Colleges and universities claim to do many things. In addition to strengthening students’ cognitive skills, many institutions promise to transform students, socially and ethically. As a result, measures of institutional success are too frequently based on subjective assessments of students’ values, orientations, and social growth. For example, colleges and universities may claim success if students develop a greater commitment to diversity, participate in volunteer work, or merely express satisfaction with their college experience. Even when objective measures of outcomes are used, they frequently focus on graduation rates, student retention, employment statistics, and successful applications to graduate school and law school. Colleges that report impressive student outcomes often start with better students, such that institutional success is merely a reflection of the ability to recruit the best and brightest applicants. In order to measure student learning, we need measures that consider change in cognitive skill over time. Using the Collegiate Learning Assessment (CLA), Arum and Roksa (2011) make a compelling case that colleges and universities fail to produce significant “value added” growth in students’ critical thinking, complex reasoning, and writing skills.

One easy criticism of Arum and Roksa’s work is that the measures of student outcomes used in their study are fairly limited. For example, the researchers find that volunteer work, group study time, and extracurricular activities do not produce gains in CLA scores and, in fact, may actually hinder student learning. One could merely reject Arum and Roksa’s conceptualization of the university’s mission and argue that these types of activities and experiences produce other valuable outcomes. It is not difficult to find arguments that colleges and universities are responsible for promoting social change. One of our own institutions claims to “affirm the values of peace, non-violence, human dignity and social justice.” (See Elizabethtown College’s mission statement, available online at http://www.etown.edu/public.aspx?topic=Institutional+Mission+Statement) Perhaps the real value of higher education is in transforming students’ values and perspectives. Yet, a number of other recent studies suggest that college education produces little change in these types of outcomes as well.

In The Still Divided Academy (Rothman et al. 2011), we provide evidence that there is very little difference in students’ social and political values between the beginning and end of their college careers. For example, college seniors and their first-year counterparts have similar attitudes on government involvement in job creation and income distribution. The two groups are also remarkably similar in their attitudes toward homosexuality (p. 76). Similarly, in their study of students after 4 years at UCLA, Sidanius and colleagues (2008) find that exposure to diversity and multicultural programming produced little notable effect on students, especially in regard to “whites’ political and racial attitudes, which seemed to be quite crystallized at college entry and were quite stable across the college years” (p. 318). While this is encouraging for those who express concern about ideological indoctrination on college campuses, it raises serious doubts about any claim that, in the absence of cognitive learning, colleges are producing significant behavioral and attitudinal outcomes.

In short, we are convinced by Arum and Roksa’s argument that students are not learning as much as they should be. But why aren’t our students learning? Critics of higher education often look to explain institutional failings by pointing to the fact that faculty are engaged in research, rather than teaching. This is not a new criticism of higher education. In fact, the distraction of research is faulted whenever people believe that academics should be doing more of something else. In The Divided Academy (1975), Everett Carll Ladd, Jr. and Seymour Martin Lipset explain that faculty research was frequently faulted for the failure of some academics to advocate for social change:

The thrust of student activism in the 1960s contained a curious mixture of the external and the internal, of criticism of national policies and the role of the university. Militant student activism was associated with sharp criticisms of the research culture of the university. Many activists who put political reform as primary saw the commitment of scholars to research and publication as self-serving careerism, leading those so engaged away from political involvement and away from devoting time to undergraduate education. For the activist, the university is or should be a “house of politics”; for the dedicated scholar, regardless of ideology, it is basically a “house of study” (p. 210–211).

But Arum and Roksa are essentially arguing that our colleges and universities ought to be a “house of study.” Students don’t spend enough time reading, writing, and studying. They learn the most when professors set high standards. If Ladd and Lipset’s conclusions still hold, we might expect that research oriented faculty would be the most likely to promote the sort of rigorous academic standards that Arum and Roksa argue are necessary. Research faculty would value intellectualism and direct students toward scholarship, and away from extracurricular activities, volunteer work, and other distractions.

Yet, Arum and Roksa argue that a focus on research is partially to blame for poor student outcomes. Clearly, if no one is teaching courses, this will have a negative impact on student learning. Additionally, there is a legitimate argument to be made that faculty research and course release have monetary costs. Faculty who do little to no teaching may detract from the educational mission of the university. But these individuals are the exception. Even at research universities, the vast majority of faculty report an interest in the education of undergraduates, which was not the case 40 years ago (Schuster and Finklestein 2006, 488). Faculty at research institutions also report that they spend, on average, 50% of their time on teaching, and only 27% of their time on research (p. 469). The myth of the university that has turned against teaching appears to be overblown. But critics are correct when they state that other institutions, liberal arts and comprehensive colleges, have seen an increase in faculty research productivity. The assumption is that this has come at the cost of high quality teaching. However, there is no compelling evidence that faculty who are seriously engaged in research are less demanding or effective in the classroom than those who espouse a deeper commitment to teaching. Nor do Arum and Roksa offer any direct evidence that students learn less when institutions value and promote faculty scholarship. In fact, their findings may suggest the opposite.

When making the argument that research undermines teaching, Arum and Roksa rely on Alexander Astin’s, What Matters in College:

While some have argued, and indeed it is possible, that faculty research and teaching can be complimentary, the empirical evidence unfortunately suggests that this tends not to be the case on most of today’s campuses. In What Matters in College? Alexander Astin constructed two scales: one of the faculty’s research orientation (defined primarily in terms of publication rate, time spent on research, and personal commitment to research and scholarship) and one of the faculty’s student orientation (reflecting primarily the extent to which faculty believed that their colleagues were interested in and focused on student development). The two scales were strongly negatively correlated, and ironically, if not surprisingly, the faculty’s student orientation was negatively related to salary compensation. After examining a range of student outcomes from academic to affective, Astin concluded that “there is a significant institutional price to be paid, in terms of student development, for a very strong faculty emphasis on research” (pp. 7–8).

The authors are correct when they claim that Astin finds evidence of a negative relationship between student outcomes and faculty research orientation, which measures faculty members’ publication rates, time spent on research, and personal commitment to research. However, the outcome measures that Astin uses are not objective measures of student learning; rather, they are the same questionable measures of student outcomes that Arum and Roksa find to be unacceptable for assessment purposes. For example, Astin finds that faculty research orientation has negative effects on students’ “orientation of the faculty”. In other words, when faculty place an emphasis on their own research and scholarship, students report that their professors show less concern for them and are less accessible. It is logical to assume that this would have a negative effect on student learning. But Arum and Roksa find no relationship at all between students’ perceptions of faculty accessibility and performance on the CLA, nor do students who meet with faculty outside of class show greater growth in cognitive skills. Astin may be correct that research oriented faculty are perceived as less accessible, but Arum and Roksa convincingly demonstrate that these measures of faculty-student relationships are not related to students’ cognitive growth.

Astin also reports that faculty research has a negative effect on a number of other student outcomes, including involvement in various types of non-curricular activities. When professors are more research oriented, students are less likely to be elected to student office, attend a recital or concert, or tutor other students. They also report less growth in interpersonal skills. In other words, students appear to socialize less and interact with others less when their professors are focused on research. Yet, Arum and Roksa find that social engagement has a negative effect on student learning. Students show the most growth on the CLA when they study alone, rather than with peers. Similarly, student clubs and extracurricular activities appear to compete with, rather than enhance, academic work. Again, we find no evidence from Astin’s work that faculty research infringes on concrete measures of student learning. To the extent that faculty research infringes upon social activities, Arum and Roksa suggest that this may actually benefit students.

According to Astin, faculty research orientation also has a negative relationship with “social change orientation.” This is a measure of institutional priorities, as perceived by students, and includes “student involvement in community service activities” and other measures of commitment to community and social change. But if faculty research hinders “social change orientation,” this may also have a positive effect on CLA scores. Arum and Roksa find that volunteer work has a negative impact on student CLA scores. Similarly, Astin finds that volunteering has a negative effect on GRE Verbal scores.

Hence, we conclude that while Astin identifies a negative relationship between faculty research orientation and student outcomes, this conclusion is based on measures of student growth that Arum and Roksa clearly find problematic (participation in clubs, volunteer work, social growth, etc.). In fact, when Astin does look at more objective measures of cognitive and academic growth, he finds a positive effect for faculty research orientation. Even after controlling for the effect of students’ SAT scores, a proxy measure of their skill upon entering college, students from research oriented institutions have higher GRE composite scores than their peers. Astin also finds a negative relationship between GRE scores and working on group projects or doing volunteer work. Thus, when looking at objective measures of cognitive performance, Astin’s work is consistent with Arum and Roksa’s findings; volunteer work and group study/projects have the same effect on GRE scores as they do on CLA scores. Accordingly, we suspect that if one were to examine the effect of faculty research orientation on CLA scores, one would find a positive effect, similar to the effect it has on GRE performance.

Arum and Roksa find that students learn the most from professors who require substantial reading and writing and who have high expectations for students. An alternative theory to the “research as distraction” model, and one we believe is suggested by the evidence above, is that high expectations and rigorous course requirements are more common among faculty who are committed to research and scholarship. While it is true that time spent on research detracts from the time one has to read and grade papers, academics who engage in serious research and scholarship differ from their non-publishing counterparts in more than their time commitments. They may also have different perspectives on the mission of higher education and the definition of scholarship.

Professors who seriously engage in research are committed to the pursuit of knowledge, rather than merely the transfer of knowledge. This commitment to intellectual inquiry translates into an institutional culture that encourages student learning and academic rigor. Astin finds that research oriented faculty are, predictably, more likely to involve students in research. Research oriented faculty, by definition, also read and write a great deal. In fact, these are the very activities that would, theoretically, take time away from students. It is logical to assume, however, that professors who engage in scholarship expect the same of students. Academics are frequently accused of attempting to clone themselves. If faculty researchers produce student researchers, cloning may be useful.

Professors who define their success in terms of research and scholarship are also less likely to succumb to pressures to appease students. Our own research in The Still Divided Academy (Rothman et al. 2011) demonstrates that, compared to research institutions, baccalaureate colleges tend to place substantially more emphasis on student course evaluations for tenure and promotion (p. 60). One survey finds that many professors admit to lowering standards and expectations in order to earn better course evaluations (Ryan et al. 1980). We might expect that professors who have substantial research accomplishments could earn tenure and promotion without having to appease students, and would be free to apply more rigorous standards in their courses. As Arum and Roksa note, one solution to this problem is to include measures of rigor and course requirements when evaluating faculty teaching.

It is also worth asking whether all professors are equally capable of teaching the skills that the CLA measures. If regular reading and writing promotes critical thinking, complex reasoning, and writing skills among college students, perhaps it contributes to the same cognitive growth and skills among college professors. It is doubtful that this learning automatically subsides the minute one earns a Ph.D. It is reasonable, then, to hypothesize that faculty who publish the most may be the best critical thinkers. Yet, seldom are specific professors or courses designated for “critical thinking and complex reasoning”. While we require professional certification and advanced degrees to certify one’s competency in chemistry, mathematics, or any other discipline, we have no formal method for evaluating expertise in critical thinking and complex reasoning. Given the rigors of the peer review process, which serves, in part, to evaluate reasoning and critical thinking, a record of publishing may be the best proxy we have for assessing faculty expertise in this area.

Although we believe that research oriented professors tend to promote the sort of academic experience that Arum and Roksa find to be most effective, it may not be the research, in and of itself, that matters. Rather, commitment to research may be an outcome of an underlying orientation that matters. If one could locate people who place a high value on intellectualism and scholarship, yet have little desire to produce any of their own work, one might be able to form the ideal university. Indeed, we can think of some colleagues who fit this model. But overall, these things are probably correlated; those who value intellectual inquiry, value it both in their own lives and in the lives of their students. Ideally, these scholars would also find teaching to be engaging.

Critics of faculty research activity often make the assumption that hours spent on research are directly and inversely related to hours spent on teaching. However, this is an overly simplistic view of faculty time. At some institutions, one can pursue alternative paths to promotion and tenure. In addition to research-intensive and teaching-intensive faculty, there is also a class that Schuster and Finklestein (2006) refer to as “administrative-intensive” faculty. These individuals also spend significantly less time in the classroom than their teaching-intensive peers (p. 203.) Even for those who don’t fall into the “intensive” category, administrative duties can occupy the time that others devote to research. For example, although professors of education spend less time on research than social science and natural science faculty, they appear to spend more time participating in faculty governance (Schuster and Finkelstein 2006, p. 485).

Additionally, although faculty at research institutions do spend a larger percentage of their time on research, they also spend more time working overall (Schuster and Finkelstein 2006, p. 79), and less involvement in faculty governance (p.485). Thus, research time need not detract from teaching. In fact, one study of faculty time finds that, although research activities at liberal arts and comprehensive institutions has increased over time, so has the number of hours that faculty spend teaching and preparing for teaching (Milem et al. 2000). While the researchers do find that there has been a decrease in student advising and informal contact time, Arum and Roska’s findings suggest that this would not have a negative effect on learning. Additionally, student demand may be driving part of this change in advising. Both authors of this article frequently sit alone during scheduled office hours, responding to student emails and text messages. In sum, every hour spent on research is not an hour taken from teaching. Sometimes it is an hour taken from the Parking Appeals Committee, The Civic Engagement Committee, or, more often, from families and leisure time. We suspect that a trend towards “institutional assessment” activities also places new demands on faculty time that may actually interfere with teaching and research.

It is also worth noting that the disciplines that are most research focused tend to produce the most student gains on the CLA. Ladd and Lipset (1975) found that faculty in the social sciences and natural sciences tend to be more focused on research than their counterparts in education (p. 348). More recently, Schuster and Finkelstein (2006) confirmed that faculty work differs by discipline, with publishing rates being highest in the social sciences and natural sciences (p. 99). Yet Arum and Roska report the weakest cognitive gains among students enrolled in education and social work courses, and significantly stronger gains among students enrolled in the natural sciences and social sciences. Students appear to be learning the most in disciplines that employ faculty who are most dedicated to research. We think it is quite possible that similar relationships between faculty research and student learning could be observed between faculty within the same discipline, or between institutions. This is an important area for future study.

We think Arum and Roksa would agree that a commitment to teaching does not inherently translate to rigor. But it is possible that a “teaching” orientation may actually detract from rigor and learning. Rewards for teaching are seldom based on rigor. Instead, institutions promote teachers because they are popular with students, or spend time with students outside of class. Faculty members are praised when they engage students in volunteer work or service-learning. Colleges reward teachers who promote social activism, because academics have collectively embraced the notion that a college education is aimed at the “whole student,” and that social/moral development is as important as intellectual development. Thus, faculty who are committed to “teaching” tend to fall prey to trends in higher education that move away from rigor and more traditional methods. A quick review of the most recent program from the American Political Science Association’s Conference on Teaching and Learning demonstrates the point. “Tracks” at the 2011 conference emphasize “Civic Engagement,” “Experiential Learning,” “Simulations and Role Play,” “Study Abroad” and “Technology in the Classroom”

(See APSA’s program online at http://www.apsanet.org/content_31632.cfm).

Yet Arum and Roksa’s research may suggest that old fashioned methods are the best. Students should read a lot and write a lot. Community service, experiential learning, and civic engagement projects may actually detract from students’ cognitive development by shifting institutional focus away from coursework. Faculty who are more oriented toward research may be getting the formula correct, simply by being oblivious to untested, experimental innovations in teaching. At the very least, when professors are engaged in research and scholarship, they serve as an example to students and help to create a culture that values inquiry, learning, and scholarship—a house of study.