|
Neosexism Scale
Please note that this section is an archive (last updated in June 2006). [disclaimer]
Sections:
Overview |
Instrument Reviews |
Construct Overviews |
Book Compendium Reviews |
Internet Site Reviews
|
|
|
|
Practical Information
|
Instrument Name:
|
Neosexism Scale
|
Instrument Description:
|
The Neosexism Scale assesses the extent of respondent neosexism, which is defined as a “manifestation of a conflict between egalitarian values and residual negative feelings toward women.” (Ref: 1) There are 11 scale items, some of which were adapted from modern racism scales previously developed. The scale’s items tend to focus on gender versus public policy and gender versus work situation issues. The Neosexism Scale has been developed and tested primarily in the context of U.S. and Canadian cultures, although some studied use of the scale has occurred in the U.K. The scale attempts to employ subtle enough questions so that a respondent is able to express existent neosexist attitudes without having to go so far as to admit that s/he holds a socially undesirable belief (e.g., that women are inferior to men). (Ref: 1-2)
|
Price:
|
Free (Items are included in literature described below.)
|
Administration Time:
|
No information found.
|
Publication Year:
|
1995
|
Item Readability:
|
Flesch-Kincaid Reading Level of 8.2. Each item contains approximately 15 words.
|
Scale Format:
|
An 11-item, seven-point Likert-type scale (1=”total disagreement” and 7=”total agreement”).
|
Administration Technique:
|
Self-administered.
|
Scoring and Interpretation:
|
Two items (items #2 and #11) are recoded inversely. (Ref: 1) Responses to all scale items are summed and divided by 11 (total number of items); total scale scores therefore range from 1 to 7, with a higher score indicating a more neosexist attitude (i.e., the presence of more gender prejudice).
|
Forms:
|
English and French versions are reported.
|
|
|
Research Contacts
|
Instrument Developers:
|
Francine Tougas, Rupert Brown, Ann M. Beaton, and Stéphane Joly.
|
Instrument Development Location:
|
University of Ottawa
School of Psychology
P.O. Box 450
Stn. A
Ottawa
Ontario K1N 6N5, Canada
|
Instrument Developer Email:
|
ftougas@uottawa.ca
|
Instrument Developer Website:
|
No information found.
|
|
|
Annotated Bibliography
|
1. Tougas F, Brown R, Beaton AM, Joly S. Neosexism: Plus ga change, plus c’est pareil. Pers Soc Psychol Bull 1995 Aug;21(8):842-9.
 |
Purpose: To investigate the nature, antecedents, and consequences of neosexism (a new, more modern type of gender prejudice); to develop and test a new scale for measuring neosexist attitudes. |
|
Sample: Two studies were conducted. The first study involved N=130 male students enrolled in introductory psychology courses in a Canadian university; ages ranged from 18 to 43 years, with a mean age of 21.6 years. The second study included N=149 male workers whose ages ranged from 26 to 60 years, with a mean age of 41.5 years. The workers had an average of 14 years of experience in a large Canadian firm where an affirmative action program for women and minorities had an implementation history of five years. |
|
Methods: The two studies were conducted in order to evaluate: 1) a path analytic predictive model of attitudes toward affirmative action; and 2) practical implications of neosexism. Study 1 employed measures of collective interest, old-fashioned sexism, and general attitudes toward affirmative action. Participants completed questionnaires at baseline and again one month later. Path analysis was used to determine the effect of neosexist attitudes on attitudes toward affirmative action. Study 2 employed all measures used in Study 1, as well as measures of (a) attitudes toward the affirmative action program in place and (b) an evaluation of the competence of women. Participants completed all questionnaires at a single point in time. Regression analysis was conducted to determine the effect of neosexist attitudes on (a) attitudes toward affirmative action, (b) attitudes toward the actual affirmative action program in place at the work site, and (c) the evaluation of women’s competence. In Study 1, both English and French versions of the Neosexist Scale were used; Study 2 was conducted in French only. |
|
Implications: Although construct validity of the Neosexism Scale has been studied with men, its validity among women must also be investigated. In addition, further work is required to determine if the Neosexism Scale is a general gender prejudice measure or whether it is more a measure intertwining such constructs as employment inequality and/or employment competition. |
2. Campbell B, Schellenberg EG, Senn CY. Evaluating measures of contemporary sexism. Psychol Women Q 1997 Mar;21(1):89-101.
|
Purpose: To evaluate and compare two measures of contemporary sexist attitudes (the Modern Sexism Scale and the Neosexism Scale) and their correlates. |
|
Sample: N=106 Canadian college students enrolled in undergraduate psychology courses at the University of Windsor, located two miles from downtown Detroit. 71 (67%) were female. Sample ages ranged from 18 to 40 years, with a mean age of 20.48 years (SD=2.58). Other sample characteristics: twenty-three percent were of non-European ancestry; 77% considered themselves “liberal” or “very liberal” in political viewpoint; 83% classified their sexual orientation as being exclusively heterosexual. Frequency of religious service attendance was also reported: sixteen percent attended weekly; 19% monthly; 48% yearly; 17% never. |
|
Methods: Participants were each given a set of self-administered questionnaires to complete. They had two sexism measures (the Modern Sexism Scale and the Neosexism Scale) and four validity measures (the Attitudes Toward Feminism and the Women’s Movement Scale, the Attitudes Toward Lesbians and Gay Men Scale–short version, the Protestant-Ethic Scale, and the Humanitarian-Egalitarian Scale), all of which to respond to in a single sitting. In addition, participants were asked to complete the Marlowe-Crowne Social Desirability Scale–short version and a demographic questionnaire. Participants were seated separately when filling in their questionnaires so as to be assured of response confidentiality. Reliability, correlation, and regression analyses were conducted to compare the two sexist scales. |
|
Implications: The Neosexism Scale seemed subtle enough to measure modern less overt prejudices based on gender. The scale allowed both male and female respondents to express their underlying gender/political/employment attitudes without being pushed to express blatant beliefs that women were, for example, inferior to men. Nevertheless, absolute separation between male and female mean scale responses was minimal (0.90); dichotomizing mean scores into agree/disagree categories showed that males and females would both disagree overall with the scale’s gender prejudicial statements. College-aged participants may not have enough work experience to be aware of modern gender-related labor practices.
|
3. Masser B, Abrams D. Contemporary sexism: The relationships among hostility, benevolence, and neosexism. Psychol Women Q 1999 Sep;23(3):503-17.
|
Purpose: To examine the relationship between the Neosexism Scale and the Ambivalent Sexism Inventory (ASI; Glick & Fiske, 1996), to measure the psychometric qualities per scale used, and to investigate scale associations with gender-related attitudes and value orientations. |
|
Sample: Total N-size across three samples was N=907. All participants were from the U.K. There were three samples, one each drawn from three populations. Sample 1-School was drawn from a population of 17-year-old students studying for university-entry qualifications (n=295, with 148 males and 147 females). Sample 2-University was drawn from a population of undergraduate university students (n=491, with 190 males and 301 females). The mean age of Sample 2 was 23 years. Sample 3-FTE was drawn from a population of full-time employees originating from a number of organizations around Britain (n=121, with 60 males and 61 females). Eighty percent of Sample 3 had received university entry or university-level education; their mean age was 35 years. Participants from all three samples were predominantly White, of British origin, and highly proficient in English. |
|
Methods: Scales were presented to participants in one questionnaire booklet; the questionnaires were all self-administered and completed in a single set time frame, specific to each sample. The presentation of scales was randomized per sample. The self-administered questionnaires were: the Neosexism Scale, the ASI, Attitudes toward Women’s Rights, Attitudes towards Lesbian and Gay Men’s Rights, the Humanitarian-Egalitarian Scale, and the Protestant Ethic Scale. All scales were presented to all participants, with two exceptions: the Humanitarian-Egalitarian Scale and the Protestant Ethic Scale were only presented to two unique subgroups of Sample 2-University. Reliability, correlation, and multiple regression analyses were then conducted to enable a comparison of the two sexism measures. |
|
Implications: The Neosexism Scale was employed for the first time in a non-North American culture (i.e., British). The scale was found to function similarly to previous North American use: Its reliability was comparable, as was its dimensionality (i.e., unitary). In addition, its ability to detect gender-related response differences and its association with other measures of gender attitude and value orientation were also comparable.
|
|
|
top
|
Factors and Norms
|
Factor Analysis Work:
|
An exploratory factor analysis was conducted on the 11-item scale; the analysis was performed using data from 130 male students from a study conducted in Canada. No definite factor structure was obtained. (Ref: 1) In a principal components analysis on the 11-item scale using data from 106 university students, a unidimensional structure was confirmed, although item #11 had a factor loading of only 0.160. (Ref: 2) Unidimensionality of factor structure was also confirmed via principal components analysis performed on data from three British samples. (Ref: 3)
|
Normative Information Availability:
|
No information found.
|
|
|
Reliability Evidence
|
Test-retest:
|
A test-retest reliability coefficient of 0.84 was reported, using a one-month retest time period with N=130 male university students. (Ref: 1)
|
Inter-rater:
|
No information found.
|
Internal Consistency:
|
Cronbach’s alphas of 0.78 (N=130) and 0.76 (N=149) were reported. (Ref: 1) A second study reported a Cronbach’s alpha of 0.81 (N=106.) (Ref: 2) Cronbach’s alpha for three samples of British participants (school-aged, university-aged, and full-time employees) ranged from 0.62-0.85. (Ref: 3)
|
Alternate Forms:
|
No information found.
|
|
|
Validity Evidence
|
Construct/ Convergent/ Discriminant:
|
A number of researchers have noted “most tests of the construct validity of sexism scales seek to determine whether women and men respond differently to these scales” (e.g., Swim et al., 1995). (Ref: 2) In separate studies seeking construct validity support for the Neosexism Scale, gender differences were found between average Neosexism Scale scores for men and women: 1) In a Canadian study, the mean Neosexism Scale score for men was 34.41 (average=3.13) and for women was 24.52, (average=2.23, p<0.0001) (Ref: 2); 2) a multivariate analysis of variance using data from three British studies involving a total of 907 participants reported a similar gender difference finding (M men=3.30, SD=0.89, M women=2.70, SD=0.76, p < 0.001). (Ref: 3) Results of a principal components analysis using 19 items from both the Neosexism scale (11 items) and the Modern Sexism Scale (eight items) and which reported a two-factor solution suggested the two scales measured comparable but not identical constructs. (Ref: 2) Finally, workers having stronger sexist beliefs than university students (Neosexism Scale mean score of 3.31 versus 3.04, p<0.05) may be supportive of the theory that stronger beliefs are more likely to arise under conditions of increasing political and/or economic pressure (e.g., employment competition). (Ref: 1)
|
Criterion-related/ Concurrent/ Predictive:
|
Sexism scale correlations: (a) The correlation between the Neosexism Scale (NS) and Old-fashioned Sexism was 0.64 (Ref: 1-Study 1); (b) the correlation between NS and the Modern Sexism Scale (Swim et al., 1995) was 0.588 (Ref: 2); (c) correlations of NS and the Hostile Sexism subscale (of the ASI) across three samples ranged from 0.52 to 0.81. (Ref: 3) Validity measure correlations:
(a) The correlation between NS and an affirmative action measure (AA) was -0.58, indicating more neosexist attitude meant less support for affirmative action policies and programs. In that study’s path analysis, the standardized direct effect of NS on AA was reported to be -0.46 (although the standardized total effects were not reported). (Ref: 1-Study 1)
(b) In a related study, the correlation between NS and AA was -0.36, between NS and Attitudes Toward the Program in Place (PRO) was -0.43, and between NS and Evaluation of the Competence of Women (EVAL) was -0.35. (Ref: 1-Study 2)
(c) The Attitudes Toward Feminism and the Women’s Movement Scale (FWM; Fassinger, 1994) and the Attitudes Toward Lesbians and Gay Men Scale–short version (ATLG; Herek, 1998) were also used to examine NS’s validity. Correlation between NS and FWM was -0.516 (more neosexist meant less supportive of the feminist movement); between NS and ATLG it was 0.411 (more neosexist meant more negative attitude toward lesbian and gay men. (Ref: 2)
(d) Another study reported correlations between NS and the Attitudes toward Women’s Rights (AWR) and Attitudes toward Lesbian and Gay Men’s Rights (ALG,) as –0.56 and –0.40, respectively. (Ref: 3) Value system measures: The correlation between NS and the Protestant Ethic Scale (PE) was 0.212 (more neosexist meant more likely to adhere to an individualistic value system; the correlation between NS and the Humanitarian-Egalitarian scale (HE) (Katz & Hass, 1988) was -0.224 (more neosexist meant less likely to support egalitarian values; these value system correlations did not differ significantly in magnitude between genders. (Ref: 2) In another study, the correlations between NS and PE and HE were 0.06 and -0.21, respectively. (Ref: 3)
|
Content:
|
No information found.
|
Responsiveness Evidence:
|
No information found.
|
Scale Application in VA Populations:
|
No information found.
|
Scale Application in non-VA Populations:
|
Yes. (Ref: 1-3)
|
|
|
Comments
|
The Neosexism Scale is based on an adaptation of the principal tenets of the modern racism model, as put forward by McConahay in 1989. (Ref: 1) It attempts to measure a new or modern type of gender prejudice, “neosexism,” that is thought to be more subtle and perhaps more disguised than typical sexism. In the present era and particularly in the West, female inferiority is, for the most part, no longer considered a socially acceptable belief to hold. Theories of modern sexism, as do theories of modern racism, suggest that, because of the powerful influence of social desirability factors, gender prejudice has not disappeared but, rather, has gone underground.
The Neosexism Scale’s items are largely concerned with employment issues. (Ref: 1) In essence, the scale attempts to measure political attitudes related to gender discrimination (Glick & Fiske, 1996) (Ref: 2-3) and may be thought of as a measure of attitudes toward gender labor policies.
There are some cautions to be associated with use of the Neosexism Scale. The field of modern sexism has proposed somewhat competing theories on what gender prejudice is and how it is manifested in present days. Therefore, a researcher must be clearly aware of how authors of the Neosexism Scale have defined the construct “neosexism,” as well as how it is that the scale attempts, by focusing on gender inequalities in employment and related gender policy issues, to measure it.
It is important to note that the magnitude of correlations between Neosexism Scale measures and other sexism measures have been primarily moderate (most from the 0.50s to 60s), while correlations with measures of gender policies and politics (e.g., affirmative action and the feminist movement) have also been, for the most part, moderate (most from the 0.40s to 50s), though of slightly lesser magnitude. There may be some inherent confusion in seeking to validate a neosexism measure using gender discrimination measures without losing track of the fact that the neosexism measure itself is a measure of gender discrimination.
A second validity issue involves differences observed between male and female responses to the Neosexism Scale. Mean gender differences discussed here, though statistically significant, ranged from 0.60 to 0.90, based on a response scale from 1-7. The mean responses suggested that males tended to “somewhat disagree” with gender prejudicial statements, while females tended to either “somewhat disagree” with the same prejudicial statements (i.e., to possess a similar level of neosexist attitude as males) or to “disagree” – expressing slightly less of a neosexist attitude than males. Thus, in practical terms, there was little observed absolute separation of neosexist attitudes between genders. Minimal gender differences could be explained by the sample presence of culture-indoctrinated females who disparage their own gender (e.g., Ref: 2), yet such an explanation undermines the determination of an expected gender-supportive (i.e., female) response level versus an expected gender-antagonistic (i.e., male) response level.
Reliability figures, for the most part based on internal consistency reliability, can be considered promising for an instrument under development but borderline inadequate for a tool seeking to determine group differences and inadequate for measuring individual differences. No evidence is reported of instrument sensitivity for detecting attitude change. Finally, there may be an issue related to respondent age appropriateness that must be factored into understanding sample responses obtained: The Neosexism Scale may be less sensitive to detecting gender prejudice when used with younger age groups (e.g., high school and college students). Younger age groups are typically less work-experienced. However, items from the Neosexism Scale may better address a more working-world experienced, labor-sophisticated audience that has seen firsthand what gender policies mean on the ground in terms of hirings, raises, and promotions. Thus, younger age groups may, on average, respond with less neosexist attitude, not because they are less gender prejudiced but because they have less real-world work experience.
|
Updates
No information found.
|
|
|
|
