An application of Bayesian measurement invariance to modelling cognition over time in the English Longitudinal Study of Ageing

摘要: OBJECTIVES Recommended cut-off criteria for testing measurement invariance (MI) using the comparative fit index (CFI) vary between -0.002 and -0.01. We compared CFI results with those obtained Bayesian approximate MI cognitive function. METHODS used function data from Waves 1-5 of English Longitudinal Study Ageing (ELSA; Wave 1 n = 11,951), a nationally representative sample adults aged ≥50. tested longitudinal (prior difference intercepts/loadings ~N(0,0.01)) in an attention factor (orientation to date, day, week, month) memory (immediate delayed recall, verbal fluency, prospective task). RESULTS Conventional found strong (CFI + 0.002) but either weak or (CFI -0.004). The also supported 9/20 intercepts thresholds were noninvariant factor. This supports rather than invariance. CONCLUSIONS Within ELSA, is suitable analysis not More generally, situations where appropriate are unclear, could alternatively be used.