In the widely accepted marital support gap hypothesis it is stated that, compared to men, women receive less support from their spouses, and that the support they do receive is less helpful than what they provide to their husbands (Belle, 1982). However, in numerous observational studies, differences have not been detected between husbands and wives in terms of the amount of emotional support (e.g., reassurance, comfort), instrumental support (e.g., advice, assistance), and unhelpful/negative types of support (e.g., minimizing problems, ignoring support seeker's view) they provide to their support-seeking spouse (see e.g., Verhofstadt, Buysse, & Ickes, 2007).
It should be noted, however, that these studies were focused exclusively on biological sex differences (male vs. female) in observed support provision, thereby overlooking the role of spouses' gender identity. It could, however, be expected that the extent to which a spouse possesses traditionally considered as masculine qualities (e.g., self-reliant, independent, competitive) or traditionally considered as feminine qualities (e.g., warm, understanding, compassionate) may have an effect on his/her method(s) of providing support to a partner in distress.
We assumed that individual differences in spousal support provision are subtle and depend on an interaction between biological sex and gender identity. More specifically, we expected differences between husbands and wives in the way they support each other, but only for spouses with traditional sex-typed orientations, that is, husbands characterized as masculine and wives characterized as feminine.
Fifty married couples participated in an observational laboratory study. They provided questionnaire data and participated in two 10-minute support interaction tasks, designed to assess each spouse's support provision behavior. The videotaped interactions were subsequently coded for emotional, instrumental, and unhelpful/negative support provision behavior and proportions were computed using the Social Support Interaction Coding System (SSICS; Bradbury & Pasch, 1992). The Bem Sex Role Inventory (BSRI; Bem, 1974) was used to determine husbands' (masculine husbands: n = 24; other: n = 26) and wives' (feminine wives: n = 23; other: n = 27) gender identity employing a double median split procedure.
A repeated measures analysis of variance revealed a significant biological sex x husbands' gender identity x wives' gender identity interaction for observed instrumental support, F(1, 46) = 3.77, p = .05. Wives provided more instrumental support to …