THE EFFECT OF REINFORCEMENT AND TRAINING ON GROUP STANDARDIZED TEST BEHAVIOR

Abstract

(often divided into groups on variables such as ethnicity, geographical location, aptitude, type of program, or sex) are frequently used to make decisions about funding, program evaluation, student placement, and curriculum selection. When test scores are used to document the occurrence of educational inequities, to compare results across groups of students, and to make educational decisions regarding policies, the results need to be valid and interpretable as indicators of student knowledge (i.e., scores must be a measure of the skills the test was designed to measure). Previous research has demonstrated that a number of variables other than knowledge can contribute to students' test scores. (Reviews by Jensen, 1980; Kirkland, 1971; Millman, Bishop & Ebel, 1965; Sattler & Theve, 1967; and Vernon, 1954 are particularly recommended.) Variables that have been shown to affect test scores include reinforcement procedures, practice tests, and test-wiseness. Previous research has demonstrated that higher test scores are associated with providing students with tangible rewards or verbal praise (Ayllon & Kelly, 1972; Baer, 1978; Edlund, 1972; Hurlock, 1925; Rasmussen, 1974), having students practice on identical or similar tests (Droege, 1966; Peel, 1951, 1952; Quereshi, 1968), and coaching students on test-wiseness skills and strategies (Gibb, 1964; Moore, 1971; Moore, Schultz, & Baker, 1966; Sherriffs & Boomer, 1954; Slakter, 1968; Wahlstrom & Boersma, 1968). There have been very little data reported in the literature on adherence to standardized procedures, but studies have demonstrated that factors associated with the actual administration of tests do affect test scores. Previous research has shown that higher test scores are associated with the use of familiar examiners (Thomas, Hertzig, Dryman & Fernandez, 1971), positive pretest interactions between examiner and student (Mandler & S. Sarason, 1952; S. Sarason & Mandler, 1952), and familiar testing locations (Seitz, Abelson, Levine & Zigler, 1975). Examiners can influence test scores by preparing students for the test. For instance, students who know the testing purpose (Carlson & Lazarus, 1953), have a positive perception of the examiner (Davis, 1951; Postman, Bronson & Gropper, 1953), take a practice test (Snow, 1974), and possess a low level of anxiety (Butler, 1980; Hill, 1967; McCoy, 1965; I. Sarason, 1973), receive higher test