Reactions Toward Affirmative Action Measures for Women

Reacciones a las Medidas de Acción Afirmativa para las Mujeres

 

 

Silvia Moscoso¹, Antonio L. García-Izquierdo², María Bastida¹

¹University of Santiago de Compostela
²University of Oviedo

Research reported in this paper was partially supported by grant nº PSI2008-03617 from the Spanish Ministry of Science and Innovation.

Address for correspondence

 

 

---

ABSTRACT

This paper examines whether there are differences between women and men in their reactions towards different types of specific affirmative action measures (AAMs) for women. 192 women and 128 men from a variety of jobs responded to a survey. We conclude that women react more positively than men towards AAMs, although the reactions of both men and women vary depending on the type of AAM. Thus, reactions become increasingly positive when personnel decisions were made on the basis of merit and increasingly negative when decisions were made on the basis of gender. We also found that reactions towards AAMs are related to the variables "unfairness perception" and "threat perceived to men", and especially to the former. Also, reactions towards the generic concept of affirmative action are more positive than towards specific AA measures. Implications of these results for research and practice are discussed.

Keywords: affirmative action measures, working women, reactions, unfairness perception, perceived threat.

---

RESUMEN

Este trabajo examina si existen diferencias entre mujeres y hombres en sus reacciones hacia diferentes tipos de medidas de acción afirmativa (MAAs) para mujeres, para lo que 192 mujeres y 128 hombres ocupantes de diferentes puestos respondieron a un cuestionario. Los resultados muestran que las mujeres reaccionan de manera más positiva que los hombres hacia las MAAs, aunque las reacciones de ambos varían en función del tipo de medida de acción afirmativa. Las reacciones son más positivas cuando las decisiones de personal se basan en el mérito y se vuelven más negativas cuando las decisiones están basadas principalmente en el género. También encontramos que las reacciones hacia las MAAs se relacionan con las variables "percepción de injusticia" y "percepción de amenaza para los hombres", y especialmente con la primera. Otro resultado es que las reacciones hacia el concepto genérico de acción afirmativa son más positivas que hacia medidas específicas de AA. Por último, comentamos las implicaciones de estos resultados para la investigación y la práctica.

Palabras clave: medidas de acción afirmativa, mujeres trabajadoras, reacciones, percepción de injusticia, amenaza percibida.

---

 

In recent decades, a large body of legislation in Europe has been dedicated to equality between women and men. This is mainly made up of various Treaty provisions and Directives concerning access toemployment, equal pay, maternity protection, parental leave, social security, and occupational social security, the burden of proof in discrimination cases and self-employment. In this context, Affirmative Actions Measures (AAM), have been the subject of a passionate debate and have been the focus of important research in many countries. For example, in the United States it is one of the most controversial topics and, in the European Union, the issue is becoming an increasingly important question (Crosby et al, 2003, Doverspike et al., 2000, 2006; Krings, Tschan & Bettex, 2007). Faundez (1994) provided an explanation for this growing interest, when he argued that AAMs have the shared objectives of eliminating discrimination and simultaneously promoting equal opportunities. However, as Faundez (1994, p.55) states "most of these objectives have a bearing on the issues of equality, it is not surprising that affirmative action is controversial, as it raises sensitive moral, economic and political questions".

European legislation defines discrimination as a treatment which, apart from being differential, is illegal because it uses forbidden criteria, and it distinguishes between direct and indirect discrimination. Direct discrimination consists of treating someone differently solely on the basis of age, sex, skin colour, national origin or disability, among other factors (art. 2 (1a) Directive. 2000/78/EC about a general framework for equal treatment in employment and occupation). EU Directive 97/80 on shifting the burden of proof in sex discrimination cases (art. 2) points out that indirect discrimination takes place when an apparently neutral provision, criterion or practice disadvantages a substantially higher proportion of the members of one sex, unless that provision, criterion or practice is appropriate and necessary and can be justified by objective factors unrelated to sex. As Directives 2000/43/EC, 2000/78/EC and 2002/73/EC state, it is possible to justify indirect discrimination if that provision, criterion or practice is objectively justified by a legitimate aim and the means of achieving that aim are appropriate and necessary by reason of the nature of the particular occupational activities concerned or of the context in which they are carried out. Such a characteristic constitutes a genuine and determining occupational requirement, provided that the objective is legitimate and the requirement is proportional. Assumptions are those of: a) professional activities for which sex is a factor because of the nature and conditions of the work, b) the protection of women, particularly in pregnancy and childbirth, and c) to promote equal opportunities between men and women (art. 2.6 EU Directive 76/207 on Equal Treatment). In the case of Spain, main legal dispositions on EEO, based on previous European legislation (Directives 2002/73, 2004/113, and 97/80) are Organic Act 3/2007, on effective equality between women and men (LOIMH), and Act 39/99 for the promotion of the reconciliation of family and work. Organic Act 3/2007 stresses differences based on sex, although Spanish case law even puts the burden of proof on the defendant, instead of the plaintiff as is usual in most cases. Moreover, Act 3/2007 distinguishes between direct discrimination and indirect discrimination (adverse impact).

With regard to Affirmative Action, these are a step beyond the principle of equal treatment. Directives 76/207/CEE and 2002/73 EC, and the European Union Treaty through the European Justice Court provides the basis for legal consideration of some national AAM favoring women. In Spain, AAMs are defined as particular measures for women in order to correct situations of inequality in comparison to men in art. 11 of LOIMH. Consequently, AAMs are promoted in the LOIMH in every circumstance, especially in those cases where women are underrepresented. AAMs are based on two assumptions: that they compensate for traditional inequalities in the society, and that they are a measure for enhancing social integration for the disadvantaged minorities. The most controversial AAMs are those of reverse discrimination, such as quotas, preferential treatment on equal merit, and merit exemption for the focal group. Spanish case law has determined the legality of AAM in several verdicts of the Constitutional Court as in the STC 34/981, STC 128/1987, STC 109/1993 y STC 187/1993, and even for reverse discrimination as STC 269/1994. AMMs are recognized now in the LOIMH. Generally speaking, AAMs must be temporary, proportionals and reasonable. A development of this issue have been made in García-Izquierdo and García-Izquierdo (2007).

It is, therefore, necessary to carefully distinguish and clarify certain legal questions, in order to improve EEO. For example, article nº. 5 of the LOIMH specifically states that difference of treatment based on a sex-related characteristic will not constitute discrimination in access to employment when, in light of the nature of the particular tasks concerned or the context in which they are performed, such a characteristic constitutes a genuine and determining occupational requirement, provided that the objective is legitimate and that the requirement is proportionate. Therefore, it becomes necessary to specify what really constitutes a genuine and determining occupational requirement is. The LOIHM include some consequences for employers in case of violation of the norms, of which the most important are compensation, rejection of the decision and subsequent application of the court decision, and the invalidity of the discriminatory dismissal. An equality plan could reverse this situation.

From the social point of view and based on research, the supporters of affirmative actions put forward several arguments. On the one hand, they suggest that AAMs allow discrimination from the past to be rectified and prevented in the future. In addition, supporters of AAMs sustain that these measures are necessary in order to ensure diversity in the workforce. They also suggest, as a third argument, that AAMs help to ensure that selection processes and personnel decisions are fairer. The opponents of AAMs, however, use the opposite argument, that is, that these measures are unfair because they give preferential treatment to one group over another. In this way, they argue, personnel decisions fail to take into account meritocratic criteria. In addition, opponents suggest that affirmative measures can be negative for the target group as they cast doubt on their competence. Responding to this reasoning, supporters of AAMs consider the arguments of the opponents to be based on prejudice (Crosby, Iyer & Sincharoen, 2006; Krings, Tschan & Bettex, 2007).

Part of this debate has been focused on how to define the concept of affirmative action, which has led to society being badly informed about what it means. For example, Crosby, Iyer & Sincharoen (2006) claimed that affirmative action occurs when an organization invests resources (including time and money) to ensure that nobody is discriminated against based on gender or ethnic group. Previously, Johnson (1990, p. 77) had already defined affirmative action as "a generic term for programmes which take some kind of initiative, either voluntarily or compulsory, to increase, maintain or rearrange the number or status of certain group members usually defined by race or gender, within a larger group." Later, Bacchi (1996, p. 17) pointed out that "affirmative action referred to proactive measures undertaken to create a non discriminatory work environment." From the legal and institutional points of view, the US Commission on Civil Rights (1977, 2004) defined AAMs as any measure, beyond simple termination of a discriminatory practice, adopted to correct or compensate for past or present discrimination or to prevent discrimination from recurring in the future. A similar definition was given by the Committee on Equal Opportunities for Women and Men (CEOWM) of the Council of Europe (2008). According to the CEOWM, affirmative action is a strategy designed to establish equal opportunities through measures which allow discrimination which is the result of social practices or systems to be rectified. This variety of definitions about affirmative action measures has caused different countries to implement different types of affirmative action policies and has meant that they even use different terminology (Pylkkanen, 2004). With regard to this last point, although the term "affirmative action" is used in the legislation in most countries (for example the USA, Australia, South Africa and Northern Ireland), some others, on the other hand, use the term "positive action" (e.g. the UK and Spain). In Holland, the terms "positive discrimination" and "positive action" are sometimes used, and the term "employment equity" has been used in Canada.

With regard to the different types of affirmative action measures, these can be classified according to the relative weighting of two variables, namely, demographic status and merit criteria (Krings et al., 2007). Thus, there are many ways in which variables can be combined, resulting in different affirmative action measures. For example, measures such as opportunity enhancement programs or education programs for women are considered to prioritize merit rather than gender. However, preferential selection and quota systems are both examples of measures which give more weight to gender than merit. Several authors, taking into account these two variables, refer to a continuum which goes from soft measures to hard measures (Krings, Tschan & Bettex, 2007; Pylkkanen, 2004). In general, by soft measures they mean those which are aimed at encouraging participation by women (or other target groups) in areas where they are underrepresented, but still taking merit rather than gender to be the main criterion. On the other hand, hard measures, by giving priority to gender over merit are considered to be positive, or inverse, discrimination (Edwards, 1995). Similarly, Kravitz (1995, Harrison, Kravitz, Mayer, Leslie, & Lev-Arey, 2006) offers another classification composed of four categories: (1) Opportunity enhancement AAMs, which consist of offering assistance to the target groups before the selection decisions, usually through focused recruitment or training. The objective is to add more target group members to the pool of qualified candidates. (2) Equal opportunities (elimination of discrimination), where undervaluing or giving a negative weight to member of target groups is not permitted. (3) Tiebreak AAMs, also known as "weak preferential treatment", where the target group receive preferential treatment only when their qualifications are equal to those of non-target group members. (4) Strong preferential treatment AAMs, where both quota systems and preferential treatment are included, that is, target group members are given preference even if their qualifications are inferior to non-target group members.

The affirmative action measures discussed above are aimed at promoting equal opportunities and increasing the representation of disadvantaged groups. However, just because an organization has a program of affirmative action, does not mean that it will necessarily achieve the desired objectives. If affirmative action programs are not well designed, it can lead to results which are contrary to those expected (Crosby, Iyer & Sincharoren, 2006; Konrad & Linnehan, 1995). In this sense, several researchers have suggested that AAMs can produce negative effects and responses. For example, AAMs could create the stereotype that those who benefit from these programs could never have succeeded on their own merits. That is to say, they could stigmatize the target group and cause their competence to be undervalued (Resendez, 2002; Ruiz & Moya, 2005; Sowell, 2004; Zelnick, 1996). Another negative consequence, according to its critics, is that it works as a type of reverse discrimination and so may increase tension between groups (Lynch, 1992).

Given the above, it seems clear that reactions toward AAMs are of vital importance for their development and success. These reactions and attitudes towards affirmative action for target groups, has been an issue that has aroused the interest of many researchers since the eighties, especially in the United States (Bell, Harrison & McLaughlin, 2000; Bobo, 1998, 2000; Harrison, Kravitz, Mayer, Leslie & Levy-Arey, 2006; Konrand & Linnehan, 1995; Kravitz & Platania, 1993) and more recently in other countries (Konrand & Hartmann, 2001; Krings, Tschan & Bettex, 2007). Many of these studies have found that women and black people are generally more favourable toward affirmative action than white people and men (e.g, Bobo & Smith, 1994; Kravitz & Platania, 1993; Matheson et al., 1994). These results were confirmed in a meta-analysis conducted by Harrison et al. (2006). However, other researchers have found that attitudes vary widely and erratically over time (Ewoh & Elliot, 1997; Schuman et al., 1997; Steeh & Krysan, 1996). Crosby et al. (2006) attribute these results to the breadth of the term affirmative action and confusion about its definition. They also believe that the variability in terms of support for affirmative action programs may be due to the way in which such attitudes have been measured. In most studies, participants have evaluated specific affirmative action practices, while in others participants were asked to assess the generic term "affirmative action".

With regard to reactions to different types of AAMs, there is some research which shows that AAMs giving priority to merit over gender have received more support and are considered fairer than those AAMs where gender is more important than merit (e.g, Kravitz & Platania 1993; Krings, Tschan & Bettex, 2007; Matheson et al., 2000). Specifically, Krings et al. (2007) assessed the reactions to eight different AAMs. Four of them involved measures which gave preference to gender over merit (evaluating women's qualifications differently, including quotas for women, organizational incentives for the hiring of a woman and preferential selection of women with equal qualifications). The other four measures were considered to give more weight to merit than gender (special continuous education programs for women, encouragement of applications from women, the financing or provision of childcare facilities and mentoring programs for women). Their results showed that attitudes become increasingly positive as more weight is given to individual merit and increasingly negative as gender takes precedence.

Research has also focused on other variables that may be related to reactions toward AAMs. In this sense, Beaton and Tougas (2001) conducted a study which examined the reactions to an affirmative action program based on the preferential treatment of three different target groups: women, disabled people and black people. Results showed that reactions varied depending on both the sex of those who responded to the questionnaire and on the target group. Women were more favourable toward affirmative action programs regardless of the nature of the target group. In addition, it was found that women and men do not give equal weight to certain social justice concerns when they evaluate affirmative action programs. The effects on support for affirmative actions of two social justice concerns were also tested: scope of justice (extending fair treatment to others) and threat perceived by the non-designated groups. The results showed that reactions to these programs are more influenced by scope of justice explanations in men than in women. Konrad and Hartmann (2001) found that the relationship between gender and attitudes to AAMs focused on women was mediated by perceptions of the impact of the action on material self-interest, belief in the existence of gender discrimination and traditional attitudes toward women. In the Harrison et al. (2006) metaanalysis, the results showed that the perceived beliefs about the fairness of the measures proved to be a powerful predictor of attitudes towards AAMs. Namely, the perception of fairness moderated the effects of the hardness of the AAM. Another variable examined was how the description given of the AAMs could affect attitudes towards them.

In summary, research on reactions to AAMs, in general, has shown that target groups (for example, women or black people) are more positive towards such measures. However, it seems that the way in which the reactions and attitudes were measured (assessing specific measures against the general concept of affirmative action), as well as the type of measure involved (preference to merit or gender) affect the results. In addition, there are a number of variables (e.g. perceived fairness of the situation or the threat perceived by non-target groups) that may influence reactions. In addition, most studies have two other characteristics, namely, that the majority were carried out with samples of students, not workers, and that they were conducted in the U.S. or Canada. In Europe there are scarcely any studies, and none have been undertaken in Spain. It therefore seems relevant to extend the studies about reactions and attitudes toward women to other countries beyond the USA or Canada, in order to confirm whether there are differences across countries.

The objective of this study is therefore to examine the reactions towards different specific types of AAMs for women and to analyze if there are differences between men and women. At the same time, we will examine whether the unfairness perception about the situation of working women and the threat perceived by non-designated groups variable (in this case men) have a relationship with the reactions to the specific AAMs. Accordingly, the following hypotheses are advanced:

H1: Gender affects reactions toward affirmative action measures for women, with women being more positive than men toward these measures.

H2: There are differences between women and men in unfairness perception about the situation of working women, with women perceiving that the current situation is more unfair than men.

H3: There are differences between women and men in threat perception to men, with men perceiving a greater threat.

H4: Both women and men show different reactions to AAMs depending on the type of measure. Specifically, the more weight merit has over gender, the more positively the measure will be valued.

H5: There will be differences in the assessment of AAMs depending on whether a description of specific measures or a general scale on attitudes toward affirmative action measures are used.

 

Method

Sample

Participants were 320 employees of public and private Spanish organizations (192 women and 128 men). Their age ranged from 19 to 68 (M= 39.34). The length of tenure ranged from two months to 30 years (M= 9 years). The majority of participants (65.4%) worked in the private sector (industry, consultancy, services, construction and IT) and the rest (34.6%) worked in the public sector. The level of studies was as follows: primary school (n=35), high school (n=120), university (n=155), other (n=5) and 5 participants did not indicate their level of studies. Regarding the type of contract, 198 participants had temporary contracts and 114 had permanent (8 participants did not indicate the type of contract). Regarding organizational level 66 participants were supervisors (38 men and 28 women).

 

Procedure and Measures

Participants received a questionnaire containing the scales that are described below, as well as several demographic questions (e.g., gender, age, education) and questions related to work (e.g., job, tenure, contract type, company). After filling out the questionnaire, participants were able to return it by email or give it back to a team member.

Attitudes toward different types of affirmative action programs

This first part of the questionnaire is composed of descriptions of five different affirmative action measures and was developed based on Krings, Tschan and Bettex (2007). Each description refers to one type of affirmative action measure, which could be placed on a continuum according to whether merit is given more weight than gender (e.g., preferential treatment for women, quotas for women) to the opposite situation, where gender has preference over merit (e.g., equal opportunities). Previously, 150 university students had filled out a first version of the questionnaire composed of eight descriptions of different AAMs. Results showed that the questionnaire was much too long and some of the eight descriptions were very similar in their content. For this reason, descriptions were combined and reduced to five. Participants had to read a description of each AAM and then rate these on four Likert-scales: effectiveness, necessity, fairness and adequacy. Scales ranged from 1 (nothing) to 5 (very much). The alpha coefficient for this scale is .72. A description of all five measures can be seen in Appendix.

Attitudes toward Affirmative Action Scale (ATAA)

The Spanish version of the Attitudes toward Affirmative Action Scale (ATAA; Parra, 1991) was used. This instrument is composed of 12 items. However, some modifications were made to the original scale. Firstly, items refer to AAMs for different target groups in the original scale, but, in our version, they focus only on AAMs for women. Secondly, Parra's Scale used a 7-point Likert Scale. In our case we decided to use a 5-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree). The alpha coefficient for this scale is .87. Some examples of items are: "Affirmative action programs for women will result in progress for everyone", "Affirmative action programs for women are unfair to men" or "Affirmative action for women helps to ensure fairness in employment practices".

Unfairness perception about the situation of working women

This scale is a combination and an adaptation of two other scales: Beaton and Tougas' Scale (2001) and Henry and Sears' (2002) Symbolic Racism 2000 scale. Beaton and Tougas' original scale was composed of nine items, but in our version we used only six items. We assessed the extent to which the principles of fairness were applied only in relation to affirmative actions for women, not for other target groups as in the original scale. Two examples of items are: "I believe that women are unfairly treated in this country" or "Because of their sex, women are already at a disadvantage when competing for jobs and promotion". The original scale by Henry and Sears consisted of eight items. We used just three of them, only those related to women and applicable in Spain. One example of an item is: "Discrimination against women in the labor market is no longer a problem in Spain". A 5-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree), was used instead of the 7-point and 4-point Likert scales in Beaton and Tougas' and Henry & Sears' scales respectively. We combined both scales in just one and conducted factor analyses to check the dimensionality of the scale and, as expected, items loaded on a single factor. The alpha coefficient for this scale is .89.

Threat to Non-designated Groups (Men)

This scale was an adaptation of Beaton and Tougas' scale (2001). In the original scale, a total of 6 items assessed the extent to which affirmative action measures could be considered a threat for members of different non-target groups (men, white people, and people without disabilities). In our case, we only asked about the threat for men. A5-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree), was used instead of the 7-point Likert scale in the original. Some examples of items are: "The government implements so many programs which promote the situation of women that it is men who suffer" or "These days, it is easier for women to get a job than men". The alpha coefficient for this scale is .84.

 

Results

Table 1 shows descriptive statistics for all variables and the results of the various variance analyses comparing men and women across the studied variables. The five descriptions of specific types of AAMs correspond to variables AAM1 to AAM5. Each one was assessed on four scales (effectiveness, necessity, fairness and adequacy), which were summed up to create a single average for each type of AAM (in Table AAM1 to AAM5). These measures can be classified in a continuum from those measures which give more weight to merit over gender to those where gender takes precedence over merit. Thus, AAM1 is a measure which relates to creating equal employment opportunities by eliminating information that might be discriminatory (e.g. marital status, number of children) or preference for one sex stated in the job offers. AAM2 refers to providing organizational incentives (e.g. subsidies) to encourage the promotion of women to managerial positions. AAM3 refers to the use of a quota system for women in the access to employment and promotions. AAM4 is focused on work-life balance, but with benefits for women only. Finally, AAM5 is a example of strong preferential treatment where women are given preference and fewer qualifications are required of them in the selection process than men. We next conducted a principal component analysis in order to explore whether there were just one or more than one component among the AAMs, and the results showed that there was only one, which explained 47.72% of variance. We therefore created a global score of reactions toward AAMs, adding up the mean scores obtained in each of the five AAM descriptions (in Table Global AAMs). As was hypothesized, women scored higher than men in all variables except in threat perceived, and the ANOVAs showed that the differences were statistically significant in all cases except for AAM5. This specific measure is the one where highest priority is given to gender over merit (evaluating women's qualifications differently). Interestingly, the mean score for women and men for this specific measure was less than three, that is, neither women nor men consider it to be a positive measure. With regard to the perceived threat variable, the results were also less than three, which suggest that AAMs are not considered to be a major threat for men. Hypotheses 1, 2 and 3 were therefore confirmed.

Hypothesis 4 stated that there would be differences in both men and women in their reactions to AAMs depending on the specific type of measure. The results show that the greater the importance given to merit over gender, the more positive are reactions to these measures, and that the differences are statistically significant (F=187.62, p<.000 and F=104.90, p<.000, for men and women, respectively).

Table 2 shows the correlations between variables for men and women separately. In both samples, we found significant correlations between the five specific types of AAMs, with the exception of the correlation between AAM1 and AAM4 in the case of men. The general scale of attitudes toward AAMs (ATAA) correlates with all variables in the sample of men. In the case of women, it correlates with all variables except for those AAMs where gender is given more importance than merit (AAM4 and AAM5). The Global AAM composite correlates significantly with the general scale of attitudes (ATAA), unfairness and perceived threat in both samples, but the correlation between the Global AAM composite and Unfairness is three times larger in the case of men than women (.52 vs. .19).

According to Hypothesis 5, there are differences in the assessment of the AAMs depending on the way in which they are measured (composite created summing up specific AAMs vs. a general measure of attitude towards affirmative actions). The results of the comparisons showed significant differences for the total sample (t=11.189, p<.000), with the same pattern being shown when the comparisons were made for the men and women samples independently (for men t=7.202, p<.000; for women t=8.490, p<.000). Hypothesis 5 was therefore supported by the data. In view of these results, we conducted two multiple regression analyses using the two measures of AAM reactions (Global AAM and ATAA) as the dependent variables, and unfairness and threat as independent variables. The multiple regression analyses were carried out for men and women separately. Results (Table 3) showed that global AAM are remarkably better predicted for men than for women (.513 vs. .208). However, regarding the prediction of ATAA, the results are very similar for both genders (.659 vs. .617). Consequently, when the reactions are measured with a general attitude scale of AAMs, both the perception of threat and the perception of unfairness are valid predictors. However, when reactions are assessed using specific AAMs, only the perception of unfairness is a significant predictor in both genders, but it is a notably better predictor in the sample of men.

 

Discussion

The objective of this study was to examine if there were differences between women and men in their reactions towards different types of affirmative action measures for working women. We also test if unfairness perception of the situation to working women and the perceived threat to men have a relationship with reactions to the specific AAMs. To assess reactions to the AAMs, we used the descriptions of five specific types of measure based on the weight given to merit or gender. Thus, each measure is placed in a different point on the continuum depending on the weight which gender is given in decisions. Some measures use individual merit as the principal criterion (e.g. equal opportunities) whereas other types of measures use gender as the principal criterion (e.g. preferential selection) in personnel decisions. In addition, we used a scale that assessed the general concept of affirmative action measures, in order to test whether there were differences in the reactions to AAMs depending on how reactions are evaluated.

Just as in previous research, the results showed that reactions varied depending on AAM type: reactions become increasingly positive as more weight is given to merit and increasingly negative as more weight is given to gender. That is, both men and women are more favourable to measures such as the elimination of discrimination or equal opportunities than to measures such as quota system or preferential treatment. However, we found that women's scores are significativily higher than men, which seems logical if we consider that women are the target group in this study, and these results are also consistent with previous literature (e.g., Beaton & Tougas, 2001; Konrad & Hartmann; 2001; Krings, Tschan & Bettex, 2007). Only in the case of strong preferential treatment were there no differences between men and women. Both considered this type of measure (women are given preference and fewer qualifications are required of them than men in the selection processes) to be very negative. These results are important as they show that just because an organization has a program of affirmative action, it does not mean that it will necessarily achieve the desired objective, especially if the AAMs produce negative reactions in their employees. Although some defenders of this type of strong measure consider them to be necessary in order to change the situation of underrepresentation of women, the fact is that, besides being illegal in some countries, they would have a negative effect for the target group (e.g. they could stigmatize the target group and cause their competence to be undervalued).

The results showed that reactions toward AAMs are related with the variables of unfairness perception and threat perceived to men, and especially with the former. The findings show that men consider this situation less unfair than women, even believing that women nowadays continue to be at a disadvantage in the workplace. Another interesting result was that this variable was more related with the reactions toward specific AAMs in the case of the men than the women. That is, the men's reactions toward affirmative action measures are affected by the unfairness perceived about the employment situation of women. However, in the case of women, this variable had much less effect on their assessments of AAMs. A similar result was found by Beaton and Tougas (2001), where scope of justice was a more important predictor of favourable attitudes toward AAMs among men than among women. In the case of perceived threat to men, our findings show that men have a higher perception that AAMs may be discriminatory to their group than women, although, in both cases, the degree of threat perceived is small. This variable is also a better predictor of reactions toward AAMs in the case of men than women. Men will have more negative reactions toward AAMs for women only if these measures are perceived to be a threat to their group. Konrad and Hartmann (2001) found similar results, which they explain by saying that men were less likely to support affirmative action measures because they may consider them a threat to their careers, because they were less likely to believe that women suffer discrimination, and because they held more traditional attitudes toward women. However, Beaton and Tougas (2001) found that perceived threat was a more important determiner of evaluations of AAMs among women than among men.

Crosby et al. (2006) considered that the way in which attitudes and reactions are measured in different studies could affect their assessment. In some researches, participants evaluated specific affirmative action measures, while in others participants were asked to assess the generic term "affirmative action". We compared both methods and found that there were significant differences. The reactions toward AAMs are more postive when are assessed in a general way than when specific measures are considered. In addition, the perception of unfairness and perceived threat are good predictors of reactions toward AAMs for men and women in the first case, whereas when specific measures are used, only unfairness perception is a valid predictor, especially for men. These results suggest differences in each case considered here. Although further research is needed about this, a consequence of these results is that organizations should be cautious if they wish to measure the reactions of employees toward AAMs in order to implement them. If AAMs are assessed in general terms, without specifying the type of measure, it could lead us to think that any measure will be accepted by employees, when, in fact, this is not the case; both men and women react negatively toward certain types of measures. Related to this, some studies (Arriola & Cole, 2001; Kravitz & Platania, 1993; Krings, Tschan & Bettex, 2007) found that, in general, there was a lack of knowledge about what AAMs are. For example, in the Krings et al., (2007) study, two thirds of the respondents did not know any AAMs or named a program that was not an example of AAMs at all, and employees that knew some AAMs did not associate them with measures where gender is the most important criterion in personnel decisions.

Results of this research also have implications for practice. In Spain, as in most countries, both public organizations and private companies need to implement equality policies and affirmative action measures in order to respond to social demands related to equal opportunities and non-discrimination in employment for different target groups in general, and for women in particular, as well as to obey the law (Pylkkanen, 2004; Myors et al., 2008). However, organizations should consider that employees react more negatively to some specific types of AAMs, and if these are not well designed it can lead to results contrary to those expected. In this sense, several authors warn of the need for care about how AMMs are implemented (Crosby, Iyer & Sincharoren, 2006; Konrad & Linnehan, 1995). Measures such as strong preferential treatment or quotas are more negatively assessed both in men and women, and particularly the latter. Moreover, in many countries they are illegal. For example, quota systems are considered illegal in countries such as the United States. It is even necessary, in the case of equal merit, to justify the selection of a target group member by showing past discrimination or underrepresentation of this group. In Spain, however, the law does allow the implementation of quota systems, but other types of measures where the priority criterion is gender rather than individual merit are considered unlawful.

For affirmative action measures to be successful, organizations should take into account several issues. Public support of management in both public and private organizations could help legitimize affirmative action programs, as well as providing the necessary resources to implement them (French, 2001; Nacoste & Hummels, 1994). Another important factor for the success of AAMs is communication. It is necessary to transmit a clear and persuasive message about what the objectives to be achieved with these programs are. Thus, in order for the AAMs to be effective, transparency of criteria used in selection and training processes, an emphasis on the potential advantages of these policies for non-beneficiaries (i.e., the more diverse teams are, the more effective they will be), and appealing to social responsibility can all be of help (Crosby, Iyer & Sincharoren, 2006). Some research has shown that interaction and communication between groups, beneficiaries and non-beneficiaries of these programs, can help the latter to be more receptive. Also, if organizations know that some factors, such as unfair perceptions about the situation of working women or perception of threat by men can affect the reactions toward AAMs, they should try to change these beliefs, for example with awareness programs.

In summary, reactions toward AAMs vary depending on the type of measure and also gender. Women gave them a more positive assessment than men, although in both cases reactions became increasingly positive as more weight is given to merit and increasingly negative as more weight is given to gender, both in women and in men. It is therefore important that, in order to ensure that such measures succeed, organizations are careful about the way in which they evaluate their employees' reactions. Likewise, organizations must take into account that variables such as unfairness perception of the situation of working women and the threat perceived to men also affect reactions toward AAMs.

 

References

Arriola, K. R. J. & Cole, E. R. (2001). Framing the affirmative-action debate: Attitudes toward out-group members and White identity. Journal of Applied Social Psychology, 31, 2462-2483.        [ Links ]

Bacchi, C. L. (1996). The politics of affirmative action. Thousand Oaks: Sage Publications.        [ Links ]

Beaton, A. M. & Tougas, F. (2001). Reactions to affirmative action: Group membership and social justice. Social Justice Research, 14, 61-78.        [ Links ]

Bell, M. P., Harrison, D. A. & McLaughlin, M. E. (2000). Forming, changing, and acting on attitudes toward affirmative action programs: A theory-based approach. Journal of Applied Psychology, 85, 784-798.        [ Links ]

Bobo, L. (1998). Race, interests, and beliefs about affirmative action. American Behavioral Scientist, 41, 985-1003.        [ Links ]

Bobo, L. (2000). Race, interests, and beliefs about affirmative action: Unanswered questions and new directions. In D. O. Sears, J. Sidanius & L. Bobo (Eds.), Racialized politics: The debate about racism in America (pp. 137-164). Chicago, IL: University of Chicago Press.        [ Links ]

Bobo, L. & Smith, R. A. (1994). Antipoverty policies, affirmative action, and racial attitudes. In S. H. Danziger, G. D. Sandefur & D. H. Weinberg (Eds.), Confronting poverty: Prescriptions for change (pp. 365-395). Cambridge, MA: Harvard.        [ Links ]

Boe nº71 de 23 de Marzo de 2007. Ley Orgánica 3/2007, de 22 de marzo, para La Igualdad Efectiva de Mujeres y Hombres, 12611-12645.        [ Links ]

Brutus, S., Parra, L. F., Hunter, M., Perry, B. & Ducharme, F. (1998). Attitudes toward affirmative action in the United States and Canada. Journal of Business and Psychology, 12, 515-533.        [ Links ]

Committee on Equal Opportunities for Women and Men of the Council of Europe (2008). http://assembly.coe.int/Main.asp?link=/committee/DocRef/EGA_E.htm.        [ Links ]

Crosby, F. J., Iyer, A., Clayton, S. & Downing, R. A. (2003). Affirmative action. Psychological data and the policy debates. American Psychologist, 58, 93-115.        [ Links ]

Crosby, F.J., Iyer, A. & Sincharoen, S. (2006). Understanding affirmative action. Annual Review of Psychology, 57, 585-611.        [ Links ]

Doverspike, D., Taylor, M. A. & Arthur, W. (2000). Affirmative action. A psychological perspective. Huntington, NY: Nova.        [ Links ]

Ewoh, A.I & Elliott E. (1997). End of an era? Affirmative action and reaction in the 1990s. Review of Public Personnel Administration, 38-51.        [ Links ]

Edwards, J. (1995). When race counts: the morality of racial preference in Britain and America. New York: Routledge.        [ Links ]

Faundez, J. (1994). Affirmative action: International perspectives. Geneva: International Labour Office.        [ Links ]

French E. (2001). Approaches to equity management and their relationship to women in management. British Journal of Management,12, 267-85.        [ Links ]

García-Izquierdo, A.L. & García-Izquierdo, M. (2007). Discriminación, igualdad de oportunidades y selección de personal en España. Revista de Psicología del trabajo y de las organizaciones, 23 (1), 111-138.        [ Links ]

Harrison, D.A., Kravitz, D.A., Mayer, D.M., Leslie, L.M. & Lev-Arey D. (2006). Understanding attitudes toward Affirmative Action Programs in Employment: Summary and Meta-Analysis of 35 Years of Research. Journal of Applied Psychology, 91, 1013-1036.        [ Links ]

Henry, P. J. & Sears, D. O. (2002). The symbolic racism 2000 scale. Political Psychology, 23, 253-283.        [ Links ]

Johnson, R. A. (1990). Affirmative action policy in the United States: its impact on women. Policy and Politics 18, 77-90.        [ Links ]

Konrad, A. M. & Linnehan, F. (1995). Formalized HRM structures: Coordinating equal employment opportunity or concealing organizational practices? Academy of Management Journal, 38, 787-820.        [ Links ]

Konrad, A. M. & Hartman, L. (2001). Gender differences in attitudes toward affirmative action programs in Australia: Effects of beliefs, interests, and attitudes toward women. Sex Roles, 34, 415-432.        [ Links ]

Kravitz, D. A. (1995). Attitudes toward affirmative action plans directed at Blacks: Effects of plan and individual differences. Journal of Applied Social Psychology, 25, 2192-2220.        [ Links ]

Kravitz, D. A. & Platania, J. (1993). Attitudes and beliefs about affirmative action: Effects of target and of respondent sex and ethnicity. Journal of Applied Psychology, 78, 928-938.        [ Links ]

Krings, F., Tschan, F. & Bettex, S. (2007). Determinants of attitudes toward affirmative action in a Swiss Sample. Journal of Business and Psychology, 21, 585-611.        [ Links ]

Lynch F.R. (1992). Invisible Victims: White Males and the Crisis of Affirmative Action. New York: Praeger        [ Links ]

Matheson, K. J., Echenberg, A., Taylor, D. M., Rivers, D. & Chow, I. (1994). Women's attitudes toward affirmative action: Putting actions in context. Journal of Applied Social Psychology, 24, 2075-2096.        [ Links ]

Matheson, K. J., Warren, K. L., Foster, M. D. & Painter, C. (2000). Reactions to affirmative action: Seeking the basis for resistance. Journal of Applied Social Psychology, 30, 1013-1038.        [ Links ]

Myors, B, Lievens, F., Schollaert, E., Van Hoye, G., Cronshaw, S., Mladinic, A., Rodríguez, V., Aguinis, H., Steiner, D., Rolland, F., Schuler, H., Frintrup, A., Nikolaou, I., Tomprou, M., Subramony, S., Raj, S., Tzafrir, S., Bamberger, P., Bertolino, M., Mariani, M., Fraccaroli, F., Sekiguchi, T., Onyura, B., Yang, H., Anderson, N., Evers, A., Chernyshenko, O., Englert, P., Kriek, H.J., Joubert, T., Salgado, J., Kö-nig,C.J.,Thommen, L., Chuang, A., Sinangil, H.K., Bayazit, M., Cook, M., Shen, W. Sackett, P. (2008). International Perspectives on the Legal Environment for Selection. Industrial and Organizationa Psychology. Perspectives on Science and Practice, 1, 206-246.        [ Links ]

Nacoste, R. W. & Hummels, B. (1994). Affirmative action and the behavior of decision makers. Journal of Applied Social Psychology, 24, 595-613.        [ Links ]

Parker C.P, Baltes B.B, & Christiansen N.D. (1997). Support for affirmative action, justice perceptions, and work attitudes: a study of gender and racial-ethnic group differences. Journal of Applied Psychology, 82, 376-89.        [ Links ]

Parra, L. F. 1991. Development of attitudes toward affirmative action scale. Un-published Master's thesis. Bowling Green State University.        [ Links ]

Pylkkanen, E.M. (2004). Words that carry meaning: issue definition and affirmative action. McGill University. Montreal, Quebec, Canada: Unpublished Doctoral Thesis        [ Links ]

Real Decreto Legislativo 1/1995, de 24 de marzo Texto Refundido de la Ley del Estatuto de los Trabajadores, pp 2.        [ Links ]

Resenedez, M.G. (2002). The stigmatizing effects of affirmative action: An examination of moderating variables. Journal of Applied Social Psychology, 32, 185-206.        [ Links ]

Ruiz, J. & Moya, M. (2005). Consecuencias psicosociales de la implementación de las políticas de acción positiva para los no beneficiarios: Efectos diferenciales según el grupo beneficiario. Revista Interamericana de Psicología, 39, 413-420.        [ Links ]

Schuman H., Steeh C., Bobo L. & Krysan M. (1997). Racial Attitudes in America: Trends and Interpretations. Cambridge, MA: Harvard University Press.        [ Links ]

Sowell T. (2004). Affirmative Action Around the World: An Empirical Study. New Haven: Yale Univ. Press.        [ Links ]

Steeh C & Krysan M. (1996). The poll trends: affirmative action and the public, 1970-1995. Public Opinion Quarterly, 60, 128-58.        [ Links ]

United States (2004). Affirmative action appropriate under Title VII of the Civil Rights Act of 1964, Washington, DC: Government Printing Office. Available from http://www.access.gpo.gov/nara/cfr/waisi_dx-00129cfr1608-00.html.        [ Links ]

Zelnick B. (1996). Backfire: A Reporter's Look at Affirmative Action. Washington, DC: Regnery Publication.        [ Links ]

 

 

Address for correspondence:
Silvia Moscoso
E.U. Relaciones Laborales
University of Santiago de Compostela
15782 Santiago de Compostela
Spain. E-mail: silvia.moscoso@usc.es

Manuscript received: 27/08/2010
Review received: 30/10/2010
Accepted: 05/11/2010