Measuring Empathy in Couples: Validity and Reliability of the Interpersonal Reactivity Index for Couples

Abstract

This article includes three studies aimed at adapting the Interpersonal Reactivity Index (Davis, 1980), a measure of general cognitive and emotional empathy, to assess empathy expressed within a romantic relationship. We examined the factorial structure of the Interpersonal Reactivity Index for Couples (IRIC). Results supported the use of the 2-factor measure (Dyadic Perspective Taking and Dyadic Empathic Concern) in 3 independent samples (individuals in heterosexual relationships, individuals in same-sex relationships, and partners of heterosexual couples). The IRIC also demonstrated good convergent, discriminant, concurrent, predictive, and incremental validity as well as adequate internal consistency and stability over 18 months.

Acknowledgments

Funding for this study was provided by the Social Sciences and Humanities Research Council of Canada. Preliminary results were presented at the annual meeting of the Société québécoise de recherche en psychologie, Montreal, Quebec, Canada, March 2006. We thank Daniel Coulombe for statistical consultations and all the graduate students at the Couple Research Lab for their assistance with data collection.

Notes

^a n = 895.

^b n = 126.

^c n = 384.

^d n = 100.

^a Due to a factor loading below .30, this item was omitted in further analyses.

^a n = 695.

^b n = 126.

¹The deletion of Items 3 and 9, which had the lowest factor loadings, yielded slightly better model estimates in all samples: individuals involved in heterosexual relationships, χ²(43, N = 695) = 125.42, p <.001; SRMR =.039; CFI =.97; RMSEA =.053; 90% CI [.042,.063]; individuals involved in same-sex relationships, χ²(43, N = 126) = 44.56, p = ns; SRMR =.054; CFI =.97; RMSEA =.041; 90% CI [.000,.080]; male partners of heterosexual couples, χ²(43, N = 192) = 88.64, p <.001; SRMR =.070; CFI =.89; RMSEA =.074; 90% CI [.052,.096]; female partners of heterosexual couples, χ (43, N = 192) = 92.6, p <.001; SRMR =.040; CFI =.91; RMSEA =.077; 90% CI [.055,.099]. Deleting Items 3 and 9, however, did not improve validity or reliability indexes. Accordingly, we kept these two items in the final IRIC questionnaire to keep the IRIC items closer to the original IRI items.

^a n = 192.

^b n = 192.

∗ p <.05.

∗ p <.05.

^a n = 154 men; 541 women.

^b n = 40 men; 86 women.

^c n = 192 men; 192 women.

∗ p <.05.

^a n = 46 men, 56 women.

∗ p <.05.

^a n = 154.

^b n = 541.