Non-normal Data in Repeated Measures ANOVA: Impact on Type I Error and Power.

BACKGROUND: Repeated measures designs are commonly used in health and social sciences research. Although there are other, more advanced, statistical analyses, the F-statistic of repeated measures analysis of variance (RM-ANOVA) remains the most widely used procedure for analyzing differences in means. The impact of the violation of normality has been extensively studied for between-subjects ANOVA, but this is not the case for RM-ANOVA. Therefore, studies that extensively and systematically analyze the robustness of RM-ANOVA under the violation of normality are needed. This paper reports the results of two simulation studies aimed at analyzing the Type I error and power of RM-ANOVA when the normality assumption is violated but sphericity is fulfilled. METHOD: Study 1 considered 20 distributions, both known and unknown, and we manipulated the number of repeated measures (3, 4, 6, and 8) and sample size (from 10 to 300). Study 2 involved unequal distributions in each repeated measure. The distributions analyzed represent slight, moderate, and severe deviation from normality. RESULTS: Overall, the results show that the Type I error and power of the F-statistic are not altered by the violation of normality. CONCLUSIONS: RM-ANOVA is generally robust to non-normality when the sphericity assumption is met.

Measurement Invariance and Validity of the Satisfaction With Life Scale in Informal Caregivers.

BACKGROUND: The number of informal caregivers within ageing population is increasing and there is a growing research interest to promote their well-being, and therefore there is a need for adequate measurement tools. We aim to provide validity evidence of the Satisfaction with Life Scale (SWLS) in a representative sample of British older adults, including measurement invariance across caregivers and non-caregivers. METHOD: Data was drawn from English Longitudinal Study of Ageing (ELSA). The sample consisted of 3,754 caregivers and 4,036 non-caregivers. The structure and measurement invariance were tested through a confirmatory factor analysis. Reliability and validity evidence based on relationships with other variables were also analysed. RESULTS: Our results supported the one-factor structure of the SWLS, CFI = .996; NNFI = .993; RMSEA = 0.081, and measurement invariance across caregiving status. McDonald´s omega was .93. Scores on the SWLS were positively correlated with quality of life, positive social support, and self-reported health, and negatively with loneliness, depression, negative social support, difficulties in activities and instrumental activities of daily living, and number of health conditions. CONCLUSIONS: These findings provide new psychometric evidence to support the use of the SWLS in research which focuses on caregivers and on the comparison with non-caregiver samples.