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

Published in final edited form as: Psychol Trauma. 2016 Jul;8(4):535–540. doi: 10.1037/tra0000157

Table 3.

Multilevel Model Analyses with Mistrust Predicting Use of Physical and Psychological IPA

	Final Model
Physical IPA Use	β	SE	t

Constant	0.10	0.10	0.98
Gender	−0.20	0.07	−3.04^**
A Mistrust	0.24	0.07	3.61^***
P Mistrust	0.25	0.07	3.75^***

Psychological IPA Use	β	SE	t

Constant	−0.02	0.09	−0.20
Gender	0.04	0.06	0.67^a.
A Mistrust	0.29	0.06	4.61^***
P Mistrust	0.33	0.06	5.21^***

Note. All IPA variables are from combined reports from both members of the couple. Any differences in statistical significance with self-reported IPA use are noted in the table. Square-root transformed variety scores were used for physical IPA, and square-root transformed frequency scores were used for psychological IPA. All analyses are based on standardized variables, with the exception of Gender, which was coded as 0 = female, 1 = male.

Abbreviations: IPA, Intimate Partner Aggression; A, Actor Effect; P, Partner Effect.

^**

p <.01.

^***

p <.001.

^a.

Significant with self-reported IPA use in negative direction