Does Congruency Between Homework and Test Problems Improve Test Performance?

Jacqueline Jenkins

Download Paper | Permalink

Conference: 2024 ASEE Annual Conference & Exposition
Location: Portland, Oregon
Publication Date: June 23, 2024
Start Date: June 23, 2024
End Date: June 26, 2024
Conference Session: Civil Engineering Division (CIVIL) Technical Session - Effective Teaching 2
Tagged Division: Civil Engineering Division (CIVIL)
Page Count: 10
DOI: 10.18260/1-2--47202
Permanent URL: https://peer.asee.org/47202
Download Count: 61

Paper Authors

author page

Jacqueline Jenkins Cleveland State University orcid.org/0000-0003-2292-7176

Download Paper | Permalink

Abstract

The purpose of this study was to understand whether the alignment of homework problems with test problems, in terms of their complexity and difficulty, can improve test performance. A post course study was conducted examining the correlation between individual student scores on online homework and on subsequent, associated tests. Data from two sections of an undergraduate statics course taught by the author at Cleveland State University during the fall 2022 and spring 2023 semesters were examined. The course content, lecture notes, in-class examples, textbook, and online learning platforms remained the same, and the test questions were similar. All homework was completed through the textbook publisher’s online learning platform. The problems for the assigned homework were different. Overall, the level of difficulty and expected time needed to complete the spring semester problems were approximately 25% greater than that for the fall semester and better aligned with the test problems. It was expected that congruency between homework and test problems would result in better test grades. Correlation analyses indicated that the relationship between the performance on the homework and the final grades in the course were stronger for the spring semester (0.877) than for the fall semester (0.712). For each of the seven summative tests and associated homework assignments, the relationship between the scores were examined. For fall semester, the correlations were in the range [0.18-0.58], whereas for the spring semester they were in the range [0.47-0.69]. However, descriptive statistics of the homework scores and test scores, suggested that the 35 students during the fall 2022 semester generally performed better than the 38 students during the spring 2023 semester on both the assigned homework and tests. That observation was confirmed through a series of comparison of means tests, which compared the performance on both the homework and tests across the two semesters. Therefore, assigning homework questions that better aligned with the test questions, albeit more difficult, appeared to strengthen the association between homework grades and test grades but did not translate to better test grades.

Citation
Format

Jenkins, J. (2024, June), Does Congruency Between Homework and Test Problems Improve Test Performance? Paper presented at 2024 ASEE Annual Conference & Exposition, Portland, Oregon. 10.18260/1-2--47202

TY - CPAPER
AB - The purpose of this study was to understand whether the alignment of homework problems with test problems, in terms of their complexity and difficulty, can improve test performance. A post course study was conducted examining the correlation between individual student scores on online homework and on subsequent, associated tests. Data from two sections of an undergraduate statics course taught by the author at Cleveland State University during the fall 2022 and spring 2023 semesters were examined. The course content, lecture notes, in-class examples, textbook, and online learning platforms remained the same, and the test questions were similar. All homework was completed through the textbook publisher’s online learning platform. The problems for the assigned homework were different. Overall, the level of difficulty and expected time needed to complete the spring semester problems were approximately 25% greater than that for the fall semester and better aligned with the test problems. It was expected that congruency between homework and test problems would result in better test grades.
Correlation analyses indicated that the relationship between the performance on the homework and the final grades in the course were stronger for the spring semester (0.877) than for the fall semester (0.712). For each of the seven summative tests and associated homework assignments, the relationship between the scores were examined. For fall semester, the correlations were in the range [0.18-0.58], whereas for the spring semester they were in the range [0.47-0.69]. However, descriptive statistics of the homework scores and test scores, suggested that the 35 students during the fall 2022 semester generally performed better than the 38 students during the spring 2023 semester on both the assigned homework and tests. That observation was confirmed through a series of comparison of means tests, which compared the performance on both the homework and tests across the two semesters. Therefore, assigning homework questions that better aligned with the test questions, albeit more difficult, appeared to strengthen the association between homework grades and test grades but did not translate to better test grades.
AU - Jacqueline Jenkins
CY - Portland, Oregon
DA - 2024/06/23
PB - ASEE Conferences
TI - Does Congruency Between Homework and Test Problems Improve Test Performance?
UR - https://peer.asee.org/47202
DO - 10.18260/1-2--47202
ER -