CFE 2019: Start Registration
View Submission - CMStatistics
Title: BFF: Bayesian, fiducial, frequentist analysis of age effects in daily diary data Authors:  Shevaun Neupert - North Carolina State University (United States) [presenting]
Jan Hannig - University of North Carolina at Chapel Hill (United States)
Abstract: Age effects in within-person slopes in daily diary data were examined with Bayesian, Generalized Fiducial Inference (GFI), and frequentist paradigms. Daily stressor exposure data across six domains were used to generate within-person emotional reactivity slopes with daily negative affect. Systematic age differences and similarities in these reactivity slopes were tested, which are inconsistent in previous research. 116 older (aged 60-90) and 107 younger (aged 18-36) adults from the Mindfulness and Anticipatory Coping Everyday study responded to daily stressor and negative affect questions each day for eight consecutive days, resulting in 1,627 total days. Daily stressor domains included arguments, potential/avoided arguments, work/volunteer stressors, home stressors, network stressors, and health-related stressors. Using Bayesian, GFI, and frequentist paradigms, results for each of the six stressor domains with a focus on interpreting age effects in within-person reactivity were compared. As an example, frequentist multilevel models and Bayesian models with main effects of arguments (reactivity slope), age, and their interaction predicting negative affect suggested no age differences in reactivity. However, the GFI solution suggested that older adults were less reactive than younger adults. GFI is a useful tool that provides additional information when making determinations regarding null age effects in within-person slopes.