Table 3.

Change in Loneliness Scores Over Time, as Modified by Social Support

	Unstandardized (standardized) estimates	Standard error	p Value
Loneliness intercept × friends	−3.35 (−.54)	0.34	<.001
Loneliness intercept × family	−0.23 (−.20)	0.06	<.001
Loneliness slope × friends	0.11 (.16)	0.08	.21
Loneliness slope × family	0.01 (.11)	0.40	.40

Notes. Latent growth curve (LGC) was used to identify how social support from friends and family modified the loneliness trajectory (slope and intercept). Refer back to Model B of Figure 1. This model assumed linear change and fit the data well, χ² (df = 14) = 15.385, p = .352; root mean square error of approximation = .018.