Nancy M. Henley

Developing a Scale to Measure the Diversity of Feminist Attitudes

This article reports two studies aimed at developing a new scale of attitudes toward women, incorporating six theoretically derived perspectives: conservative; liberal, radical, socialist, and cultural feminist; and women of color (womanist). In Study 1, an initial pool of 306 items was given to 117 respondents. Selection of 10 attitudinal and 3 behavioral items from each perspective for the final scale was based chiefly on item-total correlations. In Study 2, the Feminist Perspectives Scale was administered to 344 respondents and showed reasonable reliability and validity. For Femscore, a composite of five feminist attitudinal subscales, Cronbachs alpha was .91; test–retest correlations were .91 at 2 weeks and .86 at 4 weeks. Empirical data provided generally good support for the six perspectives, and factor analysis produced partial support for the six-perspective structure.

DEVELOPMENT OF THE SHORT FORM OF THE FEMINIST PERSPECTIVES SCALE

Development and testing of a 36-item version of the Feminist Perspectives Scale are reported. Study 1 (N = 209) found high internal consistency for Femscore (α = .85). Alpha was greater than .70 for several subscales, although some showed reliability sufficient only for testing large groups. Test-retest correlations for all but one subscale were between .77 and .87. Validity was demonstrated by correlations between short and long forms (Femscore r = .87), among subscales, and with demographic/attitudinal variables. In Study 2 (N = 95), relationships to the Attitudes toward Women Scale (AWS) and Bem Sex Role Inventory (BSRI) were examined. The AWS was correlated negatively with the Conservative subscale and positively with the Liberal Feminist and Fembehave subscales. BSRI femininity was also positively correlated with the latter two subscales.

THE LANGUAGE OF BLACK GAY MEN'S SEXUAL BEHAVIOR: IMPLICATIONS FOR AIDS RISK REDUCTION

The development of appropriate AIDS risk reduction interventions targeted at African-American gay men could be aided by an awareness of their terminology for specific sexual behaviors and types of sexual encounters. This paper explores similarities and differences between the HIV-related sexual language of Black and White gay men. While much of the vernacular is shared, differences in some terms and greater or lesser emphasis on others are apparent.

Frequency and Specificity of Referents to Violence in News Reports of Anti-gay Attacks

Two studies were conducted to test the hypotheses that: (1) news reports of anti-gay attacks would use fewer and vaguer referents to the violence than similar stories about attacks against straight persons; and (2) this milder use of referents would cause readers to perceive less harm done and to blame the perpetrator less. A content analysis of two newspapers found that one used far fewer, less specific nominals to refer to anti-gay than to anti-straight violence, whereas the other, based in a more gay-friendly community, did not differentiate significantly by sexual orientation. An experimental study in which frequency and specificity of referents were systematically varied in mock newspaper stories found that greater referent frequency, but not specificity, caused readers to perceive greater harm to victims. The results are interpreted in terms of cognitive processing and within the context of the use of linguistic variation to encode and enforce power differences.

MEASURING FEMINIST ATTITUDES

We wish to thank the authors of these commentaries for their thoughtful remarks and for the encouragement they have expressed for our endeavor. These conimentaries will help us and other readers in our efforts to understand better the measurement of attitudes about women and gender-related issues. There are several themes that are recurrent in these commentaries. Most notable is the recognition of the importance of acknowledging and learning more about the diversity of feminist theoretical positions. All three commentaries call for further research in this area, urging either further development of the Feminist Perspectives Scale (FPS) or similar instruments, application to other questions in feminist theory, or revision of feminist theory to reflect the empirical data. Both Frieze and McHugh (1998) and Spence (1998) agree with us that the Attitudes Toward Women Scale (AWS; Spence & Helmreich, 1972) may be conceived as a measure of primarily liberal feminist attitudes. All three authors take up the difficulty we encountered in measuring liberal feminist attitudes. Spence provides a detailed examination of subscale items and points to the need for more thought and development of this subscale to bring the items more in line with the philosophy described for this perspective. Hyde (1998) endorses one of the interpretations we offer in our article-that the widespread acceptance of liberal feminist attitudes by diverse persons may underlie the lack of a coherent liberal feminist factor. She also interprets this finding as meaning “that the positions and attitudes purportedly espoused by liberal feminists do not have any coherence in real people’s answers” (p. 362). As this point is made within the larger context of a discussion of the relation between feminist theory and empirical data, we address it in a separate discussion below. Both Spence and Hyde address the question of‘ the independence of the FPS subscales, although in dfferent ways. Spence relates her discussion to our presumed

Women as a Social Problem

In recent years women have become increasingly visible in social issues recognized by our society. These years have seen the successive parade of the issues of sex roles, woman battering, rape, employment discrimination, sexual harassment, incest, pornography, the aged, and homeless (“shopping bag”) women, followed by a succession of attempted remedies. As social scientists, we may wonder why these particular issues have been targeted for action as social problems; how particular responses to them are carried out; and why the intended solutions may not be lasting. In this chapter I wish to examine the definition and redefinition of social issues, with special attention to women’s relation to them, and to women as social problems themselves. (In this discussion, although I may use the term social problems more often than social issues, I will use the terms almost interchangeably.1)