of student motivation over the semester. As expected, declines in the
interest/enjoyment (IMI) and pressure/tension (IMI) subscales pre-
dicted performance negatively and positively, respectively. Surpris-
ingly, the analysis also showed that the more a student’s value/
usefulness (IMI) score declined, the better they did in the class. In
contrast to the observed IMI pre/post-survey differences, MSLQ pre/
post-survey differences did not predict the final course percentage.
The primary goal of this study was to measure biology student
motivation and see how it changed over the course of a semester.
The observed overall decrease in motivation was surprising given
the results of previous studies that showed either little change
(Tsigilis & Theodosiou, 2003) or an increase in different aspects
of motivation (Sungur & Tekkaya, 2006; Cheang, 2009; Van Vliet
et al., 2015). One possible explanation is that students simply may
not have been as intellectually stimulated by the class as predicted.
If students were not interested and engaged in the class, their motivation would be expected to decrease. However, this explanation
runs counter to end-of-the-semester evaluations in which most students indicated that the course stimulated their interest in the subject
matter (data not shown). A second possibility is that students may
enter this course with unrealistic expectations; as students become
aware of the class specifics and expectations (e.g., structure, difficulty,
work load), they may lose some of their initial enthusiasm. This is
especially true for students who are very interested in topics that are
not covered in the class. A third possibility is that students have lower
scores on the post-survey because it is given during a very stressful
time of the semester when students are turning in final projects, giving
presentations, and preparing for final exams. This final possibility
seems especially likely, as many students feel overworked and fatigued
during this time (B. Gibbens, personal observation).
Overall student scores were highest on questions related to
value and lowest on anxiety items. This indicates that students
cared deeply about the subject matter and that pressure/tension levels were manageable. While student pre-survey scores related to
intrinsic motivation, self-efficacy, and value were high for both
the IMI and the MSLQ, no ceiling effect was observed.
Biology student scores on different motivation subscales are
comparable to what has been seen in other studies done on undergraduate populations. For example, Pintrich et al.’s (1993) study
found that students score high on intrinsic goal orientation and
self-efficacy subscales ( i.e., > 5 on a 1–7 point Likert scale) and lower
on test anxiety (e.g. 3. 63 on a 1–7 point Likert scale). By contrast,
the subscale scores in the present study were markedly different
from those in studies done on primary and high school biology students. For example, scores on IMI and MSLQ subscales related to
intrinsic motivation, value, and self-efficacy were 11–42% higher
for the present study’s undergraduate cohort than they were for
sixth-grade (Sturm & Bogner, 2008) or tenth-grade (Sungur &
Tekkaya, 2006) biology students. These differences highlight the
importance of academic level, classroom context, and student cohort
differences when doing motivation studies. Because motivation is
expected to differ in different disciplines (Breen & Lindsay, 2002)
and in different classes (Crede & Phillips, 2011), it is recommended
that instructors administer motivation instruments on their own
student cohorts to establish local norms for comparative purposes
(Pintrich et al., 1991).
Student scores on comparable IMI and MSLQ subscales were
very similar (Figure 1), suggesting that either survey could be used
Figure 1. Pre/post-survey scores for the Intrinsic Motivation
Inventory (IMI) and the Motivated Strategies for Learning
Questionnaire (MSLQ) subscales (100% represents the
maximum possible score and 0% the lowest possible score
for each subscale). Error bars represent standard error.
Asterisks denote significant differences between pre/post-
survey scores as determined by paired t-tests (* P < 0.05,
P < 0.01, P < 0.001).
Figure 2. Predicting course outcomes with motivation data:
standardized beta weights from ordinary least squares (OLS)
regression analyses of student pre/post-survey data. Only
subscales that were significantly predictive of course outcomes
are shown. (A) Pre-survey motivation subscales found to be
predictive of student course outcomes. Data were obtained
from two separate OLS regression analyses, one conducted
using the dimensions of the Intrinsic Motivation Inventory (IMI)
as predictors, the other conducted using the Motivated
Strategies for Learning Questionnaire (MSLQ) dimensions.
(B) Changes in the depicted subscale scores ( i.e., post-survey
scores minus pre-survey scores) either positively or negatively
predicted student outcomes. Relative significance is indicated
(* P < 0.05, P < 0.01, P < 0.001).