Search

Summary.-The present study, conducted just prior to the 1973 U. S. Supreme Court decision on abortion, focused on effects of personality characteristics and one healch-profession choice (student nursing vs a liberal am major) upon attitudes towad abortion. Ss were 96 students from an introductory psychology class, 49 student nurses, and 46 liberal arts maiors. It was predicted that student nurses would be more negative toward legalizing abortion than would liberal arts majors. Results supported this prediction (p < .025). Personality characteristics were assessed by the 16 PF (Cattell, Eber, & Tatsuoka, 1970) and were analyzed to yield four second-order factors: anxiety vs adjustment, extraversion vs introversion, subduedness vs independence, and tenderminded emotionality vs alert poise. Anxiery, introversion, subduedness, and tendermindedness were all expected to predispose Ss to be more against abortion than the other halves of these dichotomies. Predictions were supported (ps from .05 to ,025) for all but anxiety. The 1973 U. S. Supreme Courc decision setting aside restrictive stace abortion laws undoubtedly will not represent the end point for this topic, and both intensely held public attitudes and professional concern in the behavioral sciences preceded the decision and continue. Previous research has concentrated on the relation of ind:vidual attitudes and behavior to the issue of abortion; however, none of these studies directly related the person's attitudes toward abortion to personality characteristics. The present study explored chis relation. Early effects of changes in abortion laws were studied by Char and McDermoct (1972) following the legalization of abortion in Hawaii in 1970. They briefly ueated and studied groups of nurses required by their jobs to assist with abortions and who, as a result, were described by the researchers as suffering acute identity crises regarding their nursing roles. Balakrishnan, Ross, Allingham, and Kantner ( 1972 ) , concerned with broad scale explanatory variables, sampled attitudes of over 1600 Canadian women and related individual attitudes to nine demographic variables. Religion, relative frequency of church attendance, and husband's income were the best predictors of these women's attitudes toward abortion. On the other hand, Kane, Lachenbruch, Lipton, Morris, and Baram (1973) compared college girls seeking abortions with a nonpregnant control group and concluded that motivational factors, such as guilt over contraceptive use and impulsivity, may have predisposed many of the pregnant group to become pregnant. Such findings suggest the impact of individual personality characteristics on the issue of abortion. The fact that Balakrishnan, er al. employed nine broad demographic characteristics to explain

185 Ss rated two issues on 16 semantic differential potency scales. Ss were classified for or against the issues on the basis of their responses to two unmarked 9-cm lines descriptively labeled at the end points. Semantic differential data from the four resulting classes were factor analyzed using principal components solutions with varimax rotations. Previous research (Jones, 1969) suggested that these scales might fractionate into several components of variance and that the structure would vary depending on attitudinal valence (for-against the issue). Descriptive comparisons and comparisons employing Tucker's coefficient of congruence (Harman, 1967) supported the hypothesis of dissimilar fractionation.

153 Ss responded to the Sherif-Hovland instruments for 2 issues. Ss then rated their statements judged most acceptable (MA) and most objectionable (MO) on 18 semantic differential scales. Separate factor analyses of the semantic differential scales were run, collapsing across both issues. The patterns of percentages of variance indicated that the 2 semantic structures were probably not identical. Tests of significance were performed for 2 single scales and for factor scores (8) for the first 2 components (evaluation and potency) extracted in the factor analyses. The issues were tested separately, comparing Ss' ratings of their MAs and MOs. The tests were significant and showed that MA was rated “better,” more important, and more acceptable (a check on validity of Ss' responses) than MO; however, MO was judged more potent than MA. It was concluded that the 2 positions, MA and MO, are probably not simply antithetical.

In a pretest-posttest control group design (Campbell & Stanley, 1963) with 2 experimental samples and a treated control sample, 132 Ss responded twice to a Sherif-Hovland instrument (Sherif & Hovland, 1961). The intervening treatments written by E appeared to be Xeroxed copies of newspaper articles: 1 represented primarily the evaluative (E) dimension of meaning (Osgood, Suci, & Tannenbaum, 1957); 1 was a fusion of potency (P) and activity (A), i.e., dynamism (D). A control (C) article was irrelevant to the experimental issue, viz., censorship of movies. Posttest-pretest change scores were compared for the 3 samples by 15 t tests for independent samples (Walker & Lev, 1953). E and D were not significantly different because both changed on the 5 Sherif-Hovland measures in the same directions. Compared with C, however, E was quite effective; 4 of the 5 measures changed significantly; comparisons of D and C indicated that 2 of the 5 measures changed significantly. For this issue E was more effective than D, although more research is needed to clarify the effects of dimensions of meaning of communications on attitude change.