Table 1.

Fit Indices for morning and evening growth mixture model solutions over seven assessments, with dyad as a clustering variable.

GMM	LL	AIC	BIC	Entropy	VLMRc
Morning Fatigue
1-Classa	−2729.675	5491.349	5547.820	n/a	n/a
2-Class	−2667.797	5377.594	5451.712	0.775	123.755n.s.
3-Classb	−2641.862	5335.724	5427.489	0.811	51.870**
4-Class	−2629.867	5321.734	5431.146	0.843	23.990n.s.
Evening Fatigue
1-Classd	−2837.094	5706.188	5762.659	n/a	n/a
2-Class	−2808.314	5662.627	5743.804	0.662	57.561**
3-Classe	−2793.976	5643.951	5742.775	0.716	28.116*
4-Class	−2779.614	5627.227	5747.228	0.731	31.791n.s.

^*p < .05,

^**p < .01,

^n.s.p > .05.

Abbreviations: AIC = Akaike Information Criteria; BIC = Bayesian Information Criterion; CFI = comparative fit index; GMM = Growth mixture model; LL = log likelihood; n/a = not applicable; n.s. = not significant; RMSEA = root mean square error of approximation; VLMR = Vuong-Lo-Mendell-Rubin likelihood ratio test.

^aRandom coefficients latent growth curve model with linear and quadratic components; Chi² = 60.528, 26 df, p = .0001, CFI = 0.962, RMSEA = 0.073.

^b3-class model was selected, based on its having the smallest BIC and a significant VLMR. Further, the VLMR is not significant for the 4-class model, and the 4-class model estimated a class with only 1.5% of the sample – a class size that is unlikely to be reliable.

^cThis number is the Chi² statistic for the VLMR for morning and evening fatigue. When significant, the VLMR test provides evidence that the K-class model fits the data better than the K-1-class model.

^dRandom coefficients latent growth curve model with linear and quadratic components; Chi² = 78.126, 26 df, p < .00005, CFI = 0.965, RMSEA = 0.089.

^e3-class model was selected, based on its having the smallest BIC and a significant VLMR. Further, the VLMR is not significant for the 4-class model.