The importance of the family in promoting the psychosocial development of children has been well documented in the literature (e.g., Kagan, 1980; Parish, 1987). According to Byrne (1977), the family represents a basic human support system within which various needs are met, or go unmet. Thus, degree of family happiness has been found to be significantly related to individuals' subsequent level of self-esteem, as well as regard for others within the family (see Parish, Dostal, & Parish, 1981; Parish & Nunn, 1988).
That the family serves as a basic support system potentially associated with how individuals come to see themselves and other family members is important to understand, but another important issue is, "Don't parents' actions toward one another play a similar role?" To date, Parish (1988) has reported that college students' self-concepts were significantly related to how their fathers acted toward their mothers (r = .61, p [is less than] .01), and how their mothers acted toward their fathers (r = .58, p [is less than] .01), but Parish and Necessary (1993) failed to find similar relationships with high school students. With such disparate findings already available, one wonders what relationships could prevail with younger children (i.e., middle-schoolers) of either or both sexes, and do perceived actions by parents also directly relate to how these children evaluate their parents? Finally, is their a significant relationship between how parents act toward one another, i.e., does hatefulness truly beget hatefulness? These questions were addressed in the present study.
A total of 186 sixth- to eighth-grade students (98 females and 88 males) enrolled in a large midwestern middle school voluntarily participated in the study. Each student completed the following forms:
The Personal Attribute Inventory for Children (PAIC; Parish & Taylor, 1978). This survey consists of 48 words (24 positive and 24 negative adjectives), from which each student was instructed to select exactly 15 that best describe the target in question (Form A = themselves; Form B = their father; Form C = their mother). The score on each of the three forms of the PAIC is the number of "negative" words checked.
The Love/Hate Checklist for Children (L/HCC). This instrument was derived from the Love/Hate Checklist originally developed for adult populations by Parish (1988). The L/HCC consists of 30 words (15 loving and 15 hateful adverbs), from which each student was asked to select the 10 that best describe how one parent acts toward the other parent (Form A = how father acts toward mother; Form B = how mother acts toward father). The score on each of the two forms of the L/HCC is the number of "hateful" words checked.
RESULTS AND DISCUSSION
A series of Pearson product-moment correlations revealed several significant relationships between how fathers were perceived to act toward mothers, and vice versa, and how the children evaluated themselves, their fathers, and their mothers, as well as how their parents' perceived actions correlated with each other. Notably, these correlations varied markedly as a function of the gender of the respondent.
Regarding self-concepts, the findings from this study varied substantially from the findings reported earlier by both Parish (1988) and Parish and Necessary (1993). Specifically, while Parish (1988) found highly significant relationships between college students' self-concepts and their parents' actions, and Parish and Necessary (1993) found no TABULAR DATA OMITTED such relationship with high school students, in the present study there was only a modest significant correlation overall (r = .26, p [is less than] .01), and for each of the sex groups ([r.sub.m] = .23, p [is less than] .05; [r.sub.f] = .32, p [is less than] .05) between children's self-concepts and fathers' actions toward their wives. Regarding mothers' actions, they were significantly related only to their daughters' ratings of themselves ([r. …