Accuracy of Self-Reported High School Courses and Grades of College-Bound Students

ACT Research Report Series88-1Accuracy of Self-Reported High School Courses and Grades of College-Bound StudentsRichard Sawyer Joan Lalng Mark HoustonMarch 1988 For additional copies write: ACT Research Report Series P.O. Box 168 Iowa City, Iowa 52243©1988 by The American College Testing Program. All rights reserved. ACCURACY OF SELF-REPORTED HIGH SCHOOL COURSES AND GRADES OF COLLEGE-BOUND STUDENTSRichard Sawyer Joan Laing Valter Houston i Dedicated to the memory of RALPH ASCHENBRENNER1918-1987Ralph Aschenbrenner was an educator and school adminis­trator for almost 40 years. As a consultant to ACT following his retirement, he coordinated the transcript evaluation phases of this study and of ACT's earlier research in this area. We miss him as a professional, as a colleague, and as a friend. I AbstractThe High School Course/Grade Information Section (CGIS) of the ACT Assessment registration folder collects detailed information about the courses students have taken or plan to take in high school, and the grades they have earned in courses they have completed. In this study, we compared the data provided by students on the CGIS with corresponding information from their high school transcripts. Using criteria developed for the study, we found, for the typical course, that about 10% of the students provided no information; that about 87% of the students' statements with respect to whether they took the course could be presumed to agree with their transcripts; and that about 3% of the students’ statements were inconsistent with information on their transcripts. Of the students who provided no information about a course, most, according to school records, had not taken the course. Among students who reported grades for a course, the typical rate of exact agreement between student-reported and transcript grades was 71%. About 97% of the students reported grades that were within1 letter grade of the corresponding transcript grades. ACCURACY OF SELF-REPORTED HIGH SCHOOL COURSES AND GRADES OF COLLEGE-BOUND STUDENTSMost postsecondary institutions require applicants to supply evidence that their high school background has adequately prepared them to meet the demands of college, university, or technical school coursework. Often, this evidence takes the form of an official high school transcript. Unfortunately, transcripts vary widely among schools. For example, grades may appear as numbers or as letters, based on a variety of scales; the student who earned a "B" in the honors section of English IV may not be distinguishable from the one who earned a "B " in the regular section of that same class; sometimes the name of a given course does not reflect its content; and, of course, there is no common format in which transcripts are prepared.To simplify the interpretation required, some institutions ask applicants to complete a form listing their high school coursework and the grades they earned. This form is then used, with other materials, in making preliminary admissions decisions, in counseling, and/or in placement. The official transcripts, if required, serve as confirmation of the students' self-reported information.Similarly, both major college admissions testing programs in the United States— the ACT and the SAT— ask participants to provide information about their high school coursework. This self-reported information, with the test scores and other background information, is sent to colleges designated by the students, where it typically becomes part of their admissions files.Considerable research has been conducted to investigate the accuracy of self-reported data. Such data have been found to be relatively accurate (Astin, 1965); to be unaffected by incentives to distort responses (Walsh, 1967, 1968); to be more accurate when information more readily available to respondents is requested (Armstrong, Jensen, McCaffrey, & Reynolds,1976); and, often, to have a level of concurrent and predictive validity comparable to that of test data (Baird, 1976). Pace, Barahona, and Kaplan (1985) made the point that: "The quality of questionnaire answers(reliability, validity, credibility) depends most of all on the quality of the questions"— indicating that, in general, respondents provide accurate data if they understand what is being asked.When we look more specifically at research related to the accuracy of self-reported courses and grades, we find similar results. Fetters,Stowe, and Owings (1984) asked high school seniors to report the amount of coursework they had completed in a variety of areas. Correlation coefficients between self-reported and transcript data ranged from .28 to .87, although the authors noted that, because of certain methodological characteristics of the study, the coefficients were probably underestimates. They obtaine