Administrative Decentralization of for Colleges and Universities - Research Proposal Example

Running Head: Administrative Decentralization Dissertation Proposal: Administrative Decentralization The tolerant operating environment for colleges and universities during the 1950s and 1960s allowed for the administration and non-teaching staff to grow almost unchecked while quality and continuous improvement waned. However, the decline of resources and the dramatic acceleration of inflation during the 1970s, combined with the decline of the traditional college-age cohort during the 1980s and 1990s, has made the efficient and most economical methods of administration a top concern of university leaders (Lloyd 2006, 831-854; McLaughlin, Teeter, Howard, & Schott, 1987, 6-10; Tomaney 2006, 129-135). To address the concern for efficient and economical graduate program administration, the Council of Graduate Schools has promoted centralized administration of graduate programs since the mid-1970s (Winetrobe 2006, 138-139; Council of Graduate Schools, 1976; (Mcgregor 2005, 477-494). Graduate programs in business administration, for example, which are professional terminal degree programs likened to medicine and law, are often categorized with degree programs from the arts and sciences and the humanities. The talk of centralization rarely involves the law or medical degree programs, but they often include graduate programs in business. Many of the comments to support centralized administration appear to be based on anecdotal evidence from university leaders who are currently associated with some sort of centralized operation, or on the limited research on centralized programs. No study or formal research uncovered has explored the perceptions of those professionals involved in graduate business program administration at public and private institutions. (Henig 2006, 1-3; Condor 2006, 123-158; Keating 2005, 11-15) There has been no significant study or research to determine the level of graduate program administration decentralization at accredited schools of business or analysis of how decentralization is related (if at all) to other institutional characteristics. (Royles 2006, 137-156; Pike 2006, 157-178) In light of the limited literature and research on this topic, and based on comments made by graduate program directors, the specific objectives of this national study were to (a) determine the level of decentralization (school-based administration) of graduate business programs at public and private American Assembly of Collegiate Schools of Business (AACSB) accredited schools of business; (b) determine if certain institutional characteristics such as size, program offerings, and the school funding type (public or private) are related to the degree of decentralization; and (c) investigate the degree of association, if any, between the perceptions of administrative effectiveness and quality by graduate program directors and the degree of decentralization. (Steven 2006, 606-607) Method The population chosen for this study were those institutions willing and able to achieve accreditation by the AACSB. Of the approximately 800 public and private graduate programs in business offered by colleges and universities in the United States, 269 were accredited by the AACSB. Accreditation by the AACSB means that the business programs have met established standards in admissions criteria, faculty qualifications, curriculum, and resources. The subjects of this study were the 269 graduate program directors based on the 1992-1993 list provided by the AACSB, also incorporating the five institutions added and one deletion made at the April 1993 meeting of the AACSB. The study was also limited to the perceptions of these program directors, who may or may not accurately reflect the needs of the applicants and students in the final two hypotheses, which are stated later. The directors of graduate business programs were sent letters seeking their assistance with the survey instrument. The confidentiality of the responses was guaranteed, and I offered to share the results of the study on request. Non-respondent directors were sent a follow-up questionnaire asking them to please respond. Instrumentation For this study, an instrument developed and used by Lynch, Bowker, and McFerron (1987, 1-7) to measure the degree of centralization of graduate program administration was adapted to measure the independent variable, the degree of decentralization of graduate business program administration. Permission to use this adapted instrument was obtained from Dr. Lynch. Because the study for which this instrument was originally designed measured overall university administration of graduate programs, certain changes were made. The "centralization index" was originally prepared by assigning weights to two groups of questions. The first group asked the respondent to indicate how many of four professional schools within the university were administered separately (that is, business, education, law, medicine) upon which a positive response would result in subtraction of one point in the index. The second group of variables considered the effects of the six items prescribed by the Council of Graduate Schools (CGS) as activities for which the chief graduate school officer should be responsible (Watts 2006, 16-19; Lynch, Bouker, & McFerron, 1987): Admitting applicants to graduate status Reviewing academic progress Awarding fellowships and coordinating all financial aid for graduate students Appointing committees to conduct qualifying examinations and to super vise research for masters and doctoral degrees. Certifying the completion of degree requirements and recommending granting of the degree Overseeing the adequacy of student services and promoting the welfare of graduate students (CGS, 1981). Respondents were asked to check off in columns next to each administrative function which offices perform that function. Multiple columns could be checked, for functions are sometimes shared in some universities. For each item that was stated to be a responsibility of the graduate school, the decentralization index was increased. For every responsibility stated to be carried out in a college or school, the index was decreased. The resulting index was a decentralization continuum ranging from a possible -10 to 10 from respondents. I modified the index developed by Lynch, Bowker, and McFerron by first substituting additional variables for the first four questions relating to other graduate schools. In addition, the fourth CGS variable, pertaining to committees responsible for qualifying examinations, was replaced because the AACSB does not accredit doctoral programs and many graduate business programs do not have qualifying exams as part of their curriculum. The additions to the five responsibility questions are Inquiry handling, or who has the responsibility for answering prospective applicants' questions and mailing out application packages? Student recruitment, or who has the responsibility for attending college graduate school fairs, forums, and corporate graduate business program recruitment events? Advertising, or who has the responsibility for initiating the graduate business program newspaper, magazine, radio, television, and other media advertisements? New student orientation, or who has the responsibility for formally welcoming new incoming graduate business students with a formal reception or program? Transfer credit evaluation, or who has the responsibility for reviewing requests to transfer in graduate credits from other colleges and universities? These five substitutions replaced those unusable in the original instrument and bring the total number of variables back up to 10. The additions also make the instrument more specifically suited to graduate business program administration. In addition, other independent variables, also based on Lynch, Bouker, and McFerron's instrument, were sought: The size of the school of business as determined by number of graduate business students, number of full-time faculty, administrators, and clerical staff, The status of the institution (public, private-independent, or private-church affiliated), Admissions selectivity of the graduate business programs as determined by the average undergraduate grade point average of incoming students and the average score on the Graduate Management Admission Test, and Background information on the respondents. The index range of -10 to 10 was reversed to indicate a positive number for the level of decentralization subjects of this study were the schools and colleges of business, not Council of Graduate School members as in the Lynch study. Before the respondents answered the part of the instrument pertaining to the level of decentralization and establishing a rating for the independent variables, several dependent variables on administrative effectiveness and quality of the graduate program administration were gathered. These dependent variables were based on a study of intra-institutional mergers by Somervill (1988). Permission to use these adapted variables was obtained from Dr. Somervill. Somervill's study of 541 faculty and administrators at 18 4-year doctoral-granting universities included endogenous variables that could be controlled by the administrative process. These perception variables were selected by Somervill as a result of an extensive literature review and were judged by a panel of experts to possess construct validity. The variables were presented on a five-point Likert scale ranging through I (strongly agree), 2 (agree), 3 (undecided), 4 (disagree), and 5 (strongly disagree). The questions used in this study were based on the following variables from Somervill's research: Need to improve current programs Need to develop innovative programs Meeting the needs of students Efficient use of resources One proposed addition to this list, based on the responsibility of many graduate business program directors, was meeting the needs of applicants. These five dependent variables plus institutional characteristics and the independent variables measuring the level of decentralization formed the basis for the research hypothesis presented in this article. To determine if the adapted instrument would be clear to the respondents, it was pretested with selected graduate program directors. The pretesting resulted in two minor corrections in the wording of the perception variable questions. Research Hypotheses Based on the objectives of this study and the design of variables presented here, the following hypotheses were formulated to be tested at the 95% level of confidence, using a t test and the Pearson correlation coefficient: H1: There is significant difference in the degree of decentralization between the primary funding source types (public and private). H2. There is a positive correlation between the institutional size and the level of decentralization. H3. There is a positive correlation between the admissions selectivity and the level of decentralization. H4. There is a negative correlation between the level of decentralization and the perceived need to improve programs, the need to develop innovative programs, the level of applicant and student service, and efficient use of resources. H5. The negative correlation postulated in Hypothesis 4 holds true while also controlling for institutional characteristics such as primary funding source, institutional size, and admissions selectivity. Results The overall results found that the mean level of decentralization at the AACSB-accredited institutions surveyed was 5.135. This finding shows that most graduate business schools and colleges were decentralized in the administration of their graduate business programs. The actual range varied from a low of 4 to a high score of 10. A frequency distribution shows that only 14 (6.8%) of the 207 AACSB graduate business programs were centralized in the administrative functions surveyed. Nine of the respondents (4.3%) were neutrally administered between both the school or college of business and a central office. The overwhelming majority (88.7%) of institutions had decentralized graduate business program administration, with levels ranging from 1 through 10. These results help fulfill the first objective of this study, which was to determine the level of decentralization of AACSB-accredited schools of business. Hypothesis Testing. The first hypothesis to be examined was the determination of whether the level of decentralization scores was significantly different for the two primary funding source types (private/public). A t test was carried out between the level of decentralization and the two types of institutional funding. The results, summarized in Table 1, show that the sample mean of the decentralization level for private institutions is 6.76 and for public universities the level was somewhat less at 4.46. The t values reveal that the probability of Hypotheses 1 being false was well below the significance cut-off level of .05 level. Therefore, H1 is accepted at the .05 level of significance. The population of private and public institutions differed significantly in their level of graduate business school administration decentralization. The second hypothesis states that the level of decentralization correlates positively with the size of the graduate business program administration. The coefficients were unstandardized in that they were calculated from the original values of the variables. The Pearson correlation coefficients were significant at the 95% level of confidence for five of the seven measures of institutional size: number of administrators, clerical staff, inquiries, applications received, admitted students, and registered students per year (see Table 2). In addition, a sixth variable (number of faculty) did show some measure of association and was nearly significant (.08) in association with the decentralization. Therefore, H2 is partially accepted and partially rejected. It should be noted that the coefficients listed in the tables were unstandardized, in that the original values of the variables were used in the analysis. Interestingly, the number of administrators did not show a significant level of association (.248) with the level of decentralization, as the other staffing variables did. However, overall the institutional size variables were clearly associated with a higher level of decentralization at AACSB-accredited institutions. Specifically, the more decentralized the administration was, the higher were the number of prospective student inquiries, graduate business applications, admitted students, new registered students, and clerical staff. The third hypothesis stated that the level of decentralization is positively correlated with admission selectivity of the graduate business programs. The results (detailed in Table 3) of the Pearson correlations for the two admissions selectivity variables (UGPA and GMAT) are bifurcated, yet overall associated with greater decentralization. There was a very strong association of .3588 (above the 99.9% level of confidence) between the average GMAT score and the level of decentralization. The average UGPA undergraduate grade point average for entering graduate students showed a somewhat weaker correlation (.1274), yet was still statistically significant (at the .036 level of significance). Therefore, H3 is accepted at the 95% level of confidence. This shows that there is a relationship between the admissions selectivity of an AACSB business school and the level of administrative decentralization. Specifically, the more decentralized administratively a graduate business program is, the more selective the admissions policies are relating to the average GMAT score and average entering undergraduate grade point average. The fourth and fifth hypotheses in this research attempted to determine if greater decentralization is perceived as better by the graduate program directors involved in the survey. The graduate program directors were asked five questions relating to their perceptions of the administrative effectiveness of their operations. The perception results are summarized in Table 4. The results showed that most agreed or strongly agreed that they were indeed meeting the needs of applicants and students, and that their programs were current and innovative. In addition, most felt that their administrative resources were being used efficiently. There was some variability in the answers, but the perceptions were generally positive. This was to be expected because the respondents were in effect making a self-assessment of their effectiveness. However, the information was valuable because the difference among schools for each perception was the variable correlated with the level of decentralization. The next logical step was to analyse further these data by determining if there was a relationship between these perception variables and the level of decentralization. It was predicted in the hypothesis that the level of decentralization would be associated with better perceptions of administrative effectiveness by the graduate program directors. The results listed in Table 5 were mixed in that three of the five perception variables were found to be correlated with the level of decentralization, and two were not. The perception by the program directors that they were meeting the needs of applicants was associated (.2204) with more decentralization and was significant at the .001 level. The second perception variable found that meeting the needs of current students had some measure of association (.1064) but was not significant at the .05 level (.064). These findings may show that a more decentralized graduate program administration leads to more satisfied applicants (as was supported by the increased admissions pipeline size in H3) but not necessarily to more satisfied students. This could be explained by the fact that the administration (specifically the graduate admissions function) is primarily responsible for bringing students in, and the faculty is primarily responsible for keeping them satisfied once they are admitted, through teaching and contact hours. These findings are also not consistent with some recent organizational theorists (such as Drucker, 1974, 1-7; Peters, 1992, 11-13) who in general believe that more decentralization of authority leads to greater employee and customer satisfaction. The third and fourth perception variables found that greater currency of the graduate program, as well as more innovation, was associated with heightened decentralization. Although most of this study was related strictly to administrative aspects of graduate business program administration, these two variables show that the curriculum may also be associated with the level of decentralization. The program directors' perceptions that their graduate programs are current had a Pearson correlation coefficient of .2198 (p < .001), and the perception of program innovation had a correlation of .2337 (p < .001). These findings support the organizational management literature, which states that centralization is not the best organizational structure for a changing environment (Litchfield, 1971, 1-8; Pfeffer, 1981; Peters, 1992). A decentralized administrative structure is believed to support more entrepreneurism, and may therefore be the reason for more perceived program currency and innovation by the graduate program directors surveyed. The fifth perception variable asked whether the program directors perceived that their administrative resources were being used efficiently. This may seem to have been a biased question to elicit a particular response from those who favoured either a centralized or decentralized operation, especially because the survey instrument clearly stated that the purpose of the study was to measure the level of decentralization. However, these perception variables were asked for before the decentralization level, and the confidentiality of responses was guaranteed to the respondent. Interestingly, the results showed that the perceived level of administrative effectiveness was not significantly associated with the level of decentralization. In fact, the measure of association was the weakest of the five perception variables (Pearson = .0166, I-tailed significance = .406). This finding is somewhat inconsistent with the other findings from the study, which showed that heightened decentralization led to more students in the admissions pipeline and greater admissions selectivity. These may be considered factors important to the efficient use of administration resources, but apparently the graduate business program directors may have been thinking of other types of resource efficiency. The finding also seems inconsistent with the prodecentralization literature cited in the more traditional Weberian management philosophy for the perceived administrative resources of the program directors (Weber, 1947). Thus, H4 is partially accepted and partially rejected. The fifth and final hypothesis to be examined states that the level of centralization correlates negatively with the perception variables of program directors, while controlling for seven other independent variables (seventh-order partial correlation). The variables used as controls were those that were found in the previous analyses to be statistically significant in their measure of association with the level of decentralization: the primary funding source (private or public); the size of the institution (number of clerical staff and all admissions pipeline data); and the admissions selectivity (average UGPA and average GMAT score). The analysis of this hypothesis fulfils the third and final objective of this study. Because the primary funding source variable is nominal (with two possible values--private and public) and the other independent control variables are both numeric and interval, the analysis of this hypothesis consisted of three steps. First, I did a seventh-order partial correlation, using all of the independent numeric variables as controls. This was then repeated while controlling for private and again for the public institutions. The separate analyses using the funding type must be read with concern because of the relatively low minimum number of cases (n = 32 for private institutions). This is why the partial correlation was first done for all institutional funding types (minimum number of cases n = 133). The results from the seventh-order partial correlation analysis using the independent interval numeric variables for all institution funding types are summarized in Table 6. The same three perception variables that were significant in H4 were significant here as well: meeting applicant needs, meeting student needs, and program innovation. The level of confidence was not as high as when there were no controlling variables (.001 from H4), but the measure of association was still significant at the .05 level for all three. The results from the seventh-order partial correlation analysis for private institutions are summarized in Table 7. This analysis shows a reduction in the association between the five perception variables and the level of decentralization. However, for private institutions there was still a significant measure of association (.004) in how well the program directors perceived that the applicant needs were being met. The other two variables that were significant in the two previous analyses were now insignificant at the .05 level (meeting student needs and program innovation) but still showed some measure of association (.092 and .101 significance). The results from the seventh-order partial correlation analysis using all of the independent interval numeric variables for public institutions are summarized in Table 8. For public institutions, the level of association was significant (.044) for program currency, but not for any of the other dependent variables. Interestingly, the first variable (meeting applicant needs) was insignificant for public institutions, whereas it was significant for private colleges and universities in the previous analysis. The three previous partial correlational analyses lead us to partially accept and partially reject H5. The results show that most AACSB-accredited institutions in this study were decentralized in the administration of their graduate programs, and that institution funding type, size, and admissions selectivity were associated with the level of decentralization. In addition, there was a significant measure of association for three of the five measures of administrative effectiveness surveyed. Overall, the five dependent variables showed generally positive perceptions by accredited graduate business program directors. Conclusion The decentralization index, measuring the level of business school-based administration, ranged from possible scores of -10 (very centralized) to 10 (very decentralized). The results found that the mean level of decentralization was 5.1, indicating that most AACSB schools are decentralized. This is important when considering that the first objective stated was to determine the level of decentralization. This information had not been gathered before and is now available to university decision-makers when needed. If a business school holds or plans to seek professional recognition by the AACSB, one fact now to be considered when deciding on organizational structure is that most accredited schools of business are decentralized. This finding is opposite the findings of the two previous major studies on centralization of graduate program administration. The national study by Lynch, Bowker, and McFerron (1987), which surveyed graduate school deans and found that most were centralized (88% were found to be slightly centralized to very centralized), concluded nearly the exact reverse of the findings from this study of graduate business programs. The authors of that study concluded that centralization is incomplete as a national movement. The findings from this national study of graduate business programs may suggest that this hypothesized movement may not be completed for all types of graduate programs. Another research study of graduate programs surveyed deans and chairpersons of schools of social work (Seelig, Schirtzinger, & Eldridge, 1982, 26-41). Unlike the accredited graduate business programs surveyed here, the administration of graduate programs at schools of social work was found to be more centralized. However, the results of the study presented in this article indicate that the authors of that study may have correctly predicted a shift away from centralization of student services within the university. Therefore, of the three studies now known involving the centralization level of graduate programs, two found more centralization (schools of social work and CGS members) and one found more decentralization prevalent (accredited schools of business). In conclusion, the results of this study have important implications for graduate business program directors and senior-level university decision-makers. Perhaps the most important finding is that more centralization is not necessarily associated with more students in the graduate admissions pipeline or with a higher quality of students in graduate business programs. More centralization may in fact be negatively associated with these important areas. Therefore, it may be justified to offer business schools the option of retaining more autonomy in the administration of their graduate programs. Institutional researchers and planners may now be able to make a more informed decision if they are considering basing graduate program administration on a central university office. References Condor, S., Gibson, S. and Abell, J. (2006) English identity and ethnic diversity in the context of UK constitutional change. 6 (2) Ethnicities 2006, pp.123-158. Council of Graduate Schools (CGS). (1976). Organization of work within the university. Washington, DC: Author. Council of Graduate Schools (CGS). (1981). The organization and administration of graduate schools in the U.S Washington, DC: Author. Drucker, P.F. (1974). Management Tasks, responsibilities, practices. New York: Harper and Row. Henig, S. (2006) Modernising Britain : central, devolved, federal. London: I. B. Tauris & Co., 2006. Keating, M. and Salmon, T. C. (2005) The dynamics of decentralization : Canadian federalism and British devolution. Georgetown, Ontario: McGill-Queen's University Press, 2005. Litchfield, E. H. (1971). Organization in large American universities: The administration. In J. V. Baldridge (Ed.), Academic governance. Berkeley, CA: McCutchan Publishing. Lloyd, M. G. and Peel, D. (2006) Devolution, decentralization and dispersal : asserting the spatiality of the public sector in Scotland. 14 (6) European Planning Studies, pp.831-854. Lynch, D. M., Bowker, L. H., & McFerron, J. R. (1987). Alternative models for the administration of graduate education. Planning for Higher Education, 15(2), 1-7. Mcgregor, P. G. and Swales, K. (2005) Economics of devolution/decentralization in the UK : some questions and answers. 39 (4) Regional Studies 2005, pp.477-494. McLaughlin, G., Teeter, D. J., Howard, R. D., & Schott, J. S. (1987). The influence of policies on data use. Cause/Effect, 9(1), 6-10. Peters, T. (1992). Liberation management: Necessary disorganization for the nanosecond nineties. New York: Alfred A. Knopf. Pfeffer, J. (1981). Power in organizations. Cambridge MA: Ballinger Publishing. Pike, A. ... [et al.] (2006) Devolution and the Trades Union Congress in North East England and Wales. 16 (2) Regional and Federal Studies 2006, pp.157-178. Royles, E. (2006) Civil society and the new democracy in post-devolution Wales - a case study of the EU structural funds. 16 (2) Regional and Federal Studies 2006, pp.137-156. Seelig, M., Schirtzinger, M., & Eldridge, W. D. (1982). A holistic perspective on student growth and development: A look at academic programs in graduate schools of human service. International Social Work, 25, 26-41. Somervill, C. Z. (1988). Intra-institutional mergers of academic units: Growth in the context of decline. Paper presented at the annual meeting of the Association for the Study of Higher Education, Washington, DC. Steven, M. (2006) The government of Scotland : public policy making after devolution. 29 (3) West European Politics 2006, pp.606-607. Tomaney, J. and PIKE, A. (2006) Deepening democracy and engaging civil society? 'Economic and social partners' and devolved governance in the UK. 16 (2) Regional and Federal Studies 2006, pp.129-135. Watts, D. (2006) Pressure group activity in post-devolution Scotland. 18 (2) Talking Politics 2006, pp.16-19. Weber, M. (1947). The theory, of social and economic organizations (L. Henderson & T. Parsons, Trans.). Glencoe, IL: The Free Press. Winetrobe, B. (2006) Parliament shall speak unto parliament. 345 SCOLAG 2006, pp.138-139. Tables TABLE 1. Decentralization Level With Primary Funding Source: t Test No of Variable cases M SD SE of M Private 59 6.7627 3.559 .463 Public 147 4.4626 3.132 .258 Mean difference = 2.3001 Levene's Test for Equality of Variances: F = .987 P = .322 t test for Equality of Means 95% Variances t df 2-tail SE diff CI for diff sig. Equal 4.58 204 .000 .502 (1.309, 3.291) Unequal 4.34 95.98 .000 .531 (1.247, 3.353) Table 2. Correlation of Decentralization Level With Institutional Size Correlations Coefficient Cases 1-tailed sig. Administrators .0494 193 .248 Clerical staff .3266 196 .000(*) Fte business faculty .1003 197 .080 Inquiries/year .3785 178 .000(*) Applications/year .3026 196 .000(*) Admits/year .2089 197 .002(*) Registered/year .1803 199 .005(*) Note. Minimum pairwise N of case: 178. (*) 1-tailed significance below .05. TABLE 3. Correlation of Decentralization Level With Admission Selectivity Correlations Coefficient Cases 1-tailed sig. Average UGPA .1274 200 .036(*) Average GMAT score .3588 201 .000(*) Note. Minimum pairwise N of case: 200. (*) 1-tailed significance below .05. TABLE 4. Summary of Perception Results Question M 1. We are meeting the needs of graduate 1.874 business applicants. (N = 206) 2. We are meeting the needs of graduate 1.957 business students. (N = 207) 3. Our graduate business programs are 2.246 current. (N = 207) 3. Our graduate business programs are 2.539 innovative. (N = 167) 4. Our administrative resources are being 2.198 used efficiently. (N = 207) Note. Scale for mean ranges through 1 (strongly agree), 2 (agree), 3 (neutral), 4 (disagree), and 5 (strongly disagree). TABLE 5. Zero-Order Correlation of Decentralization With Perception Variables Correlations Coefficient Cases 1-tailed sig. Meeting applicants needs -.2204 206 .001(*) Meeting student needs -.1064 207 .064 Program is current -.2198 297 .001(*) Program is innovative -.2337 167 .001(*) Administrative efficiency -.0166 207 .406 Note. Minimum pairwise N of cases: 167. (*) 1-tailed significance below .05. TABLE 6. Seventh-Order Partial Correlation of Decentralization With Institutional Size and Admissions Selectivity Variables Correlations Coefficient Cases 1-tailed sig. Meeting applicants needs -.2020 163 .005(*) Meeting students needs -.0901 163 .125 Program is current -.1784 163 .011(*) Program is innovative -.1557 133 .036(*) Administrative efficiency -.0613 163 .418 Note. Minimum pairwise N of cases: 133. Controlling for number of clerical employees, inquiries/year, applications/year, admits/year, new registered students/year, average UGPA, and average GMAT score. (*) 1-tailed significance below .05. TABLE 7. Seventh-Order Partial Correlation of Decentralization With Institutional Size and Admissions Selectivity Variables for Private Institutions Correlations Coefficient Cases 1-tailed sig. Meeting applicants needs -.4046 39 .004(*) Meeting student needs -.1882 39 .119 Program is current -.2112 39 .092 Program is innovative -.2244 32 .101 Administrative efficiency -.1297 39 .209 Read More