It is often said that "As the twig is bent, so grows the tree." But who bent the twig? Of course, children are like trees in that many factors serve to mold and shape their lives - family life, in particular. Parish and Nunn (1989) reported a significant relationship between college students' evaluations of their families and their ratings of themselves, their mothers, and their fathers. In other studies, family happiness (e.g., Parish, Dostal, & Parish, 1981) and family structures (e.g., Nunn & Parish, 1982) were likewise found to be significantly related to these same variables.
In another study by Parish (1988), college students' self-concepts were found to be significantly related to how "loving" their fathers were perceived to have been toward their wives, and vice versa. In addition, Parish and Necessary (in press) surveyed middle-school students and found that their self-concepts were related to fathers' perceived actions toward their mothers, though not their mothers' actions toward their husbands. They also reported that ratings of fathers generally varied with how loving they were toward their wives and how loving their wives were toward them. Ratings of mothers, however, varied only with how loving they were toward their husbands, but not vice versa.
These findings leave the impression that many familial relationships are associated with one another. The problem, though, is that an overall picture of these various interactions between family members is still basically unavailable. Therefore, the present study sought to examine the various correlations among all of the following variables: college students' self-concepts; evaluations of fathers; evaluations of mothers; rating regarding how respondent acts toward his/her father; rating regarding how respondent acts toward his/her mother; rating regarding how respondent's father acts toward wife; rating regarding how respondent's mother acts toward husband.
In the present study, 128 college students enrolled at a large Midwestern university, voluntarily completed the following forms in a counterbalanced fashion.
Each respondent completed three forms of the Personal Attribute Inventory (PAI; Parish, Bryant, & Shirazi, 1976). The PAI consists of 100 alphabetically arranged adjectives (i.e., 50 positive and 50 negative), from which the respondent is instructed to select exactly 30 that best describe the individual in question. The score on each form is the number of negative adjectives checked. Forms 1, 2, and 3 sought to assess each respondent's ratings of himself/herself, his/her mother, and his/her father, respectively.
In addition, each respondent completed four forms of the Love/Hate Checklist (LHC; Parish, 1988). The LHC consists of 90 alphabetically arranged adverbs (i.e., 45 loving and 45 hateful), from which the respondent is instructed to select 25 that best describe how - - - - acts toward - - - -. Forms 1 and 2 sought to assess each respondent's ratings of how he/she acts toward his/her father and his/her mother, respectively. Forms 3 and 4 sought to assess each respondent's ratings of how his/her father acts toward his wife, and his/her mother acts toward her husband, respectively.
RESULTS AND DISCUSSION
A series of Pearson product-moment correlations was computed between the scores of each of the above-mentioned scales.
Respondents' self-concepts were found to be significantly correlated with evaluations of their fathers (r = 0.28, p [less than] .001) and their mothers (r = 0.24, p [less than] .01) (as measured on the PAI), how they act toward their fathers (r = 0.30, p [less than] . …