AN ASSESSMENT OF THE USE OF WORD ANALYSIS TESTS AS A PREDICTOR OF ACHIEVEMENT IN THE LEARNING OF AN ENERGY RELATED TOPIC BY SELECTED SENIOR HIGH SCHOOL CHEMISTRY STUDENT

This study set out to determine whether or not any of the three word analysis tests or their combination showed an advantage in predicting change in achievement ((DELTA)) over conventional testing. Conventional tests included Intelligence Quotient (IQ), Preliminary Scholastic Aptitude Test (PSAT), Stanford Test of Academic Skills Form A (STA or TASK), Grades (BG--biology final grade, CG--chemistry semester grade) and Rank in Class (RC).;The research population of 80 senior high school chemistry students (N = 80) was divided into two groups. The experimental comparison group (n = 48, EC) and the control comparison group (n = 32, CC) received instruction on the interrelationship between heats of solution and the reaction tendencies related to the solution process. The EC was instructed using laboratory techniques. The CC was instructed in the lecture mode.;All students took an achievement pretest (AT(,1)) followed by three word analysis tests. The word analysis tests were designed to measure the students' specific ability to interrelate seven key concepts related to the instruction topic. The survey identification test (SWA) determined the students' familiarity with the key concepts. Students were required to respond to five questions for each key concept. The word association test (WA) allowed the students freedom to write as many concepts as they thought related to each of the key concepts. The rank interrelation test (RI) required that the students assign a numerical value for the relationship among the key concepts.;After instruction, all students took a posttest (AT(,2)). Pearson r values were calculated for each word analysis test and the total word analysis test scores with the change in achievement. Pearson r correlations were also calculated for the (DELTA) and each of the conventional test scores. The scores were also categorized into ranges to show where advantages for word analysis or conventional testing occurred.;Analysis of data showed that the Pearson r values were somewhat higher for the word analysis tests with the (DELTA) than for the conventional tests with the (DELTA). This was particularly true for the WA and WAT.;The range categorization showed that the word analysis tests had a consistent advantage in ranges 1 and 2 (lowest level conventional scores). The only conventional and (DELTA) r value which exceeded that of the word analysis tests was the RC (rank in class).