EXAMINATIONS: STUDENT FACTORS AFFECTING
A Reexamination of the Relationship of High School
Psychology and Natural Science Courses to Performance in
a College Introductory Psychology Class
Richard A. Griggs
Sherri L. Jackson
University of Florida
Many researchers have found no relationship between having completed a high school psychology course and final grades in a college introductory psychology class (e.g., Hedges & Thomas, 1980), but Carstens and Beck (1986) found that students with strong backgrounds in high school natural science (i.e., biology, chemistry, and physics) obtained higher course grades in their first college-level psychology class. Carstens and Beck's subjects were not all first-term freshmen; some may have taken or may have been enrolled in college natural science courses (C. B. Carstens, personal communication, March 25, 1988). Thus, their results may have been confounded by two factors: (a) the number of college credits completed by each subject (e.g., see Griggs & Ransdell, 1987) and (b) prior completion of or concurrent enrollment in college-level natural science courses.
Our study eliminated these possible confoundings. Subjects were freshmen enrolled in introductory psychology, but not enrolled in a natural science course. We also controlled for possible instructor effects. Our subjects were all from the same class, whereas Carstens and Beck's subjects were from five classes taught by two different instructors. In addition, we compared the performance of students whose high school psychology course covered science-oriented topics with the performance of students whose courses did not.
The subjects were 199 students (117 women and 82 men) in an introductory psychology class at the University of Florida. All were first-term students without any college credits, and none were currently registered in a college natural science course. They participated as part of the course requirements.
During the first class meeting, the students were given a 60-item multiple-choice pretEst. As in the Carstens and Beck (1986) study, the items were divided into frequent and infrequent topics subtests based on topics covered in high school psychology courses. Frequent topics were personality, disorders, therapies, assessment, learning, cognition, developmental psychology, and social psychology; infrequent topics were research methodology, statistics, physiological psychology, sensation, and perception. This classification is comparable to the one used by Carstens and Beck (1986) and is consistent with a recent survey of topic coverage in high school psychology courses (Ragland, 1987).
All items were selected from the test bank (Brink, 1986) accompanying Myers's (1986) introductory psychology text. Based on information from previous terms, we selected questions with an average item difficulty of 0.70.
During the first week of classes, students completed a questionnaire concerning their high school courses in psychology, biology, chemistry, and physics. Students who had taken a high school psychology course completed another questionnaire about topic coverage in that course. If four of the five topics—research methodology, statistics, physiological psychology, sensation, and perception—were covered, the course was classified as a more scientifically oriented course.
With the students' permission, their SAT scores were obtained from the registrar and course examination and final grades from the instructor. Course grades were coded as A = 4.0, B+ = 3.5, B = 3.0, and so on, with E = 0. Examination scores were on a scale ranging from 0 to 60 points.
The intercorrelations matrix is given in Table 1. The high school natural science variable was not significantly correlated with infrequent topics subtest scores, as it was in the Carstens and Beck study. In agreement with Carstens and Beck's findings, high school psychology was positively correlated with the frequent topics subtest scores, p < .05, and high school natural science was positively correlated with course grades, p < .01.
Because SAT total score was significantly related to the frequent topics subtest score and course grade, two multiple regressions were performed using SAT total score as a covariate.