Optimizing Measurement Reliability in Within-Person Research: Guidelines for Research Design and R Shiny Web Application Tools

• Published in: Journal of business and psychology Vol. 37; no. 6; pp. 1141 - 1156
• Main Authors: Yang, Liu-Qin, Wang, Wei, Huang, Po-Hsien, Nguyen, Anthony
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.12.2022; Springer Nature B.V
• Subjects: Behavioral Science and Psychology; Business and Management; Community and Environmental Psychology; Industrial and Organizational Psychology; Monte Carlo simulation; Organizational behavior; Original Paper; Personality and Social Psychology; Psychology; Reliability; Research design; Social Sciences; Validity; Within-person research; Validity; Level-specific alpha; R shiny web application; Reliability
• ISSN: 0889-3268; 1573-353X
• DOI: 10.1007/s10869-022-09803-5

Within-person research has become increasingly popular over recent years in the field of organizational studies for its unique theoretical and methodological advantages for studying dynamic intrapersonal processes (e.g., Dalal et al., Journal of Management 40:1396–1436, 2014; McCormick et al., Journal of Management 46:321–350, 2020). Despite the advancements, there remain serious challenges for many organizational researchers to fully appreciate and appropriately implement within-person research—more specifically, to correctly conceptualize and compute the within-person measurement reliability, as well as navigate key within-person research design factors (e.g., number of measurement occasions, T ; number of participants, N ; and scale length, I ) to optimize within-person reliability. By conducting a comprehensive Monte Carlo simulation with 3240 data conditions, we offer a practical guideline table showing the expected within-person reliability as a function of key design factors. In addition, we provide three easy-to-use, free R Shiny web applications for within-person researchers to conveniently (a) compute expected within-person reliability based on their customized research design, (b) compute observed validity based on the expected reliability and hypothesized within-person validity, and (c) compute observed within-person (as well as between-person) reliability from collected within-person research datasets. We hope these much-needed evidence-based guidelines and practical tools will help enhance within-person research in organizational studies.