Bryan K. Saville et al.
156
Of the 15 attributes/behaviors contained on Ferrett’s inventory, students in both
sections reported engaging in 10 of them more often with interteaching. For three of the
attributes (evaluating statements and arguments from readings and class, examining one’s
own assumptions and opinions about course material, and looking for supporting evidence
for problems), students in both sections reported that they did so equally with interteach-
ing and lecture. Finally, there were section differences for two of the attributes: For being
“curious about course material,” one section reported doing so more with interteaching,
whereas the other section reported doing so more during lectures; for “evaluating knowledge
of course material,” one section reported doing so more with interteaching, whereas the
other section reported doing so equally with interteaching and lecture.
Overall, these results suggest that interteaching may produce increases in the behaviors
that are indicative of critical thinking. Clearly, this is not to say that lecture-based courses,
when constructed correctly, cannot (and will not) lead to increases in critical thinking
(McKeachie, 2002). However, the very nature of interteaching—with its focus on student
discussion, peer-to-peer teaching, and frequent feedback—may be more likely to provide
a context in which students engage in these highly desired behaviors.
An integrated statistics–research methods course. Recently, Dunn, Smith, and Beins (2007)
edited the volume Best Practices for Teaching Statistics and Research Methods in the Behavioral
Sciences. Interestingly, their title reflects two ways in which instructors typically approach
their statistics and research methods courses. First, the title separates statistics and research
methods. Likewise, most instructors view statistics and research methods as distinct
courses—hence the reason they commonly appear in our psychology curriculum as such.
Second, the title places statistics before research methods, reflecting another common
pedagogical practice: teaching statistics before research methods. Indeed, at our university,
we have followed this approach for some time. Students take Psychological Measurement
and Statistics (Psyc 210), and then they take Psychological Research Methods (Psyc 211).
Although the logic for this two-semester sequence is grounded in wanting students to
learn basic statistical concepts that they can later apply in their research methods course, a
critique of this approach suggests that it may not provide a context in which students can
appreciate why they are taking statistics (Barron et al., 2007; Christopher, Walter, Horton, &
Marek, 2007). Failure to provide context for learning can severely impact student motiva-
tion (Lepper & Henderlong, 2000) and comprehension (Bransford & Johnson, 1972).
Thus, teaching statistics before research methods is akin to “putting the cart before the
horse.” Does it really make sense to spend a great deal of time teaching students about
specific tools they will use to analyze data before teaching them why we conduct research
in the first place? Should we then be surprised when students have trouble comprehending
the information they encounter in their statistics courses? And should we be surprised that
we have to spend time in our research methods courses revisiting ideas they covered in
statistics?
In addition to teaching two separate courses, we are now offering a new year-long, inte-
grated version of these courses. Specifically, students shift in and out of units on statistics
and research methods each semester. Our goal is to provide students with better context
in which they can learn about the different methodological approaches and statistical tools
that psychological researchers use to build a valid body of knowledge. To correct the “horse
before the cart” problem we mentioned earlier, students first learn about a particular