| Derek Briggs
University of Colorado
The effectiveness of admissions test preparation: New evidence from ELS: 2002
The purpose of the proposed study is to examine the extent to which commercial test preparation (i.e., "coaching") continues to have an effect on the scores from college admissions exams. This study builds upon previous research the author has conducted on this topic using data from the National Education Longitudinal Study of 1988 (NELS:88). That research found, on average, positive but small effects associated with the SAT and ACT exams (Briggs, 2001; 2002). The present study would attempt to replicate these findings using the more recent Educational Longitudinal Study of 2002 (ELS:2002). While the study using NELS data was based upon a sample of high school seniors taking admissions exams in 1992, the proposed study using the ELS:2002 would be based upon a much more recent cohort of high school seniors taking admissions exams in 2004. The NELS and ELS surveys were each administered to nationally representative samples of test-takers, and collected much of the same information on students with respect to their college admissions exams scores, test preparation activities, and other background characteristics. As a result, the proposed study provides an opportunity to evaluate whether there has been a change in the effectiveness of coaching over time.
In this study the analytic techniques of linear regression and propensity score matching will each be used to estimate effects for coaching on both the SAT and ACT. The extent to which each approach leads to consistent conclusions about the effects of coaching will be evaluated. From a policy standpoint, it is important to have timely information about the coachability of admissions tests. If the most generalizable evidence indicates that the average effect of coaching is small, and has gotten smaller since the SAT changed format in 1994, this would be positive news from an equity standpoint. On the other hand, if the effect of coaching on the SAT has increased, it may suggest that the efforts of The College Board to make the exam less coachable have failed.
Back to Funded Research Grants Page |