Discriminant analysis to predict graduation--nongraduation in a master's degree program in nursing

Area of Science:

• Health Sciences
• Educational Psychology

Background:

• Predicting academic success in graduate nursing programs is crucial for student retention and program quality.
• Traditional admissions metrics like GPA and GRE scores are commonly used but their predictive power varies.

Purpose of the Study:

• To determine if readily available admissions data can predict graduation outcomes in a master's nursing program.
• To differentiate between graduates, dropouts, and non-accepted applicants using discriminant analysis.

Main Methods:

• Discriminant analysis was employed using baccalaureate GPA and GRE-verbal/quantitative scores as predictors.
• The study included 102 graduates, 103 dropouts, and 65 non-accepted individuals.
• One-way ANOVA and Scheffé tests were used for follow-up comparisons.

Main Results:

• The discriminant analysis model significantly differentiated between the groups (p < .0001), explaining 98% of the variance.
• Baccalaureate GPA, GRE-verbal, and GRE-quantitative scores significantly predicted group differentiation (p < .05).
• Baccalaureate GPA (22%), GRE-verbal (13%), and GRE-quantitative (10%) scores showed practical significance in differentiating groups.

Conclusions:

• Admissions data, particularly undergraduate GPA and GRE scores, are valuable predictors of success in master's nursing programs.
• These findings can inform admissions strategies to improve student selection and retention.
• Further research should explore additional predictive factors for graduate nursing education.