Meta-Analytic Methods to Detect Publication Bias in Behavior Science Research

Area of Science:

• Behavioral science research
• Psychology
• Single-case experimental designs

Background:

• Publication bias is a significant concern across scientific disciplines.
• Its prevalence and evaluation methods are less documented in behavior science fields.
• Single-case experimental designs (SCEDs) present unique challenges for bias assessment.

Purpose of the Study:

• To explore and present meta-analytic techniques for evaluating publication bias in behavior science research.
• To provide guidance for researchers using or adapting these methods for SCEDs.
• To address the challenges of identifying grey literature in meta-analyses.

Main Methods:

• Description of meta-analytic techniques for examining publication bias when published and grey literature are available.
• Presentation of alternative meta-analytic techniques for situations where grey literature is inaccessible.
• Adaptation of methods primarily used for group design studies for SCEDs.

Main Results:

• The study provides preliminary guidance on applying meta-analytic techniques to evaluate publication bias in behavior science.
• Sample datasets and R scripts are offered to facilitate understanding and application of statistical analyses.
• The research aims to increase understanding of publication bias within the behavior science domain.

Conclusions:

• Evaluating publication bias is crucial for the integrity of behavior science research, particularly for SCEDs.
• The presented meta-analytic techniques offer viable approaches for bias assessment, even with limited grey literature.
• This work encourages behavior scientists to assess and address publication bias in their research.