A power analysis for fidelity measurement sample size determination.

Area of Science:

• Research Methodology
• Psychology
• Education

Background:

• Assessing treatment fidelity is increasingly vital for accurate outcome analysis.
• Current methods for fidelity measurement, often observational, lack clear guidance on sample size.
• Precisely measuring fidelity is essential for understanding intervention effectiveness.

Purpose of the Study:

• To propose a practical methodology for determining adequate sample sizes for fidelity data collection.
• To provide guidance on sample size determination for observational fidelity assessments.
• To enhance the precision and reliability of fidelity measurements in research.

Main Methods:

• A novel methodology for sample size calculation is presented, considering statistical power for treatment effects and fidelity-outcome relationships.
• The approach utilizes probability sampling from a finite population framework.
• Illustrative example provided for a time-based fidelity measure over a study duration.

Main Results:

• A practical method for calculating sample size for fidelity observation is proposed.
• The methodology integrates considerations of statistical power and fidelity-outcome linkage.
• Software for sample size calculation is made available to researchers.

Conclusions:

• The proposed method offers a systematic approach to determining sample size for fidelity assessment.
• Implementing this method can lead to more precise and reliable fidelity data.
• This contributes to stronger evidence in intervention research by ensuring adequate fidelity measurement.