Use of admissions data to predict student success in postsecondary freshman science

Abstract

The purpose of this study was to determine if significant relationships exist for any of the variables, age, gender, previous GPA, test scores (ACT, Compass), number of accumulated credits, and student success in Biology. This study strived to determine what academic/admissions data can be used to determine the likelihood of student success in Biology. A quantitative correlational study using stepwise multiple regression analysis was used for this study. The study was a retrospective study. Data was composed of a convenience archival sample from the institutional database. Multiple regression analysis was conducted to determine the effect each independent variable has on the dependent variable of student success. For the data set ACT, the variables math score, prealg score, writing score, reading score, and previous GPA were all significant. For data set CMP the variable of student's age was not significant, but the other variables were significant. For the Blanks data set, the only variable of significance was gender. Using stepwise multiple regression analysis the data sets produced regression models showing predictability based on stepwise significance. For Blanks data set, the variables previous hours earned, gender, age, and previous GPA were used. For the ACT data set, math score and reading score were used. For the CMP data set the variables included math score, writing score, previous GPA, gender, reading score, and previous hours earned. The level of predictability of the regression equation for the ACT data set and Blank data set was low. However, the predictability for the CMP data set was moderate. The highest percent of variance explained by the regression models was 11.6% of the CMP data set.