Previous Page Table of Contents Next Page

Institutional and Gender Effects on Academic Hiring Practices

Scott Hutson


Fifteen years ago, Alison Wylie (1983, Comments on the "Socio-Politics of Archaeology": The Demystification of the Profession. In The Socio-Politics of Archaeology, edited by J. M. Gero, D. M. Lacy, and M. L. Blakey. Research Reports No. 23, Department of Anthropology, University of Massachusetts-Amherst) noted that "the archaeological research enterprise can profitably be viewed as a field of struggle to produce and control . . . `scientific capital'." (See also P. Bourdieu, 1988:11 Homo Academicus. Stanford University Press, Stanford). Since then, various archaeologists have demonstrated that archaeological "capital" is unevenly distributed: male archaeologists seem to be more successful than female archaeologists, some types of research are more highly valued than others, and some research institutions are more prestigious than others.

This paper explores potential areas of inequality in academic hiring practices. Two recent articles in the SAA Bulletin have focused on this particular issue [M. Zeder, 1997b, The American Archeologist: Results of the 1994 SAA Census. SAA Bulletin 15(2):12-7; B. L. Stark, K. A. Spielmann, B. Shears, and M. Ohnersorgen, 1997, The Gender Effect on Editorial Boards and in Academia. SAA Bulletin 15(4):6-9]. In summarizing the results of the 1994 SAA survey, Zeder (1997b:15-16) noticed that male professors earn more than female professors and that men are more likely to secure tenure track positions than women. In a study conducted by a subcommittee of SAA's Committee on the Status of Women in Archaeology (COSWA), Stark et al. concluded that women are not being hired in academia in proportion to their representation among Ph.D. recipients (1997:9). In this paper, I continue to examine the "gender effect" on hiring and explore another potential source of unevenness in hiring practices--the "institutional effect." Specifically, I ask the following questions: What are the chances of getting a teaching job today if you are from an "elite" graduate department as compared to if you're not? Is the difference significant? What are the chances if you are a female as compared to if you are a male? What institutions currently hold the most control over the teaching profession?

In addition to answering these questions, I will add a diachronic approach to the investigation of the gender and institutional effects. Such time depth can illustrate how employment patterns have been changing over the past 30 years. Specifically, I will assess the degree to which elite institutions have been able to reproduce or "institutionalize" their success on the job market over the years, and describe the tempo of the glacial thaw toward women in archaeology.

Patterns of inequality can be found in hiring practices of the non-academic sectors of archaeology, so it is reasonable to ask why this paper focuses only on hiring practices in academia. The reasons are partially practical and partially substantive. On the practical level, information on anthropology faculty was the most accessible. On the substantive level, professors might have a greater influence on contemporary archaeology because they are empowered with the duty of training future archaeologists, and they may have a formative influence on students with no formal exposure to archaeology. Since most jobs in archaeology require some degree of college-level training, even those student archaeologists not at all stimulated by their professors are at least subjected to their lectures and grading.

Return to top of page


The databases of archaeology teachers and Ph.D. recipients used in this study were collected from various editions of the annual AAA Guide. Unless noted otherwise, when I refer to archaeology teachers in a certain year, I refer to the subset of faculty members in the departmental faculty listings of the Guide who identify themselves as specialists in archaeology of some sort (culture history, bioarchaeology, etc.) and who hold either a full time, part time, or visiting professorship, and I eliminate emeritus, adjunct, or research professorships.

The potential flaws in this method should be discussed before continuing. First, due to the lack of a standardized classification system, it is often difficult to determine which professors actually teach. For example, I exclude adjunct professors because in many cases they are merely affiliates who do not teach. However, in the cases where they do teach, my sample of teachers becomes less complete. Second, the research specialties listed by professors may be outdated, incorrect, or not correlated with the subjects they actually teach. Third, the Guide tends to list only faculty that are part of a formally organized anthropology or sociology/anthropology department. Archaeologists who teach in community colleges or small two- or four-year colleges without an anthropology department are often not listed. Fourth, most recipients of Ph.D.s from American departments who teach archaeology outside of the United States are not listed in the Guide, making my sample less representative. Fifth, there is difficulty in attributing gender to some first names that are indistinct or foreign. Despite these flaws, the Guide represents the largest and most easily accessible source of information for demographic study and has been successfully used in many other studies.

Return to top of page

The Institutional Effect

When looking for a teaching job, how important is it to have received a Ph.D. from an elite institution? Are the graduates of the elite doctoral programs the ones most often hired to teach? The 12 "elite" schools in this study are the schools that were ranked most highly in a 1992 survey of SAA members with Ph.D.s. The results of this survey were published in the SAA Bulletin in 1993 (Table 1). Between 1991 and spring 1996, 502 people received Ph.D.s in archaeology from U.S. programs, according to listings in the AAA Guides from these years. Of these, 80 (15.94 percent) found either full- or part-time teaching jobs by the 1996-1997 academic year, according to the 1996-1997 Guide. Of these 80 successful job-finders, 36 graduated from the 12 highly ranked schools. During this time period, these 12 schools awarded 176 degrees, which means that a Ph.D. from a highly-ranked program had a 20.5 percent chance of finding a job (36 out of 176). Those graduating from non-ranked programs had a 13.5 percent chance of finding a job (44 out of 326). The deviation of these results from the expected frequencies is significant only at the 0.10 level of statistical significance (chi square = 3.4711, df = 1). Stated differently, the highly-ranked programs awarded 35.1 percent of the Ph.D.s, but took 45.0 percent of the jobs.

Table 1

Considering all job positions in the 1996-1997 Guide, not just those with recent Ph.D.s, we find that some schools have many more of their Ph.D.s placed in jobs. Table 2 shows the 12 schools that have been most successful in finding teaching appointments for their graduates. When combined, these 12 "powerful" schools control a majority of the archaeology teaching positions (full or part time) in the United States. Of the 850 positions held by archaeologists with Ph.D.s from American universities, graduates from the 12 powerful schools occupy 457 (53.8 percent) of the positions. The other 370 remaining teaching positions are divided among recipients of degrees from 63 graduate programs. Seven programs that have awarded Ph.D.s in archaeology over the past 30 years have been completely unsuccessful in placing their graduates in teaching positions. All together, 14.6 percent of the nation's archaeology graduate programs control 53.8 percent of the teaching jobs.

Table 2

However, this does not mean very much until we control for the fact that some schools award many more Ph.D.s than others. My database of Ph.D.s awarded begins with 1967 and continues to spring 1996, with a gap for the years 1969-1970. In these three decades, U.S. graduate programs awarded 2,148 Ph.D.s. By 1996-1997, 749 of these Ph.D. recipients held teaching jobs. The 12 most powerful schools for these three decades (see Table 3) awarded 849 of the total doctorates (37.5 percent) and these Ph.D.s held 369 of the teaching jobs in 1996-1997. A graduate from one of these programs had a 44.2 percent chance of finding a teaching job, whereas a graduate from a non-elite school had a 29.3 percent chance (380 out of 1,299). The deviation of these results from the expected frequencies is significant at the p = 0.001 level (chi square = 31.5329, df = 1). Stated differently, the powerful schools awarded 39.5 percent of the Ph.D.s and took 49.3 percent of the jobs from 1967 to 1996.

Table 3

How do the results for the period between 1967 and 1996 compare with the results from 1991-1996? The deviation from expected frequencies is highly significant for the period between 1967 and 1996 but only weakly statistically significant for the 12 highly-ranked schools in the period between 1991 to 1996. Though one might conclude from this that the institutional effect is weakening, I believe that the higher level of statistical significance for 1967 to 1996 is more likely a result of the greatly enlarged sample size. Though the top schools over the last 30 years occupied a larger proportion of jobs than the top schools in the last five years (49.2 percent and 45.0 percent, respectively), the difference in proportion of Ph.D.s offered in the two cases could easily account for this discrepancy (39.5 percent and 35.1 percent).

If it seems that the institutional effect is not weakening, at least new schools are not entirely constrained from producing Ph.D.s that are successful on the job market. In 1968, the first year AAA published the doctoral affiliation of professors, the professors came from only 23 U.S. doctoral programs. Among the professors listed in the 1981-1982 Guide, 66 graduate programs were represented. By 1996-1997, 75 graduate programs had succeeded in placing one or more of their Ph.D. recipients in teaching jobs. Also, there is upward and downward mobility within the institutional effect. New schools have been able to move up to the very top in the last 30 years, while others have moved downward. For example, Arizona State University, the University of Massachusetts-Amherst, and the University of Texas had no presence in the community of archaeology professors in 1968, but have had some of the highest success rates in finding teaching jobs for their recent (1991-1996) Ph.D.s. Comparing the list of the 10 schools with the most pre-1968 Ph.D. recipients in teaching positions in 1968-1969 to the list of the 10 schools with the most post-1967 graduates in teaching positions in 1996-1997 illustrates a degree of both upward and downward mobility (Table 3). We cannot correct for the bias toward large schools in this comparison because listings of total numbers of Ph.D.s granted by each institution are not available prior to 1966.

To summarize, it appears that there is indeed an institutional effect in past and current academic hiring practices. This is perhaps what most readers suspected: It is no surprise that a Ph.D. from the nation's most respected archaeology program will have a better chance of getting a teaching job than somebody from a small, poorly known program. By looking closely at the numbers, however, I have quantified the strength of the institutional effect and suggested that it is not weakening. I have also shown that membership in the group of successful programs is never guaranteed: As new programs successfully compete for the scarce teaching positions, some traditionally successful schools have struggled to reproduce their elite status. Nevertheless, a very important question about unevenness in hiring practices remains unanswered--What does the institutional effect mean? Can we say that Ph.D.s from elite institutions are more successful in finding jobs simply because these elite institutions have better professors and resources, and therefore attract the best students? Alternatively, can we uphold the more critical conclusion reached by J. H. Bair, W. E. Thompson, and J. V. Hickey [1986, The Academic Elite in American Anthropology: Linkages Among Top-ranked Graduate Programs. Current Anthropology 27(4):410-412] in their 1986 analysis of top-ranked general anthropology programs that "no more in [archaeology] than in the rest of the world do the deserving get their just reward"? (1986:412). Perhaps limiting ourselves to a simple dichotomy of explanations--that highly ranked programs are good or that they maintain numerical and hegemonic domination--is too naive. Certainly other factors, like strong mentor recommendations of their Ph.D. students, crosscut these two concerns, though such topics are beyond the scope of this discussion.

Before continuing, however, it is instructive to note that the archaeology programs ranked highly by the 1993 SAA Bulletin survey are also the ones that have awarded the most Ph.D.s over the years. Since only archaeologists with Ph.D.s could participate in the survey of highly-ranked schools, we might take a critical look at the list of the 12 highly-ranked archaeology graduate programs and wonder if these are really the best programs, or if these programs simply have the largest faction voting for them. Either way, the 12 schools that dominate the archaeology faculties (Table 2) are similar to those that appear on the 1993 top 12 rankings (Table 1).

Return to top of page

The Gender Effect

Examining hiring practices is also a useful way to explore the chilly climate toward females in archaeology. However, before discussing the gender effect on hiring practices, it is useful to see what the Guide can tell us about underrepresentation in doctoral programs. Figure 1 presents the proportion of male to female Ph.D. recipients for the span of 1967-1996, excluding 1969 and 1970. In 1967 and 1968 women received only 7 (12.7 percent) of the total 55 degrees awarded, whereas in the last five years of the 1970s, women received 110 (26.8 percent) of 410, and from 1991 to 1995, women received 185 (42 percent) of 440. Despite these gains, the number of women receiving doctorates in archaeology is still lower than the number of female doctorate recipients in anthropology and the social sciences as a whole (A. Ford and A. Hundt, 1994, Equity in Academia--Why the Best Men Still Win: An Examination of Women and Men in Mesoamerican Archaeology. In Equity Issues for Women in Archaeology, edited by M. C. Nelson, S. N. Nelson, and A. Wylie. Archaeological Paper of the American Anthropological Association No. 5, p.142).

Figure 1

If we compared the proportion of female archaeologists completing their Ph.D.s to the proportion of females who enter doctoral programs in archaeology, we could determine whether or not there was a gender-based "weeding" process in graduate school. Although the Guide does not differentiate anthropology graduate students by subfield, the recent SAA survey provides information of this sort. Whereas 42 percent of the archaeology doctorates went to women in the early 1990s, females represented 51 percent of the student membership in SAA in 1994 (M. Zeder, 1997a:10, The American Archaeologist: A Profile. AltaMira, Walnut Creek). If this equality in gender representation in graduate school is recent, it could mean that there will be numerical equality in the gender of Ph.D. recipients once these students complete their degrees. However, if this equality of representation in graduate school is not new, then it suggests that women already enrolled in school have less chance of finishing a doctorate than men.

In terms of hiring practices, there is a slow move away from the glacial climate toward women. Women identifying themselves as specialists in archaeology in the 1968 AAA Guide held only 9.5 percent of the full- or part-time teaching positions (23 women out of 242 total positions).1 In 1981, women held 16.5 percent of the teaching positions (109 of 660), and by 1996, women held 24 percent.2 Although this slow increase in female representation in archaeology faculties suggests that the climate toward women in archaeology is thawing, it in fact says nothing about female underrepresentation in hiring practices because the figures are not compared to the proportion of women in the hiring pool--e.g., the proportion of women who receive Ph.D.s.

C. Kramer and M. Stark (1989, The Status of Women in Archaeology. Anthropology Newsletter 29(9):1, 11-12) attempted such a comparison and noted that between 1976 and 1986, women received approximately 36 percent of the Ph.D.s in archaeology, but that in 1986, women only held 20 percent of the full time teaching positions. Although this led Kramer and Stark to wonder where the recent female Ph.D.s had gone, this comparison actually says little about underrepresentation of women because a majority of the teaching positions in 1986 were held by archaeologists who received their Ph.D.s prior to 1976. The figure of 20 percent cannot be compared to the figure of 36 percent because the 36 percent represents only graduates between 1976 and 1986, while the 20 percent represents a sample of Ph.D.s that goes from 1986 all the way back to the 1930s. To make the samples comparable, professors in 1986 who received their Ph.D.s prior to 1976 should have been eliminated. Also, based on listings in the Guide, women received 32 percent of the total Ph.D.s between 1976 and 1986, not the 36 percent that Kramer and Stark present. While Stark et al. (1997) reach a similar conclusion to that of Kramer and Stark, it rests in part on an analysis that perpetuates these methodological errors.

To improve on the test for underrepresentation, I used essentially the same technique as Kramer and Stark, but I made the sample of Ph.D. recipients comparable to the sample of professors. I made comparisons for three chronological periods: the early 1990s, the late 1970s, and 1967-1996. For the early 1990s, I quantified the women listed as faculty in the 1996-1997 Guide and received their Ph.D. between 1991 and spring 1996 and compared this to the number of women who received Ph.D.s from 1991 to spring 1996. I did the same calculation for men. (This is the same analytical technique used for the institutional effect, but with the data coded by gender rather than by graduate institution. I could not assign gender to 18 Ph.D. recipients of this period and therefore excluded them from the sample.) The results show that for women, 30 (15.2 percent) of the 198 recent Ph.D. recipients found jobs, and for men, 49 (17.3 percent) of the 284 recent graduates found jobs. Stated differently, women received 41 percent of the Ph.D.s and held 38 percent of the new jobs. Though men have slightly better chances of finding teaching jobs, this difference is not statistically significant at either the p = 0.05 or p = 0.10 levels (chi squared = 0.3145, df = 1). If we eliminate part-time and consider only full-time teachers, then women are better slightly respresented, holding 40 percent (as opposed to 38 percent) of the jobs.

Conducting the same analysis for Ph.D. recipients in the late 1970s reveals that women were overrepresented in proportion to their representation among Ph.D. recipients 15 years ago. Women who received their Ph.D.s between 1976 and 1980 had a 40 percent chance of finding a teaching job by 1981 (45 jobs out of 112 Ph.D.s) whereas men had only a 30 percent chance (88 jobs out of 293 Ph.D.s). Stated differently, women received 27.7 percent of the Ph.D.s but took 33.8 percent of the jobs. Although this disparity is much more substantial than in the preceding analysis, it is still not statistically significant at either the p = 0.05 or p = 0.10 levels (chi square d = 2.3986, df = 1). Restricting the sample of professors to only those with full-time positions does not affect the results.

For 1967 to 1996 combined, 42 Ph.D. recipients were of indeterminate gender, women had a 29.1 percent chance of finding a job (199 faculty of 685 Ph.D.s) whereas males had a 37.8 percent chance of finding a job (535 faculty of 1,416 Ph.D.s). This difference is statistically significant at the p = 0.005 level (chi square = 10.0643, df = 1). Stated differently, women held 32.2 percent of the doctorates, but only 27.1 percent of the faculty positions.

In summary, the gender effect is a significant factor in hiring practices for the combined time span of 1967 to 1996, but does not appear to be significant in either of the two five-year windows within this span (the early 1990s and the late 1970s). In fact, female archaeologists actually had a better chance of finding a teaching job than men in the late 1970s. There are two ways to account for this discrepancy: (1) it is possible that hiring practices in the 1980s and early 1970s were unequal enough to overcompensate for the parity in the late 1970s and early 1990s; or (2) it is possible that women, although hired in equal proportions, do not stay in their jobs as long as men. The results of the 1994 SAA survey support the second explanation, showing that women are more likely to hold non-tenure track positions or lower paying positions and are less likely to receive employment benefits when compared to men (Zeder 1997a:75, 86, 101). In other words, I postulate that women are underrepresented because tenure discrimination or unequal compensation drives some women into other careers.

Return to top of page


The results of this study suggest that there is an institutional effect on hiring practices; however, it is currently not as strong as it has been in the past. We cannot determine if the fact that highly-ranked institutions have better employment records is due to the power of a prestigious name or due to the possibility that students receiving Ph.D.s from these institutions are "better" archaeologists. On the other hand, contrary to the conclusions of Stark et al. (1997), this study suggests that women are indeed being hired in academia in proportion to their representation among Ph.D. recipients. However, the gender effect influences hiring practices in other ways. Even though women might be hired to teach in equal proportion to the number of women receiving Ph.D.s, the types of jobs they find are not equal to the jobs held by men, perhaps causing women to abandon academia more often than men. Of equal importance is the fact that while there is gender parity in the student population of SAA, fewer females than males receive Ph.D.s. Like Stark et al. (1997), I conclude that the chilly climate toward females begins at least as early as graduate school, "where conditions both in the discipline and the wider society affect the proportion of women Ph.D. recipients." Researching these conditions, as well as the meaning of the institutional effect, will require scholars interested in the unequal distribution of archaeological capital to move toward a finer-grained ethnographic approach.


1I have included emeritus faculty in this sample, but have excluded faculty without doctorates.

2Professors receiving degrees from Canadian graduate programs and teaching in Canada were included. In 1986, female professors held 20 percent of full time jobs (Kramer and Stark 1988). In 1992 female professors held 21 percent of full time jobs (Stark et al. 1997).

Scott Randolph Hutson is in the graduate program at University of California, Berkeley.

Return to top of page

Previous Page Table of Contents Next Page