Part 2: Data Tables and Charts

Chart 1 derives from the study sample (i.e., the institutions that responded to the study survey). It reports the distribution by size (i.e., number of gross square feet) of all library projects in the sample.

Chart 2 derives from the study sample (i.e., the institutions that responded to the study survey). It reports the distribution by size (i.e., number of gross square feet) of library projects in the sample that involved 100,000 gross square feet or less.

Table 3a reports the responses to question 1 in the study survey. Question 1 identified several different possible motivators (items a–n in the first column) for library capital projects and asked respondents to indicate on a six-point scale how strongly each factor motivated the respondent’s project. The percentage of all responses to a given motivator (e.g., “growth of library staff”) that occupied a given point in the response scale (e.g., “not a factor”) is recorded, along with confidence interval for that percentage. A third number, called the chi-square factor, is also provided when the response varied in a statistically significant way from what one would expect in a random distribution of responses. The higher the chi-square factor, the more significant is the result. So, for example, Table 3a reports that 38.6% of all survey respondents who answered question 1a, about the growth of library staff, said that this was not a factor in their planning.
One can be 95% confident that responses to question 1a would fall within the range of 38.6 ±6.4% for the entire population of study projects. The chi-square factor of 5.8 indicates this response differs in a highly significant way from what one would expect in a random, or chance, distribution of responses to this question. In this case, dark shading is used to indicate that the response occurs more frequently than in a random distribution; light shading used elsewhere (e.g., question 1d) indicates a response that occurs less frequently than would be expected in a random distribution.

Caveat about Table 3b: Readers should understand that the relatively small number of institutions representing many of the Carnegie classification types in the study sample, as reported in Table 3b, makes inferences about the larger study population somewhat unreliable. This is indicated in the wide confidence intervals reported in this table.

Table 3c (on p. W-90) reports the responses to question 1 in the study survey but differs from Table 3a in sorting the responses according to the year when projects were completed.

Caveat about Table 3c: The relatively small number of projects completed in most years in the study sample, as reported in Table 3c, makes inferences about the larger study population somewhat unreliable. This is indicated in the wide confidence intervals reported in this table.

Table 4a reports the responses to questions 2–13 in the study survey. It is similar to Table 3a except that all the questions are either explicitly or implicitly yes/no questions (rather than questions that invite responses on a scale). The percentage of all respondents to a given question who answered in the affirmative is recorded, along with a confidence interval for that percentage. A third number, called the chi-square factor, is also provided when the response varies in a statistically significant way from what one would expect in a random distribution of responses. The higher the chi-square factor, the more significant is the result. So, for example, Table 4a reports that 84.8% of all survey respondents who answered question 3a, about the systematic assessment of library operations, said that such an assessment was done. One can be 95% confident that responses to this question would fall within the range of 84.8 ±4.7% for the entire population of study projects. The chi-square factor of 28.3 indicates this response differs in a highly significant way from what one would expect in a random, or chance, distribution of responses to this question. In this case, dark shading is used to indicate that the response occurs more frequently than in a random distribution; light shading used elsewhere (e.g., question 3c) indicates a response that occurs less frequently than would be expected in a random distribution.

Caveat about Table 4b: The relatively small number of institutions representing many of the Carnegie classification types in the study sample, as reported in Table 4b, makes inferences about the larger study population somewhat unreliable. This is indicated in the wide confidence intervals reported in this table.

Table 4c (on p. W-100) reports the responses to questions 2–13 in the study survey but differs from Table 4a in sorting the responses according to the year when projects were completed.

Caveat about Table 4c: The relatively small number of projects completed in most years in the study sample, as reported in Table 4c, makes inferences about the larger study population somewhat unreliable. This is indicated in the wide confidence intervals reported in this table.

Table 5 describes the distribution of projects by lead architects and of lead architects by projects in the study population.

Comments are recorded under the survey question to which respondents
attached them. This summary (on p. W-106) attempts to capture responses that are not well represented in the quantitative data or that express particularly well an idea often expressed in the comments. Where appropriate and possible, a summary analysis is offered of subjects that figure frequently in the comments.

Summary and Partial Transcriptions of the Phone Interviews with Library Directors and Chief Academic Officers

As a part of this study, phone interviews were conducted with 25 library
directors from institutions that responded to the study survey. Six presidents or chief academic officers (CAOs) at the institutions involved in the library director interviews were also interviewed. The summaries have been edited to provide anonymity. They are grouped according to the Carnegie classification of the institutions involved, as follows:

Doctoral/Research Universities-Extensive

CAO interviews 1–4
Library director interviews 7–14

Doctoral/Research Universities-Intensive

CAO interview 5
Library director interviews 15–18

Master’s Colleges and Universities I

Library director interviews 19–22

Master’s Colleges and Universities II

Library director interviews 23–24

Baccalaureate Colleges-Liberal Arts

CAO interview 6
Library director interviews 25–29

Baccalaureate Colleges-General

Baccalaureate Colleges-General

Click here (goes to p. W-116) for summaries of those interviews, with partial transcriptions of what those interviewed said.

Quantitative Findings of a Survey Conducted for the Council of Independent Colleges
Tables 6a and 6b present some of the findings of a November 2001 survey of library directors and chief academic officers (CAOs) at institutions that are members of the Council of Independent Colleges (CIC). The CIC is an association of independent colleges and universities working together to support college leaders, advance institutional excellence, and enhance private higher education’s contributions to society. To fulfill its mission, CIC provides ideas, resources, and services that assist institutions in improving leadership expertise, educational programs, administrative and financial performance, and institutional visibility. See http://www.cic.edu/ for more information about the CIC and for the original publication of the data reported here.

Description of the Summary Tables

• Questions 1 and 2 prompted respondents to agree or disagree with statements in a survey statement, using a five-point scale to register their views. Individual statements are listed in each row of the table followed by a set of columns that report (a) the percentage of all respondents using each of the prompted responses (bold face) and (b) the confidence interval for the response (italics). This is followed by two other statistics: (c) the median view among all responses, which is a crude measure of central tendency among the responses but the only one available for this kind of data; and (d) the results of a chi-square test for independence between the responses to each statement provided by CAOs and library directors. Where this test indicates that differences in the responses are statistically significant, the test result is reported in bold face followed by (e) the percentages of CAO and library director responses, provided to facilitate understanding of how they differ.Look, for instance, at question 1.04:(a) Of the 101 respondents, 70% strongly agree with the statement
  that library staff should provide in-class library instruction; 24% agree with this assertion, while 4% are undecided or have no opinion. Two percent of the respondents disagree with the statement,
  but none strongly disagrees with it.(b) Each of these response values is followed by a confidence interval. In the case of the 70% who strongly agree, for instance, one can be confident that 95% of the time between 60% and 80% (i.e., 70% plus or minus 10%) of the larger population of all CAOs and library directors at CIC institutions (represented by the sample who responded to this statement) would strongly agree with this statement.(c) The median view among all respondents indicates strong agreement that professional library staff should provide in-class library instruction.

  (d) While the median view (a relatively crude measure) of both CAOs and library directors is the same, the chi-square test for difference (a more discriminating measure) indicates that CAOs as a group hold this view less strongly than do library directors.

  (e) This difference is evident in the large percentage of “agree” responses among CAOs, compared to the small percentage of “agree” responses among library directors.

• Question 3 prompted respondents to assign priority, on a three-point scale, to several activities that might be accommodated in new library space. The rest of the table functions the same way as the table for questions 1 and 2.Look, for instance, at question 3.14. Of the 101 respondents, 25% would give high priority to providing new library space to student socializing (without food), while 41% of the respondents would give medium priority to this need and 35% would give it low priority. In the case of the 25% who regard such socializing as a high priority, for instance, one can be confident that 95% of the time between 17% and 33% (i.e., 25% plus or minus 8%) of the larger population of all CAOs and library directors at CIC institutions
  would strongly agree with this statement. The median view among all respondents is that providing new library space for student socializing has a medium priority. CAOs and library directors, however, have decidedly different views of this matter, as evident in the comparatively large percentage of library directors assigning high priority to this need and the comparatively low percentage who regard it as a low priority.