1. An Evaluation of Pipeline Interventions for Minority Scholars An Evaluation of Pipeline Interventions for Minority Scholars Roberta Spalter-Roth, Jean H. Shin, and Olga Mayorova American Sociological Association and Patricia White National Science Foundation November 2011 Photo credits: Illinois Springfield, St. Gallen SymposiumIllinois SpringfieldSt. Gallen Symposium

2. Research Questions In this presentation, we focus on evaluating the careers of American Sociological Association Minority Fellowship Program (MFP) Fellows, who receive their awards in graduate school. We investigate whether: MFP Fellows expand disciplinary knowledge in the area of race and ethnicity, and how this research impacts their careers; MFP Fellows achieve what are regarded as archetypal or “ideal” faculty careers at Research I universities or are more likely to have “alternative” careers; and whether Dissertation advisors as graduate school mentors have a direct impact on MFP Fellows’ careers. 1

3. The Three Comparison Groups We analyze careers of graduates from three groups who were awarded a PhD in sociology between the 1996-1997 and 2008-2009 academic years and were employed in the U.S. in 2010. These groups are: 1.The ASA Minority Fellowship Program (MFP) Fellows (n=108) 2.National Science Foundation (NSF) Dissertation Improvement Grant Awardees in Sociology (n=266) 3.A randomly selected group of sociology PhD graduates (n=158) 2

4. Why These Comparison Groups? ASA has been long overdue for a systematic evaluation of the Minority Fellowship Program (MFP), now nearly four decades old. Funding for the program is currently in transition. NSF desired a comprehensive evaluation of the Dissertation Improvement Grants program in Sociology for its Board of Visitors in 2010. Both funding programs are considered “interventions” but represent very different kinds of mentoring as well as the basic characteristics of the awardees. 3

5. Dataset and Unobtrusive Measures The study uses unobtrusive methods of data collection. Demographic, institutional, employment, and activities information from the ASA membership and NSF awards databases and on-line resources. Graduate school mentors (the chair or the first listed co- chair of the dissertation committee), and other dissertation information from the ProQuest database. Publication data from faculty on-line CVs and from Google Scholar and Google Book. 4

6. TABLE 1. DEMOGRAPHIC CHARACTERISTICS, EDUCATION, AND EMPLOYMENT OF 1997-2009 SOCIOLOGY PHD GRADUATES EMPLOYED IN THE U.S. IN 2010 (N=532) a Percentages were calculated based on 484 cases with 48 cases excluded because of missing information on race/ethnicity. b Group differences are statistically significant at 0.05 level based on Chi-square test. 5 Demographic Characteristics Control Group (%) MFP Fellows (%) NSF Awardees (%) Female62.061.159.8 Minority a,b 22.2100.015.6 Degree Information: PhD program at Research I University b 69.681.597.7 PhD received in 2003 - 2009 (i.e., seven or fewer years since degree) b 47.556.572.2 Dissertation Topic: Race/ethnicity b 25.269.418.1 Gender27.231.520.7 Health, healthcare issues b 20.339.87.1 2010 Employment: Tenured/tenure-track faculty position b :61.460.271.8 Research I Institution15.211.136.8 Research-intensive, doctoral, masters, baccalaureate, and associate institutions36.130.629.3 HBCU/HIS/MSI b 10.118.55.6 Non-faculty position:27.227.822.9 Administrative positions in academia1.9 2.3 Researchers (academic or other meeting)12.014.816.9 Non-academic positions13.311.13.8 Non-tenured instructor b 11.412.05.3 Number of Cases158108266

7. TABLE 2. MENTOR CHARACTERISTICS FOR 1997-2009 SOCIOLOGY PHD GRADUATES EMPLOYED IN THE U.S. IN 2010 (N=532) 6 Advisor’s Minority Status and Gender: Control Group (%) MFP Fellows (%) NSF Awardee s (%) Minority female3.811.11.5 Minority male5.113.93.4 White female31.627.831.6 White male48.738.955.6 Unknown race/ethnicity10.88.37.9 Number of Cases158108266

8. FIGURE 1. EXPECTED PROBABILITIES OF HOLDING A FACULTY POSITION AT A RESEARCH I INSTITUTION IN 2010 FOR 1997-2009 SOCIOLOGY PHD GRADUATES, BY GROUP (N=353) 7 Based on results from the logistic regression with robust standard errors. Control variables are set to sample means in these calculations and controls for missing data are set to zero. * Statistically significant differences from the control group (0.05 level, 2-tailed test). MFP Fellow, white MFP Fellow, mentor Control Group NSF awardee Expected Probability Group 0.48 0.15* 0.42 0.64*

9. FIGURE 2. COEFFICIENTS FROM THE REGRESSIONS ON THE TOTAL NUMBER OF POST-PHD PUBLICATIONS AND THE TOTAL NUMBER OF POST-PHD PUBLICATIONS IN TOP THREE SOCIOLOGY JOURNALS BY 2009 FOR 1997-2008 SOCIOLOGY PHD GRADUATES IN ACADEMIC POSITIONS IN 2010 (N=342) 8

10. FIGURE 3. EXPECTED PROBABILITY OF HAVING RECEIVED AT LEAST ONE POST-PHD NSF GRANT AWARD FOR 1997-2009 SOCIOLOGY PHD GRADUATES EMPLOYED IN ACADEMIC POSITIONS IN 2010 BY GROUP (N=353) 9 Based on results from the logistic regression with robust standard errors. Control variables are set to sample means in these calculations and controls for missing data are set to zero. * Statistically significant differences from the control group (0.05 level, 2-tailed test). Control GroupMFP fellowsNSF awardees Expected Probability Group 0.03 0.07 0.14*

11. FIGURE 4. EXPECTED PROBABILITIES OF HOLDING AN ASA SECTION'S OFFICER POSITION FOR 1997-2009 SOCIOLOGY PHD GRADUATES IN ACADEMIC POSITIONS BY GROUP 10 Control GroupMFP fellowsNSF awardees Expected Probability Group 0.07 0.18* 0.15* Based on results from the logistic regression with robust standard errors. Control variables are set to sample means in these calculations and controls for missing race/ethnicity were set to zero. * Statistically significant differences from the control group (0.05 level, 1-tailed test).

12. FIGURE 5. COEFFICIENTS FROM THE LOGISTIC REGRESSION ON GETTING TENURE BY 2010 FOR 1997-2002 SOCIOLOGY PHD GRADUATES IN ACADEMIC POSITIONS (N=140) 11

13. FIGURE 6. EXPECTED PROBABILITIES OF PURSUING ACADEMIC AND NON-ACADEMIC CAREERS FOR 1997-2009 SOCIOLOGY PHD GRADUATES EMPLOYED IN THE U.S. IN 2010 (N=532) 12 Based on results from the logistic regression with robust standard errors. Statistically significant controls for years since graduation, dissertation topic, total number of publications, and post-PhD grant awards were included in the model and were set to sample means in these calculations.

14. Conclusions MFP Fellows are significantly more likely to write dissertations on race and ethnicity. MFP Fellows are significantly more likely to have “alternative” careers, although they are less likely to do so than members of the control group. MFP Fellows who attended Research I universities are more likely to pursue archetypal or “ideal” careers. 13

15. Conclusions, cont’d For MFP Fellows, having a white male mentor in graduate school may help them on the trajectory to “ideal” careers Not attending a Research I institution for the PhD is related to a lower likelihood of publishing in “top” general sociology journals, receiving post-PhD NSF grants, and to a greater likelihood of careers at teaching-intensive schools. Having a white male mentor does not appear to have direct effects on other aspects of an archetypal career Yet, having such a mentor in graduate school is significantly related to gaining tenure on time. 14

16. Next Steps We hope to add several additional cohorts to the analysis, as well as a larger group of non-MFP minority sociologists in order to study intersectional analyses. Examine what Bonilla-Silva (2011) has called exclusion from the “web of group affiliations.” To do this, we will examine minority faculty’s positions in: Department networks; Co-authorship networks; and Professional activity networks (ASA, regional societies and sociology journals’ editorial boards) 15

17. Questions for Future Research What are other kinds of mentoring relations and how can we measure them, unobtrusively? What kinds of capital are most important for career success and how can we measure them? We have suggested three methods for measuring “webs of inclusion and exclusion”—co-authorship and co-presentation networks, department relationships, and discipline centrality. What other alternatives are there? What are some strategies for continuing to encourage work on topics of race and ethnicity? Is it important to valorize “alternative” careers, and, if so, how? 16