank order of the variousrank- being a one in four sample. All of the subjectswere male ings has tended to he high: thus the lower the ranking as no women are emploved at managerial level. index the higher the rank order. The only exception to the random choiceof the sample The alternatives were ranked as follows: was the deliberate inclusion of the threemanagers who work on the floor on which therest of the investigation 'Fable 45 was conducted and w ho are responsible for the personnel Rank order of managers' preferences of size ofclerical taking part in the previous two parts of theinvestigation %%01-king area described in this apperdix. The inclusion oftheir views and attitudes was felt to he especiallyappropriate. RankType of layout Ranking order index Method of investigation The method of investigation for thispart of the study Open general clerical area butsome 19 was exactly similar to that described for the section separate managerial and specialist offices clerks in Part Two. A different questionnairewas sent out (Supplement 61, and a different covering letter from A moderate amount of partitioning 17 the experimenter (Supplement 51, butnothing else was to produce rather smaller clerical areas in any way different. 3 Completely open office space for the 50 whole department withouteven Construction of the questionnaire private offices The questionnaire to managers is basicallysimilar to the A complex of small work cells 54 one to section clerks insofar as it is concerned primarily perhaps with each containing a with the effect of room sizeon the respondent's pro- single section fessional functions, and toa lesser extent with personally held feelings. It would not be expected that people The rankings show that the open clericalarea with some would be able to divorce one from the otherentirely, but separate managerial and specialist offices the instructions with the questionnaire was by far the asked the re- most preferred alternative. This is in fact thesystem spondent to make every effort toanswer the questions being operated by each of the respondents strictly in terms of management considerations. at the present Scope time. A moderate amount of partitioningwas the next for personal feelings was specifically allowedfor by the most acceptable alternative, but the very different alter- inclusion of the attitude scale used in the other question- natives of completely open office and complex ofsmall naires, the rationale and construction of whichis des- ones appeared to be almost equally unattractive. cribed in Part One of this appendix. The problem posed by question 2 in the questionnaire Question I called for a direct ranking of fourmethods of was: dividing a large office area, ranging from the completely Other things being equal, in which size ofgeneral open to the extensively divided. Question 2 required the office would you expect to find the best morale,and manager to assess the optimum size for a general office from which would you expect the highest relativepro- on the separately presented criteria of morale and prod- ductivity?' uctivityfive arbitrary unit sizes were presented as a Five office sizes were offered, together witha rejection checklist. Question 3, which had also been included in clause and were selected with the frequenciesshown in the section clerks' questionnaire, asked foran assess- table 46. ment of whether people working in large offices tended to be treated more as functionaries and lessas individ- Table 46 uals. Each of the three questions is empirical in nature Rank order of managers' expectation of sizeof general and, having face value only, requiresno special techni- office producing best morale and high productivity ques for construction. As the managers might be expectedto be the most Rank Office size FrequencyFrequency verbally fluent of the groups covered by thepresent sec- order chosen forchosen for tion, and as each must perforce have devoteda certain morale product- amount of thought to the layout of his department,two ivity completely open-ended questions (questions4 and 5) were included. These called for summaries of themanagerial 40 to 60 persons 5 6 advantages and disadvantages of theopen plan office. 2 60 to 100 persons 4 4 Question 7 was another open-ended question.It was 3 20 to 40 persons 3 2 provided to enable the respondentto elaborate on his 4 10 to 20 persons replies to the other questions and also for himto make 5 100 or more persons any other comment, from a non-managerial point of Size of work unit not view, on the influence of the size of officeareas. relevant to questions I I

Results It can be seen that there wasno difference between the Question I presented the respondent with the following ranked order of the alternatives for moraleand prod- problem, and asked for the answerto be presented in the uctivity, though their frequency of choice is following way : slightly different. Indeed, apart from only two exceptions, ' If you were asked to plan all the your department again respondents ranked morale and efficiency togetherin from scratch, and were applying onlythe criteria of respect of a given office size. One manager rejected the efficiency and productivity, which of the following connection of either with roam size, but the resultsshow alternative arrangements would be your first choice, the strong tendency on the part of hiscolleagues to per- second, and so on. The alternatives are conceived as ceive a connecting link between morale, efficiency,and referring to general clerical tasks, butexclusive of a given size of working group. typing pools and machine areas whichpresent special The picture of preferences presented is problems. not clear, in the sense that it does not indicate a trend from smallto large Please put a "1" in the box besideyour first choice, a or vice versa. However, the majority of the choices, "2" beside your second choice, and for so on.' both morale and productivity,were directed at the two For the purpose of summarizing the results, thenum- categories of office size containing between bers signifying the order assigned to each them 40 to alternative 100 persons. This, then, wouldseem to be the optimum were summed : this yielded a convenient index for rank- size by a majority vote. ing the items. A low score against an alternative will In reply to question 3, 'Is thereany tendency for people 124 to be treated less as individuals and inure as function- Table 47 aries in larger offices?' the distribution of replieswas as follows: Rank order of managers'response to attitude scale Deli n - Moder- Un- Rank Moder- Detin- Statement Frequency itely ately decided order ately itely chosen Yes Yes No No 2 3 () 3 7 Any disathantages of thelarger 9 The results support thegeneral proposition that people office are offset bygreater are not treated less as indi%idualsin the larger offices advantages but as Ike of the replies, which represent a third ofthe The social life ofa larger office is 8 sample, were of the oppositeopinion, a goodmeasure of likely to be better than thatof a dissension must be notedas existing. small one Question 4 was a completely open one which called for The size of the office itselfis not 8 a summary of the managerialadvantages of the open- really important plan type of office. 4 A good working atmosphereis more The number of 7 reasons given in response to thisquestion likely to be found ina large office averaged 3.5 perperson, comparing very favourably with than a small one the number of 5 reasons and comments given to questions There is an uncomfortablefeeling 5 and 7 which averaged 4 only 1.8 and 2.0per person re- of being watched all thetime in a spectively. large office The following nine advantagesof open-planningwere The larger offices makeone feel extracted: 3 6 relatively unimportant 0) Easier overall supervision as all stall* arc normally A large office is definitelybetter to 3 in the view of the supervisor (. work in than a smallone (ii) Greater economy and 8 flexibility of space The large office creates lessfeeling 2 (iii) Facilitates transferof work and stall* betweensect- of permanence and securitythan ions a small one (iv) Most common recordscan be kept within the one area thus reducing excuses to leaveroom than the managerial (v) More opportunity for better one. type staff to become Most of the ifnswers given acquainted with all aspects ofoffice routine were vdriations of the ad- (vi) Better layout of the vantages or 'disadvantages of theopen office which have sections permitted already been summarized (vii) Uniform standardsof discipline can be operated, for questions 4 and 5.The and be seen to operate range of the additionalcomments was therefore rather restricted but the sentiment ( viii) Wasted journeyscan be saved if one can see wheth- expressed most frequently er someone is at their desk was the one crystallized in the followingquotation from or not a reply form : (ix) Office maintenanceand cleaning is simple and cheaper 'The calibre of themanager is more important in de- Question 5 was the termining the attitude of thestaff to their job than is converse of the previousone and the size of office'. called for asummary of the managerial disadvantages of the open-plan office. This gives, perhaps,a clue to a puzzling result in the The interpretation of response to question 6 where, althoughthe statements any such question can only be selected in the attitude scale done impressionistically. Afrequency ranking can seldom were predominantly favour- be applied rigorously able to large offices, thestatement that 'the size of the because of the way in whicha number of fairly similar office itself is not reallyimportant' was ranked injoint statements neverthelessexpress second place. It subtle, and sometimesvital, differences. In this seems therefore that themanager re- case gards office size asa factor very much subordinate many of the replies were similar butto sum and rank to the them would be to give quality of the managementand supervision. a spurious degree of accuracyto The question of the effects findings. Instead, fourstatements are quoted, direct upon the individual of work- from the reply forms, ing in a large officearea produced two distinct points of as being representative of widely view. The first held views. Of these, the saw the open plan as a possible threatto first one expresses themost the individual; the second frequently stated opinion: either saw no such threator (i) There are too saw only positive personaladvantages. The following many distractions with somany respondent is typical of the people, and any distractionaffects larger numbers minority which saw theopen (ii) A friendlier plan as a possible threatto the individual: group work more usefully anda vast 'Too much emphasis plan doesn't promote this, cannot be placed on the vital only pockets of friends importance of making who are a distractionto all others when it is their every effort to let the staff inan turn to relax and chat open (and large) office know thatthey are individuals whose opinions andsuggestions for the betterment (iii) The morale of themore ambitious staff tends to be of lowered by the absence of the office are both soughtand acted upon by the visible 'status' symbols management wherever possible. and a feeling that theirqualities will be submerged Without this interest in the unidentifiable by the management thestaff morale and efficiency mass surrounding them. This will deteriorate, and could mean the loss ofsome of the better staff absenteeism increase. (iv) Managers tend The ideal officearea would seem to beone in which to become involved in routine about 50 staff are employed, since this numberappears matters, particularly those relatingto staff. to be the maximum capable The statements comprisingthe attitude scale to office of assuming an integral identity to whicheveryone feels they contribute size were selected withthe following frequenciesand some- may be ranked as in table 47. thing; and as a result theywork in harmony with each other and take pride in the Taken in the aggregate, thepattern of response of each efficiency of their " unit" individual represents compared with others. Thelarger the "unit" themore a moderately favourable attitude impersonal does it become on the part of managers towards theopen plan office. and the members tendto Question 7, like questions belittle the value of theirown particular work'. 4 and 5, was completelyopen. The following quotation is It drew attention to therestrictions imposed by the taken from the reply formof specific nature of the a member of the group who feltthat the open plan offers previous questions, and invited clear psycho-social advantages. elaboration of any of thereplies given. It also invited The quotation couldnot the respondent to be described as typical,but adds a littlemore to the dis- comment on the problem of the closed cussion : versus the open plan office from a point of view other ' I think large offices can have a "humbling'' effecton

125 the cocky or pompous but do not necessarily retard course, question 5 which expressly sought thedisadvant- the indiv idualist. Indeed, the dominant personality agesof open offices) showed that the managerswere, as seems to prosper and good examples of dress and be- a group, well disposed to large office units,and an ex- ha% iour seem to rub-off" onto the new entrants'. tensivelist of managerial advantages was putforwardin That the tendency exists for people to prefer smaller reply to question 4 in support of their view. areas1as referred to, butithout it apparently being However, a smaller number of direct reservations about regarded as a %Aid determinant of organizationalpract- the completely open office were recorded in replyto ice: question 5 and were implicitin the replies to otherquest- The "enclosure" system is historical in England and ions. In comparing the results of question 8 withthose boundaries are often (seemingly inadvertently) mark- of questions I and 2, it was concluded that themanagers' ed by clerks themselves in their distribution of office position is best reflected by a preference for moderately machinery. If no partitioning is provided, and if there large rather than very large orvery small office areas. is no positive action taken by management,one will Such a conclusion would seem to be congruent with the often find "fencing" by filing racks. The urge not to range of comments made in reply to questions4,5 and7. get mixed up with other people doing different work Thediscussion of results and their relationshipto the is strong 1 would say in British bones,open plan or hypotheses tested in this, and the other parts composing not, but I am strongly against people setting up their this appendix, will he treated fully in theconclusions own shop in too small groups'. which follow. Table 48 gives the results for the paired comparison choices: it shows the frequency with whicha given office layout, as designated by the letter at the head or the CONCLUSIONS column, was preferred to the layout designated by the This passage draws together the research findings ofthe row. three previous parts of the appendix in orderto test the formally stated hypotheses and draw the conclusions Table 48 from them. Its form is firstly a consideration of the Managers: results of paired comparison study of office hypotheses and the extent to which theyare confirmed layout preferences by the data; secondly a summary of the findings.

L MNOP Totals Part oneGeneral clerical workers Hypothesis 1: The smaller workingarea will generally 6 I 4 8 19 be preferred to the larger. M 9 4 7 9 29 The relevant results for the testing of this hypothesisare N 14 1 1 1 3 1 1 49 those from items two, three and five, all of whichpro- 0 11 8 2 12 33 vide clear confirmatory evidence. P 7 6 4 3 20 Item one, the attitude scale, showed that the generally held attitude towards the large open plan officewas un- Totals 41 31 II 27 40 150 favourable. Additionally, a rank correlation of the order Preferred of the statements selected by the clerksas representing their views with the order in which the statements had rank order I 3 5 4 2 been ranked by the judges used in constructing the scale, --15 proved to be positively significant. In other words, the trend was for statements whichwere progressively less A simple expressionofthe preferred order may be favourable to large offices to be progressivelymore freq- schematized as follows, in a descending order of prefer- uently chosen. ence from left to right : Item three, the paired comparison preferences, showeda clear rank ordering of the various sizes of officearea in a P M 0 N descending order of preference from the smallestto the largest office sizes. Item five, the choice of workplace, also showeda clear preference for the smaller size of office. The results from each of these separatesources yielded the same result: the smaller workingarea was preferred to the larger. The hypothesis was therefore confirmed. Perhaps the most interesting feature of the results is their relationship to those of question 2. The number of Hypothesis 2: People working in largeopen areas will people who could work in each of the various office sizes be relatively better disposed towards largeareas than portrayed in the model were, taking the first two rank- those who work in smaller areas. ings, 50 and 100 respectively. This is closely in accord The data for such comparisons dependsupon contrast- with results of question 2 in which office sizes of between ing the results derived from people working inthe three 40 and 100 people were selected by the majority ofre- small semi-partitioned areas with those frompeople spondents as promoting the best morale and product- working in the remainingopen areas. The results bear- ivity. The smaller unit sizes were unpopularon both ing on the hypothesis are contained in itemstwo, three rankings, as were they also unpopular in quc.,:ion I. and five. The completely open office was ranked thirdon the In response to the attitude scale composing itemtwo, it paired comparisons, but fifth when rankedon the crit- was found that the scale position of the respondents in eria of promoting morale and efficiency in question 2. the smaller areas was very much lower than that ofthose However, in reply to question 1 the completelyopen in the open ones. Both groups held negative office ranked third attitudes almost tying with the small units towards large open offices, but the attitude ofthe people for fourth (and last) place. The relatively small changes in the small areas was considerablymore negative than in position are perhaps due to the criteria used for the that of people in the open areas. A chi-squaredtest was rankings. But, whatever the explanation of these differ- performed on the frequency distribution withwhich the ences, a moderate amount of partitioning was found to items composing the scale were checked bythe people in be preferred to either a completely open or a completely each of the two types of area. The respondentsfrom the closed office plan. small areas reported, with a greaterfrequency than would be expected from the whole table,that the large Summary of results office makes one feel unimportant, and that itcrates The individual findings of all the questions (except, of less feeling of permanence and security thana small one. 126 Conversely, respondents in the largeareas reported w ith therefore confirmed. greater than the expected frequency thata good working Column 10 shows the number andpercentage of isolates atmosphere was more likely to be found in a large office found in the large and smallareas. Again, the differences than a small. The differences were significantat the 0.02 approach the order of 2:1i.e. 5 percent in theopen areas level of probability. The conclusion derivedin present- compared with 9 percent in the smaller.The numbers in ing the results of item twowas that, though both groups both cases were small but the oppose large offices, those working in them differences marked. The were less second part of the hypothesiswas also confirmed, and opposed to them than were those working in small areas. thus the entire hypothesisas stated. The results of item three, the paired comparisons,failed to show any such effect. The frequency tables prepared Hypothe.sis 6: Time will be marked separately for the respondents in large and differences between small areas the attitudes of clerks, supervisors andmanagers to open gave a similar ranked preference of the different office office planning. sizes, both consistently preferring thesmaller to the Differences in the attitudes of clerks, larger. supervisors and managers towards open office planning However, the results for item live, were manifest in the choice of work- all of the results in which comparisonis possible. The place, again revealed a difference in the response patterns most direct comparisons may be made withthe replies of those respondents in the large andthose in the small to the attitude scale and paired comparison working areas. A chi-squared ranking of test of the data revealed preferred office layouts, whichwere common to all three significant differences at the 0.05 levelof probability. groups. Both groups preferred the two smaller working areas In reply to the attitude scale, itwas found that the group but, of the two, the respondents working inthe smaller scale scores of clerks, section clerks andmanagers were areas chose the smallest area in the photograph signifi- 0.907, 0.750, and 1.307 respectively.The neutral point cantly more often than the respondentsworking in the on the scale was 1.125, and it is clearly large areas. The converse also held. apparent that section clerks hold a relatively lessfavourable personal Three sources have been referredto as providing the re- attitude towards open planned offices than sults against which the present hypothesis do either of can be tested. the other groups, and that the attituleof clerks falls One only of these, item three, failedto yield any evid- within the unfavourable half of the scalewhereas the re- ence that respondents working in largeopen areas are plies of the managers are favourable. comparatively better disposed towards large offices than The paired comparison photographswere ranked by are respondents from the smaller areas. The result isper- clerks and section clerks alike in descending haps not altogether surprising though order of as both groups preference with increased size of officespaces. Managers were opposed to the large office and, as the differences differed from both in the form of their replies, in their attitudes were opting for ones of degree and not kind, it moderately large and completelyopen areas in prefer- might be expected that some items wouldbe relatively ence to small ones. insensitive in reflecting smaller attitudinaldifferences. Further differences between the attitudes of However, as the result of this supervisors one item is neutral rather and managers were apparent from thereplies to the than contradictory to the hypothesis, and as items two question of whether it was felt that therewas a 'tendency and five both yielded positive results, thestated hypo- for people to be treated less thesis is held to be confirmed. as individuals and more as functionaries in larger offices'. Seventeenout of twenty section clerks agreed with thisstatement compared with Hypothesis 3: The number of friendshipchoices made five out of fifteenmanagers. and received between people will be progressivelyfewer Section clerks opted for small enclosed with increased distance. work units and gave strong supporting reasons. The The totals row of table 31 summarizes all managers, by con- the d-:a nec- trast, opted consistently for moderatelyopen areas. essary to test this hypothesis. The results showa clear In summary, the most marked differences trend for the number of socio-preferential of attitude to- choices to wards large clerical areaswere those between supervis- decrease with the increase of distance betweenpeople. ors and managers. However, the attitude of supervisors One exception to this was noted and discussedbut as (as measured by the attitude scale) this does not appear to contradict the hypothesis, was shown to be con- it is siderably more negative towards largeoffices than was therefore held to be confirmed. the attitude of the clerks. The hypothesis was therefore confirmed. Hypothesis 4: The number of inter-sectionpreferences expressed will be greater in theopen areas than in the Part two Section clerks small partitioned ones. Hypothesis: Supervisory staffs will chooseto have their The data needed to test this hypothesis is contained in own groups separated off from those of othersuper- column 6 of table 33. Of the total number of Named pref- visors. erences given, 64 percent of those made by respondents A question asked directly whethersupervisors would working in the open areas were madeto other members choose to have sections in of the same section. Thus 36 percent of separate screened-off areas*, the socio- the results were strongly in favour ofthis. Replies to preferential choices must have been directed towards other questions made it clear thatthe screening of sect- members of some other section. By contrast, 81percent ions was regarded as very desirable and of choices made by respondents working in underlined the the small strong preference of supervisors for smallareas rather areas were directed towards members of theirown than large open ones. section and thus only 19 percent to members ofother The hypothesis was therefore confirmed. sections. The hypothesis was thus confirmed. Part threeManagers Hypothesis: Managers will tend to prefer Hypothesis 5: There will be a higher proportion a small num- of re- ber of open office areas toa larger number of smaller ciprocal choices in the smaller workareas than in the ones. larger ones, though a greater number of isolates. The most direct confirmation for thishypothesis came Column 8 and 10 of table 33 contain thedata necessary from the paired comparison photographs, to test this hypothesis. where the layouts divided into four units of fiftypeople and two Column 8 shows the number and percentage ofrecipro- units of one hundred were ranked in first cated choices made within a single section, and and second compares places respectively. The completelyopen plan was rank- the large open areas with the small. The differencesare ed in third place; the six-part division in gross : fourth, and the 38 percent in the large areas compared with66 eight-part division in last place. percent in the small. The first part of the hypothesis is The attitude scale showed that themanagers tended to

127 prefer large office units to small. and the results ofother Four main conclusions may be draw n from the results questions showed that their preferencewas for the med- of the study reported in this appendix: ium-large rather than the very large or the%co, small I. Despite any differences in the pattern of results for areas. sub-samples, all groups except managers prefer to The h> pothesis was therefore confirmed. have small general clerical areas rather than large ones. This confirms one of the results of the first CIS Summary of findings questionnaire. which is discussed in chapter 10. The broad frame of reference w ith which this study 2. There is a considerable dissonance between the started was to examine variously the attitudes of clerks, views of managers, supervisors and clerksas to what supervisors, and managers to large offices, andto de- constitutes the most satisfactory type of officearea. scribe some of the objectiveconsequences of different The several bases for these views, and their actualorg- size groupings for the formation of social relationships. anizational. personal and social consequences, have It was found that clear differences existed between the not yet been the subject of experimental investigations. attitudes of the carious occupational categories. Mana- However, they present problems of considerablecon- gers were found to prefer fairly large general offices, sequence and should therefore he assigned a priority whereas clerks preferred small ones, and supervisors in subsequent management and user-requirement were more extreme than the clerks in their preference for studies. small offices and their disapproval of largeones. 0 It was 3. The results showed that people working in large also found that, though both groups of clerkswere opp- open areas are better disposed towards them than are osed to large office areas, those in the smallones were people working in small semi-enclosedones. This much stronger in their opposition than those actually seems to suggest that some form of prejudice is opera- working in them. tive such that people are ei0-er better disposed to- The results thus raise an important question from the wards what is familiar, or les.. well disposed to what management or user-requirement point of view of the is unfamiliar. On either explanation, the findingcon- relative weight to be given to the attitude expressed by veys the implication for user-requirement studies that the managers, supervisors and clerks. Managers offered they should only be conducted with subjects having good managerial reasons why a course oppositeto the first hand experience of the conditions being investi- course advocated by supervisors should be followed in gated. the division of space and supervisors similarly justifieda 4. Friendship patterns were shown to develop in quite course opposite to the managers. The attitudes of the different ways in different types of working area. A clerks must also be given some consideration, if only to working group, housed in a small semi-partitioned ascertain whether to ignore them would haveany re- area, was found to be more internally cohesive than a percussions in terms of lowered working effectiveness. similar working group surrounded by others ina large So far these questions have nowhere been dealt with, area. Conversely, groups working in the small areas though buildings continue to be built, and the decisions were found to have fewer friendship links with other concerning the size of working spaces, taken. parts of their department. Such matters are like!y to The objective consequences of different office sizesare have considerable effects on the working effectiveness contained in the results of the socio-preferential study. of the small and large unit and to have occupational, These revealed marked differences in the socio-prefer- social and personal consequences for theoccupants. ential response patterns of people working in the small I t therefore seems that this topic also should be made semi-partitioned areas and those working in there- the subject of further studies in management and ind- maining open areas. ustrial psychology. It was found that the pattern of choices dependedupon three factors: age, sex and inter-personal distance. The results also showed that people in the largerareas made a greater average number of choices, that very many more of these were directed outside of the section in which they were made, and slightly more went outside of the department. It was also apparent that people in the small areas reciprocated very many more of the choices within their own section, and had fewer reciprocated choices with members of their department other than those in their own section. It was found that therewas a higher proportion of isolates in the smallerareas than in the large, though the numbers were small in eachcase. The above, then, were the objective differences foundto result from locating working sections in large and small areas. The existence of these differences raises the management question of which one is to be preferred. The intra-group cohesiveness of the sections in the small areas meets many of the criteria for small group effectiveness. On the other hand, the sections in theopen areas have better interconnections with other parts of the department and may therefore constitutemore effective total working units than thesum of individual sections. The results so far have shown differences in socialgroup formation which have resulted from the conditionspro- vided by the building. To attempt to establish which social or working organization is the most desirable would be to go beyond the scope of thepresent report and involve considerations wholly divorced fromthe design problem. Suffice it to say that considerable social consequences have been found to result from different 44. It has been said that there is an element ofstatus seeking in the types of space provision, and that these differences justi- supervisors' desire for small spaces and that theymay not like to fy further investigation of the buildingas a management feel that they arc under the direct supervision ofa Chief Clerk, though experience has shown that, in the main, suchsupervision tool. is necessary 128 Appendix 5.1 Supplement 1

THE UNIVERSITY OF LIVERPOOL DEPARTMENT OF BUILDING SCIENCE THE UNIVERSITY. LIVERPOOL 3 IELENIONE : ROYal 6022Ext. 353

Dear Sir or Madam,

This document is being circulated by staff members of Liverpool Universityas part of our continuous study of the design of modern office buildings, and with particular referenceto user requirements. The replies to the questionnaire will, of course, be treated absolutely confidentiallyand dealt with in terms of generalisations, not specific cases. The work is supported both by themanagement of the CIS and by the Guild, and therefore represents no sectional interest whatever.

We should be very grateful for your co-operation in this study: after all, designis only successful insofar as it meets people's needs and preferences. We are, therefore,once again referring our problem to you as users for assessments of some of the matters we are investigating connected with office planning.Our broad intention is to assess the conditions which your experience indicates lead to the most satisfactory workingconditions. We can then relate this information to the technical design problems and attempt to describe theway in which the most favourable working conditions might be created. The precision with whichwe can attain this goal can, of course, be no greater than the precision with which you complete this document. We therefore hone thatyou will feel that the trouble involved in making such a careful appraisal will be well worth while interms of the improved conditions that you are helping to devise, and that you will be kind enough to help us in thisway once again.

Yours faithfully,

B. W. P. Wells Appendix 5.1 Supplement 2

PART ONE Personal DetailsPlease put a tick in box beside appropriatereply

Name (Initials) (Surname)

Sex: MALE FEMALE Grade: BASIC A L1 BASIC A Li BASIC B BASIC B Li GRADE 3 Li BASIC C GRADE 2 ri MACHINE OPERATOR L_J GRADE 2A LI TYPIST GRADE I GRADE 2A P OTHER (please write) GRADE I OTHER (please write)

Age group: 15- 19 15- 19 Li 2024 2024 2529 2529 3034 3034 35 -39 3539 LJ 40 or over 40 or over

Department

Name of Section Clerk in charge ofyour Section PART TWO Spatial Considerations The following eight statements each refer to the size of general office workareas and to personal feelings about them. The context of these statements is thecontrast between the open planning of the general andthe subdivision of the office (' the large office') same area by partitioning ('the small office'), conceivedas being different unit sizes of the same company. The small office'may be thought of as containing about 25 people; people. The large office' about 75 Please put a tick in the box nextto the statement or statements with which beside the statement with which you are in agreement, and a cross you disagree, thus putting either a tickor a cross against every statement. Will you also put a second tick in the box against the statement with which you feelmost stronglyin agreement, ie, es 1. A good working atmosphere is more likely to be found in a large office thana small one. 2. The larger offices makeone feel relatively unimportant 3. A large office is definitely betterto work in than a small one 4. The size of the office itself isnot really important 5. The large office creates less feeling ofpermanence and security than a smallone 6. There is an uncomfortable feeling of being watched all the time ina large office 7. The social life of a large office islikely to be better than that ofa small one 8, Any disadvantages of the largeroffice are offset by greater advantages PART THREE Layout Preferences

The pictures that are to be projected each showtwo alternative layouts of the same office floorarea. (Example) You will see that one layout is placed abovethe other and that each has been designated bya letter written below it. In the Reply Section below, will you please draw a ring around the letter for the layout thatyou feel would be the office arrangement in whichyou would prefer to work.

(Example)

The choice may sometimes be difficult to make, butplease try to make one for every case. Reply section

Slide I 0 Slide 6 N

Slide 2 Slide 7

Slide 3 --N Slide 8 M

Slide 4 L ' Slide 9 0

Slide 5 P Slide 10

N _J

PART FOUR Social Considerations

From the individual's point of view,no less important than occupational satisfactions are the socialones. It therefore follows that good working area design should also meet legitimate social needs. These needsare served in different ways by different spatial arrangements, andit is the purpose of this part of the studyto get some information about this. One way of getting a 'bird's-eye view' of the social climate of an organisation is in terms of what is knownas sociometry. The method is simply to ask all the members of a given group to specify those other members of thegroup with whom they are particularly friendly. We appreciate that this information is rathermore personal than that which is normally asked for, butour justification for asking must be that the informationis valuable and will be of practicaluse to the planning of new office blocks. Once more we should like to stress that these documents are absolutely confidential.The contents will be available to absolutely no one on the staff of the CIS, and the completed document will be collectedpersonally by a member of the University Staff The data willbe used for statistical generalisations only andonce it has been coded and transferred to punched cards, the questionnaire willbe immediately destroyed. For the present study we should like you to list, in order of preference, those people who also workon the floor, whom you would like to work close to. There is no question of any action resulting from thesechoices: they will only be used to test certain theories about worksatisfaction.

Absolutely confidential

Order Surname Christian name Mr/Mrs/Miss

1 2 3 4 5 6 7 8 9 10

*Floor specified on original

131 Additional Information Do you have any particularly good friends or acquaintances on other floors? If so, wouldyou be so kind as to mark the number in the boxes beside thefloors. Number I. * *

I 3. Fifth to tenth U 4. Eleventh to twenty-second I I

* Floors specified on original

PART FIVE (Sho slide) As you can see, this slide is a model of an office layout partitionedin various ways (indicated). The model is meant to resemble the CISbuilding in general terms, having conditioning. permanent supplementary lighting and air Will you please try to imagine that you are going to work somewhere in this office andthat you have complete freedom to choose your own work place(with the exception of the supervisor's indicated). desks which face into the room On the print of this slide which is given below, please put a bold ' Xon the desk position at which you would choose to work.

imi irmutinurinis

Largemon 41011111101 Also grim

Please state reasons for choosingyour work place Appendix 5.1 Supplement 3

THE UNIVERSITY OF LIVERPOOL DEPARTMENT OF BUILDING SCIENCE THE UNIVERSITY, LIVERPOOL 3 TELEPHONE : ROYal 6022Ext. 353

QUESTIONNAIRE TO SECTION CLERKS Dear Mr

We are, in the Department of Building Science at the University of Liverpool,still looking intensively at the question of the design of modern office buildings. Ourconcern is particularly with user requirements, and at the present time specifically with the way in which the interior layout of the buildingaffects the supervision of workinggroups, and here we are thinking in terms of comparisons between large clericalareas of 75 or more persons in any open unit and smaller areas of about 25 people.

At the present time there is very little information about the effects oflarge open plan offices compared with smaller areas on the work of section heads. It may prove to bea consideration of either fundamental or slight importance, but however high or low it comes in the scale of environmental needs,one wants to be clearly aware of the advantages or penalties involved.

In considering your replies it would be very much appreciated ifyou would give them careful attention and try to tick the box which exactly fits your opinion, as the issues involved also affectsuch other matters as lighting, ventilation, circulation spaces, structural form, anda host of others. It therefore follows that a reply which is either favourable or unfavourable to a given arrangement should be seriously heldin terms of your working experience. Almost any arrangement can be allowed for, but the datawe need is about the actual supervisory consequences. After all, the building is not an end in itself, but ameans to an end, and that end is to produce the most suitable working arrangement possible.

The replies to the questionnaire will, of course, be treated confidentially anddealt with in terms of generalisations, not specific cases. As you are no doubt aware, the work is supported both by theManagement and the Guild, and therefore represents no sectional interests. Thepurpose is to assess the conditions which your experience indicates leads to the most satisfactory conditions for work. We then have theproblems of design to consider in order to optimise physical conditions in the light of your experience. The precisionwith which we can obtain this goalcan, of course, be no greater than the precision with whichyou complete this document, and we therefore hope that you will be kind enough to give us this help once more.

Yours sincerely,

B. W. P. Wells

P.S. The final section of the questionnaire cannot be completedat this stage as the slides to be shown with it will not be presented until the coffee room session. Perhaps, therefore,you would be kind enough to complete the questionnaire as far as possible and bring the whole thing along withyou to the coffee room.

133 Appendix 5.1 Supplement 4

Will you please make the effort to think about the problemstrictlyin terms of supervisory considerations, rather than such connected ones as the general level of office noise,ventilation and lighting; matters which will receive separate attention. As with the other questionnairesyou have been kind enough to complete, we should be very grateful if you would avoid discussing the questions untilafter you have returned the questionnaire. (Please place a tick in the box below the statement whichmost exactly represents your opinion for each of the alternatives).

1. Assuming that a large office space was provided forsay 4 to 6 sections totalling perhaps 100 to 150 people, how would you rate the following types of arrangement:

(i) Sections not physically Very Moderately Indifferent Moderately Very separated from one another favourably favourably unfavourablyunfavourably El El El (ii) Five-foot high panel parti- Very Moderately Indifferent Moderately Very tioning between sections favourably favourably unfavourablyunfavourably El El El (iii) The use of filing cabinets, Very Moderately Indifferent Moderately Very telephone and postal tables, etc as favourably favourably unfavourablyunfavourably the boundaries between sections

(iv) Floor to ceiling panel parti- Very Moderately Indifferent Moderately Very tioning between sections 0favourably favourably unfavourablyunfavourably

(v) Wide gangways separating Very Moderately Indifferent Moderately Very sections favourably favourably unfavourablyunfavourably

(vi) Clear glass partitioning Very Moderately Indifferent Moderately Very between sections favourably favourably unfavourablyunfavourably

2. Are the members of a section mom aware of themselvesas a member of a particular section if they are physically separated from other sections? Definitely Moderately Undecided Moderately Definitely Yes Yes No No El

3. Is there any tendency for people to be treated lessas individuals and more as functionaries in the larger offices? Definitely Moderately Undecided Moderately Definitely Yes Yes No No

4. Do you feel that the actual size of clericalareas - that is the number of people working in anyone space - has any real importance from the supervisory point of view? Definitely Moderately Undecided Moderately Definitely Yes Yes No No E El

134 11

5. In your experience is it easier cm less easy to carry out the specifically supervisoryduties of a section which is situated within a large clericalarea or a small one? LARGE SMALL Definitely Moderately No Moderately Definitely Easier Easier Difference Easier Easier

OR No satisfactory basis for decision

6. If it was a question of choice,would you choose to have area? your section located by themselves ina screenedoff Definitely Moderately No Moderately Yes Definitely Yes Preference No No El Cl 7. Spatial Considerations The following eight statements each refer to the size of general office workareas and to personal feelings about them. The context of these statements is thecontrast between the open planning of the and the subdivision of the general office (' the large office') same area by partitioning (' the small ot7'), conceivedas being different unit sizes of the same company. 'The small office'may be thought of as containing about 25 people; people. 'The large office' about 75 Please put a tick in the box nextto the statement or statements with which the statements with which you are in agreement, and a cross beside you disagree, thus putting either a tickor a cross against every statement. Will put a second tick in the box against the you also statement with which you feel most stronglyin agreement, ie,

A good working atmosphere is more likely to be found in a large office thana small one The larger offices make one feel relativelyunimportant A large office is definitely better to workin than a small one The size of the office itself is not reallyimportant The large office creates less feeling of permanence and security than a smallone There is an uncomfortable feeling ofbeing watched all the time ina large office The social life of a large office is likelyto be better than that of a small one Any disadvantages of the larger officeare offset by greater advantages

8. The questions included in thisquestionnaire were, necessarily, somewhat of personal response. It would arbitrary and limiting in their freedom however be exceedingly helpful ifyou would comment on any matters arising from the supervisory problems of theopen compared with the smaller office plan. Elaboration in the questionnaire would be of your replies to questions very welcome, as would your commenton the size of office areas from a point of view other than the supervisoryone.

r)) 135

.41111111. I PLEASE IGNORE THIS SECTION FOR THE TIME BEING BUTBRING IT ALONG, TOGETHER WITH THE COMPLETED QUESTIONNAIRE TO THE COFFEEROOM SESSION

OFFICE LAYOUT PREFERENCES 9. The pictures thatare to be projected each show two alternative layouts of the same office floorarea. (Example)You will see that one layout isplaced above the other and that written below it. In the Reply Section each has been designated bya letter below, will you please drawa ring around the letter for the layout that feel would be the officearrangement which you tend to recommend you any of them. if there were no organisationalobjections to

(Example)

Full air conditioning is envisaged,together with evenly distributed fluorescent fittings. permanent illumination from ceiling mounted The choice may sometimes bedifficult to make, but pleasetry to make one for every Reply Section case.

Slide I 0 Slide 6 1 P

Slide 2 Slide 7 M

Slide 3 N Slide 8 M

Slide 4 Slide 9 0

Slide S P Slide 10 N

136

ti .

Appendix 5.1 Supplement 5

THE UNIVERSITY OF LIVERPOOL DEPARTMENT OF BUILDING SCIENCE THE UNIVERSITY, LIVERPOOL 3 TELEPHONE : ROYal 6022Ext. 333

QUESTIONNAIRE TO MANAGERS Dear Mr As you will no doubt be aware, the survey of environmental conditions that we of the University of Liverpool have been making at the CIS since the summer of 1962 is still continuing. The brief questionnaire attached to this letter is one of the concluding items of the project and is concerned with one of the most fundamental matters in office layout design, the relative merits of the open plan and the smaller unit office. The relative merits of the two competing layout plans will, of course, depend upon the point of view from which they are judged. We have sought other points of view and now need to know the answers to some questions related to productivity and man management. The only way we can get these answers is by referring them directly to you, and we therefore hope that you will be kind enough to give us this help.

Yours sincerely,

B. W. P. Wells

P.S. The final section of the questionnaire cannot be completed at this stage as the slides to be shown with it will not be presented until the coffee room session. Perhaps, therefore, you would be kind enough to complete the questionnaire as far as possible, and bring the whole thing along with you to the coffee room.

137 Om.

Appendix 5.1 Supplement 6

Except where otherwise specifically requested,will you please make the effortto think about the problem strictly in terms of management considerations, rather than such connectedones as the general level of office noise, ventilation and lighting; matters which will receive separate attention. As with theother questionnaires you have been kind enough to complete,we should be very grateful if you would avoid discussing have returned the questionnaire. the questions until after you

1. If you were asked to plan your department again from scratch, andwere applying only the criteria of efficiency and productivity, which of the followingalternative arrangements would be The alternatives are conceived your first choice, second, and so on. as referring to general clerical tasks, but exclusive oftyping pools and machine areas which present special problems. Please put a 1 ' in the box beside your first choice,a '2' beside the second choice, and so on. Rank (a) Completely open officespace for the whole department withouteven private offices (b) Open general clericalarea but some separate managerial and specialist offices (c) A moderate amount of partitioningto produce rathe smaller clericalareas (d) A complex of small work cellsperhaps with each containinga single section

2. Other things being equal, in whichsize of general office wouldyou expect to find the best morale, and from which would you expect the highest relativeproductivity? Please tick Please tick for morale for productivity 10 - 20 persons 20 - 40 persons 40 60 persons 60 -100 persons 100 or more persons ID Size of work unit not relevant to questions Eli

3. Is there any tendency for people to be treated less as individuals andmore as functionaries in the larger offices? Definitely Moderately Undecided Moderately Yes Definitely Yes No No

4. Can you please summarise the managerial advantages of the open-plantype of office:

5. Will you please now summarise the managerial disadvantages of the open-planoffice:

138 6. Spatial considerations The fnllov.ing eightstatements each refer to the size of general office workareas and to personal feelings about The context of thesestatements is the contrast between the them. and tlw subdivision of the open planning of the general office ('thelarge office') same area by partitioning (' the small office'),conceived as being different unit same company. 'The small office'may be thought of as containing about sizes of the people. 25 people; The large office'about 75

Please put a tick in the boxnext to the statement or the statements with which statements with which you are inagreement, and a cross beside you disagree, thus putting eithera tick or a cross against every also put a second tick in the statement. Will you box against the state.nent withwhich you feel most strongly in agreement, ie,177

A good working atmosphere is more likely to be found ina large office than a small one The larger offices makeone feel relatively unimportant El A large office is definitelybetter to work in thana small one The size of the office itself isnot really important C The large office creates less feeling of permanence andsecurity than a small one There is an uncomfortable C feeling of being watched all thetime in a large office The social life of a large office is likely to be betterthan that of a smallone Any disadvantages of the largeroffice are offset bygreater advantages 0

7. The questions included in this questionnairewere, necessarily, somewhat arbitrary and of personalresponse. It would however be exceedingly limiting in their freedom helpful if you wouldcomment on any additional matters arising from themanagement problems of the questions in the questionnaire open compared with the smaller officeplan. Elaboration ofyour would be very welcome,as would your comment on the size of point of view other than themanagerial one. office areas froma

139 , PLEASE IGNORE THIS SECTION FOR THE TIME BEING BUT BRING IT ALONG WITH THE COMPLETED QUESTIONNAIRE TO THE COFFEE ROOM SESSION

OFFICE LAYOUT PREFERENCES 8. The pictures that are to be projected each show two alternative layouts of the same office floor area. (Example)You will see that one layout is placed above the other and that each has been designated by a letter written below it. In the Reply Section below, will you please draw a ring around the letter for the layout thatyou feel would be the office arrangement which you tend to recommend if there were no organisational objections to any of them.

(Example) Y

Full air conditioning is envisaged, together with evenly distributed permanent illumination from ceiling mounted fluorescent fittings. The choice may sometimes be difficult to make, but please try to make one for every case. Reply section

Slide 1 0 Slide 6 r--N

P 1_.

Slide 2 L Slide 7

Slide 3 N Slide 8

LM L

Slide 4 L Slide 9 M

0 P

Slide 5 Slide 10 0

140

4 Appendix 6.1 Surveys of daylight in offices

In spite of the difficulties mentioned inchapter 6, measurements of daylight factor were made intwo office buildings: in two rooms on the fifteenthfloor of a side- lit building and overa representative area of an extensive, open-plan, top-lit, single-storey officebuilding. The instrument usedwas a BRS EEL daylight factor meter. This includes two photocells, the ' indoor ' cellbe- ing cosine corrected and the ' outdoor 'reference cell enclosed in a blackened tubular shield. The' outdoor ' was placed on the roof of the building and elevatedat an angle of 42° from the horizontal. (Fora CIE sky the luminance at this altitude is numericallyequal to the total illumination on an unobstructedhorizontal plane). The indoor cell was placed in thecentre of the working area of each desk in the room or area surveyed. Measurements in the side-lit officewere made on one day and repeated on the next. The skywas completely overcast during the time of bothsurveys. In spite of the apparently similar sky conditions, the daylightfactors at the same position differed fromone day to the next (table 49).

Table 49 Daylight factors in one office measured on two successive days daylight factor (percent) Date Desk no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 29th January 8.0 8.0 6.9 6.9 5.6 9.8 8.8 2.6 2.7 2.7 2.6 2.0 2.8 2.7 30th January 7.0 6.4 5.6 5.5 4.7 8.2 7.4 2.8 2.0 2.1 1.9 2.6 2.5 2.3 Commence Sky luminance at 42° from horizontal Finish 29th January 200 to 260 ft.-L 30th January 320 to 400 ft.-L Average daylight factor 29th January-4.8 percent 30th January-4.1 percent

To compute the daylight factors whichwould be obtained under a CIE sky, a detailed recordof the actual sky distribution should be obtained.However, for such steps to be worthwhile the sky luminance distribution must be recorded at the same timeas the measurements are taken and this was not possible in thissurvey. Only one set of measurementswas made in the single- storey office building. The average ofseventy-one readings of daylight factor was 4.6percent and the range 2.4 to 8.2 percent. The readings were taken in theafternoon of 3rd March 1963; during the periodin which measurements were made the sky luminance variedbetween 600 and 680 ft.L.

141 Appendix 6.2 Surveys of electric light in offices

Surveys of electric lighting were made in six offices. A typical floor, or section of a floor, of a building was chosen; the colours and condition of the floors, walls, ceilings and desks were noted, and in some cases their reflectivities were measured. The position, type andcon- dition of the lamps were noted and the room dimensions (including its height) and the positions of desks, were measured. Illumination surveys were made when the external illumination did not exceed one lm. /sq.ft. A portable photoelectric photometer (an EEL ' Lightrnaster I was used. The cell is cosine and colour corrected but the measurements obtained must be multiplied by a correction factor which depends on the type and colour of the lamps. Where there was more than one colour of lamp in a surveyed area, the number of lamps of each colour was counted and a weighted average value of the correction factor adopted. The photocell was placed in the middle of the working area of each desk top and the value of ilie-tination at that point read from the meter. All the of...-..es except one were unoccupied when the measurements were taken. In the one office where people were seated at their desks they were asked not to lean over the cell.As the lighting installation in this building produced an even and fairly shadow-free illumination, the reduction caused by the shadowing effect of the people was probably small; a comparison of individual results for occupied and unoccupied desks indicated that it was about 4 percent of the average. Measurements of surface reflectivities were made using a reflectometer designed by the Paint Research Station. This has a photocell with a hole in the middle through which a beam of light from a small lamp is directedonto the specimen, which then reflects the light back to the cell. The test head was first placed on a freshly prepared surface of magnesium carbonate and the meter calibrated to this basis. The head was then placedon the specimen and the meter deflection noted. Several readings were taken o'- each surface and theaverage values found. In some cases it was not possible tomeasure ceiling reflectivities without a great deal of difficulty, and estimates werf: made with the aid of shade cards of known Munsell vallie. To assess the accuracy of these estimates, the reflectivities of several test surfaceswere estimated aimd then checked by measurements. The estimated values were within 15 percent of the measured values.

142 Appendix 6.3 Scale for estimating relative percentage of daylight and electric light used in experimental study of satisfactory lighting conditions for clerical workers THE UNIVERSITY OF LIVERPOOL DEPARTMENT OF BUILDING SCIENCE THE UNIVERSITY, LIVERPOOL 3 TELEPHONE : ROYal 6022Ext. 353

As illumination plays such an important part in clerical work, weare especially concerned with its evaluation. The next item is particularly difficult, estimating the amount of illumination due to daylightsources and the amount due to electric lighting. Please attempt this judgement even though you may find it very hard. Holding the sheet of paper flat on the desk, will you please try to estimate the relativeamounts of daylight and electric light illuminating it. The following scale is marked from 0100 % of daylight. Will you please put a cross on the scale at theappro- priate point.

PERCENTAGE OF DAYLIGHT No daylight: 0% 10 20 30 40 50 60 70 80 90 100 % All daylight: entirely electric lightLII11111_111 no electric light

Floor Position Meter reading

143 Appendix 7.1 Surveys of the thermal environment in offices during winter

Surveys of the winter thermal environmentwere made within the offices of six organisations(in five buildings). The purpose was to learn howclosely modern office buildings provide the standards of thermalenvironment currently recommended. Wherever possible measurementswere made in large open-planned offices, but theareas of buildings chosen did not always cover a whole floor.The areas surveyed amounted to about 7,000 sq.ft in theairconditioned offices and 3,000 to 4,000 sq.ft in thenaturally ventilated offices. In each office building between 16 and 20globe thermometers were positioned so that conditions representative of the occupied zones could bemeasured. The usual arrangement was to place the globes inrows across the building at a spacing of between 8 and16 feet. An exactly symmetrical layout ona grid was never possible because of the placing of the desks. The globe thermometerswere located 4ft above the floor because this is about head-height for seated officeper- sonnel. Previous research workers havetaken temperature measurements at various heightsranging from 2ft 6ins to 5ft. 1'2'3'4 The measurements ofair temperature, relative humidity and air movementwere made at approximately the same heightas the measurements of globe temperature. Measurements of airmovement were made with kata thermometers andwere extremely time- consuming, so that they could be made onlya few times a day. Air and globe temperatures and relative humidity were measured at the same time as air movement and intermediately. The majority of readingswere taken between 1020 and 1140 hours and between 1420 and 1540 hours. Vertical air temperature gradientswere measured in four buildings having different systems ofheating. The measurements were made with a portable, manually- balanced potentiometer and thermocouple. Records of the external climatewere obtained from the nearest meteorological stations. Three officesare located in a city centre within two-fifths ofa mile of a meteorological station whose instrumentsare sited on the roof of a city building; another office issituated in a suburb of a large provincial city, withinone mile of another meteorological station. The meteorologicaldata for these four offices should thereforebe reasonably representative of the climate around them. Thenearest meteorological station to the two other offices is six I. HIckish, D. E: Thermal sensationsof workers in light ind- miles distant. Since the officesare in a city centre, some ustry in summer..1 Hygiene, 1955, 33 (1) differences between the readings obtainedfrom the 2. Black, F. W: Desirabletemperatures in offices. .1 Inst Heat station and the actual values around thebuilding are Vcnt Engrs. 1954, 22 (11). certain. 3. Bruce, W: Man and his thermalenvironment. Technical paper no. 54 of the Division of Building Research,(NRC 5514). These surveys of winter thermal conditionswere made National Research Council (of Canada).(The Council) Ottawa, during an extended spell of cold weatherin the first 1960 (February) three months of 1963. 4. Institution of Heating and VentilatingEnemas: Guide to current practice. London (The Institution), 1959 144 Appendix 7.2 Surveys of the thermal environment in offices during summer

Surveys of the thermal environment were made during the summer of 1963 within three buildings. The method of measurement was the same as that already described in appendix 7.1 for winter thermal surveys. Records of the external climate were obtained in the same way as those described in appendix 7.1. Summer thermal surveys were restricted to days when the meteorological stations forecast sunny conditions with shade air temperatures of 70°F or higher. Unfort- unately, there were only two warm spells in the summer of 1963, the first at the beginning of June, the second at the end of July. Appendix 8.1 Surveys of noise levels in office buildings

Surveys of noise levels were made in a number of offices within nine buildings. The purpose of these surveys was to compare typical noise levels in some provincial office buildings with currently recommended criteria. The surveys were made in general and private offices, typing rooms, machine rooms, conference rooms anda boardroom. Most of the general offices contained add- ing machines or typewriters. The number of occupants in the general offices ranged from three up to several hundred; some of the private offices and both the conference rooms and the boardroom were unoccupied. The occupants of the various offices were asked to carry on with their normal duties while the surveys were being made so that the noise levels measured represented the usual conditions. In order that measurements should be representative of normal conditions, they were made only during the middle of the morning and the middle of the afternoon. Peak traffic periods were avoided. The instrument used was a Bruel and Kjaer precision sound level meter and octave filter set. A condenser microphone was used with the meter. The meter was supported on a tripod with the microphone 4ft above floor level. It was regularly calibrated with an acoustic calibrator. Since the sound level varied considerably over short Periods of time in all the offices, the range of readings for an interval of approximately fifteen seconds was obtained for each octave band and weighting network, but noises which were unlikely to occur frequently were not included in the measurement. In the smaller offices measurements were made near the centres of the rooms, ie, between four and six feet from the windows. In most of the larger offices preliminary readings showed that there was little variation in the range of noise levels in different parts of the offices and so readings were taken near the centre of the rooms. In some of the offices measurements were made with all the windows closed and then, if there were no objections from the occupants, with one window open.

146 Appendix 10.1 Supplement 7

THE UNIVERSITY OF LIVERPOOL DEPARTMENT OF BUILDING SCIENCE THE UNIVERSITY, LIVERPOOL 3 TELEPHONE : ROYal 6022Ext. 353

15th October, 1962. Dear Sir or Madam, You will find, attached to this letter, a questionnaire in which you are invited to make a number of observations about your building. The questionnaire has been devised and circulated by members of the research staff of the University of Liverpool. In this, we are indebted to the management of the CIS and the Guild of Insurance Officials for their support and encouragement. Through the medium of questionnaires and interviews, we hope to collect the sort of information that will be in- valuable in the planning of work places. In this study we are concerned with the individuals' personalresponse to the conditions created by the architects, planners, and engineers. The timing of this questionnaire was planned to coincide as closely as possible with themove into the new building. This is because the conditions in the previous buildings should still be fresh in peoples' minds, and they should therefore be able to make clearer comparative judgements. However, there are drawbacks in obtaining reactions at this stage. One of these is the time required by the engineers to master the heating and ventilating plant and to maintain the conditions that were planned. One cannot, of course, evaluate ultimate conditions until they have been attained, but much of value can be learned from the study of existing conditions. This creates the need for further research, and it is proposed to repeat some parts of this survey in 1963, after a suitable settling-in period. I know that the questionnaire will require a good deal of time and thought to complete, but I hope thatyou will agree with me that the cause is worthwhile. Any large office building, built for perhapsa hundred years, will contain in its lifetime many thousands of people for much of theirs. It is thereforea matter of considerable practical importance to learn as much as possible about the effects on the individual of various working conditions. All the information thus collected will becompletely confidential toa small number of staff members in the university. No member of the CIS will have access to any of these documents. All replies will be quiteanonymous, and the results dealt with in statistical terms. I am sorry that it is not possible for me to talk to you personally, though I have talked tosome people at the stage of drafting these questions, and look forward to seeing many more. Iam hoping, however, that even in this short letter, I have been able to convey enough of the purpose and value of the research aims toencourage you to complete the questionnaire with care and judgement.

Yours faithfully,

Brian Wells.

147 01111111....1-

Appendix 10.1 Supplement 8

UNIVERSITY OF LIVERPOOL Confidential Department of Building Science Will you please answer each of the following questions by puttinga tick in the brackets on the right hand side of the statement which is most nearly appropriate toyour own case. ie, ( i) The figures to the right of the brackets eg 3:18, are merely the punching instructions for the statistical analysis so please ignore them. Try to answereveryquestion, and please don't discuss your answers with other people until thequestionnaires are completed. With the best will in the world, it is hard to avoid influencing others,and the value of the entire project will depend upon getting as many independent assessmentsas possible. When you have completed the questionnaire, wouldyou please put it in the envelope provided and hand it to the Section or Chief Clerk in your Department sometime before Thursday.

1. Were you previously employed by the CIS in premises other than thepresent ones? ) Yes ( No () 1:11 If yes, at which of the following:- 1. 109, Corporation Street ( ) 1:0 2. 113-115, Portland Street ( ) 1:1 3. 117-119, Portland Street ( ) 1:2 4. 22, Long Millgate ( ) 1:3 5. Crown Buildings, 96 Dantzic Street ( ) 1 :4 6. 77 Dantzic Street ( ) 1:5 7. 79 Dantzic Street ( ) 1 :6 8. Trevalyan Buildings, 52 Corporation Street ( ) 1:7 9. New Elizabeth Street, Cheetham ( ) 1:8 10. Ormes Buildings, 14, The Parsonage ( ) 1:9 11. Other CIS Building (Please write)

( ) 1:01

2. On which floor do you now work? Basement Document Store( ) 25 13 ( )3 :13 Ground( ) 00 14( )3 :14 1 ( )2 :01 15( )3 :15 2( )2 :02 16( )3 :16 3( )2 :03 17( )3 :17 4( )2 :04 18 ( )3 :18 5( )2 :05 19( )3 :19 6( )2 :06 20( )3 :20 7( )2 :07 21 ( )3 :21 8( )2 :08 22( )3 :22 9( )2 :09 23( )3 :23 10( )2 :10 24( )3 :24 11 ( )2 :11 12( )2 :12

3. Please turn to the back of this Questionnaire. You will find four floor-plansof this building printed on a single sheet of foolscap. Each individual will be concerned with onlyoneof these diagrams, and this is the one for the floor on which you work. When you have located the appropriateone, will you please mark with an ' X ' the place where you normally work, or have your desk or machine. Please put onlyone 'X'on the plan. The small black squares on the plans represent the pillars inside the building. They should prove useful in fixingyour position. (Please note that Miller St is the street in which the main entrance is situated, andDantzic St is the street that runs between this building and the new Conference Hall). 148 Appendix 10.1 Supplement 8

UNIVERSITY OF LIVERPOOL Confidential Department of Building Science Will you please answer each of the following questions by puttinga tick in the brackets on the right hand side of the statement which is most nearly appropriate toyour own case. ie, (J) The figures to the right of the brackets eg 3:18, are merely the punching instructions for the statistical analysis so please ignore them. Try to answer every question, and please don't discussyour answers with other people until the questionnaires are completed. With the best will in the world, it is hard to avoid influencing others,and the value of the entire project will depend upon getting as many independent assessmentsas possible. When you have completed the questionnaire, wouldyou please put it in the envelope provided and hand it to the Section or Chief Clerk in your Department sometime before Thursday.

1 Were you previously employed by the CIS in premises other than thepresent ones? Yes () No() 1:11 If yes, at which of the following: 1. 109, Corporation Street ( ) 1:0 2. 113-115, Portland Street ( ) 1:1 3. 117-119, Portland Street ( ) 1:2 4. 22, Long Millgate ( ) 1:3 5. Crown Buildings, 96 Dantzic Street ( ) 1:4 6. 77 Dantzic Street ( ) 1:5 7. 79 Dantzic Street ( ) 1 :6 8. Trevalyan Buildings, 52 Corporation Street ( ) 1:7 9. New Elizabeth Street, Chatham ( ) 1 :8 10. Ormes Buildings, 14, The Parsonage ( ) 1 :9 11. Other CIS Building (Please write)

( ) 1:01

2. On which floor do you now work? Basement Document Store( ) 25 13 ( )3 :13 Ground( ) 00 14( )3 :14 1 ( )2 :01 15( )3 :15 2( )2 :02 16( )3 :16 3( )2 :03 17 ( )3 :17 4( )2 :04 18( )3 :18 5 ( )2 :05 19( )3 :19 6( )2 :06 20( )3 :20 7( )2 :07 21 ( )3 :21 8 ( )2 :08 22( )3 :22 9( )2 :09 23( )3 :23 10( )2 :10 24( )3 :24 11 ( )2 : 11 12( )2 : 12

3. Please turn to the back of this Questionnaire. You will find four floor-plansof this building printed on a single sheet of foolscap. Each individual will be concerned with onlyone of these diagrams, and this is the one for the floor on which you work. When you have located the appropriate one, will you please mark withan ` X ' the place where you normally work, or have your desk or machine. Please put only one 'X' on the plan. The small blacksquares on the plans represent the pillars inside the building. They should prove useful in fixingyour position. (Please note that Miller St is the street in which the mainentrance is situated, and Dantzic St is the street that runs between this building and the new Conference Hall). 148

1 4. To which age group do you belong: (Please place tick against appropriate group) 15 19 )14 :0 40 44 ( )14 :5 2024 )14:1 45 49 ( )14 :6 2529( )14 :2 50-54 ( )14 :7 3034 )14 :3 5559 ( )14 :8 3539( )14 :4 60 or more( )14 :9

5. What is your sex: MALE( )15:1 FEMALE( )15 :2 What is your rank or grade: What is your rank or grade: Basic ' A' ( )16 :01 Basic 'A' ( )17 :01 Basic 'B' ( )16 :02 Basic ' B' ( )17 :02 Grade 3 ( )16 :03 Basic ' C' ( )17 :03 Grade 2 ( )16 :04 Machine punch operator ( )17 :04 Grade 2a ( )16 :05 Other machine operator ( )17 :05 Grade 1 ( )16 :06 Stencils group leader ( )17 :06 Official ( )16 :07 Comptometer operator ( )17 :07 Other (Please write) Copy typist ( )17 :08 ( )16 :08 Audio typist ( )17 :09 Shorthand typist ( )17 :10 Grade 2a ( )17 :11 Grade 1 ( )17 :12 Other (Please write) ( )17 :13

6. Do you feel that the lift service is:

Completely inadequate ( )18:1 Fair, but not quite adequate ( )18 :2 Quite adequate ( )18 :3

7. Do you feel that it is important to be able to see out of the officeeven if you have plenty of artificial light to work by? Definitely yes ( ) 19: 1 Moderately yes ( ) 19 : 2 Unimportant ( ) 19 : 3 Moderately not ( ) 19 : 4 Definitely not ( 1 19 : 5

8. Please place a tick beside any of the following words which you feel describes the colour scheme ofyour office: (You may tick as many words as you feel appropriate)

Cold ( ) 20 :1 Bright ( ) 21 :1 Restful ( ) 20: 2 Tasteful ( ) 21 : 2 Mediocre( ) 20: 3 Warm ( ) 21 : 3 Cheerful ( ) 20 : 4 Depressing( ) 21 :4 Beautiful( ) 20: 5 Glaring ( ) 21 : 5 Soothing( ) 20 : 6 Dull ( ) 21 : 6 Ugly ( ) 20 : 7 Distracting( ) 21 : 7 Others, please state

9. Do you ever feel uncomfortably cold whilst working?

Never ( ) 22 :1 Sometimes ( ) 22: 2 Often ( ) 22 : 3 Always ( ) 22 : 4

10. Do you feel that the height of the ceiling is: Too high ( ) 23 : 1 About right( ) 23 : 2 Too low ( ) 23 : 3

11. On balance, do you enjoy your work more or less, in the new surroundings?

Definitely more ( ) 24 :1 Moderately more ( ) 24 : 2 No change ( ) 24 : 3 Moderately less ( ) 24 : 4 Definitely less ( ) 24 : 5 149 12. Does the air in the office normally seem fresh? Yes ( ) 25 : 1 No ( ) 25 : 2

13. Do you ever feel uncomfortably hot whilst youare working? Never ( ) 26: I Sometimes ( ) 26: 2 Often ( ) 26: 3 Always ( ) 26 : 4

14. Do you feel that the amount of toilet accommodation is adequate?

Definitely inadequate ( ) 27 :1 Somewhat inadequate ( ) 27 : 2 About right ) 2277 43 Rather more than needed ( Definitely more than needed ( ) 27 : 5

15. Please put a tick against any of the following whichyou feel might be improved. (You may tick as many as you like) Colour scheme ( )28 : 11 Office furniture ( )28 : 6 Ventilation ( )28 : 0 Noise ( )28 : 7 Lighting ( )28 :I Draught prevention ( )28 : 8 Lift service ( )28 : 2 Space around desks ( )28 : 9 Toilets ( )28: 3 Heating ( )28 : 4 Canteen ( )28 5 Anything else not listed above 29 :

16. Did you ever suffer from headaches in the premises you have just left? Never ( ) 30 :1 Seldom ( ) 30 : 2 Often ( ) 30 3 i

i 17. Do you ever suffer from headaches in this new building? Never ( ) 31 :1 Seldom ( ) 31 : 2 Often ( ) 31 : 3

18. Taking average conditions, do you find that the available daylight, togetherwith the supplementary electric lighting is

Too bright for comfort( ) 32 :1 Comfortable ( ) 32 : 2 Too dull for comfort ( ) 32 : 3

19. In what sort of office would you prefer to work? A large open one ( ) 33 :1 A. smaller partitioned area ( ) 33 : 2 No preference ( ) 33 : 3

20. In what way have your personal relationships with colleagues been affectedby moving into the new building? More friendly ( ) 34 : 1 Unaffected ( ) 34.: 2 Less friendly ( ) 34 : 3

21. Does the temperature of the office vary noticeably? Never ( ) 35 :1 Sometimes ( ) 35 : 2 Often ( ) 35 : 3 Regularly ( ) 35 : 4

22. Did you ever suffer from eye-strain in the premises you have just left? Never ( ) 36 :1 Seldom ( ) 36 : 2 Often ( ) 36: 3

23. Do you ever suffer from eye-strain in this new building? Never ( ) 37 :I Seldom ( ) 37 : 2 Often ( ) 37:3 24. Flow do you feel about workingin a tall office block? Very much like it 38: I Moderate liking )38: 2 No particular feeling )38: 3 Moderate dislike )38: 4 Very much dislike it )38: 5

25. Tick any of the following wordswhich you feel to be descriptive you feel appropriate) of your office (youmay tick as many as Friendly ( ) 39 : 0 Warm ( 40: 0 Austere ( ) 39: 1 Efficient ( 40: 1 Comfortable( ) 39: 2 Unfriendly ( 40: 2 Dark ( ) 39: 3 Pleasant ( 40: 3 Stimulating ( ) 39 : 4 Cold ( 40 : 4 Modern ( ) 39: 5 Relaxing ( 40: 5 Gloomy ( ) 39 : 6 Impersonal ( 40: 6 Gay ( ) 39 : 7 Stuffy ( ) 39 : 8 Other(s) 41 : Light ( ) 39 : 9

26. Do you find that thereare any unpleasant draughts whereyou work? Never ( ) 42 :1 Seldom ( ) 42 : 2 Frequently ( ) 42: 3 Always ( ) 42 : 4

27. Comparing buildings, do you prefer working in this buildingor the one you have recently left? Strongly prefer this ( ) 43 : I Moderately prefer this ( ) 43 : 2 No preference ( ) 43 : 3 Moderately prefer previous office( ) 43 : 4 Strongly prefer previous office ( ) 43 : 5

28. Do you feel that it is as good for your eyes to work by artificial lightas by daylight?

Definitely yes ( ) 44 :1 Moderately yes ( ) 44 : 2 No difference ( ) 44:3 Moderately no ( ) 44 : 4 Definitely no ( ) 44 : 5

29. Do you ever findyour workplace distractingly noisy? Yes ( ) 45 : 1 If yes, is the source of the noise No ( ) 45 : 2 outside or inside the building? Outside ( ) 46: 1 Inside ( ) 46 : 2 If inside, please say what is especiallydistracting 46 :

30. Do you like thenew arrangement of working hours?

Definitely yes ( ) 47 :1 Moderately yes ( ) 47 : 2 No preference ( ) 47 : 3 Moderately no ( ) 47 : 4 Definitely no ( ) 47 : 5

31. Is there ever an unpleasant amount of glare from the windows thatyou cannot personally control? Yes ( ) 48 : 1 No ( ) 48 : 2 If yes, is this felt Seldom ( ) 49 :1 Frequently ( ) 49 : 2 Most of the time( ) 49 : 3

32. Are there any other observationsabout conditions in the new building which you would like to make,and i'or which scope has not been given inthe questionnaire?

WE, FOR OUR PART, WOULD LIKE TO THANK YOU VERYMUCH FOR THE TIME AND THAT YOU HAVE TAKENIN FILLING IN THIS EFFORT VERY COMPREHENSIVEQUESTIONNAIRE

151 Appendix 10.1 Supplement 9 FLOOR PLANS OF BUILDING

HANOVER Sr HANOVER ST. Leap.a.m.. *

25---a-- 4

1 , a a a al a MILLER STREET ROUND FLOOR Ela La--

a HANOVER ST. a

MILLER STREET FL 004

HANOVER ST. 11 0

111 -a 11-111-

MILLER STREET FLOORS S 24. _ um IF

MILLERSTREET FLOORS 2.3.4.

01111.111111111111111411B-- Appendix 10.2 Supplement 10

THE UNIVERSITY OFLIVERPOOL DEPARTMENT OF BUILDING SCIENCE THE UNIVERSITY, LIVERPOOL 3 TELEPHONE ROYal 6022Ext. 353

August, 1963.

Dear Sir or Madam, As an extension of the questionnaires and physical measurements made since lastOctober we are again looking in detail at the conditions for personal comfortas determined by environmental conditions. We should therefore be grateful ifyou would complete the attached check-listsome time during the morning. Instrument readings will also be made duringthe course of the day and I shall look forward the morning for more information. to visiting you during It is only by virtue of the replies that we get to this questionnaire that we shall be ableto know what standards to aim for. We therefore hope thatyou will be kind enough to consider as possible. your answers to the questions as carefully Sincerely, B. W. P. Wells

153 Appendix 10.2 Supplement 11

THE UNIVERSITY OF LIVERPOOL DEPARTMENT OF BUILDING SCIENCE THE UNIVERSITY, LIVERPOOL 3 TELEPHONE : ROYal 6022Ext. 353

Confidential Will you please answer each of the following questions by putting a tick in the brackets on the right hand side of the statement which is most nearly appropriate to your own case, ie, ( I). The figures to the right of the brackets, eg, 23 : 8, are merely the punching instructions for the statistical analysis so please ignore them. Try to answer every question, and please don't discuss your answers with other people until the questionnaires are completed. With the best will in the world, it is hard to avoid influencing others, and the value of the entire project will depend upon getting as many independent assessments as possible. When you have completed the questionnaire, would you please put it in the envelope provided and hand it to the Section or Chief Clerk in your Department sometime before lunch.

1. Floor 17:

2. Position

3. To which age group do you belong? (Please place tick against appropriate group) 15-19( )22 :0 4044 ( )22: 5 2024( )22 :1 4549 ( )22: 6 2529( )22 :2 5054 ( )22: 7 30- 34( )22 :3 5559 ( )22: 8 3539( )22 :4 60 or more( )22: 9

4. What is your sex? MALE ( ) 23 :1 FEMALE( ) 23 : 2 What is your rank or grade? What is your rank or grade? Basic 'A' ( )24 :1 Basic 'A' ( )25:1 Basic B' ( )24 :2 Basic B' ( )25: 2 Grade 3 ( )24 :3 Basic 'C' ( )25: 3 Grade 2 ( )24 :4 Machine punch operator ( )25: 4 Grade 2a ( )24 :5 Other machine operator ( )25: 6 Grade 1 ( )24 :6 Stencils group leader ( )25: 7 Official ( )24:7 Comptometer operator ( )25: 8 Other (please write) Copy typist ( )25: 9 ( )24 :8 Audio typist ( )26:1 Shorthand typist ( )26: 2 Grade 2a ( )26: 3 Grade 1 ( )26: 4 Other (please write) ( )26: 5 154 5. Do you ever feel uncomfortably :old whilst working? Never ) 27 :1 Sometimes ( ) 27: 2 Often ) 27 : 3 Always ) 27 : 4

6. Does the temperature of the officevary noticeably? Never ) 28 :I Sometimes ( ) 28 : 2 Often ) 28 : 3 Regularly ) 28 : 4

7. Do you ever feel uncomfortably hot whilst working? Never ( ) 29 :1 Sometimes ( ) 29 : 2 Often ( ) 29 : 3 Al ways ( ) 29 : 4

8. Do you find that there are any unpleasant draughts where you work? Never ( ) 30: 1 Seldom ( ) 30: 2 Frequently ( ) 30: 3 Always ( ) 30: 4

9. Does the air in the office normallyseem fresh? Yes ( ) 31 :1 No ( ) 31 : 2

10. Do you ever suffer from headaches in this building? Never ( ) 32 : 1 Seldom ( ) 32: 2 Often ( ) 32: 3

11. Do you feel that the height of the ceiling is: Too high ( ) 33 :1 About right ( ) 33 : 2 Too low ( ) 33 : 3

12. Taking average conditions, doyou find that the available daylight, together with the supplementary lighting is: electric Too bright for comfort )34: 1 Comfortable )34: 2 Too dull for comfort )34: 3

13. In what sort of office would you prefer to work? A large open one ( )35: 1 A smaller partitioned area( )35: 2 No preference ( )35: 3

14. Do you feel that it is as good foryour eyes to work by artificial light as by daylight? Definitely yes ( )36: 1 Moderately yes ( )36: 2 No difference ( )36: 3 Moderately no ( )36: 4 Definitely no ( )36: 5

15. Tick any of the following words whichyou feel to be descriptive of your office (you you feel appropriate) may tick as many as Friendly ( )37: 1 Light ( )38 : 2 Austere ( )37: 2 Warm ( )38: 3 Comfortable( )37 : 3 Efficient ( )38 : 4 Dark ( )37: 4 Unfriendly ( )38: 5 Stimulating ( )37: 5 Pleasant ( )38 : 6 Modern ( )37 : 6 Cold ( )38 : 7 Gloomy ( )37: 8 Relaxing ( )38 : 8 Gay ( )37 : 9 Impersonal ( )38: 9 Stuffy ( )38 : 1

16. Do you feel that it is important to be ableto see out of the office even if you have plenty of work by? artificial light to

Definitely yes ( )39 :1 Moderately yes ( )39 :2 Unimportant ( )39:3 Moderately not ( )39 :4 Definitely not ( )39 :5 155 17. Is there ever an unpleasant amount of glare from the windowsthat you cannot personally control?

Yes ( )40:1 No ( )40: 2 If yes, is this felt Seldom ( )41:1 Frequently ( )41: 2 Most of the time ( )41: 3

18. Do you ever suffer from eye-strain in this building? Never ( )42:1 Seldom ( )42: 2 Often ( )42: 3

19. Please place a tick beside any of the following words whichyou feel describes the colour scheme of your office: (You may tick as many as you feel appropriate)

Cold ( )44 :1 Restful ( )44 : 2 Mediocre ( )44 : 3 Cheerful ( )44 : 4 Beautiful ( )44: 5 Soothing ( )44: 6 Ugly ( )44 : 7 Bright ( )44: 8 Tasteful ( )44: 9 Warm ( )45 :1 Depressing ( )45: 2 Glaring ( )45 : 3 Dull ( )45 : 4 Distracting ( )45 : 6

WE SHOULD LIKE TO THANK YOU VERY MUCH FOR YOURKIND CO-OPERATION ONCE AGAIN AND TO ASSURE YOU THAT YOUR EFFORTS WILL BE EXTREMELYVALUABLE IN HELPING US TO ESTABLISH OPTIMUM WORKING CONDITIONS FORCLERICAL OCCUPATIONS

156

111111111111118111111111.11wPwwwww.--- Index

Acoustics (see Noise) tiles 72 light on 43 Central Office of Information (COI) Aesthetics 92 lighting surveys 48 ceiling height 43 noise surveys 68 lighting 49, 51 space surveys 39, 40 sensibilities 25 space 39 thermal environment surveys 62 user satisfaction surveys 25, 29 tall buildings 38 Civil Service 29 Airconditioning 18, 21, 61, 64, 65, 66, 67 lighting surveys 48 CIS 81, 84, 85, 87, 89, 90 noise surveys 68 noise of 69, 87 space surveys 39, 40 working position, choice of 110 thermal environment surveys 61, 63 Amenities 21, 74, 75 Cleaning 94 Appearance (see Aesthetics) light fittings 56, 57 Town Planning, and 28, 38 rooms 34 Appraisals 22, 90 windows 34, 53 Architect 25 Cleanliness 87, 88 CIS 80, 81 Clerks 82, 86-88, 104, 106-121, 126-132 research by 93, 94, 95 Client 19, 28, 74 Attitudes Climate attitude scale 108-111, 121, 123, 124, 125, 126, 127 atmospheric pollution 18, 52 questionnaires 129-140 location and 25, 36 to the CIS 86-90, 104-128 micro-climate 25, 36 to environment 95, 96 sunlight 52, 53, 54, 55 to open plan 88, 89, 104-128 weatherstripping 53 Automatic data processing (ADP) wind 53, 54, 65 machine heat 64 Clothing machine operators 45, 89 airconditioning and 87 user surveys of 25, 29, 30, 40, 48, 51, 62, 68, 74 open-plan office and 126 storage of 34 Barrow-In-Furness thermal environment and 61, 63, 64 accommodation surveys 29, 30 Colour 51 lighting surveys 48 ceilings 48, 57, 142 noise surveys 68 floors 84, 142 space surveys 40 furniture 48, 57, 142 thermal environment surveys 61 Munsell values 51, 142 ventilation surveys 68 physical effect of 51 Blinds 33, 54, 67 subjective effects of 51, 75, 87, 89 venetian 33, 54, 64 walls 48, 57, 142 vertical 67, 84, 87 Comfort 25 British Standard Codes of Practice thermal 61, 62, 63, 64 daylight 52 Construction heating and thermal insulation 63 components 92, 94 sound insulation 70, 71 costs of 93, 97 ventilation 63 curtain walling 54 Building bye-laws 28 modules 46 ceiling height 43 noise reduction and 70, 71, 72, 73 daylighting 18 partition 71, 92, 94 Building Research Station (BRS) lighting surveys 48 thermal environment and 54, 61, 63 Co-operative Insurance Society (CIS) 25, 79-90,104-140, 147-156 London Noise Survey 68, 70 airconditioning 81, 84, 85, 87, 89, 90 noise surveys 30, 68 aural environment 85, 89 psychological surveys 30, 74 design process 80, 81, 82, 83 space surveys 40 interior design 81 thermal environment surveys 62 interviews in 86-88 user surveys 25, 29, 30 labour supply 80, 90 Bfirolandschaft 19, 42 O and M 80 open-plan offices 81, 82, 88, 89, 104-140, 150 Car parking 20, 28, 34, 38 private offices 82, 83 legislation 28 questionnaire surveys in 88, 89, 90, 106-140,147-156 Catering 21 site choice 80 CIS 83 staff tea room 34 spatial environment 83 thermal environment 84. 85, 87, 89, 90 Ceilings total environment 79-90 acoustic 71, 72, 73 view 51 finishes 57, 72, 92, 94 height 43, 87, 89 visual environment 52, 55, 57-60, 83, 84,87, 88, 89

157 Costs Feedback analysis, cost 93, 97 inadequate 19, 74 annual 34, 92, 94 Fire 94 annual, per worker 34, 35 capital 92 window design, and 53 Flexibility 21, 125 ceiling height, and 43 cleaning 34 Floors 70, 71, 72, 73, 84, 87, 92, 94 construction 93, 97 Friendship patterns 105, 106, 113-118, 127, 128, 131,132 age, and 115, 116, 117 electricity 34, 97 between floors 115, 116, 117 external construction 97 foundation 97 inter-personal distance, and 115, 117, 118 intra-departmental 118 heating and ventilating 34, 65, 95, 97 sex, and 115, 116, 117 insurance 34, 97 shifts, and 117 internal construction 97 office size 118 lease, and 34 Liverpool 34 work efficiency 106 Furniture 41 moving-in 35 chairs 18, 73 rates 34 colour of 48 redecoration and maintenance 34 desks 42, 43, 57, 84 rentals 34 services 34, 97 space requirements of 40 site selection, economics of 20, 38, 80 structure 97 Geographer 25, 93, 94 Glare 33, 50, 51, 52, 57, 87, 89 Day lighting 18, 29, 32, Ch.6 (48-60), 83, 87,95, 141, 143 CIS 84, 87, 89 daylight factor 48, 50, 52, 55, 141 index 84 estimation of 57-59, 143 Gowers Report 28, 30 glare from 33, 50, 52, 84, 87, 89 need for 32, 52, 57, 58, 59, 60, 88, 89, 95 Health obstruction of 18, 32, 38, 51, 84 City of London, in 29, 30 open-plan, and 18, 21, 104 environment, and 29, 30, 63, 74 plan-form, and 18, 32, 45 Liverpool, in 29, 30 preference for 18, 33, 52, 57, 58, 59, 89, 95, 120,121 occupational health service 30 questionnaire surveys of 143, 149-151, 155, 156 office size, and 29 side-lit building 55, 141 noise, and 68 sky luminance 141 Height 27, 45 standards of 48, 49, 50, 51 airconditioning, and 84 sunlight 33, 52, 54, 55, 67, 84, 87 ceiling 43 surveys of 29, 30, 31, 32, 33, 48, 49, 55, 141, 143 daylight, and 33 top-lit building 55, 141 economics of 43, 45 town planning, and 18, 28, 38 location, and 20 view 18, 21, 33, 44, 51, 52, 84, 87, 88, 89, 95,104, 120 noise, and 33, 68 Decorations 48, 57, 83 Design subjective reaction to 33, 88, 89 designers 28 temperature variation with 33, 64, 84 town planning, and 28 environmental 17, 18, 19, 25 view, and 33 principles of 81 process of 19, 80, 81, 82, 83 team 80, 81 Imperial Chemical Industries Limited (ICI) space standards 40 town planning, and 28 Desk-plan 41, 42, 43 thermal environment 63 Individuality 105, 122, 124, 125, 126, 127, 134, 138 Dimensions Interior design height 43, 48 CIS 81 modules 21, 46, 47, 83 Isolates 114, 118, 127, 128 room depth 45, 48, 51, 59 Japan Efficiency 29, 74, 7), 92, 106, 124, 125, 126, 128, 138 planning of space 41, 42 Hawthorne stud, of 75 Electric lighting 18, 29, 32, Ch.6 (48-60), 84, 87, 92, 95,142, 143 Labour dependence on 48 supply 80, 90 economics of 34, 95, 97 costs 38 estimation of 57 - 59, 143 Lavatories 21, 34, 88, 89 fittings 46, 55, 56, 57 flicker 51 Lay-out preference 111-113, 121, 123, 124, 125, 126,127, 131, 132, 136, 140 fluorescent 33, 55, 56, 57, 84 age, and 1 1 1, 112 glare from 50, 51, 57, 87 department, and 112, 113 lamps, replacement of 56, 57 office size, and 113 permanent 32, 84 sex, and 112 PSALI 51, 52, 57, 59, 95 Legislation 19, 28 questionnaire surveys of 143, 149-151, 155 Building Byelaws 18, 28 standards 48, 49, 50, 51, 56 definition of office 27 surveys 29, 30, 32, 33, 48, 49, 56, 142, 143 Factories Act 19, 28, 63 tungsten 55, 56, 57 lighting 18, 49 variation in intensity 48 'Marsh Act, 1960' 28 Environment noise 68, 69 appraisals of 22, 90 aural 18, Ch.8 (68-73), 85 Offices, Shops and Railway Premises Act, 1963 19, 27, 28 definition of 25 space 39 thermal environment 62, 63 efficiency, and 29, 74, 75 Town Planning Acts 28, 45 external Ch.4 (36-38), 92 Lifts 34, 89 good 20 health, and 29, 30, 74 Lighting (see also, Daylighting, Electric lighting)18, 29, 30, 32, Ch.6 (48-60), 83, 92, 95, 141-143 inter-relation of environmental factors 22, 92 aesthetics of 51 spatial 18, Ch.5 (39-47), 83 stabilised 92 brightness patterns 49, 57, 59 British Lighting Council 55, 60 staff turnover, and 29 ceiling 43 subjective and social aspects of 22, 25, Ch.9 (74-76), 86-90,92 thermal 18, Ch.7 (61-67), 84 daylight/electric light compromise 18, 51, 52, 57-60, 143 efficiency, and 75 total 17-19, 20, 21, 22, 25, Ch.10 (79 -90) horizontal plane, on 50 visual 18, Ch.6 (48-60), 83 Illuminating Engineering Society 48, 49, 50, 51, 52, 55, 56,60 windowless 22, 48, 60, 92 Legislation 18, 49 External environment Ch.4 (36-38), 92 open plan, and 18, 21, 104 airconditioning, and 84 requirements of 49, 50, 51 micro-climate 25, 36 satisfaction with 48, 87, 89 thermal environment, and 84, 85 subjective response to 75 vertical plane, on 50 f.

Liverpool office space 99 airconditioning, and 21, 65 ceiling height, and 43 PRU survey in 25, 30-35 deep blocks, and 21, 104 size of offices 26 disadvantages of 125, 138 Wattleworth survey 29, 30, 40, 48, 61, 68 economics of 104 Location 20, 36-38, 98-103 flexibility of 104 Location of Offices Bureau (LOB) 37, 38 London, City of health, and 105 noise levels, and 85, 92 Noise Survey 68, 70 team spirit in 123 office space 98 thermal environment of 144 Robinson survey 29, 30, 40, 48, 61, 68 Organisation 104, 106, 123 O and M 21, 27, 80 Maintenance 34, 94 Orientation 33, 54, 62, 64, 67, 85 Management 106, 124, 125, 126, 128, 137, 138,139 Ownership 19, 28 Managers 104, 106, 123-126, 127, 128, 137-140 Manchester government 28 CIS 79, 80, 81 owner-occupier 18, 19, 20, 21, 28 office space 99 speculative developer 19, 20, 28 PRU survey 25, 30-35 Paired comparisons 108, III, 112, 113, 121,123, 126, 127 Mechanised offices (see, Automatic Data Processing) Partitioning Modules 21, 46, 47, 92 CIS 82, 122 CIS 83 flexibility of 21 Morale 88, 106, 124, 125, 126, 128 modules, and 46 noise insulation of 21, 70, 71 Noise 18, 33, Ch.8(68-73) 85, 92, 96, 146 Personal relationships (see also, Friendship air-borne 69 patterns) 44, 45, 76, 88, 89 airconditioning 69, 87 staff/management 88 ceiling height, and 43 Physical environment 18, 45, 90 construction type, and 54, 68, 71 surveys of 25, 29, 30, Ch.5, Ch.6, Ch.7, Ch.8 construction work 33, 69 Physicist 25, 94, 95, 96 corridor 33, 69 Plan-form 18, 45, 46 efficiency, and 68 core block 21, 46, 92 external 18, 33, 38, 68, 69, 71, 72 deep buildings 21, 44, 51, 57, 59, 60, 65,66, 92, 104 floor-to-floor 33, 69 lighting, and 18, 32 furniture 18 open-plan 18, 21, 43-46 health, and 68 slab block 21, 45, 46, 92 height, and 68 small office, and 18, 21, 43-46 insulation 70, 71 tall buildings, and 38, 65 internal 18, 33, 68, 69, 71 Plan-ratio 46 legislation 68, 69 Prestig,37, 45, 46, 90 levels 18, 68-73, 85, 86 Privacy\40, 69, 120, 123 loudness level (LL) 69, 70 Private office 20, 21, 44-46, 104, 105 masking noise 69 CIS 83, 89, 104 mechanised office 30, 33, 68, 71, 85 noise in 69-70, 72, 86 noise criteria (NC) 69, 70, 72 temperature in 33 open offices 72, 85, 92 Productivity 29, 124, 126, 137, 138 over shops 33 PSALI 51, 52, 57, 59, 95 privacy from 69 Psychologist 25, 95, 96 private office 72, 86, 92 Psychology reverberation time 70, 71 building design, and 105 room-to-room 33 daylight, and 57 site survey of 69 openable windows, and 53 size of space, and 18 psychological studies 17, 21, 25, 44. 57-60, sound level meter 70, 146 74-76, 86-90, 92, 104-140, 143, 147-156 sound pressure level (SPL) 69 space, and 45 sources 18, 33, 69 view, and 51 speech interference level (SIL) 69, 70, 71 standards of 70, 71 Qualitative aspects (see, Aesthetics) 74, 92 structure-borne 33, 69 surface treatment 18, 70, 71, 72 Recreation space 21 ...urveys 68, 70, 71-73, 85. 86, 146 Reflectivities 57, 84, 142 telephone 33, 69, 73 Research teams 17, 21, 25 town planning, and 18, 38, 68 Sex type of work, and 68 typewriter 33, 69 Analysis of attitudes to CIS building, 45,89, 109 -120 ventilation and 68, 69 differing thermal requirements 33, 61, 63 Sites waste-bins 18 cheaper in suburbs, etc. 20, 38 windows, and 18, 52, 53, 68, 69, 71 of CIS 80 North America 19, 60, 69, 80, 105 legislation 28 Office and shape of building 45 building 27, 28 Social environment Ch.9 (74-76), 90, 96,104-140 communications 21, 117 group size 44, 45, 75, 76, 88, 89,92, 104-140 definition of 27 interest in workers as people 75 employment 29, 37 Sociometrie studies 18, 96, 104-140 floor area 26, 27, 45 Socio- preferential choice (see, Friendshippatterns) Solar radiation 33, 52, 54, 64, 66, 67, 87 functional requirements 21, 90, 92 future 19 Spatial environment 18, 31, Ch.5(39-47), 83 group size 44, 45, 88 aesthetics of 39 CIS 83 height of building 27, 45 daylighting 39 machine operators 45 division of space 82,104-140 mechanisation of 20, 105 j furniture 39, 40 occupancy, size of 19, 21, 28, 92 gross office space 27, 83 organisation and method (0 & M) 21, 27, 80 legislation 39 rented 19, 20, 21, 28, 34 shape 45, 46 net office space 27, 83 prestige 39 size 88, 89, 92, 104-140, 150 statistics 27, 29, 36-38, 93, 98-103 and psychology 45, 75, 105 types of staff 29 questionnaires 130, 135, 139 types of user 28 and shape 39, 45 types of work 21, 29 space per person 39, 40, 83 typists 45 standards 39, 40 status 39 windowless 48, 60 Open-plan 18, 21, 43-46, 74-76, 104-140 subjective effects 104-140 surveys of 39-40 acceptability of 44, 45, 82, 88, 89, 104-128 types of office space 27 advantages of 125, 138 and ventilation 39

159 Standard industrial classification 28, 30, 92 CIS 83, 84 Statistics complex environmental problem 18, 33, 52-54, 65. 68, 69 Office space 36, 98-103 double glazing 53, 54, 71 ' Ails 39, 45 fixed 53, 73 Subjective aspects of environment 44,45, Ch.9 (74-76), 86-90, glare from 50, 52, 87, 89 104-140 heat-absorbing 54 colour 51, 87, 89 heat loss 52, 53 inter-relation of factors 22, 92 modules, and 46 lighting 57-60, 86, 87, 89, 104 noise, and 18, 52, 53, 68, 69, 71 surveys 30, 57-60, 74, 86-90, 104-140 openable 53, 54 thermal environment 61, 62, 64, 66, 86, 87, 89, 90 screening devices 54 Supervision 88, 125 solar radiation 18, 52, 66 Superv: ors 82, 88, 104, 106, 121-123, 127, 128, 133-136 ventilation, and 18, 52, 53, 54, 65, 68, 69 Surveys view 18, 21, 33, 44, 51, 52, 84, 87, 88, 89, 95, 104, 120, 121 interviews 86, 87, 88, 95 windowlessness 22, 48, 60, 92 measurements 95, 141, 142, 143, 144, 145, 146 working position choice 120 of noise 68, 71, 72, 73, 146 Working hours 31, 89 of space 39, 40 Working position, choice of 118-121, 126, 127, 132 of thermal environment 61, 63, 65, 66, 67, 84, 85, 144, 145 age 119, 120 of visual environment 48, 55, 56, 57, 141-143 airconditioning 120 other surveys published 29, 30 daylight 120, 121 postal (of location of offices) 36-38, 98-103 office size 119, 120, 121 questionnaires 88, 89, 90, 106-140, 147-156 privacy 120 social psychological 104-140 sex 119, 120 survey sample 30-35 view 120, 121 visits 30, 93, 94 Survey sample large 25, 30, 93 restricted 25, 38 sampling 93

Texture and subjective response 75 Thermal environment 18, 33, Ch.7(61-67) 84, 85, 86-90, 92, 95, 96, 144, 145 air change rates 39, 62, 84 air movement 62, 63, 64, 65, 66, 85, 144 Code of Practice 63 comfort zones 61-64 cooling 65 corrected effective temperature 63, 64, 65, 66, 67, 85 and construction type 54, 61, 63 diff!rences in response to, due to age 33, 63, 89 sex 33, 61, 63, 89 draughts 61, 62, 63, 65, 87, 90, 96 equivalent temperature 63, 64, 65, 85 globe thermometer 144 health 63 heat balance 61 heaters 33, 61, 64, 92 heating season 61 kata thermometer 144 legislation 62, 63 lighting installation 64 machine heat 64 odours 62, 66, 87 relative humidity 62, 63, 64, 65, 66, 84, 85, 144 standards of 62, 63 stuffiness 62, 66, 87 summer 33, 61, 62, 63, 64, 66, 85, 96, 145 supplementary heaters 61 surveys 61, 63, 65, 66, 67, 84, 85, 144, 145 temperatures 18, 61, 62, 63, 64, 65, 66, 67, 84, 85, 90, 144 corrected effective 63, 64, 65, 66, 67, 85 equivalent 63, 64, 65, 85 gradients 62, 65, 66, 85, 144 local control of 33 radiant 61, 62, 63, 64, 65-67, 84, 85, 144 variations of 33 thermal indices 63 ventilation 18, 61, 62, 63, 64, 65, 66, 67, 144 winter 61, 62, 63, 65, 66, 84, 96, 144 Total environment 17-19, 20, 21, 22, 25 in CIS building 79-90 Town planning Act 28 aesthetics 28, 38 daylight 18, 28, 38 noise 18, 38, 68

User studies 18,19, 25, 29, 30, 45, 48, 51, 57-60, 61, 62, 66, 68, 72, 74-76, 86-90, 104-140

Ventilation 18, 30, 33, 61-67, 86, 87, 144 mechanised office 30 noise, and 68, 69 space standards, and 39 windows, and 18, 52, 53, 54, 65, 68, 69 Visual environment(see also,Daylighting, Electric lighting, Lighting) 18, Ch.6 (48-60), 83, 84

Walls 57, 71, 72, 92, 94 Windows 18, 52, 53, 54 allocation of 51 area of 48, 53 160 ."

The designof roofs for single-storey general-purpose factories A reportupon an investigation

The Pilkington Research Unit'sfirst report upona building-type study describedan investigation into the design of the most widely usedindustrial building and was published in 1962 under the titleabove. The report was in fiveparts. In Part One the principal characteristics of existing Britishfactories were examined, and the effects ofcurrent legislation upon design reviewed. The components ofthe total problemwere considered singly in Part Two: thusthe economic background to factory designwas discussed, and separate chapters dealt with whatwas already known and what was discovered during the investigationabout such structural matters as the buildingframework, structural fire protection and buildingmaintenance, and suchen- vironmental factors as lighting, heatingand ventilation. The neglected subject of thermalproblems which arise during summer was considered.In Part Threean attempt was made to assess the total effectof the various structural and environmentalcomponents upon the economic choice of roof shapeand structural module with particular emphasison the choice between day- lighted and `windowless' factories.Current tendencies in the design of single-storeyfactories were reviewed in Part Four; matters considered includedAmerican and `windowless' factories, andstandardisation of factory structures. The conclusions to be drawnfrom the whole report were summarized in Part Five.The report contained extensive referencesto the ..:xisting literature, and a series of appendices concerned withthe research methodology. The report is now out of print.