. Author manuscript; available in PMC: 2017 Oct 11.

Published in final edited form as: Emerg Adulthood. 2017 Apr 25;5(4):268–279. doi: 10.1177/2167696817704118

Table 3.

Model estimates for Aim 1 analyses (whole analytic sample)

	Model 1a: Positive affect	Model 1b: Loneliness	Model 1c: University activities	Model 1d: Alcohol use
Fixed Effects
Means
Intercept (γ₀₀)	2.48^***	1.31^***	−4.01^***	−4.30^***
Female (γ₀₁)	0.07	−0.05	−0.53^***	0.07
Location (percent of days spent on-campus; γ₀₂)	−0.13	0.03	0.16	−0.17
Single (γ₀₃)	−0.13	0.13	0.35^*	0.14
GCDR (γ₀₄)	−0.13	−0.03	0.19	0.50
Slopes
Weekend day (γ₁₀)	0.08^***	0.00	−0.33^***	1.70^***
On-campus (γ₂₀)	−0.25^***	0.32^***	0.89^**	0.60^***
On-campus*Single (γ₂₁)	0.17^**	−0.26^***	–	–
On-campus*GCDR (γ₂₂)	0.19	−0.27^*	–	–
Random Effects
Level-1 effect (ε_ij)	0.29^***	0.42^***	–	–
Level-2 effects Intercept (ζ_0i)	0.42^***	0.48^***	1.93^***	2.41^***

Note.

p < .05

^**

p < .01

^***

p < .001 Gender is coded 1 = male, 0 = female. GCDR = Geographically close dating relationship. Weekend day is coded 1 = weekend day, 0 = weekday. Within-person location is coded 1 = on-campus, 0 = off-campus. University activities is coded 1 = university activity participation, 0 = university activity participation. Alcohol use is coded 1 = one or more drinks, 0 = no drinks. Single and GCDR are dummy coded with LDDR as the reference group. Models 1c and 1d were modeled via a logit equation, and thus, do not provide a residual error term.