How Accurate Are Self-Reported High School Courses, Course Grades, and Grade Point Average?

ACT Working Paper Series ACT working papers document preliminary research. The papers are intended to promote discussion and feedback before formal publication. The research does not necessarily reflect the views of ACT. How Accurate Are Self-Reported High School Courses, Course Grades, and Grade Point Average? Edgar Sanchez Richard Buddin WP-2015-03 April 2015 How Accurate Are Self-Reported High School Courses, Course Grades, and Grade Point Average? Edgar Sanchez and Richard Buddin April 2015 Abstract Education researchers and college admissions officials often rely on self-reported coursework and grade information provided by high school students. This study examines the accuracy of self-reported course and grade information from the ACT registration with high school transcript data from a Midwestern state. The evidence shows that students’ self-reported information is an accurate representation of their high school experience. About 94% of students accurately reported their coursework. The correlation between self-reported and transcript course grades was 0.66 with 96% of self-reported grades within a single letter grade of their transcript grade. High school grade point average computed from self-reported course grades was highly correlated with transcript grade point average (r=0.83). The accuracy of coursework and grades differed little by gender, race/ethnicity, and low-income status. The results indicate that self-reported coursework and grade are reasonably valid measures for education researchers and for preliminary screening of students by college admissions officials. Keywords: self-reported student data, high school grades, high school coursework 2 Introduction Education researchers often study the relationship between high school coursework, grades, and admissions test scores on college enrollment and persistence, grade point average (GPA), and on-time degree completion (Adelman, 2006; Schmitt et al. 2009, Radunzel & Nobel, 2012; Sanchez, 2013; Mattern, Patterson, & Wyatt, 2013). Researchers seldom, however, have access to high school transcript data and typically rely on student self-reported course and grade information. One of the premises of this post-secondary research is that the self-reported information adequately reflects student background and high school preparation. Student self-reported information is also used by college recruiters and postsecondary administrators to identify and encourage potential applicants with suitable preparation to apply to their institution. Ultimately, colleges obtain high school transcripts for applicants, but self-reported information provides an efficient way to identify students adequately prepared for college. While transcripts are the official record, course titles vary considerably from district to district, so postsecondary administrators may have difficulty assessing and categorizing high school coursework. Having a short list of discrete core courses may help postsecondary administrators efficiently identify qualified applicants. Given the widespread use of self-reported coursework and grade information, this study focuses on whether self-reported information is an accurate indication of a student’s high school experience. We obtained high school transcript data from a Midwest state and compared this data with course and grade information that students self-reported when registering for the ACT test. The registration form asks for course enrollment and grade information for 30 high school courses in English, math, science, social studies, foreign languages, and the arts. 3 Literature Review Valiga (1987) examined the accuracy of self-reported course-taking and grades for about 1,000 students from 26 Kentucky and Illinois high schools that registered to take the ACT in April or June of 1983. In this study the correlations between self-reported and transcript grades ranged from 0.75 for Computer Science to 0.92 for English 11 with a median correlation of 0.86. Overall, about 80% of students accurately reported their grades. Sawyer, Liang, and Houston (1988) assessed the accuracy of self-reported coursework and grades using data from about 1,000 ACT-tested students in the 1985-1986 academic year from 53 high schools. Across 30 courses, they found that the rate of accurately reporting coursework taken ranged from 65% for American Government to 97% for English 9 and 10. The median accuracy rate across subjects was 87%. They also found that about 97% of students reported course grades that were within one letter grade of their transcript grade (e.g., A and B). Moreover, the correlations between self-reported and transcript reported grades ranged from 0.53 for Drama to 0.89 for Trigonometry with a median correlation of 0.80. The authors showed that these findings held across race/ethnicity, gender, and ACT Composite (ACTC) score. The exceptions to this trend included greater accuracy of self-reported grades for