College and Research Libraries ARL Directors: Two Deca~es of Changes Marcia J. Myers and Paula T. Kaufman This study examines a wide range of changes among ARL directors from 1970-1989 on such variables as gender, previous positions held, and volumes in the collections. In 1970, there were no female directors of ARL libraries; in 1989 there were 32 female directors. This increase has changed the demographic characteristics of ARL directors. As a group, female directors are younger, have fewer years of service, earn higher average salaries, and have received higher increases in their libraries' budgets when they assumed their jobs than male directors. uring the past two decades, higher education has. been subject to many stresses and changes. In 1973, Arthur M. McAnally and Robert B. Downs pub- lished their classic article on the chang- ing role of the university library director.1 McAnally and Downs expressed con- cern about the recent high turnover rate in this position and predicted that if the trend continued, the average span of service for directors might drop to only 5 or 6 years. As librarians enter the 1990s, it seems appropriate to reexam- ine the tenure and other characteristics of directors of large research libraries. Are many directors leaving their posi- tions to retire early or to go into teaching or other fields, as Anne Woodsworth recently suggested?2 Have the reasons for leaving the directorship or the age and educational backgrounds of direc- tors changed over the years? What changes, if any, have occurred in the characteristics of library directors in light of the increasing number of fe- males assuming top positions? What, if any, new trends can be predicted for the 1990s? LITERATURE REVIEW There has been a fair amount of re- search about the characteristics of re- search library directors during the last half-century. William L. Cohn studied 254 U.S. Association of Research Library (ARL) directors covering the period 1933 to 1973.3 Of the 7 4 ARL libraries in 1973, 34 ( 46%) had named a new director in the previous 4 years. Although there had been a large increase in the number of library doctorates awarded, this trend was not reflected among the ARL direc- tors, whose most common highest de- gree was the M.L.S. In the period 1933 to 1973, 54% of the directors died in office or retired; 14% went into teaching. Jerry L. Parsons contrasted the charac- teristics of directors of U.S. ARL libraries for the years 1958 and 1973.4 He found that there was not a trend toward youn- ger directors: the average age in 1953 was 51 years, while the average age in 1973 was 53 years. There were no women Marcia J. Myers is the Associate Dean of Libraries for Administrative Services and Paula T. Kaufman is the Dean of Libraries at the University of Tennessee, Knoxville, Tennessee 37996. Special appreciation goes to Loretta McCuiston for her assistance with data entry. 241 242 College & Research Libraries directors in 1953; in 1973, there were 4 female directors. The 1953 group held 26 doctorates; the 1973 group held 25 doc- torates. In 1958, the majority of the direc- tors had been promoted from within. In contrast, the majority of the 1973 direc- tors were recruited from other institu- tions. Parsons also found that the tenure of ARL directors appeared to be devel- oping into a short-term assignment. In 1958, the average tenure of a director was more than 8 years, with a median of 9 years. In 1973, the average tenure of the directors was less than 8 years with a median of only 5 years. In a letter to the editor of College & Re- search Libraries Louis Kaplan provided his- torical data on ARL directors who withdrew from administration before age 65.5 The percentage of those who left administra- tion increased from 10% to 20% in the 1926-1942 period to 26% in 1968-1975. Kaplan thought the reason that more di- rectors left between 1968-1975 was be- cause of the decline in fiscal support that began to be felt in 1967. Janice Fennell, noting that librarian- ship is 84% female, studied 11 female directors of large academic libraries in 1978 to develop a composite career pro- file.6 She concluded that if a woman be- comes director of a large academic library in the United States, she is likely to be middle-aged (an average age of 48), married with no children, and from a middle-class background. The profile in- dicated that ~he typical female director possesses no higher than a master's de- gree; however, she is interested in both formal and informal continuing educa- tion. The profile also indicated that the female director has held her current po- sition for less than 10 years and entered the position directly from another ad- ministrative position. Ronald Dale Karr studied directors of 90 U.S. ARL libraries in 1966 and 1981.7 He excluded libraries with vacancies, acting directors, or nonlibrarians serv- ing as directors. One of the more notable differences between the directors in the two groups was gender: there was only 1 female library director in 1966 and 12 in 1981. Of the 1966 directors, 15% lacked May1991 the library degree, but by 1981, every director was a graduate of a certified graduate library program. The propor- tion of directors with doctorates fell from 44% in 1966 to 33% in 1981. Although the 1981 directors had taken longer to reach their present positions, they were youn- ger (an average of 51.2 years) than the 1966 directors (53.6 years). The 1981 group had served an average of 6.7 years in their current positions, while the 1966 group had served an average of 11.4 years. WilliamS. Wong and DavidS. Zubatsky studied the tenure rates of 91 ARL and 80 non-ARL library directors in 1983.8 They excluded vacancies and acting directors from the study. Seventy-five of the ARL directors were male and 16 were female. Twenty-two directors of ARL libraries (19 males and 3 females) were promoted from within their institutions to the director- ship. Results of the survey support pre- . vious findings that the number of individuals holding the doctorate has not increased despite the fact that there has been an increase in the number of individuals holding the doctorate in li- brary science. Thirty-three ARL directors (32 males and 1 female), or 36% of the 91 respondents, had held their position for 9 or more years. Sixty-seven percent had held their positions for 5 or more years. These findings did not support McAnally and Downs' prediction of short tenures for library directors. Women were more likely than men (62.5% versus 46.7%) to hold their positions for 6 or less years, but this may be attributable to their more recent attainment of such positions as a group. In 1989, Anne Woodsworth reexamined the McAnally and Downs article.9 She in- dicated that the situation regarding the ten- ure of library directors has not changed. In the past 3 or 4 years, more than half of the U.S. ARL libraries have seen a change in directors. The turnover predicted by McAnally and Downs seems to have be- come the norm. Woodsworth perceived that while some of the attrition is normal, human factors contribute to a search for career alternatives. These factors include boredom, reaching a structural plateau in jobs, lack of stability in the university administration, and overwork. The cur- rent tempo and demands on the director's job will not change soon, Woodsworth asserted, and the onus is on the individual library director . to de- velop alternative career paths and es- cape routes. Woods worth's article and conclusions inspired us to examine trends in contemporary ARL university library directorships and to analyze the changes among the group over a 20-year period (1970 to 1989). . Purpose and Methodology No attempt was made to analyze the directorship of an institution before the library became an ARL member; there- fore, the population of this study ranged from 75 university libraries in 1970 to 107 in 1989. One library, which was an ARL member in 1970 but not 1989, was excluded from the study. Statistical data on the number of volumes, volumes added, materials expenditures, total ex- penditures, and number of staff for each library were extracted from the annual ARL Statistics. Throughout this study, the title "direc- tor" is used to denote the top executive of an ARL library, regardless of the exact title. The latest ARL membership list was used to locate the name of the current director. 10 Acting directors were ex- cluded from this study; Canadian ARL library directors were included. Bio- graphical directories were used to locate information about the date of appoint- ment, previous position, and reason for leavingY Directors who stepped down at age 60 were classified as retired if no other reason could be located. In order to check the trends noted by previous re- searchers for directors in the beginning of 1970, the beginning of 1979, and the end of 1989, a further search was made in biographical directories to locate in- formation on age and highest degree held. ARL' s Newsletter and issues of Col- lege & Research Libraries News were exam- ined for announcements of retirements, recent position changes, and months of appointment for newly hired directorsY The American Library Directory was also ARL Directors 243 used to verify the names of directors and years of service. The gender of the director was consid- ered the critical variable for analysis, and our personal knowledge was used to complete gender information not lo- cated elsewhere. Every effort was made to locate missing information for other variables. However, when we could not locate the director's age or other infor- mation in published sources, we did no personal follow-up to the individual or the director's library to supply the data. Uncollected data were treated as missing for all statistical analyses. The available data collected were coded, keyboarded, verified, and analyzed using the Statisti- cal Package for the Social Sciences for personal computers (SPSS/PC+). FINDINGS: THE DIRECTORS Gender and Area We began by looking at the distribution · of directors by gender in seven geograph- ical areas, as displayed in table 1.13 Of the 1,798 cases we observed, 1,538 (85.5%) rep- resented libraries with a male director, and 260 (14.5%) represented libraries with a female director. Table 1 shows that male and female directors were fairly evenly distributed by geographic areas, ranging from a high of 23.8% in the Midwest for male directors to a high of 27.7% in the East for female directors. Major findings will be presented by years and gender for ARL directors. Because the Canadian ARL di- rectors proved to be not significantly dif- ferent from other areas on the key variable of gender, data on Canadian ARL libraries were analyzed along with data from the U.S. ARL libraries. Interesting characteris- tics peculiar to Canadian directors are re- ported as appropriate. Position Titles Although we have considered each top executive to be a director, we also were interested in changes in official ti- tles used, as these might provide clues about the changing role or perception of libraries and librarians on research uni- versity campuses. Although the use of the title "director" has declined from 244 College & Research Libraries May 1991 TABLE 1 OBSERVATION BY GENDER AND AREA East Mid-Atlantic South Midwest West (Mountain) Pacific Canada Total 65.3% of all directors in 1970 to 49.5% (51.4% of the males and 45.5% of the females) in 1989, "director" is still the most common title used. "Librarian," with 29.9% in 1989 (28.3% of the males and 33.3% of the females), and "dean," with 12.1% (13.5% of the males and 9.1% of the females), follow it. · These findings are consistent with those of Wong and Zubatsky in 1983 when 50.7% of the 75 male directors and 31.3% of the 16 female directors held the title "director"; 29.3% of the males and 18.8% of the females in 1983 used the ti tie "librarian," and 13.3% of the males and 12.5% of the females were called "dean."14 Gender Issues No contemporary discussion of the characteristics of ARL directors can fail to focus on the changes resulting from an increased number of female top execu- tives in many professions and indus- tries. Our analysis reveals that there has, indeed, been a significant increase in the number of female ARL directors, from none in 1970 to 32 (31.7% of the 101 permanent directors) in 1989. Are any observable patterns attributed to this change? Replacements. Recently, a colleague remarked in passing that it appeared that male directors are being replaced by females, and female directors are being replaced by males. We took up the challenge of verifying this casual obser- vation and examined the rate of turnover Female Male Directors Directors Totals # % # % # % 198 12.9 72 27.7 270 15.0 272 17.7 52 20.0 324 18.0 310 20.2 28 10.8 338 18.8 366 23.8 20 7.7 386 21.5 113 7.3 4 1.5 117 6.5 167 10.9 30 11.5 197 11.0 112 7.3 54 20.8 166 9.2 1,538 100.0 260 100.0 1,798 100.0 as well as any significant patterns re- lated to turnover. We also thought it im- portant to confirm recent remarks by Woodsworth and others, who refer to observably shorter tenures for contem- porary ARL directors. Our analysis reveals that there has, indeed, been a significant increase in the number of female ARL directors, from none in 1970 to 32 in 1989. Table 2 displays the gender of the newly hired director and the gender of the director replaced by the new hire, classified by time periods in 5-year inter- vals. The number of replacements has increased slightly over the past 20 years, from 45 in 1970-1974 to 50 in 1985-1989. Some may interpret these data to justify the contention that turnover is accelerat- ing. However, we do not think that a growth from an average of 9 per year to 10 per year over a 20-year period repre- sents a significant increase. Time will tell if this is the start of a general trend. Table 2 shows interesting data about the genders of new hires and the incum- bents they replaced. In the period 1970 to 197 4, no males replaced female direc- tors, but 4 females replaced males. By 1985 to 1989, 54% of all new hires were of a gender different from the incumbent they replaced. Of the 50 replacements in this period, 8 males replaced female di- rectors, and 19 females replaced male ARL Directors 245 TABLE2 NEW HIRES AND REPLACEMENTS BY PERIOD 1970-74 1975-79 1980-84 1985-89 # Male replacing female 0 Female replacing male . 4 Male replacing male 41 Female replacing female 0 Total 45 directors. The data tend to indicate that a female applicant has a better chance of being offered a position in an institution that previously had been directed by a male than by an institution that had a female director. Male applicants appear to have an equal chance of replacing a male or a female. The data indicate that a female applicant has a better chance of being offered a position in an institution that had previously been directed by a male than one that had a female director. Tenure. We also looked at the tenure of each individual who stepped down from the directorship. In the period 1970 to 1974, the average years of service for the male directors who left their posi- tions were 12.8 years; there were no fe- male directors replaced during that time period. From 1975 to 1979, the average years of service for male directors leav- ing their positions was 10.3 years; female directors being replaced had served an average of 4 years. During 1980 to 1984, the average years of service for replaced directors were as follows: male directors, 10.2 and female directors, 5.8. There con- tinued to be a significant difference be- tween the tenure of departing male and female directors in 1985 to 1989, when departing male directors averaged 11.7 years of service in contrast to replaced female directors, who averaged 6.5 years of service. The average tenure of the di- rectors who left their positions has de- % # % # % # % 0.0 3 7.0 2 5.0 8 16.0 8.9 7 16.3 9 22.5 19 38.0 91.1 33 76.7 26 65.0 19 38.0 0.0 0 0.0 3 7.5 4 8.0 100.0 43 100.0 40 100.0 50 100.0 dined slightly for male directors over the last two decades, from 12.8 years in 1970 to 1974 to 11.7years in 1985 to 1989. The average for all directors replaced was 10.3 in 1985 to 1989, compared with 12.8 in 1970 to 1974. Table 3 details the tenure of incumbent directors for 1970 through 1989 in 5-year intervals. Three ARL libraries were under the leadership of the same male director for the entire 20-year period covered by this study. Because female directors are newer to directorships as a group, it was expected that they would average fewer years in their positions. The 11 female di- rectors observed in 1979 had an average of 4 years in their positions. By 1989, the average number of years for 30 female directors had increased to 5.2 years. Male directors have shown a decline in the num- ber of years in the position, from 10.3 years in 1970 to 8 years in 1989. The average tenure for all incumbents in 1989 was 7.1 years. The inclusion of Canadian directors in this study had little influence on the tenure variable. When Canadian librari- ans were excluded, the average tenure fell by only .01 %. The results of this study show that the male directors, at least, ex- ceed the average 5 or 6 years of service that McAnally and Downs predicted, although it must be noted that the trend appears to be toward shorter tenures.15 The difference between the average tenure of replaced directors and the tenure of the incumbents indicates that directors are staying in their positions for shorter lengths of time. Turnover. In 1989, 10 positions, or 9.3% of the 107 directorships, were vacated. To put this into perspective, we looked at sim- ilar data in another profession. Turnover 246 College & Research Libraries May 1991 TABLE 3 TENURE OF INCUMBENT DIRECTORS Male Year # Avg. 1970 72 10.3 1974 73 1979 78 1984 81 1989 71 rates among ARL directorships compare favorably with a recently reported study that showed the average turnover rate for the chief executive officers of Utah hospitals to be 16.6%.16 A fruitful area for future study is how the turnover rate and the reasons for turnover compare between library executives and executives in other fields. Richard B. Dwore and Bruce P. Mur- ray found that a minority of turnovers were due to involuntary reasonsY The highest percentage of turnovers (44%) were attrib- utable to voluntary reasons, such as pro- motion. Glenn M. McEvoy and Wayne F. Cascio also found evidence that turnover is lower among good performers.18 Reasons for ARL directors leaving will be discussed below. Reasons for Departing Why do directors leave their posts? Has this changed over time? Do career patterns include movement from one ARL director- . ship to another? Table 4 shows the directors' reasons for leaving their posi- tions. Throughout the four time periods shown, the most frequent reason for male directors' leaving was retirement. The per- centage of male director retirements was fairly consistent, ranging from a low of 47.4 % of all reasons for leaving in 1975- 1979 to a high o£.59.3 % of all reasons for leaving in 1980-1984. The pattern is less clear for female directors because we could observe a reason for leaving for only 17 females during the entire 20 years under study. Across all time periods, 5 of the 17 female directors (29.4%) left for director- ships in other ARL libraries, 4 (23.5%) re- tired, 4 (23.5%) went into teaching, 2 7.5 7.2 8.5 8.0 Gender Female Total # Avg. # Avg. 0 0.0 72 10.3 4 2.3 77 7.3 11 4.0 89 6.8 20 4.7 101 7.7 30 5.2 101 7.1 (11.8%) took positions in other academic libraries, and 2 (11.8%) left for jobs out- side academia. Another frequent reason for leaving was to assume a directorship at another ARL university library. The percentages of indi- viduals choosing this course of action ranged from 23.7% of the male directors in 1975- 1979 to 37.5% of the female directors in 1980- 1984. These large percentages may reflect the fact that directors are looking for opportuni- ties to direct libraries with greater resources or in different settings, or to revitalize them- selves while staying within a directorship, or to compensate for the lack of alternative po- sitions within their institutions. Although upward mobility beyond the directorship is uncommon, some individuals do find career opportunities in their own institutions. In 1970 to 197 4, no director left to accept another position in the same institution or library, and only one accepted a nonlibrary, non- university position elsewhere. By 1985 to 1989, 16% of the male directors who left during that period, in fact, accepted other positions in their own institutions. Also, the number who leave for nonlibrary, non- university positions is increasing. These nonlibrary positions included such posts as the executive director of a government entity and the presidency of a private, nonprofit organization. In 1985--1989,4% of the male directors and 28.6% of the female directors moved into nonlibrary positions. In the period 1970-197 4, the highest num- ber of observations where a reason for leav- ing was known (44 male directors) was observed. No female director, for whom we could identify a reason for leaving, departed from a permanent directorship during this ARL Directors 247 ~ 0 Ct) \0 \0 0 \0 0 period. In 1970 to 1974, 10 (22.7%) of the c::i ~ oO cx:i c::i cx:i c::i ...... ('.1 ('.1 ('.1 0 male directors leaving went into teach- Q) ...... tU ing; by 1985 to 1989 this reason had de- E Q) dined to 3 (12%) of the males and 1 1.1., =II: 0 ...... ('.1 ('.1 0 ('.1 ['... 0\ (14.3%) of the females. Teaching has ap- 00 I lO parently become a less attractive alterna- 00 0 0\ 0 0 0 0 tive. Also, the large portion of the directors ,..... ~ C'i C'i C'i 0 "' 0 c::i ~ ~ 0 ~ 11') ...... .,... ...... ...... who do not have doctorates may not find -< ~ E ...... ...... other ARL libraries than ever before. Ob- dl & o:t4> 0\ =II: ('.1 0 0 0 0 0 ('.1 viously, current ARL directors of smaller I-Ll~ ['.. institutions are likely candidates for di- ~J:..l..4 th "< ['.. 0\ ~ .:-.: c:i cr) c::i c<) u) c::i rectors of larger institutions; however, ~~ ,..... " u rn ·a complex for the scope of this study, we did "E ·g d' ::l Q) :a < ·.8 ~ consider the positions individuals held im- E u 1-t Q) 0 . ·.c .8 0 - j;.LI>< ,_;jp::) ~Cl ~~ ::c: ~ gz R C) 0\ ,..... ~ Cl ~ ~ ::c: C) 1-4 ::c: College & Research Libraries ~~ ,...... N ..... oor-.. 00 0 ocic-.i ,....;~C"') c:ic:i "<:f' N ..... ..... 0 ~ ..... (IS s & '*0 (1) \0 0 t-.. (1)"<:f' ..... ..... N 0 0\ "<:f' 0\ 0\ ll1 lt')O ~C"') o\C'i r...: t..: ,...... t...:c:i N N ..... ..... 0 Q) ..... iii ::E '* N 0 It) N N ..... lt')r:-.., N,.... ..... ..... \0 0 0 0 0 0 0 0 0 ~ c:i c:ic:i c:ic:ic:i c:io Q) ..... \0 ..... ..... ..... 0 iii ..... s & \0 ,...... '* ..... ..... ...... 0 0 0 ...... \0 00 0 t-.. It) 0 (1)0 ~ N -.cit.r) o\ ~c:i ,....;o (1) N ...... ...... 0 Q) ...... iii ::E '* N 00 0\ lt'),.....O ...... ~ N,...... ............ 0 0 0 0 0 0 0 0 ~ c:i c:ic:i c:ic:ic:i c:ic:i Q) iii s & '* 0 0 0 0 0 0 0 0 ...... 00 N Nt--..0 ,...... 0 ~ -.o Lr)oc) ocic-.icr) -.Dei N,...... ,...... N 0 ~ ...... (IS ::E '* "<:f' r:-... N N It) N "<:f' \0 ...... ...... ...... ...... \0 May 1991 in 1989. Female directors as a group are younger than their male counterparts . There were none in 1970; in 1979, they had an average age of 47.9 years and in 1989, an average age of 50.9 years. The inclusion of Canadian directors had little influence on the average age. For exam- ple, when Canadian directors are ex- cluded, the average age in 1989 for male directors was 54.7 and 51 for female direc- tors. Previous studies have found that the director was generally in the early 50s. It appears that other career paths have opened up for female directors since Cohn made his observations in 1976. While the difference in ages between male and female directors was signifi~ant, the results should be viewed with some caution. Some current biographical direc- tories do not list ages, and we could not locate the birth dates of many of the newer directors (both male and female). We think that the actual ages are probably somewhat younger than those found, even though the results of this study are consistent with the results of previous studies. For example, we could locate ages for only 86 of the 101 permanent directors in 1989. Of the re- maining 15 directors, 6 were females and 3 were males, whom we believe are in their 40s. FINDINGS: THE LIBRARIES In 1970, the 75 ARL libraries included in this study contained a total of 136 million volumes, with an average of 1.8 million volumes and 242 staff per library. By 1989, the membership had grown to 107libraries. The 106libraries that com- pleted the 1989 ARL statistics survey held 295 million volumes, with an aver- age of 2.8 million volumes and 322 staff per library. During the tremendous growth of the past two decades, have changes in library size affected the rela- tionship between library directors and libraries? This study collected selected descriptive statistics about ARL libraries-statistics that might also relate to the characteristics of the library directors. The statistics in- cluded the number of volumes in the collection, gross number of volumes added to the collection, materials and binding expenditures, and total operat- ing expenditures. Two additional vari- ables were calculated on materials and binding expenditures and on total oper- ating expenditures. These variables showed the increase in expenditures from the previous year so that percent- age increases in expenditures could be separately analyzed. We thought that these change variables were likely to be the descriptive variables first influenced by a change in directors. For 1989 and the population as a whole, no significant differences were found for the various descriptive statistics analyzed by the geographic areas classified for this study. For example, in 1989, the average number of volumes in the libraries ranged as follows: 3.3 million, East; 2.9 million, Mid-Atlantic; 2.3 million, South; 3.2 mil- lion, Midwest; 3 million, Pacific Coast; 2.2 million, Canada; and 2.1 million, West (Mountain Standard Time). While there were some differences among the descriptive variables and gen- der, these were slight. Table 7 shows the average descriptive statistics by gender for 1989. The 70 male directors controlled a slightly higher average number of vol- umes, volumes added, and total expendi- tures than the 30 female directors. Female directors in 1989 had control over a slightly TABLE 7 DESCRIPTIVE STATISTICS AVERAGES BY GENDER FOR 1989 Gender Male Female Directors Directors Statistics (N=70) (N=30) Volumes in library 2,830,581 2,568,562 Volumes added 81,096 74,824 Total staff 316 321 Material expenditures $3,901,315 $4,048,114 Total ex:eenditures $5,963,470 $5,647,779 ARL Directors 251 higher average expenditure for materi- als and a larger average staff. In 1970, there were no female directors for com- parison on the descriptive statistics. In 1979, there were 11 female directors and 78 male directors. In that year, the male directors controlled a slightly higher av- erage number of all the descriptive sta- tistics, from volumes (2.3 million versus 1.9 million) to staff (304 versus 289). Female directors have increased their control over resources in the past two decades. Although the male directors generally control larger resources than their female counterparts, the differences are slight, and female .directors have increased their control over resources in the past two de- cades as indicated by the selected descrip- tive statistics. In fact, male and female directors were fairly well matched in 1989 on two important variables: volumes in the collection and total expenditures. Li- braries with more than 3 million volumes in 1989 were controlled by 27.1% of the male directors and 23.3% of the female directors. Fifty percent of the male direc- tors controlled libraries with more than $10 million in expenditures; 53.3% of the female directors' libraries had more than $10 million in expenditures. Those figures indicate that there has been considerable progress for female directors since 1970, when they were not even represented among ARL directors. Expenditures Table 8 details the percentage change in materials expenditures and total ex- penditures related to a change in direc- torship by gender. When the percentage expenditure variables were controlled by gender and analyzed for a change or no change in directorship, the male directors' averages on the two percent- age increase variables were lower if a new hire had occurred than the female directors'. When controlled for gender and new hires, the average percentage of change in total expenditures was 8.5% 252 College & Research Libraries May 1991 TABLE 8 AVERAGE PERCENTAGE INCREASE OF MATERIALS AND TOTAL EXPENDITURES BY GENDER AND CHANGE IN DIRECTORS Gender Male Directors Female Directors Change No Change Change No Change Expenditures # % # % # % # % Materials Total 132 8.8 1,290 8.8 45 11.3 209 9.1 132 8.5 1,290 for male directors and 10% for female directors. The materials expenditures percentage increase during change was 8.8% for male directors and 11.3% for female directors. In contrast to male di- rectors, female directors apparently achieve a greater percentage increase in their budgets during their first year, as measured by percentage increase in mate- rials expenditures and total expenditures. · These increases might reflect part of the recruitment packages used to attract fe- male directors, who apparently are in de- mand more than ever. In fact, as documented by the ARL Annual Salary Sur- vey for the past three years, female directors earned an average salary higher than the average salary of male directors.24 How- ever, we are reluctant to draw conclusions from this observation because the factors relating to budget changes are complex. Tenure and the Descriptive Variables To investigate further the relationship between tenure and the descriptive vari- ables, we used stepwise multiple regres- sion analysis. Tenure was the dependent ) variable, and the independent variables in- cluded gender of the directors, volumes, volumes added, materials expenditures, total expenditures, percentage change in materials expenditures, and percentage change in total expenditures. For 1970 and 1979, no variables entered the equation using the .OS level of significance. For 1989, gender entered the equation first (Multiple R = .23), followed by percentage change in materials expenditures (Multiple R =.31 ). No other variable entered the equation at the .05 level of significance. Gender of the director and percentage change in materi- als expenditures account for very little of the variation in tenure (R Squared= .09). 8.6 45 10.0 208 8.6 While there is some relationship between gender, materials expenditures, and tenure, other factors outside the scope of this study apparently account for the length of tenure of library directors. CONCLUSIONS In examining the data about ARL direc- tors over the past two decades, it is appar- ent that the most significant change has been the large increase in the number of female directors. Their ranks have grown from a handful in the 1970s to nearly 30% of the total number of library directors in the late 1980s. Female directors are youn- ger, have fewer years of service, and are not quite as educated as their male counter- parts. Although there were no female ARL directors in 1970-and, in fact, very few during much of that decade-today, the resources controlled by both genders are fairly evenly matched. We find it most in- teresting that the average salary of female ARL directors now exceeds that of male ARL directors, and we also find it interest- ing that this fact has gone largely unno- ticed. Therefore, even though there are still many fewer female than male directors, the data indicate that females have, by and large, achieved parity with their male counterparts, at least in terms of salaries and average resources controlled. It is tempting to speculate about the ap- parently faster rise of females to ARL di- rectorships and whether it has resulted from the influence of the women's move- ment, affirmative action pressures, quali- tative differences in capabilities, or intense mentoring, to name only a few possible reasons. However, any conclusions we could draw at this point would be purely speculative and without basis in any solid research. We leave it to other researchers to investigate this complex yet import- ant subject. In the last five years, more than 50% of all new positions were filled by male directors replacing females or female di- rectors replacing males. This interesting trend has implications for job hunters. We would expect this phenomenon to decrease and eventually disappear as more females are recognized to be viable candidates for directorships. We think it fair to predict that, judging from the last two decades, we will enter the twenty- first century with more female directors than ever before. Another important measure of parity will be that of average tenure. Our study is necessarily inconclusive on that point, primarily because female directors are still relatively new to their positions. Data in the future will be most interest- ing. Earlier studies, as well as at least one recent commentator, predicted that the average tenure of all directors would fall to about 5 or 6 years. Our study showed that, overall, the average tenure was 7.1 years in 1989. From 1985 to 1989, depart- ing male directors logged an average of 11.7 years of service in contrast to de- parting female directors, who averaged 6.5 years of service. Turnover rates (i.e., the number of directorships vacated) dur- ing the four 5-year periods under study remained fairly constant. Research libraries are growing ever more complex, and it appears as if they will be under increasingly strong financial pressures in the 1990s, as higher education tries to deal with declining enrollments and skyrocketing costs. Previous authors, however, have pointed to similar factors in the past as the basis for both their observa- tions about shortened tenure and their pre- dictions of even shorter terms of service. Putting this all in perspective, we expect that studies done toward the end of the next decade will not show a significant drop in the length of service of the average ARL director, male or female. This study confirmed many findings . of previous studies of ARL directors. A typical director is still male, in his 50s, ARL Directors 253 and with a doctorate degree. Before as- suming his current directorship, he was likely an associate director at another ARL library. He probably will stay in an ARL directorship until he retires. But, among the emerging trends we identi- fied is the growing tendency for ARL directors to move from one ARL direc- torship to another; therefore, the future typical director may end his career after stints as director of more than one ARL library. We live in a society in which retire- ment ages are either increasing or be- coming purely voluntary, and given that fact, we would ordinarily expect to see ARL directors retiring at later ages or after longer tenures. In much of the profes- sional academic work force, administra- tors move among levels of responsibility for widely varying operations and fre- quently return to teaching positions well before the official end of their profes- sional working lives, without stigmas of failure. Directors of ARL libraries, as Woodsworth pointed out, have very few and limited career options. Teaching is usually limited to persons with doctor- ates, and returning to such positions as reference librarian or cataloger is often im- possible without a considerable amount of retraining and loss of face. Librarians are often perceived by universities to be inap- propriate or unqualified for general ad- ministrative positions, and the number of nonprofit or nonlibrary institutions for which librarians are seen to be qualified is also very small. We frequently hear of di- rectors who are burned out or who have become ineffective within their institu- tions. Yet, their options are indeed limited. Until the profession finds alternative ca- reer paths for ARL directors, we will con- tinue to see them hanging on in their current positions (with average tenure rates remaining fairly stable or even in- creasing) or leaving the profession en- tirely. The first alternative leaves us with little opportunity to bring vitality and cre- ativity to the leadership of our institutions. The second represents a major loss of tal- ent and contributions to our libraries. 254 College & Research Libraries May 1991 REFERENCES AND NOTES 1. Arthur M. McAnally and Robert B. Downs, "The Changing Role of Directors of University Libraries," College & Research Libraries 34:103-25 (Mar. 1973). 2. Anne Woodsworth, "Getting off the Library Merry-Go-Round: McAnally and Downs Revisited," Library ]ournalll4:35-38 (May 1, 1989). 3. William L. Cohn, "An Overview of ARL Directors, 1933-1973," College & Research Libraries 37:13T-44 (Mar. 1976). 4. Jerry L. Parsons, "Characteristics of Research Library Directors, 1958 and 1973," Wilson Library Bulletin 50:613-17 (Apr. 1976). 5. Louis Kaplan, "Letter to the Editor: Directors of University Libraries," College & Research Libraries 38:521 (Nov. 1977). 6. Janice Clinedinst Fennell, "A Career Profile of Women Directors of the Largest Aca- demic Libraries in the United States: An Analysis and Description of Determinants" (Ph.D. diss., Florida State Univ., 1978). 7. Ronald Dale Karr, "The Changing Profile of University Library Directors, 1966-1981," College & Research Libraries 45:282-86 (July 1984). 8. WilliamS. Wong and DavidS. Zubatsky, "Research Notes: The Tenure Rate of Univer- sity Library Directors: A 1983 Survey," College & Research Libraries 46:69-77 (Jan. 1985). 9. Anne Woodsworth, "Getting off the Library Merry Go Round," 35-38. 10. Association of Research Libraries, Minutes (Washington, D.C.: ARL, 1969-1988). The membership list (including the name of the director) is given as an appendix in the minutes and also issued as a separate mailing to each member library. 11. Dorothy Ethlyn Cole, ed., Who's Who in Library Service (New York: Grolier Society, 1955); Lee Ash, ed., Who's Who in Library Service (Hamden, Conn.: Shoe String, 1966); Joel M. Lee, ed., Who's Who in Library and Information Services (Chicago: American Library Assn., 1982); Directory of Library and Information Professionals, published in collaboration with the American Library Assn. (Woodbridge, Conn.: Research Publications, 1988); directors not found in the library science directories were searched in the following: Who's Who in America 1970- 71 (Chicago: Marquis Who's Who, 1970); Who's Who in America 1988-89 (Wilmette, Ill.: Marquis Who's Who, 1988). 12. The people/profile section of issues of College & Research Libraries News was examined from January 1969 to February 1990 for major items on people in the news; because they are just lists of announcements, the long lists of appointments were not examined in detail. 13. The states were classified into the following areas: East-Connecticut, Delaware, Maryland, Massachusetts, New Hampshire, Rhode Island, Washington, D.C.; Mid-At- lantic-New Jersey, New York, Ohio, Pennsylvania; South-Alabama, Florida, Geor- gia, Kentucky, Louisiana, North Carolina, South Carolina, Tennessee, Texas, Virginia; Midwest-Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, Oklahoma, Wisconsin; West (Mountain Standard Time)-Arizona, Colorado, New Mexico, Utah; Pacific Coast-California, Hawaii, Oregon, Washington; Canada. 14. Wong and Zubatsky, "The Tenure Rate," p.71. 15. McAnally and Downs, "The Changing Role," p.307. 16. Richard B. Dwore and Bruce P. Murray, "Turnover at the Top: Utah Hospital CEOs in a Turbulent Era," Hospital & Health Services Administration 34:333-51 (Fall1989). 17. Ibid. 18. Glenn M. McEvoy and Wayne F. Cascio, "Do Good or Poor Performers Leave? A Meta-Analysis of the Relationship between Performance and Turnover," Academy of Management Journal30:744-62 (Dec. 1987). 19. Cohn, "An Overview of ARL Directors, 1933-1973," p.143. 20. Ibid., p.143-44. 21. Wong and Zubatsky, "The Tenure Rate," p.75. 22. Karr, "The Changing Profile," p.283-84. 23. Ibid., p.283. . 24. Association of Research Libraries, ARL Annual Salary Survey (Washington, D.C.: ARL, 1987, 1988, 1989). Build Your Future with ACRL The library association that • contributes to the total professional development of over 10,000 academic and research liorarians • improves service capabilities of academic and research librarians • promotes and speaks for the interests of academic and research librarianship • promotes study and research relevant to academic and research librarianship Membership benefits include free subscriptions to College & Research Libraries. ACRL's official journal, and College and Research Libraries News. ACRL's news magazine for the profession - a $50 value; reduced rates for conference registration and continuing education courses, discounts on ACRL publications -and much more. The Association of College and Research Libraries is a division of the American Library Association (ALA). ALA membership is prerequisite to ACRL membership. ~,----------------------, • I Please rush information about joining ACRL to: I I ~~ I I Address I I City State Zip I L ______________________ ~ Mail to: Association of College and Research Libraries • American Library Association • 50 East Huron • Chicago, IL 60611 • 312/944-6780 •