skip to main content skip to footer

An Investigation of the Feasibility and Practical Outcomes of Pre-Equating the SAT Verbal and Mathematical Sections IRT SAT

Author(s):
Eignor, Daniel R.
Publication Year:
1985
Report Number:
RR-85-10
Source:
ETS Research Report
Document Type:
Report
Page Count:
109
Subject/Key Words:
College Board, Equated Scores, Item Response Theory (IRT), Pretesting, Scholastic Aptitude Test (SAT), Test Disclosure

Abstract

Test disclosure legislation, enacted and presently being considered in several states, has serious implications for testing programs relying on conventional equating methods to put new forms of exams on scale. Certain of the problems presented by this legislation can be circumvented by applying item response theory (IRT) based equating methods. Many of the problems can be circumvented if the IRT equating performed is pre-equating, that is, establishing conversions from raw to scaled scores, through the use of pretest data, before the time the new test is administered operationally. The purpose of this study was to determine the extent to which item parameters estimated on SAT-verbal and SAT-mathematical pretest data can be used for equating purposes in a situation where intact final form SAT testing data has normally been used. More specifically, the items that appear in any final SAT form come from multiple pretests, and to the extent that the item statistics are sensitive to the context or position in which the items appear, there may be a lack of fit between the equating based on pretest data and final form data. In this study, the verbal and mathematical sections of two SAT forms were calibrated from pretest data, pre-equated to existing SAT forms, and then the results of the pre-equating compared in a number of ways to final form IRT equating and conventional linear equating methods. (109pp.)

Read More