. Author manuscript; available in PMC: 2017 Jun 1.

Published in final edited form as: Traumatology (Tallahass Fla). 2016 Jun;22(2):85–93. doi: 10.1037/trm0000063

Table 3.

Estimated Magnitudes of Indirect and Direct Effects at Low, Average, and High Levels of Constructive Thinking

	Indirect effect via belief violations		Indirect effect via intrinsic goal violations		Indirect effect via extrinsic goal violations		Direct effect

	Coeff.	95% CI	Coeff.	95% CI	Coeff.	95% CI	Coeff.	95% CI
Low	.0130^*	[.0030, .0285]	.0469^*	[.0225, .0783]	.0027	[−.0117, .0204]	.0692^*	[.0252, .1131]
Average	.0110^*	[.0034, .0231]	.0363^*	[.0179, .0596]	.0024	[−.0099, .0172]	.0416^*	[.0090, .0742]
High	.0090^*	[.0001, .0235]	.0257^*	[.0120, .0453]	.0020	[−.0080, .0158]	.0140	[−.0283, .0562]

Note. Coeff. = b coefficients (not betas); all confidence intervals were computed based on bootstrap estimates; note that although different estimates of the indirect effect at low, average, and high values of constructive thinking are estimated and listed in this table for belief violations and extrinsic goal violations, the test of moderated mediation showed that these indirect effects did not significantly vary as a function of constructive thinking (only the indirect effect via intrinsic goal violations significantly varied).

p<.05