Societies in transition: are they more sexist? A comparison between Polish, South African and British samples

Abstract

This study investigates ambivalent sexism to women in student samples from two under-researched transitional countries, Poland (PL) and South Africa (SA), in comparison with the United Kingdom. Based on ambivalent sexism theory (AST), and in light of socio-economic context, it was hypothesised that: (1) the sample in PL and SA would be more hostile- and benevolent-sexist than the sample from the UK, (2) males would exhibit more hostile attitudes than females irrespective of country and (3) males would outscore females on benevolent attitudes in the relatively liberal UK but underscore them in relatively conservative SA. The Ambivalent Sexism Inventory was used to measure benevolent and hostile sexism. The findings largely supported the hypotheses. The participants in SA and PL were more sexist than in the UK and men were more hostile-sexist than women in all three countries. However, males outscored females on benevolent sexism not only in the UK but also in SA and PL. Moreover, the sample from PL was observed to be more sexist than the sample from SA. The findings are discussed in light of AST and the countries' transitional context.

Keywords::

• ambivalent sexism
• gender attitudes
• transition to democracy
• Poland
• South Africa
• United Kingdom

Acknowledgements

We would like to thank the University of Winchester for sponsoring this research through a Research and Knowledge Transfer Grant. We would also like to express our gratitude to the University of Gdansk in Poland and the University of Cape Town, South Africa, for their cooperation. Last, but not least, our thanks go to Dr Blu Tirohl and the anonymous reviewers for their helpful comments on earlier version of this manuscript.

Notes

^1. It should be noted here that full women's suffrage was not achieved until 1928 as compared to 1918 in PL. In SA suffrage was initially granted to White women in 1930 and was fully extended to include Black women in 1994.

^2. For the interdisciplinary reader: 3 × 2 between subject design refers to a research design where 6 groups (3 × 2 = 6) of different people (hence ‘between subject’) are compared on a given measured variable (called dependent variable – here level of sexism). This format contains information about the number of factors (here two: country and participants' sex) on which the measured scores may depend and the number of levels each factor is represented by (here 3 countries and 2 sexes). Interested readers may find more information about research designs in Richardson et al. (2011).

^3. For the interdisciplinary reader: validity of a scale (or of a measurement tool) refers to an assessment of whether the scale measures what it was designed to measure (e.g. sexism and not racism). Support for discriminant (or divergent) validity is obtained if the measure does not correlate with (returns different scores than) a measure of an unrelated concept. Convergent validity is the opposite: evidence for it is obtained if the measure in question correlates with a different measure of the same or similar concept (see Howitt and Cramer 2011).

^4. The data was tested using recognized methods in this field: a reliability analysis was carried out for both HS and BS scales using Cronbach's α test. The test assesses how well the scale measures what it is designed to measure (i.e. sexism) by examining its internal consistency (i.e. if responses to each statement/item on the scale are related). Low item-total correlations indicate that a given item/statement generates responses which do not relate well to the remaining items. Satisfactory Cronbach's α scores emerged for each country in the current study for both HS (α_PL = 0.79, α_SA = 0.83 and α_UK = 0.84) and BS (α_PL = 0.75, α_SA = 0.75 and α_UK = 0.72) (see Field 2009).

^5. The means for the UK in our study were as follows: for HS M_F = 1.76 versus M_M = 2.13 and for BS M_F = 1.92 versus M_M = 2.02 while our means for SA were as follows: for HS M_F = 2.41 versus M_M = 3.44 and for BS M_F = 3.45 versus M_M = 3.20.

^6. Although more recent data sets exist, these, unfortunately, are not comparable to those of Glick et al.'s (2000). For example, Viki and Abrams's (2003) report more recent findings concerning the UK, and Forbes et al. (2004) for Polish females but these authors adopted modified HS and BS subscales with seven-point, as opposed to the original six-point, response format.

^7. Construct equivalence (i.e. referring to whether the concept of sexism is understood and functions in the same way in all three countries) is operationally defined as factorial or metric invariance. This is tested through replicatory factor analysis – a statistical procedure which assesses if the responses to the items/statements in a given scale cluster around a number of a priori identified factors in the same way as in the original scale. Here existence of two factors (i.e. benevolent and hostile sexism) was assumed and confirmed. Level of similarity between the obtained factorial solutions (i.e. clusters of items) from each country was also assessed: Tucker's (1951) phi coefficient of congruence indicated that Polish and South African factors where highly similar to the British ones exceeding the recommended cut off point of 0.90 (Tucker 1951): 0.94 for HS in SA, 0.93 for BS in SA, 0.95 for HS in PL and 0.91 for BS in PL (see Ben-Porath 1990).

Additional information

Notes on contributors

Magdalena Zawisza

Magdalena Zawisza, Ph.D., is a senior lecturer in the Department of Psychology at the University of Winchester. Her research interests fall into three general areas: consumer, gender and applied social psychology. More specifically, she is interested in gender issues in advertising such as the effectiveness of (non)traditionally gender advertisements. In her more recent research, she examines explicit or implicit gender attitudes and stereotype content and their performance in advertising context. Another strand of her research focuses on attitudes to both men and women in countries undergoing transition to democracy. Her further interests include elaboration of gendered advertisements by men and women as well as the effects of gendered advertisements on audiences. Her work is grounded in positivist approach and uses mainly experimental methodology.

Russell Luyt

Russell Luyt is a senior lecturer in the Department of Psychology at the University of Winchester. He is particularly interested in identity processes as well as debates concerning research philosophy and qualitative methodology. His work is grounded within a critical approach and has, to date, focused primarily on the social psychology of gender. This has included, for example, the critique of traditional masculinity, femininity and gender measurement; media representations; masculinities and aggression; as well as the intersection of gender and other social categories. He is currently extending these lines of research and developing his interest in radicalisation among marginalised communities from a gender perspective.

Anna Maria Zawadzka

Anna Maria Zawadzka is a professor at the University of Gdańsk (Poland). Her research focuses on self-regulation in the context of power, consumption, advertising and preferred values and their relationship with well-being.