The public's response to NCTQ's recent report, Training Our Future Teachers: Easy A’s and What’s Behind Them, was generally positive. Clearly, many teacher prep graduates (or people who know graduates) agree with our central finding: teacher prep is not nearly rigorous enough to prepare candidates for the challenge of teaching. However, a commentary by Dean Donald Heller of Michigan State University published in the Chronicle of Higher Education, lists several points of criticism about the report and the ratings of institutions that accompanied it. We address each point below.
- Our analysis uses commencement brochures, which
often identify graduating students as having earned honors based on meeting or
exceeding a grade point average (GPA). Given that our data source offers this
proxy for GPAs (see page 2 of the report),
Dean Heller dislikes our use of the term “GPA differential” to define the
difference between the proportion of teacher candidates earning honors and that
of all undergraduate students earning honors.
The question is whether using honors as a proxy for GPA rather than GPA itself makes a difference. Heller contends that it does, noting that all we know is whether a group of students fell below a certain GPA cut-line, not how far below they fell. He suggests that the difference could be as small as a hundredth of a point. But given that our analysis generally aggregates hundreds or even thousands of students, this is implausible. Student GPAs are distributed across a range. It's virtually impossible that education school students would cluster just above the honors cut line, while most students in an institution cluster just below it. Moreover, while we grouped all levels of honors (generally Latin honors) together, we know that many students earned higher levels of honors (e.g., summa cum laude) which require much higher GPAs (often a 3.9 or higher). In short, our proxy measure is indeed tracking a meaningful underlying GPA differential between education students and all students.
Given the structure of many education programs, secondary teacher candidates often have content majors outside the department of education. When we rate institutions based on the difference in the proportion of honors among teacher candidates compared with all undergraduates, in some cases it is clear that we cannot identify or include those secondary candidates who are housed outside the department of education. To address this limitation, we use a more generous scoring rubric to differentiate between meeting and failing the standard for those institutions with less detailed commencement brochures. Dean Heller claims that our approach for dealing with these less detailed commencement brochures “fudges the issue.” To the contrary, we tested these two approaches with 50 institutions (not 29, as Dean Heller stated) to see if the institutions with less detailed information were at a disadvantage. In fact, the vast majority of ratings were the same or better, and only four institutions had lower scores when rated on less detailed information (whereas Dean Heller incorrectly stated that the rating was different – and implicitly, worse – in six cases). (For more information, see page 3 of the methodology)
Among many possible explanations that would account for their higher grades at graduation, we considered the prospect that teacher candidates are academically stronger than their non-candidate peers at the same institution. If this were the case, those teacher candidates should also earn higher grades in the general education coursework they, like all students on campus, take in the first few years of college. Our report considered four studies dating back to the 1980s that looked at whether future teacher candidates earned higher grades than other students. The findings were mixed, but even those that found that future teacher candidates earned higher grades found no more than a tenth of a letter grade point difference. A more recent and larger data source, a National Center for Education Statistics survey of over 16,000 students, found that teacher candidates’ grades in their first year of college are roughly equivalent to those of their peers. In short, those studies suggest that the GPA differential only appears after teacher candidates enter preparation programs, indicating that the cause lies in the programs themselves. Given this research, it’s unclear why Dean Heller states that we have not disproven that teacher candidates are academically stronger than their peers.
The central thesis of our report focuses on the evidence that education professors are systematically assigning more of a different kind of work – criterion-deficient assignments. Our extensive statistical analysis of course assignments and course grades, an analysis completely unmentioned by Heller, shows that these overly broad or subjective assignments are associated with higher grades – and that they are twice as common in teacher preparation as in other academic areas.
Rather than address this large topic of assignment type, Dean Heller makes much of an endnote in our report which, admittedly, was poorly phrased. This note said that the cause of higher grades in teacher prep was not lax grading standards – which Heller takes to mean that we don’t believe the main point of our own report. By “lax grading standards” we meant that education professors would look at student work that other professors would award a C and give it an A. As we explain in the report, we do not believe that lax grading standards are more prevalent in teacher preparation than in other coursework. We assert that it is the type of assignments given, rather than the grading standards applied, that reduces the rigor of preparation programs.
As we noted above, teacher candidates and non- candidate students enter their junior year with roughly the same grades, with both groups having taken a wide variety of courses in academic disciplines outside of teacher preparation. But in the remaining two years of their college careers, the teacher candidates' GPAs suddenly rise to a point that their rate of earning honors is roughly 50 percent higher than that of all graduating students. The more courses that teacher candidates might have to take outside of education during their last two years, the more likely it is that high grades in education courses must be the explanation for any large differential in honors awarded. Heller makes the rather baffling claim that our methodology does not consider that many education majors take courses outside the college of education. To the contrary, to the extent that candidates who take such courses still earn honors, it only underscores our point about the significant impact that higher grades in education courses can have on overall GPAs.
In conclusion, Dean Heller’s analysis is long on criticism but short on accuracy. He ignores the most important feature of the study about the nature of assignments in teacher prep and misrepresents many of the key points of Easy A’s, including how we tested the rating of programs with less detailed information, and the implication of teacher candidates taking courses outside the college of education. Certainly, the Easy A’s report opens the door to more questions and future research. However, the conclusions it reaches and the recommendations it makes are solid and resonate with the many teachers who have read the report.