Table 4.

Lagged multilevel model results.

	Positive mood	Negative mood	Disclosure	Support	PCE-diabetes	PCE-general	Self-care
Intercept	1.11***	1.14***	2.45***	1.66***	2.09***	2.47***	2.40***
CGM	0.01	−0.01	0.13	0.00	0.01	−0.05	0.02
Openness	0.17**	0.04	−0.03	−0.07	0.17*	0.15	0.02
Conscientiousness	0.03	−0.01	0.06	0.04	0.01	−0.01	0.15*
Day	0.00	−0.01**	− 0.02**	−0.01+	−0.01+	0.01*	0.00
Previous day DV	0.38***	0.21***	−0.02	0.30***	0.25***	0.21***	0.33***
WI play	0.11***	− 0.07***	0.18***	0.16***	0.07*	0.04	−0.01
BW play	0.25***	−0.10*	0.77***	0.62***	0.05	0.02	0.07

Note. N = 199; PCE = perceived coping effectiveness; CGM = continuous blood glucose monitor; DV = dependent variable; WI = within-person variability; BW = between-person variability. For self-care, the the analysis indicated that the final Hessian matrix was not positive definite although convergence criteria had been satisfied. We removed the random slope for within-person variability in play, and changed the random effect covariance matrix to identity. When this was done the models converged and ran without errors. Results presented in the table reflect the model without the random slope for within-person variability in play.

⁺p ≤ .10;

^*p ≤ .05;

^**p ≤ .01;

^***p ≤ .001.