Academia’s Most Wanted: The Characteristics of Desirable Academic Job Candidates in Criminology and Criminal Justice

Abstract

Although personal opinions and anecdotal evidence abound, there is a dearth of systematic research into the characteristics that academic criminology and criminal justice departments seek in job candidates. The current study uses a nationally representative sample of faculty members to assess the extent to which a wide array of attributes affects whether an applicant for an assistant professor position would be invited for an on‐campus interview. Using hierarchical linear modeling, we find that several applicant characteristics are related to the likelihood of an interview and that some of these relationships are conditioned by the characteristics of the faculty who decide whether to extend an invitation. We discuss the implications of our findings for the nature of the discipline, and for mentoring doctoral students, designing PhD program experiences, and hiring new colleagues.

Notes

1. Although criminology and criminal justice may be regarded as distinct disciplines, we chose not to separate them for this study. There appears to be considerable overlap in employment—criminal justice programs hire criminologists and vice versa—and curricula. Southerland’s (2002) analysis, for example, showed that 53 percent of undergraduate academic programs identified as exclusively “criminal justice” required coursework in criminology for their majors. Moreover, our results revealed no statistically significant differences in the desirability of academic job candidates between faculty employed in departments that offered a degree in criminal justice versus those who offered a degree in criminology.

2. Due to a problem in the email program used to administer the surveys, approximately 100 of the sample members received the prenotice email and the first mailing of the survey at nearly the same time. The approach of using email to administer the survey resulted in other technological obstacles. Two sample members refused to participate because they stated they could not be sure the attachment was not a computer virus. One person declined to participate because he found it too difficult to attach and send the completed survey by email, fax a copy, or send a printed copy by postal mail, stating, “If I can’t click and send its [sic] a problem.” Presumably, she or he would have preferred a web‐based questionnaire, which was beyond our technological capabilities given the underlying structure of the study design. Three other participants requested, were sent, completed, and returned paper copies of the survey. Finally, several respondents replied to follow‐up contacts to say that they had already returned the survey. Generously, they were all willing to resend their completed questionnaires when we informed them that their responses had not been received.

3. We did not systematically collect data on the discipline in which a doctoral degree was offered. Many respondents’ departments, however, offered a PhD in some discipline other than criminology or criminal justice (e.g., sociology).

4. Demographic profiles of the full memberships of ACJS and ASC are not available. As a very rough assessment of representativeness, however, our sample can also be compared with the results from a 2003 survey that ACJS conducted of a sample of its members (Cullen, Blevins, Pealer, Daigle, and Coleman 2004). Their response rate was somewhat higher than ours (57.1 percent), but their sample was very similar to ours in terms of percent White (89 percent) and age (mean = 46).

5. Because the Carnegie Foundation for the Advancement of Teaching has moved to a complex, multidimensional method of classifying universities (http://www.carnegiefoundation.org/classification), we elected not to sort the respondents’ schools into the now obsolete categories of “research I,” “research II,” and so on. Still, the prior literature and our own experiences led us to expect that involvement in graduate education and the size of the institution might matter in hiring decisions. Notably, university size was positively correlated with offering a master’s (.43) or doctoral (.47) degree. Even so, sensitivity analyses of our models revealed that university enrollment did not obscure any relationships between the dependent variable and whether the respondent worked in a department with a master’s or doctoral program.

6. US News and World Report published rating scores only for the 10 top programs (Butler 2005). It would have been informative to include all 35 programs as possible levels of University; cell sizes, however, would have been too small for reliable analyses.

7. We also sought to select journals that had been rated by at least 50 percent of Sorensen et al.’s sample, but this proved impossible for the journal of low prestige. Criminal Justice Studies was rated by 31.4 percent of their sample, which was more than any other general‐interest journal that scored at approximately the tenth percentile.

8. Because we include variables that describe the respondents’ departments and universities, it would be theoretically possible to estimate three‐level (applicant, respondent, department) or four‐level (applicant, respondent, department, university) models. However, we did not collect data in a way that would allow department or university to be treated as separate levels for estimation.

9. Recognizing that Gender may have appeared to be nonsignificant if female or minority respondents were influenced differently from male or nonminority respondents, we examined this possibility. Analyses not shown verified that respondent race and gender were not significant predictors of the slope for applicant gender.

10. To compute this differential, we set Enrollment to its mean and simplified the remainder of the equation by assuming that the respondent was from a nondoctoral granting department (Doctorate = 0), and that the applicant was White or Asian (African American = 0, Hispanic = 0) and had no prior professional experience (Practice = 0).