Table 1.

EAR-derived daily interaction variables: Reliabilities and correlations with well-being

		How was participants’ well-being related to their daily interactions?			Did the effects differ on weekdays and weekend?		Did personality differences account for the effects?
EAR-derived variables	Intercoder reliability	Well-being index	SWLS measure	Happiness measure	Well-being index		Well-being index resid. for Big Five
					Weekdays	Weekendb
Alone	.97	-.35**	-.36**	-.27*	-.29**	-.35**	-.40**
Talking to others	.95	.31**	.31**	.26*	.30**	.30**	.39**
Small talk	.76	-.07	-.03	-.10	-.01	-.09	.08
% of all conversations	---a	-.33**	-.25*	-.35**	-.30**	-.34**	-.17
Substantive conversations	.84	.31**	.26*	.30**	.27*	.31**	.36**
% of all conversations	---a	.28**	.20	.31**	.28**	.27*	.22*

Note: N = 79

^*p ≤ .05 (two-tailed)

^**p < .01 (two-tailed)

the variables are expressed as proportions of the total number of sampled sound files; %-variables are expressed as proportion of total number of conversations (i.e., the variable 'Talking to others'); intercoder reliabilities were computed from a training set of 221 EAR sound files which were independently coded by all coders (ICC[2,k]); the happiness measure is the average of self- and informant-ratings of happiness; the well-being index is the aggregate of participants’ self-reported life satisfaction (SWLS) and the happiness measure; resid. = residualized; SWLS = Satisfaction with Life Scale; Big Five = composite of self- and informant-reports of all five Big Five dimensions

^ano reliability reported because the variable is a quotient of two coded variables

^bdefined as Friday 6:00 pm to Sunday at midnight.