Using Reading Skills as a Predictor of Success on the Fourth Grade S-PET Test

Abstract

In the following study, the Program Evaluation Test in Science (S-PET) scores for students in the fourth grade at PS #44 within the Rochester city School District, were compared to their scores on the Degrees of Reading Power (DRP) test. The tests are administered to all fourth grade students every May. Two groups of students were included: the first group was composed of those fourth grade students who took the S-PET test in May 1992. The second group was made up of the fourth grade students who took the S-PET test in May 1993. These same students also took the DRP test in May 1992 and May 1993, respectively. This study examined the relationship between the DRP reading test scores and the S-PET science scores to see how strong a relationship there was between them. Further, it compared the level of achievement between the two groups on the S-PET test. The group who took the S-PET test in May 1993 had had extensive class time devoted to teaching specific vocabulary pertinent to the S-PET tests. If reading test scores on the DRP are an accurate predictor of success on the S-PET test, then one would expect the latter group to perform better on the S-PET test. Finally, the S-PET was evaluated for reading difficulty using the Fry Readability Graph. Results from a series of t-tests indicated that the DRP scores correlated strongly with the S-PET test in both the 1991-92 test group and the 1992-93 test group. Further calculation of the point biserial coefficient of determination (rpb2) indicated that class time devoted to teaching vocabulary relevant to the S-PET test did in fact have a significant effect upon the test scores of the 1992-93 test group. Finally, according to the Fry Readability Graph the S-PET has an estimated fourth grade level of reading difficulty.