An effect size measure and Bayesian analysis of single-case designs

Area of Science:

• Behavioral Science
• Psychology
• Research Methodology

Background:

• Single-case designs (SCDs) are crucial for evaluating interventions.
• Existing effect size measures in SCDs often do not fully capture complex changes (level and trend).
• Challenges exist in analyzing SCDs with serial dependence and combined level/trend shifts.

Purpose of the Study:

• To propose an alternative effect size measure for SCDs.
• To develop an integrated Bayesian procedure for analyzing SCDs with level and trend changes.
• To provide a standardized effect size comparable across different study designs.

Main Methods:

• Development of a novel effect size measure incorporating slope and intercept changes.
• Application of a Bayesian procedure for estimation and inference.
• Integration of a multilevel model for standardized effect size calculation.

Main Results:

• The proposed effect size measure effectively handles level and trend changes.
• The Bayesian approach provides a unified framework for SCD analysis.
• A standardized effect size measure comparable to between-subjects designs is achieved.

Conclusions:

• The novel effect size measure and Bayesian approach enhance the analysis of single-case designs.
• This method offers a more comprehensive understanding of intervention effects in SCDs.
• The integrated procedure facilitates robust inference and cross-design comparability.