Bryan K. Saville et al.
154
and then analyze the results (e.g., Stedman, 1993). For example, students might conduct
an experiment on counterfactual thinking in which they imagine that they had recently
taken an exam in one of their courses (see Medvec & Savitsky, 1997) and that they had
either just made a B+ or just missed an A−. Typically, students who imagine that they just
missed an A− are less satisfied with their grade than those who imagine that they just made
a B+. Students then analyze their data using a t test or analysis of variance (ANOVA),
which tends to make the experiment and the statistical analysis more relevant and gives
them a “legitimate feeling of ownership for the data” (Thompson, 1994, p. 41). Moreover,
these types of activity help students understand how statistics and research methods are
closely linked.
Course Format Changes
As most psychology teachers will agree, unfamiliarity with course material can hinder
critical thinking in any class, a notion that might be especially true in statistics and research
methods courses. For instance, in statistics courses, the use of complex mathematical
equations, which can be daunting even for the “math-oriented,” may exacerbate students’
feelings of unfamiliarity—especially when instructors focus on computational equations.
To illustrate, consider the computational formula for the Pearson product-moment
correlation:
r
NXYXY
NXXNYY
=
−
−() −()
∑ ∑ ∑
∑∑∑ ∑
22
22This formula is familiar to statistics instructors, but does it really make students think
critically about correlation? Probably not. More than likely, using this formula to discuss
correlations becomes an exercise in calculator work. In such cases, simple “one-shot”
course activities, like the kind we discussed in the previous section, may do little to assuage
our students’ fear of math or increase the likelihood that they will think critically about
what a correlation is. Instead, statistics instructors could alter the format of their courses.
Rather than using computational formulas to teach students about statistics, we suggest
focusing on conceptual formulas. For example, because a correlation refers to a relation
between two variables, instructors could focus on getting students to think conceptually
about correlations. To do so, instructors could instead use the following conceptual
formula, which is based on z-scores:
r
XY=
∑
ZZ
N
With this equation, students begin to see that “correlation” refers to the average relation
between an individual’s scores on two different variables. When z-scores are plotted
in a scatter plot, the negative and positive product quadrants become more obvious.