Table 5.

Multilevel Linear Models Predicting Participants’ Mood Throughout the Day From Closeness of Friendships

	Positive mood		Negative mood
Variable	B	SE	B	SE
Fixed effects
Intercept	3.41***	0.05	1.21***	0.02
Within-person effectsa (assessment-level)
Social convoy friend	0.02	0.02	0.01	0.01
Non social convoy friend	0.08***	0.01	−0.02*	0.01
Romantic partner	0.02	0.02	0.01	0.01
Family member	0.05***	0.01	0.02*	0.01
Other social partner	−0.02	0.01	0.01	0.01
Between-person effectsb
Social convoy friend	−0.12	0.25	−0.10	0.10
Non social convoy friend	0.14	0.23	0.06	0.09
Romantic partner	0.09	0.14	0.03	0.06
Family member	0.09	0.15	0.15*	0.06
Other social partner	0.03	0.21	0.07	0.09
Covariates
Agec	−0.01	0.01	0.00	0.00
Genderc	−0.06	0.09	0.06	0.04
Educationc	−0.03	0.03	0.00	0.01
Healthc	0.12**	0.04	−0.08***	0.02
Minority statusc	0.09	0.11	−0.07	0.04
% of friends in social convoyc	0.26	0.26	0.09	0.10
% of in-person contactd	0.32	0.22	−0.17	0.09
Diverse activityd	0.02***	0.00	0.00	0.00
Random effects
Intercept VAR (Level 2: Participant)	0.47***	0.04	0.08***	0.01
Residual VAR	0.17***	0.00	0.06***	0.00
−2 log-likelihood	7570.3		1420.8

Notes. Assessments n = 6,262 from 313 participants. Positive and negative mood were continuous outcomes, 1 (not at all) to 5 (a great deal).

^aThe predictor was binary variables representing any encounters with four different types of social partners.

^bThe person mean variables representing the proportion of assessments across study that each participant had encounters with different types of social partners.

^cParticipant-level covariates.

^dAssessment-level covariates.

*p < .05. **p < .01. ***p < .001.