Evaluation of the physicalclassroom by students andprofessors: a lens modelapproachDarren Douglas and Robert Gifford, Department of Psychology,University of Victoria, PO Box 3050,Victoria, British Columbia, Canada V8W3P5
SummaryA method for linking classroom evaluations to speciŽ c physical properties and forcomparing the evaluations of different groups is described and illustrated.Thirty-Ž ve college classrooms were photographed and shown to 20 professors and 51undergraduate students, each of whom evaluated the friendliness of and theiroverall preference for all the classrooms. Seven physical properties of the class-rooms were reliably assessed by independent observers. Using a modiŽ edBrunswik lens model, the relations between the physical properties and the evalu-ations by the two groups were established and compared. Between 40 and 57 percent of the variance in the evaluations could be explained from only three class-room properties: view to outdoors, seating comfort and seating arrangement.Evaluations by the students and professors were surprisingly similar, an encour-aging sign for classroom designers.Keywords: classrooms, evaluation, professors, students, lens model, Brunswik
Students spend thousands of hours in classrooms, and therefore classrooms auto-matically are among the most important physical structures in society. Althoughmuch has been written about classroom environment in the sense of organiz-ational structure or social environment (e.g. Byrne, Hattie and Fraser, 1986;Yuen-Yee and Watkins, 1994), much less has been written about the physicalenvironment of the classroom.That is not to say the topic is untouched; especially during the open-classroomand open-education debates, the physical setting of the classroom was examined
Address for correspondence: Robert Gifford, Professor, Department of Psychology,University of Victoria, PO Box 3050, Victoria, British Columbia, Canada V8W 3P5. Tel:250-721-7532; Fax: 250-721-8929; E-mail: RGifford@UVic.ca
Educational Research Vol. 43 No. 3 Winter 2001 295–309
Educational Research ISSN 0013-1881/print/ISSN 1469-5847 online © 2001 NFERhttp://www.tandf.co.uk/journalsDOI: 10.1080/0013188011008105 3

(e.g. Proshansky and Wolfe, 1974). Others have investigated the physical settingof the classroom in terms of student privacy (e.g. Ahrentzen and Evans, 1984;Pedersen, 1994) or performance (e.g.Wollin and Montagne, 1981), the in uenceof the classroom’s physical arrangement on visitors’ impressions of the teacherand students (e.g. Weinstein and Woolfolk, 1981) or how the physical classroomin uences student and teacher behaviour (e.g. Hooper and Reid, 1985; Lough-lin, 1977).However, an important goal of classroom design (among others, of course) isto create a space that students and educators like. This rather obvious goal hasnot received the attention it deserves, with some exceptions. Sommer and hisassociates (Sommer and Olsen, 1980; Wong, Sommer and Cook, 1992) createdand have continued to monitor a ‘soft’ university classroom (with fabric wall dec-orations, covered seats and carpeting). Students found it pleasing and their par-ticipation in classroom discussions was enhanced, compared to a traditional,‘hard’ classroom. Axia, Baroni and Peron (1990) reported that greater experiencewith classrooms was related to more favourable reactions to different kinds ofclassroom physical arrangements.So far, however, surprisingly little published research has appeared on the ques-tion of which (if any) speciŽ c elements of classrooms are linked to student andteacher preferences. One exception (Cohen and Trostle, 1990) was a study ofkindergarten and Ž rst-grade classrooms.The present study examines student andprofessor evaluations of university classrooms, and also illustrates a method bywhich the speciŽ c physical bases of these evaluations might be identiŽ ed.Although this study focuses on university classrooms, professors and undergrad-uate students in one particular region, the method could be applied to classroomsat any educational level in any culture.
The Brunswik lens modelLong ago, Brunswik (1956) developed the lens model as a paradigm for study-ing environmental perception. The model is a probabilistic representation of theway perceivers use environmental cues to draw inferences about the environment.Only recently has the lens model been applied to judgements of architecture (e.g.Gifford et al., 2000). The present study makes use of the lens model by measur-ing classroom physical cues such as whether the room has a view of the outdoorsand the type of seating arrangement and relating these objective features tostudent and professor evaluations.The traditional lens model includes ecological validity on its left side, that isa set of links between a ‘true’ measure of the environment and some observableproperties. On the other side of the lens model is cue utilization, that is the linksbetween the observable properties and individuals’ evaluations. For this study,the lens model is modiŽ ed so that both halves depict cue utilization, one side foreach of two groups whose evaluations are being considered. This is because thereis no ‘true’ score in the present circumstances; instead there are two sets of evalu-ations. The modiŽ ed lens model permits a clear, graphic method of comparingthe cue utilization policies of two groups, in this case, undergraduate studentsand professors (see Figure 1).Although lens models might be constructed for groups who are unfamiliar withthe setting in question, this study deliberately examined the reactions of twogroups of evaluators who have an important stake in the quality of university
296 Educational Research Volume 43 Number 3 Winter 2001