Assignment type and student performance in college economics courses

Journal of Instructional Psychology,  March, 2008  by Dmitriy Chulkov

• E-mail
• Print
• Link

The choice of assignment types in a course is shown to have differential effect on performance of specific types of students. Using data from six semesters of college Economics classes, the study demonstrates that male and non-traditional students are more likely to have higher performance on multiple choice questions. Students whose major field of study is related to the course are more likely to do better on the essay questions. This non-uniform effect implies that choosing only one assignment type in a course may provide an advantage to certain types of students. As a result, utilizing a combination of assessment instruments may be recommended.

Keywords: Assignment Type, Student Performance, Assessment Instruments

**********

A variety of tools is commonly used in higher education to assess student performance. The most prominent examples of the alternative assessment tools are the multiple choice question exams and the essay examinations. This study compares student performance on multiple choice question examination and on essay questions and examines whether the performance of the same students on each type of assignment is equivalent. Furthermore, the study reports the student demographic characteristics that are associated with better performance on each of these two assessment measures. If there are systematic differences between the types of students that do better on the multiple choice questions, and the types of students that do better on essay questions, that may have important implications for the design of course assessment tools. Sole use of a single assessment mechanism in a course may put certain students at a disadvantage.

There is limited examination in the academic literature of the relationship between assignment type and student performance. One study that attempts to compare examinations based on multiple choice questions with essay examinations comes from Scouller (1998). The focus of this study, however, is not on student performance but on the students' learning approach. Based on survey evidence in a sample of 206 students, Scouller demonstrates that students are more likely to employ deep learning approaches when preparing for an essay exam, while surface learning approaches are more likely to be associated with a multiple choice exam. In terms of performance, poorer performance in essay exams was linked with the use of surface learning strategies.

Chulkov (2006) demonstrates that the likelihood a student is motivated with grading incentives to complete coursework varies with the demographic characteristics of the student. The results of that study show that grades do not motivate all students in the same fashion. In particular, female students, students whose major field of study is related to the course, and traditional-age college students were influenced by the grading incentives in a course to a larger extent. While these differences in student types were shown to affect student effort in completing assignments, the effect on student performance was not examined. In contrast, the current study looks at the importance of student characteristics in predicting performance in the two different types of exams.

Several studies have looked at the determinants of student performance on exams, and identified a number of relevant variables. For instance, Rayburn and Rayburn (1999) demonstrate that students that systematically submit homework perform better on exams. In addition, students enrolled in longer classes perform better overall. In contrast, Ewer et al. (2002) examined the link between class length and student performance on exams in introductory Accounting classes. Specifically, they compared compressed 4-week classes and the regular 16-week classes and did not find significant differences in performance. Aldosary (1995) demonstrates that class attendance improves student performance on exams.

This study attempts to complement the existing literature in two ways. First, it tests whether student performance on multiple choice questions is different from performance of the same students in an essay question exam section. Second, the dataset used in this research presents a variety of student demographic information. There is available data on student gender, major field of study, year of study, whether the student is traditional or adult, as well as the final grade achieved. Thus, the study can establish patterns in student performance on multiple choice exam questions and on essay questions, identifying which types of students are likely to perform better with each assessment tool used.

Description of Data

The data used in this research comes from a dataset that includes 189 students enrolled in a required course--Introduction to Macroeconomics--taught by the same instructor. The data were collected from nine course sections over six semesters. This dataset was used in Chulkov (2006) to evaluate the response of different types of students to grading incentives in the course. Specifically, it was shown that getting a grade induces only a specific set of students to expend the effort and complete homework for submission.