. Author manuscript; available in PMC: 2021 Jul 21.

Published in final edited form as: Journal Mass Commun Q. 2017 Jan 19;94(2):486–514. doi: 10.1177/1077699016687724

Table 7.

Estimated CIs of Standardized Betas of Main Predictors.

			CI^b (95%)		50% overlap (two-tailed)^c	80% overlap (one-tailed)^d
DVs^a	Main predictors	β	Lower	Upper	50% overlap (two-tailed)^c	80% overlap (one-tailed)^d
Δ QoL	ES from survivors	.245	.009	.468	.181	.273^*
	ES from new patients	.006	− .214	.204
Δ CES-D	ES from survivors	− .322	− .571	− .108	− .145	.072^*
	ES from new patients	.007	− .267	.221
Δ CIC	IS from survivors	.173	− .092	.445	.176	.247
	IS from new patients	.260	.057	.441
Δ BCK	IS from survivors	− .114	− .420	.180	.066	.154
	IS from new patients	.211	− .081	.536

Note. DV = dependent variables; CI = confidence interval; QoL = quality of life; ES = emotional support; CES-D = Center for Epidemiologic Studies Depression scale (depression); CIC = cancer information competence; IS = informational support; BCK = breast cancer knowledge.

DVs are the gain scores of variables between the pre- and posttest.

CIs were estimated via a bias-corrected bootstrap (1,000 resamples).

The values were calculated by adding half of 50% of the overlapping CIs to the lower bound estimates of the variables with the greater standardized betas. In the event that the upper bound estimates of the variables with the smaller standardized betas are smaller than these values, the overlapping CIs are smaller than 50%, confirming that the betas are statistically different from each other (p < .05, two-tailed).

The values were calculated by adding half of 80% of the overlapping CIs to the lower bound estimates of the variables with the greater standardized betas. In the event that the upper bound estimates of the variables with the smaller standardized betas are smaller than these values, the overlapping CIs are smaller than 80%, confirming that the betas are statistically different from each other (p < .05, one-tailed).

p < .05.