Binomial effect size displays and gain-probability: Alternative ways to interpret hierarchical regression findings,...

Area of Science:

• Psychometrics
• Statistical Modeling

Background:

• Hierarchical regression analysis is a common statistical technique.
• Delta R-squared (ΔR²) is the traditional measure for assessing the contribution of variables in hierarchical regression.

Purpose of the Study:

• To explore the extension of binomial effect size displays and gain-probability analyses from zero-order correlation coefficients to multiple correlation coefficients in hierarchical regression.
• To demonstrate that these alternative interpretation methods can yield different conclusions compared to ΔR².

Main Methods:

• The study presents an exposition and tutorial on applying binomial effect size displays and gain-probability analyses.
• It contrasts these methods with the traditional ΔR² interpretation within a hierarchical regression framework.

Main Results:

• Binomial effect size display and gain-probability interpretations can lead to different conclusions than each other and from ΔR².
• These alternative methods provide a more comprehensive understanding of the data's implications.

Conclusions:

• Multiple interpretation methods for hierarchical regression, including binomial effect size displays and gain-probability analyses, offer a richer understanding than relying solely on ΔR².
• Researchers should consider employing diverse interpretive approaches for a thorough data analysis.