Do Graduate Student Teacher Training Courses Affect Placement Rates?

Transcription

1 Journal of Political Science Education, 10: , 2014 Copyright # 2014 Taylor & Francis Group, LLC ISSN: print= online DOI: / Do Graduate Student Teacher Training Courses Affect Placement Rates? JOHN ISHIYAMA CHRISTINE BALAREZO TOM MILES University of North Texas We investigate whether the existence of a required graduate course on Teaching in Political Science is related to overall job placement rates reported by graduate political science programs. We examine this in light of evidence from 73 public PhD-granting political science departments across the country. We find that the existence of such courses has a marginally positive effect on placement rates, particularly with regards to the placement rates of lesser ranked research departments at primarily teaching institutions. However, the most consistent predictor of placement rates across all models is the research productivity of the graduate department from which the students received their degree. Keywords PhD programs in political science, placement rates, teacher training Does requiring a teacher training course correlate with higher job placement rates generally, and more specifically, at primarily teaching institutions? In this article, we investigate whether the existence of a required graduate course on Teaching in Political Science is related to overall job placement rates. To be sure, several recent noteworthy studies have examined PhD placement rates (such as Fowler, Grofman, and Masuoka 2007; Masuoka, Grofman, and Feld 2007; Oprisko 2012; Oprisko, Dobbs, and DiGrazia 2013). Oprisko (2012) and Oprisko, Dobbs, and DiGrazia (2013), for instance, have suggested that the key to placement is affiliation with the highest profile universities (particularly the Ivy League schools). Others have, in part, examined via regression analysis some of the factors that affect placement as well (Ishiyama et al. 2012). However, these studies have either examined only placement rates at other PhD-granting institutions as opposed to placements at all institutions, including non-phd-granting programs or have focused on only a limited number of variables in explaining placement (as was the case with the Ishiyama et al. 2012). Given that the largest proportion of political scientists are employed at non-phd-granting departments, which often very much value teaching skills and look for these skills in recruiting faculty, one might hypothesize that offering a required course on teaching in a graduate political science program would Address correspondence to John Ishiyama, Department of Political Science, University of North Texas, 1155 Union Circle #305340, Denton, TX John.Ishiyama@ unt.edu 273

2 274 J. Ishiyama et al. increase the overall probability of placement of its PhD graduates, particularly at teaching institutions. Or does research productivity better correlate with placement rates in all cases? This article seeks to examine these questions in light of an original dataset collected by the authors. Why would we expect that teaching courses offered at the graduate level have an impact on departmental placement rates? First, offering a teaching course may signal that the graduate department is committed to teaching and, hence, may reflect more intentional training of teaching skills to their graduate students (a marketable skill that teaching institutions presumably value). Further, the presence of such courses signals to potential employers the teaching ability of the program s graduates, hopefully making them more employable. Although we acknowledge that the content of teacher training courses varies widely (as we pointed out in an earlier review of graduate-level teacher training courses; see Ishiyama, Miles, and Balarezo 2010), here we focus on whether the presence of such a course is related to the overall job placement of program graduates. The Role of Teaching Preparation in the Discipline There have been many calls over the past decade or so to improve the quality of graduate education in the social sciences (see Ishiyama et al. 2012). Indeed, many studies have indicated that, in the social sciences, graduate programs have generally failed to adequately prepare doctoral students for the changing character of higher education, especially in terms of training for careers at institutions that emphasize undergraduate teaching. Notably, there is considerable evidence that most available faculty positions are located at institutions that focus primarily on teaching. It has been generally estimated that only 26% 35% of faculty positions in political science are located at doctoral-granting departments (Lopez 2003; Nerad and Cerny 2003). 1 Thus, only about one third of doctoral graduates in political science can expect to become faculty members at research universities similar to their graduate institution, where research and publications are the predominant requirements for earning tenure. The remaining positions are located at other types of institutions, where teaching, professional, and community service roles are of equal or greater importance than research and publication. Further, most graduate programs generally do not adequately prepare graduate students for the realities of faculty life at the institutions where most will begin their careers (Gaff et al. 2003). As Gaff and colleagues note: [B]etter preparation for academic careers includes understanding the missions, faculty roles and rewards, and academic culture of the various institutions. Preparation should also allow students to experience the full range of roles that faculty play in these institutions and to develop the skills that will allow them to compete for and succeed in faculty positions (2003, 2). In particular, as they note, graduate programs in political science do not adequately prepare students for careers as teachers. Indeed, although many graduate students have an opportunity to teach sometime during their experience as doctoral students, these are often not structured experiences that prepare them to deal with issues such as assessment, different types of student learning, the pedagogy of the discipline, curricular innovations, the impact of technology on education, or the variety of teaching styles that may be helpful with students from different racial, ethnic, or cultural backgrounds (Gaff et al. 2003, 3).

3 Graduate Student Teacher Training Courses 275 In political science, an American Political Science Association (APSA) Task Force Report on Graduate Education (2004) argued that with the growing number of employment opportunities for graduates of political science PhD programs at primarily undergraduate institutions (PUIs) and community colleges, many departments should not simply prepare students to be political scientists but also to be teachers of political science. This does not mean the abandonment of scholarship, but the creation of teacher-scholars, or those who are trained to conduct independent and innovative scholarly work but who also have the skills by which to effectively impart knowledge to undergraduate students. Other efforts to improve teacher training and professional preparation for careers at primarily undergraduate institutions included the past Preparing Future Faculty (PFF) programs (PFF 4 included the APSA and four PhD-granting institutions in political science Indiana University, University of Colorado-Boulder, Howard University, and the University of Illinois- Chicago). Some political science departments have already fashioned their programs to explicitly prepare students for careers as teacher-scholars. The political science department of Miami University (Ohio), for example, established a College Professor Training Program for its doctoral students, which includes a mentor program to train students for independent teaching and course work on teaching political science to undergraduates (unfortunately, the Miami University program is now defunct). Similarly Baylor University inaugurated a PhD program in political science (in 2005) that is designed specifically to train teacher-scholars for the new market (for a description, see Ishiyama, Miles, and Balarezo 2010). However, there has been only limited systematic literature that has examined the extent to which teacher training programs actually enhance the likelihood of academic employment for PhD recipients in political science (with the single exception being the Ishiyama et al. [2012] book chapter). However, in this article, we expand the scope of analysis beyond what appeared in that book chapter and test the proposition that teacher-training programs that are available at PhD programs that are not the most highly ranked research productive departments would improve the likelihood that graduates from these departments would find employment. Data To begin, we used the data from Ishiyama, Miles, and Balarezo (2010) that were collected on 122 political science PhD programs in the United States, as identified by the APSA (APSA 2012). Of these departments, 41 offered a graduate-level course on teaching political science. Of these 41 departments, 28 required the course be taken by at least some of their graduate students. For instance, many departments required that the teaching course be taken if the graduate student was slotted to teach an independent section of a larger lecture course or an independent course itself. However, students that were not slotted to teach an independent section or course were not required to take a course on teaching. In 13 cases, the course was optional or listed as an elective. The largest majority (38 of 41) were located at public universities as opposed to private ones. 2 Given this disparity, in this study, we focus primarily on the public institutions (N ¼ 73). In terms of the dependent variable, we construct a measure of placement rates based upon two sources, derived from the measure developed by Schmidt and Chingos (2007). The first was obtained from a list of current PhDs employed at

4 276 J. Ishiyama et al. academic institutions in the United States from the APSA. These data have been widely used to impute the performance of graduate departments in terms of graduate student placement and has been used to calculate rankings for PhD-granting departments (see, for instance, Schmidt and Chingos 2007). The database recorded not only the current institution at which the individual was employed but also where the individual received their degree and in what year, as well as the gender of the individual. We restrict our data to faculty who were awarded a PhD between 1999 and 2007 (Schmidt and Chingos limited their study to 2004 to avoid statistical noise ), because this is the last year in which the Integrated Postsecondary Education Data System (IPEDS) reports PhDs awarded by university. Also, only faculty holding tenure-track professorial positions are included; lecturers, adjunct professors, postdoctoral fellows, and those in other positions are excluded from the analysis. From this information, we focused only on faculty who received their PhDs after 1999 (for a total of 1716 faculty) because most teacher training programs were adopted fairly recently (most were adopted sometime in the 1990s), so we wanted to examine those who were most likely to have benefited from such programs. We identified where individuals received their PhDs and the type of institution at which they were currently employed (either a PhD-granting department or otherwise). We then derived a total of employed PhDs by department that granted their degree, which became the numerator for the dependent variable. We calculate the denominator (the number of PhDs the departments produced during the same time period, ) based on data from IPEDS from the National Center for Education Statistics (NCES) (NCES 2012). IPEDS reports PhDs by year from 1999 to 2007 for each university in the country. We collected the number of PhDs produced for each of our 73 cases in political science and international relations (as long as there was no separate international relations department at that university). 3 IPEDS reported that after 1999, five thousand PhDs were granted by these public departments. Although there are potential problems with our measure of placement, we agree with Schmidt and Chingos that ideally information collected would track the school at which graduates are placed three to five years after they leave graduate school (any earlier would not give careers sufficient time to stabilize, but any later might allow faculty to drift from the career path they established in graduate school) but such data are not readily available (2007, 526). Further, we are also cognizant of the potential shortcomings of these data as it pertains to this particular project, which examines placement at all institutions (this was not necessarily a problem for Schmidt and Chingos, who focused on placement at research-intensive departments). The database, for instance, does not report placement in nonacademic jobs (such as in government, research think tanks, and the like). This does not present a problem for our analysis, since we are primarily interested in whether requiring a teaching course helps place PhDs in teaching positions (i.e., as university or college faculty). Two other limitations, however, are more problematic. The APSA database is based almost entirely on APSA members. This likely does not cover a large number of political science faculty who have allowed their memberships to lapse or have never been members of the association (this would also be a problem for Schmidt and Chingos study as well). This would be especially true for many scholars in international relations and comparative politics (who are often members of other

5 Graduate Student Teacher Training Courses 277 learned societies such as the International Studies Association or a variety of area studies organizations rather than members of APSA). Second, faculty members at primarily undergraduate teaching institutions (particularly two-year institutions) are likely to be very underrepresented in the APSA database. Thus, there is a source of potential systematic bias in the database, in that the data exclude departments that specialized in producing international relations or comparative politics scholars, or departments that generally produced PhDs who teach at primarily undergraduate institutions. Nonetheless, we agree with Schmidt and Chingos (2007) when they justify the use of these data that the database provides the advantage of accessibility and convenience, as well as being validated by the leading professional learned society in the field. Further, it is the only database currently available to conduct such analyses. Although admittedly flawed, it provides a very good starting point to calculate placement rates. Thus, our first dependent variable is the ratio of PhDs employed over PhDs produced for each of the 73 public graduate departments in our study. 4 Our second dependent variable is the number of PhDs employed at teaching institutions (i.e., non-phd-granting departments) over the total number of PhDs produced by that school. We use this measure because it is possible that the presence of teacher training courses provides advantages for faculty seeking jobs at primarily teaching departments. We also employ several control variables, more than that used in the Ishiyama et al. (2012) study. For the first variable, research productivity, our measure is the international ranking of political science departments as reported by Hix (2004). The Hix ranking of political science programs assessed 1,255 programs (U.S. programs and across the world) and ranked the 400 best based on faculty productivity in the top 63 political science journals in the world (making it the most comprehensive ranking based on research productivity in existence). We used the Hix index as opposed to other commonly used rankings such as those that use peer evaluations, for example, the U.S. National Research Council and the U.S. News and World Report, because of the many cited problems with such rankings and because they are often based on other characteristics beyond research productivity. 5 Further, we use the Hix index rather than other rankings based on content analysis of leading political science journals (Ballard and Mitchell 1998; Garand and Graddy 1999; McCormick and Rice 2001; Miller, Tien, and Peebler 1996; Welch and Hibbing 1983). Most indices only include the top institutions, relegating the others to a common lower ranking. Only the Hix index examined all political science programs (including primarily undergraduate programs) and used 63 international political science journals to measure research productivity. It is the most complete ranking of research productivity currently in existence. The second control variable measured the size of the program, where we listed total number of full-time equivalent (FTE) faculty reported by the department. 6 In addition to this, we also used as a control variable the average number of PhDs produced by the department over the 10-year period as an additional control for the size of the program. Finally, there is some suggestion that there is a gender component to placement as well. As the APSA report Women s Advancement in Political Science (APSA 2005) suggests, there are gender differences in placement and hiring, as well as career patterns, with women less likely to land tenure-track jobs and more likely to find employment at primarily teaching institutions.

6 278 J. Ishiyama et al. In terms of some summary statistics, this study includes 73 public PhD-granting political science departments, of which 28 required that at least some of the graduate students were required to take a teaching course. Of these, 16 were listed in the top 40 political science programs by the Hix index. Analysis To examine whether the existence of a required teacher preparation course enhanced the employability of PhD graduates from institutions that required such courses, in Table 1 we regressed the dependent variable, the overall placement rate, against our primary independent variable and our two controls, using all 73 cases. As indicated, the best predictor of placement was the ranking on the Hix index. The more highly ranked a department in terms of research productivity, the larger the proportion of PhD graduates who were in faculty positions as reported by the APSA. Whether the graduate program required some level of teacher preparation for its students was slightly positively related to the proportion of PhD graduates after 1990 who were in faculty positions, but the relationship was not statistically significant. The size of the graduate department as measured by FTE was generally unrelated to the placement rate. However the departments that produced more PhDs were less likely to place their students in employment at a faculty position (although that is likely due to the fact that it is more difficult to place 100% of graduates when the number of graduates is larger). In terms of gender balance, there was no relationship between the proportion of graduates who were women, PhDs produced by a department, and overall placement rate. Finally, as indicated by the Variance Inflation Factor (VIF; a commonly used measure for multicollinearity) scores, the independent variables were not collinear with one another (with VIF scores less than 4). Table 2 examines a subset of the data, those programs that were not ranked in the top 40 of the Hix index (i.e., departments that were not the most highly ranked research-productive departments in the country). These are presumably departments Table 1. Coefficient estimates and collinearity statistics for dependent variable (PhDs placed over total number of PhDs granted ) Variable Coefficient (Standard Error) VIF Was course on teaching required in department from which the faculty member received their degree? (.055) Ranking on Hix index (.000) Size of graduate department in FTE (002) Size of department in terms of number of PhDs produced (.001) Gender balance (.289) Note: Adj R 2 ¼.299. p.10. p.05. p.01. p.001.

7 Graduate Student Teacher Training Courses 279 Table 2. Coefficient estimates and collinearity statistics for dependent variable (PhDs placed over total number of PhDs granted for PhD programs not in the top 40 on the Hix Index) Variable Coefficient (Standard Error) VIF Was course on teaching required in department from which the faculty member received their degree? (.066) Ranking on Hix index (.001) Size of graduate department in FTE (.004) Size of department in terms of number of PhDs produced (.003) Gender Balance (.339) Note: Adj R 2 ¼.248. p.10. p.05. p.01. p.001. that may seek to produce graduates who are prepared to teach. This left 57 public PhD-granting departments. Table 2 reports the results of regressing the overall dependent variable, placement rates, against the five independent variables in our model. Again, as with the results reported in Table 1, even when considering PhD programs that are not in the top 40 on the Hix index, the more highly ranked a department in terms of research productivity, the larger the proportion of PhD graduates who were in faculty positions (as reported by the APSA). Thus, it appears that even for less research-intensive graduate departments, the more research intensive the degree-granting department is, the greater the likelihood that the department places its graduates in faculty positions, even when controlling for the size of the department (as measured both in terms of the number of FTE faculty and number of PhDs produced). As with the results generally, whether the graduate program required some level of teacher preparation for its students was unrelated to our measure of placement rates. The size of the graduate department as measured by FTE and number of PhDs produced were also generally unrelated to the placement rate. Thus, larger departments were not more likely to place graduate students than smaller departments. Further, departments that were not in the top 40 did not produce relatively more women candidates and were not more likely to place PhDs in employment (as was the case overall as reported in Table 1). Table 3 regresses a slightly different measure of placement the number of PhD graduates placed at non-phd-granting departments over the total number of PhDs produced. In Table 3, we examine whether requiring a teacher preparation course enhances the ability of less research-intensive departments to place their graduates at primarily undergraduate teaching institutions. As indicated in the results reported in Table 3, the more research intensive the department is, the greater the likelihood that the department places its graduates in faculty positions at teaching institutions. In addition, the results indicate that

8 280 J. Ishiyama et al. Table 3. Coefficient estimates and collinearity statistics for the dependent variable, PhDs placed over total number of PhD degrees granted (placement at non-phd-granting departments from PhD programs not in top 40 on Hix Index) Variable Coefficient (Standard Error) VIF Was course on teaching required in department from which the faculty member received their degree? (.059) Ranking on Hix Index (.000) Size of graduate department in FTE (.004) Size of department in terms of number of PhDs produced (.002) Gender Balance (.305) Note: Adj R 2 ¼.066. p.10. p.05. p.01. p.001. the presence of a teaching course at a lesser ranked PhD program does not appear to raise the likelihood of placement at a primarily teaching institution. This is rather surprising given that it is widely believed that departments that have teacher training programs may better prepare their graduates for a teaching position. However, again it appears that affiliation with a more productive research department affects placement, even at primarily teaching jobs. The other control variables are not statistically significant, so the size of the department and the proportion of PhDs produced who were women were also unrelated to job placements, even at primarily teaching institutions. Conclusion In this article, we examined the question as to whether having a required teacher training course enhances the PhD placement rates of political science graduate departments. In general, the existence of such courses had a marginally positive effect on placement rates for lesser ranked PhD departments at primarily undergraduate teaching institutions. However, what does seem to affect placement rates across all models is the extent of the research productivity of the graduate department. This may be because such departments afford their graduate students the opportunity to engage in research themselves (and more importantly to get published, which enhances the graduate students marketability) or that even primarily teaching institutions base their hiring decisions on the research reputation of the faculty candidate s department or the research record of that candidate. In other words, research productivity still remains the best predictor of job placement. The fact that the research productivity of a department best explains placement rates at all levels (including at non-phd-granting departments) may be also a function of other features of highly ranked departments. For instance, it is quite possible that, since research productivity is closely associated with activity and visibility in the discipline, higher levels of faculty productivity facilitates graduate student

9 Graduate Student Teacher Training Courses 281 networking and enhances the prospects of applicants to secure interviews. Higher levels of research visibility might render recommendations more valuable to committees, further improving candidates prospects. Although we are not in a position to directly test these propositions, they do potentially explain why research productivity is still the best explanation for placement. This conclusion should be tempered, however, by three caveats. First, given the shortcomings of our measure of placement (discussed above), one should be cautious as to interpreting the results, which should be seen only as preliminary. Only with better data (perhaps after a labor-intensive search of all faculty from all political science programs in the country) can we get a more accurate picture of placement. Second, it should be noted that there are clearly ways in which college teacher training can be incorporated into the graduate curriculum beyond simply offering a course. For instance, the PhD programs at the Baylor University program and the now defunct Miami University (Ohio) program, offer(ed) alternative cost effective ways to achieve such integration, although using very different models one based on the apprentice system and the other using more traditional resources to supplement their program as often found at the faculty development or teaching and learning centers found at most universities (for a discussion of such programs, see Ishiyama, Miles, and Balarezo 2010). Neither offer an explicit graduate teacher training course but use innovative techniques to provide better teacher training opportunities for their graduate students. Further research must take such alternative measures into account when examining the impact of graduate teacher training on placement rates. Third, although requiring a teacher training course may not enhance hireability, it should be remembered that taking a course on teaching is not the same as having teaching experience. Indeed, in our experience, teaching experience is a much better indicator of preparedness for teaching than is one s course work and is likely to impact more on a hiring decision than what appears on a transcript. However, our data cannot currently capture whether or not a department offers teaching experience as part of their graduate training. Another direction for future research is related to the probability of receiving tenure and promotion. In other words, are graduates of programs with teacher training more likely to be successful in the tenure review process at teaching-intensive institutions? The answers to these questions will have to wait until the next iteration of this project. Notes 1. The lower figure is presented by Nerad and Cerny (2003), who report positions at Carnegie classification Research 1 institutions; Lopez (2003) reports a figure of 35% in doctoral granting departments that may include departments in universities that are not classified as Research 1 institutions. 2. In our previous study, we coded the department as a public institution if it was state supported in some way or an exclusively private institution. This included institutions such as the University of Pittsburgh, as well as Miami University, which, although independently funded an in many ways much like a private institution, is supported in part by state funding and hence is classified in this study as a public institution. 3. The exceptions were University of Southern California, Claremont Graduate School, and University of Denver, where the departments generally produce PhDs who find employment in political science departments. 4. We have deliberately omitted naming the institutions so as to not unduly embarrass certain departments. All results are thus reported in the aggregate.

10 282 J. Ishiyama et al. 5. There have, of course, been several noteworthy problems with this approach. In particular, such assessments are largely subjective. The biases of this approach have been investigated and it has been argued that, because the sample of academic judges has only very limited information about the output of departments, they are forced to base their judgments on other reputations, which favors already established programs at the expense of up-andcoming ones (Katz and Eagles 1996). Further, the reputation of the department may be confounded by the reputation of the institution as a whole, which some researchers have referred to as the halo effect (Lowry and Silver 1996; Jackman and Siverson 1996). Second, the peer assessments are quite costly to conduct and, hence, are updated infrequently. 6. If this was not explicitly reported, we counted the number of Full, Associate, and Assistant Professors in a department (we excluded any faculty that were listed as visiting or adjunct or temporary or lecturer from the count). References American Political Science Association (APSA) Women s Advancement in Political Science. Washington, DC: American Political Science Association. < org/imgtest/womeninpoliticalscience.pdf> (April 2012). American Political Science Association Institutions that Grant Ph.D.s in Political Science. Washington, DC: American Political Science Association. < org/content_6947.cfm> (July 2012). American Political Science Association Task Force on Graduate Education Report to the Council. PS: Political Science & Politics 38(1): Ballard, Michael J. and Neil J. Mitchell The Good, the Better, and the Best in Political Science. PS: Political Science & Politics 31(4): Fowler, James H., Bernard Grofman, and Natalie Masuoka Social Networks in Political Science: Hiring and Placement of PhDs. PS: Political Science & Politics 40(4): Gaff, Jerry G., Anne S. Pruitt-Logan, Leslie B. Sims, and Daniel D. Denecke Preparing Future Faculty in the Social Sciences and Humanities. Washington, DC: Council of Graduate Schools and Association of American Colleges and Universities. Garand, James C. and Kristy L. Graddy Ranking Political Science Departments: Do Publications Matter? PS: Political Science & Politics 32(1): Hix, Simon A Global Ranking of Political Science Departments. Political Studies Review 2(2): Ishiyama, John, Alexandra Cole, Angela Nichols, Kerstin Hamann, and Kimberly Mealy Graduate Student Teacher Training Courses, Job Placement, and Teaching Awards in the United States. In Improving Teaching in Higher Education: The Challenges of Creating Effective Instructional Development Programs, eds. Eszter Simon and Gabriela Pleschová. London: Routledge, Ishiyama, John, Thomas Miles, and Christine Balarezo Training the Next Generation of Teaching Professors: A Comparative Study of PhD Programs in Political Science. PS: Political Science & Politics. 43(3): Jackman, Robert W. and Randolph M. Siverson Rating the Rating: An Analysis of the National Research Council s Appraisal of Political Science Ph.D. Programs. PS: Political Science & Politics 29(2): Katz, R. and M. Eagles Ranking Political Science Departments: A View from the Lower Half. PS: Political Science & Politics 29(2): Lopez, Linda Placement Report: Political Science PhDs and ABDs on the Job Market in PS: Political Science & Politics 36(4): Lowry, Robert C. and Brian D. Silver A Rising Tide Lifts All Boats: Political Science Department Reputation and Reputation of the University. PS: Political Science & Politics 29(2): Masuoka, Natalie, Bernard Grofman, and Scott Feld Ranking Departments: A Comparison of Alternative Approaches PS: Political Science & Politics 40(3):

11 Graduate Student Teacher Training Courses 283 McCormick, James M. and Tom W. Rice Graduate Training and Research Productivity in the 1990s: A Look at Who Publishes. PS: Political Science & Politics 34(3): Miller, Arthur H., Charles Tien, and Andrew A. Peebler Department Rankings: An Alternative Approach. PS: Political Science & Politics 29(4): National Center for Education Statistics Integrated Postsecondary Education Data System (IPEDS). Washington, DC: National Center for Education Statistics. < nces.ed.gov/ipeds/> (July 2012). Nerad, Maresi and Joseph Cerny Career Outcomes of Political Science PhD Recipients: Results from the PhDs Ten Years Later Study. Seattle, WA: Center for Research & Innovation in Graduate Education. Oprisko, Robert L Superpowers: The American Academic Elite. In, The Georgetown Public Policy Review. < (August 2013). Oprisko, Robert L., Kirstie L. Dobbs, and Joseph DiGrazia Pushing Up Ivies: Institutional Prestige and the Academic Caste System. Georgetown Public Policy Review. < (August 2013). Schmidt, Benjamin and Matthew Chingos Ranking Doctoral Programs by Placement: A New Method. PS: Political Science & Politics 40(3): Welch, Susan and John R. Hibbing What Do the New Ratings of Political Science Departments Measure? PS: Political Science & Politics 16(3):