Planned missing-data designs in experience-sampling research: Monte Carlo simulations of efficient designs for...

Area of Science:

• Psychological Science
• Quantitative Psychology
• Research Methodology

Background:

• Experience-sampling methods (ESM) balance survey length, frequency, and duration.
• Reducing items per signal in ESM can improve participant compliance.
• Traditional ESM designs require extensive data collection, posing logistical challenges.

Purpose of the Study:

• To introduce and evaluate planned missing-data designs for experience-sampling research.
• To demonstrate how to reduce the number of items per signal without compromising overall data.
• To assess the performance of these designs under various sample sizes and response rates.

Main Methods:

• Combination of planned missing-data designs and multilevel latent variable modeling.
• Illustration of different designs using real experience-sampling data.
• Two Monte Carlo simulation studies to examine design performance across varying parameters.

Main Results:

• Planned missing-data designs yielded unbiased parameter estimates.
• Slightly increased standard errors were observed, but generally manageable.
• Designs performed well even with extensive missing data in realistic sample sizes.

Conclusions:

• Planned missing-data designs are a viable strategy to optimize experience-sampling research.
• These methods allow for reduced item burden per signal, enhancing participant experience.
• The approach is robust and recommended for researchers utilizing ESM.