ERIC Number: ED446615
Record Type: RIE
Publication Date: 2000
Admission Models for At-Risk Graduate Students in Different Academic Disciplines.
Nelson, C. Van; Nelson, Jacquelyn S.; Malone, Bobby G.
In this study, models were constructed for eight academic areas, including applied sciences, communication sciences, education, physical sciences, life sciences, humanities and arts, psychology, and social sciences, to predict whether or not an at-risk graduate student would be successful in obtaining a master's degree. Records were available for 916 graduate students who were considered at risk because they did not meet the minimum admission requirements but were given the opportunity to begin coursework. Success was measured in two ways: (1) whether or not the student completed the degree; and (2) categorization of the graduate grade point average obtained after 9 hours of graduate coursework as either 3.5 or above or below 3.5. The predictor variables included Graduate Record Examination (GRE) scores, undergraduate grade point average, Carnegie classification of the student's undergraduate institution, and personal variables. GRE scores were combined with undergraduate grade point averages as compensatory and conjunctive to determine the best predictor for each area. The accuracy of prediction using the logistic regression models ranged from 63 percent to 84 percent. (Author/EV)
Publication Type: Reports - Research
Education Level: N/A
Authoring Institution: N/A