Accuracy of Self-Reported Activities and Accomplishments of College-Bound Students

ACT Research Report Series87-6Accuracy of Self-Reported Activities and Accomplishments of College-Bound StudentsJoan Laing Richard Sawyer Julie IMoble 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. Joan Laing Richard Sawyer Julie NobleACCURACY OF SELF-REPORTED ACTIVITIESAND ACCOMPLISHMENTS OF COLLEGE-BOUND STUDENTS I ABSTRACTThis paper examines the accuracy of student-reported demographic charac­teristics, activities, and accomplishments provided through the ACT Assessment Program. It was found that students generally reported such data with a high degree of accuracy. The typical rate of incongruent responses when school- and student-reported data were compared was about 10%; of these, only about 6% of students claimed credit for an activity or accomplishment that the school did not confirm. The results suggest that, for most purposes, colleges can be fairly confident that the reports of activities and accomplishments provided on students’ ACT Assessment records are accurate. ACCURACY OF SELF-REPORTED ACTIVITIESAND ACCOMPLISHMENTS OF COLLEGE-BOUND STUDENTSIntroductionA substantial amount of educational and social research is based on self- reported data. Subjects may, for example, be asked about their demographic char­acteristics (age, nationality, sex), their accomplishments (degrees, activities, awards), or their plans (education, career, community service). Naturally, researchers are concerned about the accuracy of such data and have conducted a variety of studies comparing self-reported data with data obtained from "offi­cial" sources such as school records, census data, and supervisors. In this paper, we examine the accuracy of data reported by college-bound students when they registered to take the ACT Assessment. In particular, we focus on the accuracy of student-reported demographic characteristics, activities, and accom­plishments.Accuracy of self-reported data by college-bound students has practical, as well as theoretical, importance. Not only are self-reported data used in research, but they are an important part of the college admissions materials at most, if not all, institutions (Breland, 1981). Such information can help these institutions identify potentially successful students (Willingham, 1985).As colleges often rely on self-reported data to identify applicants with certain characteristics, there is a large body of literature related to accuracy of self-report by college-bound students. One of the first major studies on this topic was that of Astin (1965). He reported (p. 14) that ’’Student reports of ’factual' items (about high school class size, father's occupation, extracurricu­lar achievements in high school, etc.) are highly reliable; reporting of ’non- factual’ items (about future plans and aspirations) is somewhat less reliable."Later studies tended to confirm Astin’s findings. Walsh (1967, 1968) con­ducted experiments in which subjects were presented with social and financial incentives to distort their responses. He found that accuracy of self-reported data was high and that, in general, the accuracy of experimental groups did not differ from that of control groups; that is, the incentives offered did not lead to greater discrepancies between self-reported and "official" data. These find­ings are particularly important because they imply that incentives such as college admission and scholarship eligibility may not adversely affect the accuracy of self-reported data as much as might be expected.Further evidence that college-bound students do not, in general, tend to "fake good" on self-reported items was presented by Maxey and Ormsby (1971). They found that the median level of agreement between student-reported data and school records was about 90% across items. In the remaining cases, students claimed an achievement not confirmed by the school about 6% of the time, and schools credited the students with accomplishments that they themselves did not claim about 4% of the time. As the participants in this study reported their achievements as part of registration for a college admissions test, it appears that they did not, as a rule, exaggerate to appear more attractive to the colleges of their choice.Maxey and Ormsby further noted that, when discrepancies occurred, it was not always clear that the student had reported inaccurately. In some cases, it appeare