THE EFFECT OF FACTOR SCORES, GUTTMAN SCORES, AND SIMPLE SUM SCORES ON THE SIZE OF F RATIOS IN AN ANALYSIS O F...

Area of Science:

• Social Sciences
• Psychometrics
• Statistical Analysis

Background:

• Social science research often compares group means using analysis of variance (ANOVA) F tests.
• Data from questionnaires or inventories typically requires a scoring or data reduction procedure before ANOVA.
• The impact of different scoring methods on the F test distribution is often unclear to researchers.

Purpose of the Study:

• To investigate if scoring procedures affect the magnitude of the F ratio in ANOVA.
• To provide researchers with empirical evidence regarding the choice of scoring methods.

Main Methods:

• Generated Guttman, Saaotor, and simple sum scores from item responses.
• Utilized item responses from a large sample of high school seniors.
• Analyzed the resulting F ratios from each scoring method using ANOVA.

Main Results:

• No statistically significant differences were detected in the F ratios across the three scoring methods.
• The analysis indicated that scoring method did not influence the outcome of the ANOVA F test.

Conclusions:

• The choice of scoring method (Guttman, Saaotor, or simple sum) has minimal impact on ANOVA F ratios.
• Simple sum scoring can be as effective as more complex methods for data reduction in these types of analyses.