I have been reading Adam Grant’s Hidden Potential, and Grant referred to a research study about teacher effectiveness in introductory college courses that sent me off on a deep dive into what type of college teacher is most effective. Grant’s topic was really about whether one should connect with the most expert person possible as a mentor, but his example and conclusions were based on a study focused on higher education, which raised an issue and possible stereotype I am sensitive to as a retired professor. In a section titled “What those who can do can’t teach”, Grant highlighted a research study that suggested contingent faculty (full-time, nontenure track faculty) are more effective with beginning students than tenure-track faculty. Right away, I worry that his argument suggests researchers don’t care or can’t teach. This was not really his intent, but just his way of framing the issue (the section title) was sufficient to motivate my investigation.
Grant’s position on expertise and guidance was based on a research study I will describe in greater detail after I explain his conclusion. The study (Figlio and colleagues, 2015) showed that students in Introduction to Economics performed better with full-time teachers than with tenure-track teacher/researchers, as demonstrated by higher performance in the subsequent related course. Grant explained that greater expertise results in greater difficulty communicating about the basics. Experts sometimes lack insight into their own skills, which makes it challenging to communicate effectively. They may ignore important information they assume others have or no longer recognize as present in their own thinking. Their explanations jump around with side examples that lack connections novices need to understand and put the ideas they are receiving together. I admit that some of this hit a nerve, as I sometimes received criticism in my own “intro course” teaching for spending time telling stories that some students thought were off target. To me, the stories were concrete examples I used to illustrate the concepts and principles I presented in practice. I was trying to demonstrate relevance.
So, off I went to read the original study responsible for this argument. Figlio and colleagues, who are economists, studied the students of contingent and tenure-track instructors enrolled in the first economics course. Their dependent variable, which I found strange, was the performance of these students in the following relevant course. This was argued to reflect the “value added” of the quality of instruction in the preceding course. My explanation for this approach was that this was archival data. Without equivalent performance instruments to evaluate performance in the initial course, the grade in the next course was a better dependent variable. These researchers didn’t speculate about their results in the same way as Grant. It is easiest to explain their speculation without reference to their statistical data. They produced a figure charting the distribution of the value-added variable for the contingent and tenure-track faculty and the distributions were nearly identical over the top 75% of both groups, but were different with fewer lower-performing contingent faculty with smaller value-added scores. These authors speculated that over time, more contingent faculty members who were less effective were rehired. I don’t know whether this explains why they were less effective so perhaps both sources are compatible.
How might administrators respond to these findings? Should they at this point? I guess my thought is that much more work is needed before decisions are made about staffing. First, just which institutions should be informed by these findings? The study was performed in one department of one institution. The institution, Northwestern, has a mixed mission including both research and instruction. The issue of tenure is likely heavily influenced by research and grant activity at this institution, but tenure as a practice is also what happens at many institutions with a primary instructional mission. The tenure vs. ongoing reappointment difference is a different issue.
Second, the economic focus on estimating value, as demonstrated by future benefit, is interesting but I think unusual when it comes to evaluating instructional skill. Most educational research focused on the impact of a practice examines achievement in the course where the manipulation or observed assumed causal variable is studied. The gold standard for such research (e.g. see Stroebe, 2020) would involve multiple sections of the same course, using a common method of evaluation, taught by different instructors, with students assigned to a section at random. This methodology is difficult to arrange, but it has been used often in addressing another critical question for administrators – whether student ratings predict instructional quality (the answer is no). When it comes to student achievement and whether instruction is provided by tenure track or contingent faculty using this method, I found one example (Carrell and West, 2010) that demonstrated better immediate achievement, but lower future performance associated with contingent faculty. So, there is more to be learned from additional studies.
Personal comment: I commonly taught the Introduction to Psychology course and I have considered the Grant argument about expertise as it applies to my experience. Introduction to Psychology is a survey course offered to both majors and nonmajors. Suppose you want to understand the content of such courses. In that case, you might imagine a typical textbook that contains chapters on such subjects as Research Methods in Psychology, Biological Bases of Behaviors, Learning, Memory, Sensation and Perception, Developmental Psychology, Social Psychology, Abnormal Behavior, Therapies and Treatments. These chapter titles could also be used for full courses available to students later in their studies. My expertise and the advanced courses I taught were in Educational Psychology, so Learning, Memory, Developmental, and Motivation were areas in which I propose I had some expertise. Despite the typical reaction when I am introduced to others as a psychologist, that I must be a clinician treating those with mental disorders, I did not study these areas and knew far less about them than graduate students with that focus in our own department. What would Grant’s argument suggest about my instruction in various topics covered in Intro, given the great variability in my actual expertise? I would suggest that pedagogical skills and content knowledge are both critical, yet somewhat independent, in impacting student learning. I have no data to evaluate my own contributions, but I would bet I provided a superior experience in explaining human learning compared to human pathology.
These are interesting issues with real consequences. Some suggest that administrators seriously consider hiring for two categories of faculty members in multi-mission institutions. I think there are related problems with this approach such as prestige and whether those focused exclusively on teachers are actually prepared for this role.
Why do I present this information? I suppose I assume some are just interested. I also propose that public perceptions of higher education are often poorly informed and I hope that indicating those involved are committed to exploring multiple productivity issues will help.
References:
Carrell, S. and West, J.(2010). Does professor quality matter? Evidence from random assignment of students to professors.” Journal of Political Economy. 118(3), 409-432.
Figlio,D., Schapiro, M. and Soter, K. (2015). Are Tenure Track Professors Better Teachers? The Review of Economics and Statistics, 97(4), 715–724.
Stroebe, W. (2020). Student evaluations of teaching encourages poor teaching and contributes to grade inflation: A theoretical and empirical analysis. Basic and applied social psychology, 42(4), 276-294.
