potthoff.p65 An Evaluation of Patron Perceptions of Library Space 191 An Evaluation of Patron Perceptions of Library Space Using the Role Repertory Grid Procedure Joy K. Potthoff, David L. Weis, Dale S. Montanelli, and Matthew M. Murbach The evaluation of library space for its effectiveness and efficiency in meeting patron needs is an important part of the process in planning library facilities. However, techniques for obtaining such information are not widely available in the library science literature. The Role Repertory Grid Procedure, a technique derived from personal construct theory in the behavioral sciences, was evaluated for its practicality and validity as a way of gathering patron perceptions about the effectiveness and effi­ ciency of library space. The grid did provide useful information on pa­ trons’ perceptions of library space. However, although the Role Reper­ tory Grid Procedure was found to be fairly easy to administer, questions about its validity and reliability remain to be answered. The authors be­ lieve that further work to simplify the analysis will be necessary before the grid can be widely used in evaluating library space. This project was funded in part by the Council on Library Resources, in Washington, D.C. he evaluation of library space for its effectiveness and effi­ ciency in meeting patron needs is an important part of the process in planning library facilities. However, techniques for obtaining such information are not widely available in the library science literature. The Role Repertory Grid Procedure, a technique derived from personal construct theory in the behavioral sciences, was evaluated for its practicality and validity as a way of gathering patron perceptions about the effectiveness and efficiency of library space. The grid did provide useful infor­ mation on patrons’ perceptions of library space. However, although the Role Rep­ ertory Grid Procedure was found to be fairly easy to administer, questions about its validity and reliability remain to be answered. The authors believe that fur­ ther work to simplify the analysis will be Joy K. Potthoff is an Associate Professor in the School of Family and Consumer Sciences at Bowling Green State University; e-mail: jpottho@bgnet.bgsu.edu. David L. Weis is a Professor in the School of Family and Consumer Sciences at Bowling Green State University: e-mail: weis@bgnet.bgsu.edu. Dale S. Montanelli is an Associate Professor in the Department of Human and Community Development at the University of Illinois at Urbana-Champaign; e-mail: montanel@uiuc.edu. Matthew M. Murbach is a Technical Support Specialist at Software Solutions in Duluth, Georgia; e-mail: mmurbach@sofsol.com. 191 mailto:mmurbach@sofsol.com mailto:montanel@uiuc.edu mailto:weis@bgnet.bgsu.edu mailto:jpottho@bgnet.bgsu.edu 192 College & Research Libraries May 2000 necessary before the grid can be widely used in evaluating library space. The evaluation of library space for its effectiveness and efficiency in meeting patron needs is an important part of the process in planning new, remodeled, or realigned library facilities. Lamar L. Veatch discussed the importance of envi­ ronmental design in the planning of li­ brary space.1 He stressed the need for li­ braries to assess how their environments meet the social, physical, psychological, and behavioral needs of patrons and staff before beginning new construction, reno­ vation, or rearrangement projects. This type of assessment of library environ­ ments has become increasingly important as libraries change their activities to meet the needs of both modern technology and patrons.2 As the computer terminal re­ places the card catalog as the primary access point for information about the collection, the requirements for space, lighting, and even heating and/or air- conditioning change. It is only by exam­ ining how patrons and staff actually use and view the library facility that histori­ cal design errors can be identified (and occasionally corrected) and the impact of technological change on patron needs can be assessed. In truth, such empirical analysis should precede building plan­ ning, wherever practical, whether for new or renovated space. Joy Potthoff and Dale S. Montanelli reviewed research methodologies as tools for library space planning.3 Library stud­ ies tended to rely on literature reviews, interviews, and questionnaires but rarely used more than one technique or ad­ dressed questions about the validity of respondent reports. Studies from social science disciplines, which were mostly of classroom and office environments, also used questionnaires or interviews. How­ ever, these studies employed multiple techniques for gathering information and took steps to reduce potential self-report bias.4,5 Potthoff and Montanelli concluded that when considering library space prob­ lems, librarians need to use a broader range of assessment techniques to deter­ mine not only how many square feet are needed for a library function, but also how patrons and staff perceive that space.6 Because direct input from library patrons is important, it seems desirable to develop alternative methods of gath­ ering patron input that are not prone to the biases found in typical self-report methodology. Personal construct theory, originally developed in psychology, has been used successfully to gather perceptual data in areas of urban planning.7 Martha S. Wysor investigated attitudes of students toward environmental concerns such as pollution control and recycling.8 She com­ pared self-report methods with the per­ ceptual choices of students using the Role Repertory Grid Procedure. More recently, personal construct theory, with the cor­ responding Role Repertory Grid, was used to study aesthetic judgments about visual art and choice of shopping cen­ ters.9,10 Each of these researchers reported that the Role Repertory Grid Procedure was an effective tool for assessing indi­ vidual attitudes. The procedure was developed by George Kelly as a means of testing his per­ sonal construct theory.11 Kelley maintained that individuals use mental images of their environment or “personal constructs,” which are developed through experience, to guide their behavior. For Kelly, the task for researchers is to get individuals to de­ scribe these constructs without the biases inherent in self-reporting. John Harrison and Philip Sarre, a Brit­ ish geographer and a British social scien­ tist, applied the Role Repertory Grid Pro­ cedure to the study of mental images of the physical environment.12 Their most famous study was an attempt to measure the general image of the city of Bath held by a group of residents. Harrison and Sarre were interested in the applicability of the methodology, not just the specific outcomes.13 Although they focused on environments of a much larger nature (cit­ ies), the techniques they employed also may be appropriate for libraries. http:outcomes.13 http:environment.12 http:theory.11 An Evaluation of Patron Perceptions of Library Space 193 In the present research, the Role Rep­ ertory Grid Procedure was tested for its practicality as a way of gathering patron perceptions about library space. The test actually consists of two procedures. In the first, respondents are asked to identify relevant areas in a given space. In the sec­ ond, they are asked to make forced-choice decisions about the similarity and/or dif­ ference of triad groupings of photographs of areas identified in the first procedure, what Kelly called the method of triads.14 The triads are composed of three elements (or objects to which judgment is applied). The judgments reflect the personal con­ structs people use to assess the environ­ ment. Some evidence suggests that this procedure of asking respondents to select the photographs that are most similar is equivalent to asking them to indicate a preference.15 One component of this study was the gathering of data necessary for a practi­ cal solution to an existing space problem in a library at a state university in the Midwest. By using an existing space, the researchers were able to gain substantial information about an actual library envi­ ronment that is in need of rearrangement and renovation. The library is a part of the university’s Architecture Building. Constructed in 1926, it is Georgian revival in style, with tall, arched windows; cus­ tom-carved, stained pine woodwork; and plaster friezes on the walls. Consistent with the Role Repertory Grid Procedure, patron perceptions of space within the library were investigated in two independent studies. Responses from the participants in study 1 were used to construct the triads used in study 2. Study 1 A total of sixty-five respondents, in three separate groups of library patrons at a state university in the Midwest, were asked to respond to a questionnaire con­ cerning their knowledge and impressions of the library space. The groups consisted of twenty-three graduate students in li­ brary and information science (LIS), nine­ teen graduate students from the School of Architecture, and twenty-three under­ graduate students studying interior de­ sign. All sixty-five students reported that they were familiar with the floor plan of the library, although fourteen (21.5%) in­ dicated that they had never visited the library before. These three groups of pa­ trons were chosen for this preliminary phase because they were thought to use the library regularly. Differences between groups ap­ peared to occur as a reflection of the way in which the people who study different curricula use the library. Questionnaire The questionnaire contained a series of eleven questions and a floor plan dia­ gram, drawn to scale, of the library. The first question asked the patrons to circle and label areas on the floor plan that they could identify (e.g., journal area, modern architecture area, circulation desk). A sec­ ond question asked them if they under­ stood the floor plan. Possible answers were yes and no. The students then were asked to rate a series of seven features of the library on 5-point semantic differen­ tial item. The seven features were: (1) spa­ cious/crowded, (2) clean/dirty, (3) neat/ messy, (4) easy-to-locate materials/hard­ to-locate materials, (5) quiet/noisy, (6) well lit/dimly lit, and (7) comfortable temperature/uncomfortable tempera­ ture. The students then were given a group of open-ended questions asking them to name the most important and unimportant areas of the library, the fea­ tures of the library they most and least liked, and what they regarded as the library’s most prominent architectural feature. They also were asked how often they had visited the library previously. And finally, they were asked to identify any other library that had impressed them. The questionnaire took each stu­ dent about ten minutes to complete. Results 1. Areas of Library the Students Could Identify http:preference.15 http:triads.14 194 College & Research Libraries May 2000 Sixty-one of the sixty-five students indi­ cated they could identify twenty-two ar­ eas within the library. Areas cited by more than ten students included the main stacks, the periodicals area, the circulation desk, and the study area. In general, there was consistency across the three groups. The circulation desk was identified by approxi­ mately three-quarters of all subjects in each group, whereas the stacks and the periodi­ cals area were identified by more than one- half of each group. Differences between groups appeared to occur as a reflection of the way in which the people who study different curricula use the library. For example, 63.8 percent of the architecture students identified the reserve area, but few other students did. This may be the result of the architecture class reserves being held in this particu­ lar library, whereas LIS and interior de­ sign class reserves are held in other librar­ ies in the university system. Likewise, the high level of identification of the refer­ ence area by LIS students may be the re­ sult of their class assignments requiring use of this particular library. The fourteen areas that were identified consistently by more than five students in each group were selected for inclusion in study 2 of this project. In addition, study 2 included a ceramic tile mosaic of the library’s name. 2. Semantic Differential Items The students then were asked to rate a series of seven features of the library space, using a 5-point semantic differen­ tial format. The means and standard de­ viations for these seven semantic differ­ entials are presented in table 1. Overwhelmingly, the students re­ garded the library as crowded. Although responses were somewhat mixed, as a group, the students felt that the library was slightly dirty and messy and that it was difficult to locate materials. On the other hand, most of the students regarded the library as quiet and well lit. A principal component factor analysis of these items failed to yield any clear underlying factor structure, showing that responses to these various features of the library were relatively independent (re­ sponses to any given item were not strongly correlated with responses to the others). Moreover, tests of internal con­ sistency (Cronbach’s alpha) also showed that the responses were relatively uncorrelated. A few individual compari­ sons were exceptions and are worth not­ ing. Perceptions that the library is clean were correlated with perceptions that the library is neat (r = .48, p < .01). Percep­ tions that the library is quiet were corre­ lated with perceptions that it is well lit (r = .32, p < .01), and perceptions that the library is well lit were correlated with perceptions that it has a comfortable tem­ perature (r = .38, p < .01). 3. Most Important Area TABLE 1 Library Patrons Perceptions of Features of Library Space Feature Mean* SD Crowded/spacious 1.78 0.86 Dirty/clean 2.83 0.95 Messy/neat 2.72 1.02 Hard to locate materials/easy to locate materials 2.69 1.12 Noisy/quiet 3.12 1.16 Dimly lit/well lit 3.75 0.91 Uncomfortable temperature/comfortable temperature 3.60 0.93 Note: Individual items were scored on a five-point scale, ranging from I, negative quality, to 5, positive quality. An Evaluation of Patron Perceptions of Library Space 195 In response to the question about which feature of the library the students consid­ ered most important (six missing), nine areas were identified. More than ten per­ sons named the periodical area and the study area. Five to nine students cited the main stacks, the circulation area, and the reference area. 4. Least Important Area No area of the library was mentioned by more than ten persons in response to a question about which area was least im­ portant (seventeen missing). The study area, the vertical files, and the folios area were mentioned by five to nine persons as the least important library feature. Stu­ dents identified a total of eighteen areas in response to this question. 5. Feature Liked Most Respondents also were asked which fea­ ture of the library they liked most (nine missing). Nineteen areas were named. The entrance, the study area, the wood­ work, and the lighting were cited by five to nine persons; no feature was men­ tioned by more than ten persons. 6. Feature Liked Least In response to the item asking students to name the feature of the library they liked least (seven missing), sixteen areas were identified. Five to nine persons men­ tioned lighting and noise, and more than ten persons noted that the library was crowded and poorly organized. 7. Architectural Feature Liked Most When students were asked to name the architectural feature they liked the most (two missing), they named a total of ten features. Five to nine persons cited the windows, and more than ten students named the entrance and the woodwork. 8. Most Prominent Architectural Fea­ ture In response to a question asking students to name the library’s most prominent ar­ chitectural feature (nine missing), the stu­ dents named a total of eleven features. Five to nine persons cited the windows and the woodwork, and thirty-three stu­ dents (58.9%) named the entrance. The mention of the entrance as the most prominent architectural feature was, by far, the most common response to this entire series of open-ended questions about the library. 9. Number of Visits to Library The students then were asked to indicate how often they had visited the library. A summary of their responses is presented in table 2. Although a majority reported that they had visited the library more than ten times, it is worth noting that a sizable group (21.5%) reported that they had never been to this library. The number of visits students had made to the library was significantly correlated with a num­ ber of ratings of library features. Those students who reported having visited the library more frequently were more likely to rate it as difficult to locate materials (r = -.28, p < .05), dimly lit (r = -.30, p < .05), and having an uncomfortable tempera­ ture (r = -.33, p < .01). 10. Impressed by Other Libraries Finally, the students were asked if they had ever visited another library that had impressed them. Forty-seven (77.0%) re­ sponded yes. The responses to this item were not significantly correlated with any of the perceptual ratings of the library fea­ tures. Discussionf orfStudyf1 The first study was designed to ask a group of students to evaluate the library using fairly standard questionnaire items and to identify areas of the library for in- TABLE 2 Frequency of Student Visits to the Architecture Library (N = 65) No. of Visits No. of % Respondents None 14 21.5 One time 7 10.8 3-5 times 2 3.1 8-9 times 8 12.3 More than 10 times 34 52.3 Note: The above categories were in response to an item asking respondents to indicate the number of times they had visited the library. 196 College & Research Libraries May 2000 clusion in the Role Repertory Grid Pro­ cedure to be used in study 2. Many of the students participating in study 1 indi­ cated that they were familiar with several areas of the library. Differences in the ar­ eas cited by students seemed to be asso­ ciated with their major area of study. A total of fourteen library areas were iden­ tified by the students for inclusion in the procedure in study 2. Although the library was a positive feature for a large group of the students, a sizable minority had negative feelings about it. Clearly, the outstanding features were the entrance, the woodwork, and the windows. In general, most of the students found the library to be well lit and quiet, but also crowded, dirty, and messy. At­ tempts to develop a single scale for these various evaluations of the library space failed, indicating that these assessments were independent of each other. Interest­ ingly, the more frequently students used the library, the more likely they were to evaluate it negatively. Study 2 In study 2, a short questionnaire and an interview were given to an independent group of forty patrons of the library. A research confederate approached persons who were in the library and asked if they would be willing to participate. Those who agreed were escorted to a table, where they were seated and asked to in­ dicate their staff or student status at the university, their academic department or major, and their gender. They also were asked how often they used the library fa­ cilities. Of the forty who completed the survey, thirty were undergraduates, three were graduate students, six were staff members, and one was a faculty member of the university. The overwhelming ma­ jority of these patrons (n = 28, 70%) were majors in or members of the architecture program. However, the sample also in­ cluded majors in anthropology, architec­ ture and history, art history, electrical en­ gineering, interior design, and the library staff. Twenty of these respondents were male, and twenty were female. Thirteen (32.5%) reported daily use of the library, twenty-two (55%) reported weekly use, two (5%) reported bimonthly use, two (5%) reported monthly use, and one (2.5%) reported using the library rarely. Role Repertory Grid Procedure (Method of Triads) The fourteen areas of the library most fre­ quently named by the study 1 participants as areas they could identify were selected for inclusion in study 2. In addition, study 2 included a ceramic tile mosaic of the library’s name. Each of these areas was then photographed and the photographs numbered, and the identifying numbers were randomly selected without replace­ ment to produce five sets of three pictures each. This sampling was done three times to create a total of fifteen triads of photo­ graphs. Participants were asked to indi­ cate which two pictures in each triad were most similar and why, and which two were most different and why, a procedure that has been found to generate useful judgment data.16 The interviewer re­ corded the choices for each triad and the exact phrases used to explain the choices on a specifically prepared report sheet. Each participant required an average of thirty minutes to complete the interview. The reasons given for the selection of similar and different pictures were sub­ jected to a form of content analysis known as thematic analysis.17,18 Following proce­ dures established by Timothy Perper and David L. Weis, the researchers identified all the themes given by respondents as reasons for choosing the photographic pairings they did.19 All responses were read several times, first to identify themes and later to obtain theme frequencies. Every reason given was either assigned a theme or recognized as unique. For a rea­ son to be identified as a theme, it had to be explicitly named by at least two respon­ dents. By definition, the themes are com­ ments made explicitly and overtly by the respondents. No implicit categories de­ veloped by the presuppositions of the re­ searchers were coded as a theme. Thus, An Evaluation of Patron Perceptions of Library Space 197 the respondents were able to express their own thoughts rather than having their thoughts interpreted by the researchers. Moreover, this approach allowed multiple themes to be identified for any particular respondent. In fact, many respondents gave more than one reason for making a particular choice. This ranged from a low of one theme to a high of seven. After several readings to develop a complete catalog of themes, the research­ ers made a final calculation of theme fre­ quencies (which represents the percent­ age of respondents mentioning each theme). This coding was performed by two of the researchers who initially agreed on 96.8 percent of the similarity codings and 94.3 percent of the difference codings. Because, by definition, a theme had to be stated explicitly in order to be counted, ambiguous statements failed to meet the criteria. No formal measures of interrater reliability were calculated be­ cause such measures presuppose inde­ pendent ratings of a priori categories. In this study, the researchers worked to­ gether to develop the theme categories. Only those statements the researchers agreed had been made explicitly were coded as themes. It will remain for a fu­ ture researcher to use this system to ad­ dress the question of interrater reliability among independent judges. Results Explanations of the reasons for choosing particular pairs of photographs as simi­ lar or different ranged from brief, terse comments to highly detailed, elaborate analyses. Ultimately, the researchers iden­ tified and coded a total of ten themes: 0. Do not use library: The respondent explicitly stated that he or she did not use the library. Although this might imply that no other themes would be men­ tioned, some respondents did proceed to offer further reasons for the choices they had made. 1. Learning materials: The respondent explicitly stated that books, folios, shelves, or other learning materials were either similar or different. 2. Space concerns: The respondent explicitly described the space in the pho­ tographs as messy, neat, open, closed, proximal, cluttered, confined, accessible, organized, or the like. 3. Architecture: The respondent cited architectural features as similar or differ­ ent in the photographs, such as windows, doors, arches, woodwork, or metal, or described the library as new or old. 4. Lighting: The respondent explicitly mentioned lights or lighting, or described an area as bright, dark, well lit, or poorly lit. 5. Furniture: The respondent explic­ itly cited furniture such as chairs, tables, or computers as the basis for concluding that photographs were similar or differ­ ent. 6. Acoustics: The respondent explic­ itly described an area as noisy or quiet. 7. People: The respondent explicitly mentioned people, crowds, or social ac­ tivities, or described the area as busy or lonely. 8. Aesthetics: The respondent explic­ itly described the aesthetics or beauty of the area using words such as ugly, pretty, colorful, good view, casual, formal, or in­ viting. 9. Function: The respondent chose photographs as similar or different be­ cause of the ways people used the space or because of the functions the space served. Examples included statements about using an area, doing something in the area, looking up information, study­ ing in the area, not using the area much, and stating that it was difficult to use the area. This theme also was coded if the respondent stated that he or she sat or walked in the area. For most of the photograph triads, the same pair was picked as similar by most respondents, and the reasons for similar­ ity, although expressed differently, tended to focus on the same concept, such as learning materials or the architecture of the space. There appeared to be less con­ sistency in the selection of photograph pairs as different. In some cases, the pair in each triad that was selected as most 198 College & Research Libraries similar also was selected as most differ­ ent. A review of the theme frequencies for similarities and differences showed that the principle themes were learning ma­ terials, architecture, and function. Themes that were mentioned by few respondents included do not use, space concerns, light­ ing, furniture, acoustics, people, and aes­ thetics. Because of the relatively small sample and the large number of poten­ tial cells, the only themes included in sub­ sequent chi-square analyses were learn­ ing materials, architecture, and function. A summary of those analyses is presented in tables 3 and 4. The Role Repertory Grid Procedure did yield useful information about the universe of themes that library patrons use in making judgments about library space. The results displayed in tables 3 and 4 establish that a large number of the respon­ dents (n > 20, % > 50) used the learning materials, architecture, or function themes to explain their choices of which photo­ graphs in the fifteen triads were similar or different. The tables suggest that there was greater uniformity in response to the simi­ larity reasons than the difference reasons, but these three themes dominated both sets of responses. The tables also indicate that there were some significant differences in the reasons provided for similarity and difference choices. Six of the similarity comparisons yielded significant differ­ ences, but only two of the difference com­ parisons were significant (p < .01). This alpha level was chosen because of the large number of tests conducted to reduce the probability of type-I errors. Unfortunately, there did not appear to be any clear un­ derlying pattern to those differences in the series of 3 x 3 tables. Subsequent chi-square analyses indi­ cated that there were no significant differ­ ences (p > .01) between males and females; among faculty, staff, graduate students, or undergraduate students; or varying fre­ quencies of library use in the making of similarity and difference judgments. May 2000 Discussionf orfStudyf2 Study 2 was designed to test the useful­ ness of the Role Repertory Grid Proce­ dure, a methodological approach devel­ oped to offset many of the problems associated with standard self-report methods used in the evaluation of space. The procedure did prove to be easy to ad­ minister. Respondents were readily able to make choices about which of fifteen tri­ ads of photographs were most similar and most different, and they were able to pro­ vide reasons for the choices they made. Moreover, the Role Repertory Grid Pro­ cedure and the thematic analysis em­ ployed in this research to analyze re­ sponses seemed to be a valuable way to identify the entire spectrum of possible responses. The researchers’ analysis of themes that were relevant to the respon­ dents’ assessments of the library space did not impose arbitrary factors on the persons participating in the research. Rather, this method allowed respondents to speak for themselves. This thematic analysis succeeded in identifying ten themes that respondents cited for the judgments they made. The most prominent of these were learning materials, architecture, and function. Some respondents mentioned other themes, but these were less common. Additional research on the reliability and validity of this set of themes will be nec­ essary. On the other hand, this approach also seemed to possess several limitations. The thematic analysis was time-consum­ ing and highly complex, perhaps too com­ plex for many librarians to use. In addi­ tion, the small sample used in this re­ search prevented a more systematic sta­ tistical analysis of the less frequently cited themes. General Discussion The present research was developed on the premise that assessment of patron per­ ceptions is an important component of plans for the utilization and design of li­ brary space. This project attempted to test a new approach, the Role Repertory Grid Procedure, to collecting data on patron An Evaluation of Patron Perceptions of Library Space 199 TABLE 3 Frequency of the Use of the Learning Materials, Architecture, and Function Themes Cited in Similarity Judgments, with Corres(onding Chi-square Analyses Set Pair Selected as Similar Learning Architecture Function X2 p Materials 1 Bound journals/stacks 0 2 0 (005) Bound journals/picture files 0 2 0 23.71 .0000* Stacks/picture files 1 21 0 2 Reserves/card catalog 0 7 1 (007) Reserves/circulation 0 0 2 23.47 .0000* Card catalog/circulation 13 2 2 3 Periodicals/online catalog 4 1 8 (009) Periodicals/circulation 0 0 1 7.13 .1291 Online catalog/circulation 1 6 5 4 Vault/entrance 23 0 3 (011) Vault/window folios 3 0 0 33.44 .0000* Entrance/window folios 0 4 0 5 Mosaic/study table 1 1 4 (013) Mosaic/copier 4 2 1 12.60 .0124 Study table/copier 1 0 12 6 Reserves/stacks 0 2 1 (015) Reserves/circulation 1 0 2 11.65 .0202 Stacks/circulation 13 2 2 7 Window folios/journals 5 18 7 (017) Window folios/vault 1 0 1 5.62 .2292 Journals/vault 0 0 1 8 Study table/picture files 4 1 2 (019) Study table/arch. ref. 0 4 3 8.65 .0704 Picture files/arch. ref. 1 4 5 9 Copier/card catalog 0 2 0 (021) Copier/periodicals 0 24 0 27.00 .0000* Card catalog/periodicals 0 0 1 10 Entrance/online catalog 1 1 6 (023) Entrance/mosaic 8 2 2 15.68 .0035* Online catalog/mosaic 0 4 2 11 Periodicals/picture files 7 9 1 (025) Periodicals/circulation 6 1 0 6.66 .0904 Picture files/circulation 1 2 1 12 Entrance/online catalog 8 1 6 (027) Entrance/window folios 3 2 0 4.81 .0904 Online catalog/window folios 0 0 0 13 Vault/reserves 1 3 3 (029) Vault/card catalog 3 3 1 2.08 .7213 Reserves/card catalog 4 7 5 14 Copier/circulation 1 1 1 (031) Copier/bound periodicals 2 12 1 15.34 .0040* Circulation/bound periodicals 0 0 4 15 Mosaic/study table 0 5 0 (033) Mosaic/stacks 1 14 0 0.35 .5536 Study table/stacks 0 0 0 200 College & Research Libraries May 2000 TABLE 4 Frequency of the Use of the Learning Materials, Architecture, and Function Themes Cited in Difference Judgments, with Corres(onding Chi-square Analyses Set Pair Selected as Similar Learning Architecture Function X2 p Materials 1 Bound journals/stacks 4 8 1 (035) Bound journals/picture files 0 2 0 3.45 .4861 Stacks/picture files 1 0 0 2 Reserves/card catalog 0 5 0 (038) Reserves/circulation 2 10 2 4.09 .3935 Card catalog/circulation 0 1 1 3 Periodicals/online catalog 1 4 1 (041) Periodicals/circulation 0 1 7 10.49 .0330 Online catalog/circulation 0 0 4 4 Vault/entrance 1 0 0 (044) Vault/window folios 2 12 4 4.26 .3723 Entrance/window folios 3 3 1 5 Mosaic/study table 2 0 3 (047) Mosaic/copier 0 1 8 9.13 .0580 Study table/copier 2 4 3 6 Reserves/stacks 0 4 3 (050) Reserves/circulation 3 14 0 15.63 .0036* Stacks/circulation 0 0 2 7 Window folios/journals 0 0 0 (053) Window folios/vault 1 2 5 3.27 .1947 Journals/vault 3 7 3 8 Study table/picture files 0 1 7 (056) Study table/arch. ref. 0 2 3 12.38 .0147* Picture files/arch. ref. 2 1 0 9 Copier/card catalog 0 12 3 (059) Copier/periodicals 0 0 0 0.95 .8297 Card catalog/periodicals 0 4 0 10 Entrance/online catalog 1 3 4 (062) Entrance/mosaic 1 2 3 0.88 .9281 Online catalog/mosaic 0 2 3 11 Periodicals/picture files 0 0 0 (065) Periodicals/circulation 1 3 0 0.28 .5967 Picture files/circulation 4 6 0 12 Entrance/online catalog 0 0 1 (068) Entrance/window folios 1 1 2 5.31 .2574 Online catalog/window folios 0 7 4 13 Vault/reserves 0 4 4 (071) Vault/card catalog 0 5 3 0.73 .6930 Reserves/card catalog 0 5 2 14 Copier/circulation 1 3 3 (074) Copier/bound periodicals 0 1 2 2.02 .7320 Circulation/bound periodicals 2 4 2 15 Mosaic/study table 0 6 3 (077) Mosaic/stacks 1 1 0 13.54 .0089* Study table/stacks 0 9 0 An Evaluation of Patron Perceptions of Library Space 201 perceptions and to compare this approach to more standard self-report question­ naire methods. The preliminary step of asking library patrons to identify areas of the library they knew seemed to be an effective device for constructing the set of fifteen photograph triads used in study 2. The Role Repertory Grid Procedure did yield useful information about the uni­ verse of themes that library patrons use in making judgments about library space. Additional research is needed to deter­ mine the validity and reliability of this approach, whether the themes identified in this project can be used in a priori analyses of library space, and whether the Role Repertory Grid Procedure can be converted into a more statistically usable and “librarian-friendly” format. However, the results of study 1 and study 2 also illustrate the importance of developing such a methodological ap­ proach in the study of patron perceptions of library space. The typical evaluation items in study 1 indicated that students found the library to be messy and dirty (space concerns), well lit (lighting), quiet (acoustics), and crowded (people). Al­ though these themes were identified in the Role Repertory Grid Procedure in study 2, they were not prominent or com­ mon reasons given for the similarity or difference choices made. Instead, the themes of learning materials, architecture, and function emerged as most prominent. This suggests that direct, self-report ques­ tioning and less overt decision-making methodologies may yield divergent re­ sults, a finding that may have great rel­ evance for the design of building spaces in general. As just one example, it raises questions about which themes or factors people actually use in making judgments about space. Perhaps a lens model meth­ odology could be used to determine such social judgments.20 Ultimately, the researchers were con­ cerned about perceptions of space in the library in order to improve the efficiency and effectiveness of the delivery of library services. This research demonstrated that different methodological approaches do seem to yield different sets of results. The project also has identified a number of problems, in both the library itself and the methodological tools available to re­ searchers. Although numerous questions remain about how to assess patron needs and perceptions accurately, the results do provide a challenge to develop more so­ phisticated and more accurate assessment tools. Notes 1. Lamar L. Veatch, “Toward the Environmental Design of Library Buildings,” Library Trends 36, no. 2 (1987): 361–76. 2. Michael C. Matier and C. Clinton Sidle, “Developing a Strategic Plan for Library Space Needs through 2010” (paper presented at the Annual Spring Conference of the Society for Col­ lege and University Planning, Philadelphia, Apr. 8–10, 1992). 3. Joy Potthoff and Dale S. Montanelli, “Use of Library Facilities: Behavioral Research as a Tool for Library Space Planning,” Journal of Library Administration 12, no. 1 (1990): 47–61. 4. James S. Kidston, “The Validity of Questionnaire Responses,” Library Quarterly 55 (Apr. 1985): 133–50. 5. Bernard S. Phillips, Social Research: Strategy and Tactics (New York: Macmillan, 1971). 6. Potthoff and Montanelli, “Use of Library Facilities,” 58. 7. George A. Kelly, A Theory of Personality (New York: W. W. Norton, 1963). 8. Martha S. Wysor, “Comparing College Students’ Environmental Perceptions and Atti­ tudes: A Methodological Investigation,” Environment and Behavior 5, no. 5 (1983): 615–45. 9. Rosamund Joyce Osbourn, “The Aesthetic Response: An Application of Personal Con­ struct Theory to the Perception and Appraisal of Visual Art” (master’s thesis, Univ. of Exeter, Dorset, Eng., 1988). 10. Harry Timmermans, Rob Van Der Heiden, and Hans Westerveld, “The Identification of Factors Influencing Destination Choice: An Application of the Repertory Grid Methodology,” Transportation 11 (1982): 189–203. 11. Kelly, A Theory of Personality. http:judgments.20 202 College & Research Libraries May 2000 12. John Harrison and Philip Sarre, “Personal Construct Theory in the Measurement of Envi­ ronmental Images: Problems and Methods,” Environment and Behavior 3, no. 4 (Dec. 1971): 251– 374. 13. ———, “Personal Construct Theory in the Measurement of Environmental Images,” En­ vironment and Behavior 7, no. 1 (Mar. 1975): 3–58. 14. Kelly, A Theory of Personality. 15. Timbermans, Van Der Heiden, and Westerveld, “The Identification of Factors Influencing Destination Choice,” 193. 16. Ibid., 196. 17. Bernard Berelson, Content Analysis in Communication Research (Glencoe, Ill.: Free Pr., 1952). 18. Thomas F. Carney. Content Analysis: A Technique for Systematic Inference from Communica­ tions (Winnipeg, Can.: Univ. of Manitoba Pr., 1972). 19. Timothy Perper and David L. Weis, “Proceptive and Rejective Strategies of U.S. and Ca­ nadian College Women,” Journal of Sex Research 23 (Nov. 1987): 455–80. 20. Ray W. Cooksey, Judgment Analysis: Theory, Methods, and Application (San Diego, Calif.: Academic Pr., 1996).