College and Research Libraries Career Patterns of Academic Lib_rary Administrators Barbara B. Moran A survey examined the career progression of a group of academic library administrators over a period of ten years in an attempt to identify the professional and personal characteristics associ- ated with success in attaining a directorship in an academic library. The findings suggest that two career progression patterns have existed in academic librarianship. The first pattern, typi- cal of males, required the acquisition of professional credentials and geographic relocations to achieve a directorship. The second pattern was typical of the females whose likeliest chance of attaining a directorship was found in internal promotions and rising through the ranks in one institution. lthough librarianship is catego- rized as a "women's" profes- sion and nearly 80 percent of li- brarians are women, the profession exhibits the same pattern of sex distribution as the male professions, that is, with females concentrated in.the lower positions of the library hierarchy. Survey after survey has documented that women are far less likely than men to be appointed as library directors and there is a general awareness in the profession that females have a difficult time attaining a director's position. 1 Although various conjectures have been made to explain the small num- ber of female directors, little research ex- ists to explain why women obtain top ad- ministrative positions in libraries so much less frequently than men. This study focused on two categories of factors associated with success in becom- ing a library director. The first category was composed of professional qualifica- tions, including items such as level of edu- cation, number of publications, activity in professional organizations, and years of administrative experience. The second category was composed of personal fac- tors such as marital status, number of chil- dren, and ease of geographic mobility. These two groups of variables, the profes- sional and the personal, were chosen as the focus of this study because explana- tions of female underrepresentation in top administration usually advance these fac- tors as the probable causes of the small number of female administrators. These · two explanations are often found in li- brary literature as well as in the literature of other professions. 2·1t was the purpose of this study to determine the relationship between these factors and attainment of a director's position. The librarians studied were all academic librarians who had held positions as either an-assistant or an associate library director in a four-year college or university library during the academic year, 1970-71. Indi- viduals who had held these titles were chosen as the focus of this study for sev- eral reasons. First, a brief preliminary study revealed that the position of library director was often preceded by a position as an assistant or an associate library direc- tor. Second, since the function of an assis- tant or associate library director is typi- cally general administration, and often involves formulating library policy, it is reasonable to assume that incumbents of these positions are likely to try for and at- Barbara B. Moran is assistant professor in the School of Library Science, The University of North Carolina at Chapel Hill. 334 tain a director's position as a logical step in their career patterns. This research exam- ined a group of individuals one step away from the top administrative position in an academic library, and then compared those who became directors with those who did not. Thus, the study was con- cerned only with individuals who had proven administrative competence dem- onstrated by the fact that they had all at- tained an assistant or associate library di- rector's position. METHODOLOGY The primary method of data collection used was a questionnaire mailed to a group of librarians who, in 1970, had ei- ther been an assistant or associate library director in a four-year college or univer- sity. There were 695 such individuals listed in the 1970-71 American Library Di- rectory. During the fall of 1980 current ad- dresses were obtained for as many of these individuals as possible, using vari- ous library and professional association directories. Either current or tentative ad- dresses (within the last five years) were obtained for 535 individuals. After send- ing out the questionnaires, it was discov- ered that of these 535, 91 people were ei- ther deceased or had moved without any forwarding address. These individuals were excluded from further study result- ing in a research population of 444. A review of the research on career pro- gression in libraries produced no existing instrument which could be applied to this study, therefore, a questionnaire was con- structed. Questionnaire items fell into three main categories: (1) questions con- cerning the progression of an individual's career in librarianship, (2) questions con- cerning professional qualifications, and (3) questions concerning the personal characteristics of the individual . Career progression questions were in- tended to gather information about the in- dividual's current position (or last posi- tion, if retired), the type of library where employed, number of volumes in the li- brary, number of professional librarians employed, and the number of students served. Respondents were asked to list all full-time library positions held and the Career Patterns 335 dates and institutions where they held those positions. Questions concerning professional qualifications asked about academic de- grees, the years these degrees were ob- tained, and undergraduate majors. Re- spondents were asked to list how many professional conferences they had at- tended in the last twelve months, and the number of professional or scholarly orga- nizations to which they belonged and their activity with each. Finally, they were asked to list the publications they had in various writing categories. Questions relating to personal charac- teristics gathered data about the respon- dent's age, sex, current marital status, number of children, age of children, em- ployment of spouse, and number of geo- graphic relocations. There were 320 usable responses, pro- ducing a final response rate of just over 72 percent. Of the 320 respondents, 193 (60 percent) were males and 127 (40 percent) were females. The sexual composition of the respondents was not representative of academic librarianship, which is approxi- mately 35 percent male and 65 percent fe- male, but is fairly representative· of aca- demic library administration. 3 The mean age of the respondents was 54.7 years with the average female (mean age, 58 .5) older than the average male (mean age, 52.2). In terms of marital status there was a striking difference between the males and females in the survey. A large proportion of the females-nearly 70 percent-were unmarried at the time of the survey, and over 50 percent of the fe- males had never been married. In con- trast, over two-thirds of the males were married at the time of the survey and fewer than 20 percent of the males had never been married. The respondents to the survey, both males and females, did not have large families. Indeed, 27.2 per- cent of the males and 65.1 percent of the females had no children at all. Looking only at that portion of the population who reported having children, the mean num- ber of children was 2.6 for the males and 2.4 for the females. There was no difference in the average age at which both males and females as- 336 College & Research Libraries sumed their first full-time professional li- brary position-for both the average age was 29. The males had become assistant or associate library directors at a younger age than the females, however. On the aver- age, males had assumed such a position at the age of 38, while females had become assistant or associate library directors at an average age of 42. The population in this study was a well- educated one, with more than 85 percent of the respondents possessing at least one master's degree, nearly 25 percent pos- sessing a second master's or a sixth-year degree, and nearly 10 percent possessing a doctorate. A larger percentage of males held a second master's or a Ph.D., and, probably reflecting the older average age of the females in the survey, a larger per- centage of females had the fifth-year bach- elor's degree in librarianship as their high- est degree. In terms of professional activities, librar- ians in the population were frequent con- ference attenders and the overall pattern of conference attendance was fairly simi- lar for males and females with the largest percentage of both reporting that they had attended five or more conferences, semi- nars, or workshops of at least a day's du- ration during the last year. The males and females were also very similar in terms of membership in professional organizations with males reporting membership in an average of 2.75 and females in an average of 2.89. In addition to looking at the num- ber of organizations each individual be- longed to, an attempt was made to gauge how active each individual was in profes- sional organizations by having the re- spondents indicate the number of and types of activity they performed in each organization, for example, whether they served as officers, committee members, presented papers, or merely attended meetings. An index was constructed to categorize each individual's activity into a low, moderate, or high pattern of organi- zational activity. More males than females reported having a high level of activity September 1983 within professional organizations while the percentage with moderate activity was about the same for each sex. The last measure of professional qualifi- cation examined in this study was publica- tions. Respondents to the survey were asked the number of publications they had in various categories and to list these publications. It must be emphasized that the approach used here was purely quantitative-no attempt was made to judge the quality of the publications. The responses showed a marked tendency to- wards greater publishing activity on the part of the males throughout all categories of publication. Although only one-third of the males reported having no publications at all, nearly 60 percent of the females re- ported no publications. This conforms with the pattern of fewer publications pre- viously reported for women, not only in li- brarianship, but also among faculty mem- bers.4 All respondents had been either an as- sistant or an associate library director dur- ing 1970 but the type of institution in which they were employed differed among the respondents, with women concentra:ted in college libraries, while over one-half of the men were found in re- search and doctoral granting institutions. Distribution of the individuals in 1970 can best be seen by classifying their institu- tions using the Carnegie Cassification of In- stitutions of Higher Education, which di- vides institutions according to criteria of size and type of degree offered* (see table 1). From this classification it is clear that the origins of the male and female assistant and associate library directors were differ- ent. The females were most commonly found as assistant or associate directors in comprehensive or liberal arts colleges while the males were more often found as assistant or associate library directors in universities. There was also a difference in the distri- bution of the males and females according to whether the institutions were privately *The subcategories used in the Carnegie Classifications (e .g., Research University 1, Research Uni- versity 2) have not been used in the tables presented here. Career Patterns 337 TABLE 1 CARNEGIE CLASSIFICATION OF EMPLOYING LIBRARY, 1970-1971 Total Males Females No. % No. % No. % Research university 89 27.8 75 38.9 Doctoral granting university 38 11.8 28 14.1 Comprehensive college or university 98 30.6 49 25.4 Liberal arts college 80 25.0 31 16.1 Professional or specialized institution 15 4.7 10 5.2 14 11.0 10 7.9 49 38.6 49 38.6 5 3.9 TABLE2 PROPORTION OF DIRECTORS FROM POPULATION Directors Non directors Total No . % 107 33.4 213 66.6 or publicly supported, with females more often in private institutions (61 percent), while the males most often were em- ployed by public institutions (again, 61 percent). Smaller institutions were mea- sured both in terms of student enrollment (mean, 4,839 for females and 11,402 for males) and in terms of number of volumes in the library (mean, 365,000 for females and 862,070 for males). Although this study examined the ca- reer progression of all the respondents, major emphasis was on the individuals who had become library directors in the decade between 1970 and 1980. As table 2 shows, there were 107 members of the original research population who had be- come library directors. t At first glance, it appears that males and females proceeded from assistant or asso- ciate directorships at roughly comparable rates; approximately 37 percent of the males became directors and 28 percent of the females. However, there were signifi- cant ways in which the male and female directors differed. The males who became · directors assumed this position at an ear- lier age than did the females who became directors. The males became directors at an average age of 42.5 years, the females at 50.3 years. Eight (22.2 percent) of the fe- males were appointed directors after the age of 61 compared with 2 (2.8 percent) of Males Females No . % No . % 71 36 .8 36 28.3 122 63.3 91 71 .6 the males. The heaviest concentration of females as directors of libraries· was in pri- vate institutions. Of the female directors, nearly 70 percent were found in private in- stitutions while only 26.8 percent of the male directors were found in private insti- tutions. In contrast, the males were found most frequently as library directors of public institutions. Just over 8 percent of both the male and female library directors had left the academic library fjeld to be- come directors of another type of library, most commonly a public library. Four of the female directors (11 percent) were at the time of the survey, or had been before retirement, directors of an ARL library, 18 (25.3 percent) of the male directors had held such a position. A better understanding of the locations of the directorships can be gained by plac- ing these directorships in the Carnegie Classifications used earlier, as is done in table 3. It is clear that the female achieving a directorship was most likely to be found in a small liberal arts college; a male was more than twice as likely to be found in the research or doctoral granting university category. One of the most surprising findings about the female directors was the high percentage who had become directors in the same library where they had previ- ously served as assistant or associate di- *This figure does not include that portion of the population ( n = 14 males (7 .3 percent) and 14 females (11 percent) who reported serving as an acting director but who never received a permanent appoint- ment. It does include those who served as a library director but who have subsequently retired or moved to another position. 338 College & Research Libraries September 1983 TABLE 3 CARNEGIE CLASSIFICATION OF DIRECTORS' LIBRARIES No. Research university 18 Doctoral granting university 15 Comprehensive college or university 27 Liberal arts college 28 Professional or specialized institution 9 Nonacademic institution 10 rector. Of the 36 females who had become directors, only 4 became directors of li- braries where they had been external can- didates for the position. In addition, of this 4, 3 had left the field of academic li- brarianship to become directors of nonac- ademic libraries, so only one female from the population was found as a director of an academic library where she had not previously been employed as an assistant or associate library director. The pattern ran counter to the pattern of the males as can be seen in table 4. DETERMINANTS OF SUCCESS The next section of this article examines the factors distinguishing the individuals who became directors from those who did not. It will identify those personal and professional characteristics most often as- sociated with the attainment of a direc- tor's position. To investigate the association between the dependent variable, achieved success, and the various independent variables, multivariate cross tabulations were per- formed. First, the relationship between an independent variable and dependent vari- able was examined and then elaborated by introducing a third variable, sex. This · elaboration analysis allows us to see if the original relationship between the inde- pendent variable being tested, and suc- cess in becoming a library director, still re- mains after controlling for sex,br if the variable affects one sex differently from Total Males Females % No . % No . % 16.8 15 21.2 3 8.3 14.0 12 16.9 3 8.3 25.1 17 23.9 10 27.8 26.1 12 16.9 16 44.4 8.4 8 11.3 1 2.8 9.3 7 9.9 3 8.3 the other. The chi-square (X2) test of statis- tical significance was used to determine if a systematic relationship existed between the variables . The Effect of Personal Characteristics on Becoming a Director The relationship between each of the personal characteristics and success in at- taining a directorship was examined. For the entire population there was a relation- ship found between marital status and be- coming a director, with married individ- uals more likely to become directors. The chi-square for that computation was 14.88, which is statistically significant at the . 002 level, and is a strong indication that a relationship does exist between an individual's marital status and becoming a director. When the relationship was elab- orated by looking at the sexes separately, the relationship remained strong for the male sector of the population. A chi- square of 11.57 was produced for this group which is statistically significant. But when the relationship between marital status and attained position in the female sector of the population was examined, the relationship between the two factors was not sustained. This indicates that for men, marriage is associated with becom- ing a library director. Married men have become directors in numbers dispropor- tionate to their representation in the pop- ulation. For women, however, there is no TABLE4 INTERNAL AND EXTERNAL SUCCESSION TO DIRECTORSHIP Internal succession External succession Total No . % 53 49.5 54 50.5 Males No . % 21 29.6 50 70.4 Females No. % 32 88.9 4 11.1 TABLE 5 RELATIONSHIP OF MARITAL STATUS AND RANK CONTROLLING FOR SEX Sin~le Married ~~~!~ Males Directors 5 58 8 Non directors 33 76 13 x2= 11.57 with 2 df p < .001 Females Directors 16 15 5 Non directors 48 29 14 X2 = 1.11 with 2 df p > .50 association between marital status and at- tained position. The frequencies of the chi-square showing the relationship be- tween marital status and attained position are displayed in table 5. The positive association between num- ber of children and attained position was the next relationship studied. For the en- tire population individuals with children were more likely to be directors. When the analysis was conducted separately for males and females, however, the relation- ship between number of children and at- taining a directorship remained statisti- cally significant for males, but fell short of statistical significance for females as shown in table 6. Variables associated with geographic mobility were also examined. A move of at least thirty miles was considered as a geo- graphic move. The number of geographic moves was related to becoming a director for the entire population, but when each sex was examined separately, the relation- ship remained strong for males but virtu- ally disappeared for females, as shown in table 7. Career Patterns 339 An examination of the personal charac- teristic variables claimed to cause fewer fe- males to succeed in becoming directors shows that, for females, none of these var- iables are associated with becoming a li- brary director. Neither a female's marital status, number of children, number of li- braries worked in, nor number of geo- graphic moves had a bearing on her suc- cess. For males, however, these variables were associated with attaining a top ad- ministrative position. A married male with children, having made a number of geographic moves and worked in a large number of libraries, was statistically more likely to become a director than an unmar- ried male with no children remaining in the same geographic location. The Effect of Professional Characteristics on Becoming a Director As reported earlier, the data collected were in agreement with the results of pre- vious studies showing that, as a whole, fe- males scored below males on these qualifi- cations. Females had fewer advanced degrees, fewer publications, and lower levels of activity in professional organiza- tions. Males and females scored approxi- mately the same in terms of number of conferences attended, and number of memberships in professional organiza- tions. One of the professional qualifications thought to be associated with attaining high rank in an academic library is the pos- session of advanced degrees. Therefore, the relationship between educational level and a directorship was the first relation- ship examined. For the whole population there was an association between ad- TABLE 6 0 Directors 8 N ondirectors 44 RELATIONSHIP OF NUMBER OF CHILDREN AND RANK CONTROLLING FOR SEX Males Number of Children 1-2 3-4 34 23 47 21 X2= 15.72 with 3 df p < .001 Sor More 0 6 23 10 59 Females Number of Children 1-2 3-4 6 6 22 8 X2= 2.15 with 3 df p > .50 5 or More 1 2 340 College & Research Libraries September 1983 TABLE 7 RELATIONSHIP OF NUMBER OF MOVES AND RANK CONTROLLING FOR SEX Directors Non directors 0 11 31 Males Number of Moves 1-2 3-4 8 26 36 40 x2 = 21.27 with 3 df vanced degrees and directors, and the as- sociation also existed when looking at the male subsample, but a similar relationship between education and a directorship was not found for the female subsample. The relationship between these variables is shown in table 8. Males having higher degrees were more likely to have become directors than the females having higher degrees. For exam- ple, from the portion of the sample with Ph.D.s, 64 percent of the male doctorates became directors compared with only 25 percent of the female doctorates. In like manner, amount of publication was strongly associated with becoming a director for the entire population. This as- sociation was maintained for the male subsample with a chi-square of 12. 90, with two degrees of freedom, which is signifi- cant at the .01level. With the female sub- sample the relationship disappeared. The analysis produced a chi-square which was not statistically significant as is sho"Vn in table 9. The amount of organizational activity 5 or More 26 15 17 35 Females Number of Moves 1-2 3-4 6 11 20 26 X2 = 1.67with3df 5 or More 2 10 was associated with success in becoming a director for the population as a whole, and for the males. In what was becoming a fa- miliar pattern, the relationship failed to be maintained for the females, as demon- strated in table 10. A positive association between the number of conferences attended and be- ing a library director was also found for the whole population, but when exam- ined separately, the relationship held up for the males, but was not maintained for the females, as seen in table 11. No association was found between the number of professional organizations an individual belonged to and being a library director. This relationship was the same for both males and females, as well as the total group. Thus, it seems that professional qualifi- cations thought to affect female career progression have little influence on which females become directors. Females with higher degrees, more publications, or greater organizational activity did not be- come directors more frequently than the TABLES RELATIONSHIP BETWEEN EDUCATION AND RANK CONTROLLING FOR SEX BA or 5th- Highest Degree 2dMAor6th- YearDe~ee MLS YearDe~ee Ph .D. Males Director 1 38 18 14 Nondirector 11 60 42 8 X2= 12.232 with 3 df p < .01 Females Director 8 22 4 2 Nondirector 22 48 15 6 x2 = . 91083 with 3 df p > .80 Career Patterns 341 females who did not have these character- "0 1::: OS istics. For the males these external profes- ..,.. sional qualifications were clearly related to \0 Cf) I I ~ N being library directors. ui ], Personal and Professional 1::: 0 ·..: Characteristics in Different OS .~ Types of Libraries :0 ;:S J,!....:., 0.. As previously reported, women in the ... o....: 1: coo 2 3:;~0 I I research population were more com- _os~ t:--.00 s-g~ ..... monly located at smaller academic li- cE OS 0 ci ~ braries and the males in larger ones. Be- >< ~ cause of this distribution it was possible 1:-Ll 0 c:J) :t:: that type of library (i.e., the larger re- ~ ~ search library in contrast to the smaller 0 -"0 ;:S ~ ... !:::....: 0.. ]:g 0 college library) was confounded as a vari- C) \0 00 NM 1::: 9 . ~ '1. 342 College & Research Libraries September 1983 TABLE 10 RELATIONSHIP BETWEEN ORGANIZATIONAL ACTMTY AND RANK CONTROLLING FOR SEX Males Directors Non directors X2 = 20 .766 with 3 df p < .0001 Females Directors Non directors X2 = 2.364 with 3 df p > .50 No Memberships 3 18 2 10 Organizational Activity Low Moderate Activity Activity 4 17 8 28 17 45 13 30 High Activity 47 40 12 22 TABLE 11 RELATIONSHIP BETWEEN NUMBER OF CONFERENCES ATTENDED AND RANK CONTROLLING FOR SEX Males Number of Conferences 0 1 2 3 4 Director 4 3 10 19 9 Non director 14 18 23 25 11 X2 = 13.327 with 5 df p < .02 large research libraries were more similar on a given personal or professional char- acteristic to the other females in the study than to the male directors of large research libraries. No statistical tests were performed on the data because the number of males or females in the subgroups was consistently too small to satisfy the assumptions of the otherwise applicable analyses such as chi- square. Nonetheless, the data do provide the best available indication of the extent to which the relationships previously dis- cussed exist in each of the types of li- braries. In brief, this examination showed that for the variables of marital status, number of children, number of geographic moves, educational level, and publications and activity in professional organizations, the relationships that had been found for the population as a whole seemed to exist for each of the types of libraries. Likewise, it was noted that the female directors of the large research libraries were more similar 5 or More 25 22 Females Number of Conferences 5 or 0 2 3 4 More 1 4 2 6 3 15 4 12 15 16 13 16 x2 = 9.286 with 5 df p > .09 on these six variables to the other females in the population than to the male direc- tors of the large research libraries. Conse- quently, it was concluded that the previ- ously reported relationships between sex, achieved success, and these six personal and professional characteristics were not by-products of the higher concentration of males in the large research libraries. 5 CAREER PROGRESSION PATTERNS FOR MALES AND FEMALES All of the variables examined in this study were ones that had been suggested as possible causes for the smaller number of female directors in academic libraries. From the findings of this research, there was no relationship between any of the variables and achieved success for the fe- males. On the other hand, a large number of the variables were associated with suc- cess for males. The elimination of these variables as fac- tors in the career progression of females forces one to look elsewhere for an expla- nation of the varying career progressions of male and female academic librarians. On the basis of the career progressions of the population in this study, it is possible to speculate upon the existence of two separate career patterns tha:t exist in aca- demic librarianship-one pattern existing for the males and a separate pattern pre- vailing for the females. For those males seeking to advance their careers, there seems to be a clear-cut sce- nario to follow, enabling them to rise to a directorship. If an aspirant accumulates . the professional credentials of advanced degrees, participation in library organiza- tions, some publishing, and makes a se- ries of career moves to work in different academic libraries, the chances are good that the would-be director will attain such a position. But one caution based on the experiences of the males in this study-the directorship must be attained before the age of fifty, or the chances for reaching a top position are reduced considerably. For females, again based on the experi- ences of the ones in this study, the male game plan would not be successful. Even if they take the time and the trouble to amass the professional qualifications that seem required of a director, they will not necessarily be rewarded for their invest- ment. None of those qualifications were associated with being a director for fe- males. For the females in this study, the only constant in becoming a director was to stay in the institution where they had been an assistant or associate library direc- tor. As previously noted, only one female in the study became a director of an aca- demic library where she had not worked as the assistant or the associate director, and that seems strong evidence that for a female, the likeliest route to a directorship is to try to get an administrative appoint- ment in the best academic institution pos- sible and to stay put. If this pattern of administrative success for females exists, it explains many of the discrepancies in the association between the variables and becoming director. For instance, professional credentials would not be so important a factor in the hiring of a female, who, as an internal candidate, Career Patterns 343 applied for the directorship. An internal candidate's strengths and weaknesses are well known, and proven administrative competence is apt to be the deciding fac- tor. Professional credentials are presum- ably weighed more heavily in the case of an unknown external candidate as a mea- sure of suitability for the job. This pattern of female succession to the directorship would also explain why factors such as marital status seem to have no bearing on which of the females became directors, be- cause if almost the only way a female can become a director is to stay in the same li- brary for a good number of years, family constraints such as lack of geographic mo- bility would not be a factor associated with success. The older age at which females become directors (average age, 50.3 years in this study) would also help explain why there was no significant difference in at- taining a directorship between women with children and childless women, since it is unlikely that few women of that age would have young children requiring ex- tensive child care. Metz reported that libraries hiring a di- rector preferred to hire both an outside candidate for the job and preferred to hire a male, with his data indicating that nei- ther women, nor inside candidates, were preferred for such positions. He also pos- ited that there is an interaction between the variables of sex and external succes- sion, so that a "wrong" combination of the two is especially unacceptable, that is, a female, external candidate would be the least acceptable type of individual to be hired as a director. In his study Metz found women were three times as likely as men to have been hired to directorships from within the li- brary. The data gathered in the present study are similar to the findings of Metz, in that 32 out of 36 females who became di- rectors reached that position via the inter- nal succession route while only 29 out of 71 of the males who became directors did so by internal progression. The rate of internal succession for both males and females in Metz's study was lower than the internal succession rate of the librarians in this study, probably be- cause of the populations studied. Metz 344 College & Research Libraries studied individuals who were already di- rectors, and their positions before assum- ing a directorship. It is likely that internal succession is a more common career move for someone already in an assistant or as- sociate position. The findings in this study corroborate the earlier work of Metz, and show that fe- males are most likely to be hired as direc- tors from within a library where they are already employed, and that the adminis- trative opportunities open to women are more restricted than those open to men. If the career patterns of the respondents to this survey are an accurate depiction of the career patterns of female library ad- ministrators, they serve as an indictment both of the hiring and promotion practices prevalent in librarianship. If a profession that is 80 percent female has such a fear and distrust of females as directors that only a female who has been a long-term success as an administrator is likely to be September 1983 hired as a director, then something is seri- ously wrong with the perception of fe- males as library administrators. If the pat- tern described above still prevails in librarianship, the ambitious woman wish- ing to become a director has many ave- nues of advancement closed to her, and has to passively await an opening in her present institution. The problem of female underrepresen- tation in administrative librarianship posi- tions remains a serious one. The 75,000 women librarians in the United States de- serve to have this problem explored, and definitive explanations produced, so that remedies can be made. This study has be- gun to explore some of the reasons for the differential career progressions of males and females, but there are many questions that remain to be answered. The need for further research is clearly indicated and the position of women in libraries justifies the necessity for such research. REFERENCES 1. For instance, see L. S. Estabrook and K. M. Heirn "A Profile of ALA Personal Members," American Libraries 11:654-9 (Dec. 1980). 2. For use of the explanations in librarianship, see J. T. Carey, "Overdue: Taking Issue with the Is- sues," Wilson Library Bulletin 45:593-94 (Feb. 1971) orR. L. Simpson and I. N. Simpson, "Women and Bureaucracy in the Semi-Professions" in A. Etzioni (ed.) The Semi-Professions and Their Organi- zations.(New York: Free Pr., 1969). For use of the explanations in a field other than librarianship, see S. E. Esther "Women as Leaders in Public Education," Signs 1:363-81 (1975). 3. R. J. Talbot and A. Lippe, Salary Structures of Librarians in Higher Education for the Academic Year 1975-76 (Chicago: Association of College and Research Libraries, 1976). 4. P. D. Morrison, The Career of the Academic Librarian (Chicago: American Library Assn., 1969); ]. A. Centra, Women, Men and the Doctorate (Princeton: Educational Testing Service, 1974). 5. B. B. Moran, "Career Progression of Male and Female Academic Library Administrators" (Ph.D. diss., State University of New York, Buffalo, 1982). 6. P. Metz, "Administrative Succession in the Academic Library," College and Research Libraries 39: 358-64 (Sept. 1978).