With the register-based index, OGGI, we could not find any association between equality in companies and self-rated health. However, the odds for good health were 2.8 times higher for women who rated their company as gender equal; we could not find any statistically significant association for men. Why, then, is perceived gender equality in the workplace associated with good self-rated health for women but not for men?
We would like to put forward two possible explanations. As already stated,
men have a shorter life expectancy than women,
and according to the convergence hypothesis
] more gender-
equal conditions in both the private and public spheres would decrease male mortality but increase male morbidity
]. The theory of convergence should in this case mean that the differences in health outcomes, mortality and morbidity, which today show different patterns for men and women, would converge with increased gender equality – i.e. men would live longer and women would feel better. The gender differences we see today have various explanations, sometimes biological, sometimes more lifestyle related. Men’s shorter life expectancy is often explained by risk-taking, such as excess alcohol consumption, smoking, driving at speed and so on
]. It is well established that men seek help later than women and that this negatively affects their health. The social constructions of traditional masculinity are usually considered a disadvantage to men’s health
]. Further, it has been argued that women are taught to be more aware of their bodily functions and therefore more likely to discover ill-health and hence seek help more often and sooner than men
]. Women’s greater ill-health and higher levels of sickness absence are hence explained in many ways: some explain it as a way of handling the dual burdens of paid and unpaid work; others view it as constituted by the gendered structures of society; and yet other explanations cite the biological differences between men and women. The adjusted convergence theory implies that when gender equality increases and men and women have greater equality in their possibilities to shape their lives and society, more gender-equal health will follow – men will live longer and women will live healthier lives
Further, men are considered to benefit more from an unequal situation at work, which might explain why they do not perceive differences between men’s and women’s conditions as a threat to health. If men lose privileges in their working life, it does not necessarily affect self-reported health positively.
Individuals’ experiences of gender (in)equality are most likely linked to health
]. To be overlooked, or discriminated against, in career and/or salary based on sex most likely increases the risk for poor health. Marmot
] argues that socioeconomic position is an important determinant of health. His results hold even if controlled for the effects of income, education and lifestyle risk factors such as smoking. Marmot’s causal relations are explained by the benefits of “being in control” of one’s own life. This could be part of the explanation for the differences we see in this study between men and women.
The external assessment and individuals’ own perceptions of gender equality or inequality seem not to coincide. One possible explanation could be that in large organizations employees are not likely to be aware of the overall structural conditions but rather base their assessment on “micro” experiences, i.e. the gender climate in their immediate working environment. One could reasonably argue that employees are familiar with their own working conditions and those of their closest colleagues, but less so for the company as a whole. Companies that deviate positively from the general gender pattern should perhaps develop an internal information strategy, as this kind of data is not easy accessible to individual employees. Being aware of gender equality at work may even positively influence the micro climates in the company. The index can thus be used to reveal facts that employees are not aware of. This is also a reason for conducting a more formal analysis from an external perspective – it can reveal patterns not easily detected from the individual perspective.
Another important finding of our study is the discrepancy between the OGGI and self-rated equality in companies. We found a significantly larger proportion of women in gender-unequal companies who rated their company as equal. Though not significant, almost 40% of men in gender-unequal companies gave the same rating. An explanation might be that, in judging equality at their company, respondents in our study used other criteria than the indicators used for OGGI. The index is based on an interpretation of official policy, and it is known that the practices are paternalistic in one respect – how parents do share parental leave and temporary parental leave. All the major political parties in Sweden agree on the goal that parents should share parental leave equally. However, there are very considerable differences of opinions on how to achieve this: some parties campaign for a couple’s unrestricted right to decide the division themselves, thus hoping for voluntary change
], while others argue for more or less far-reaching legislation
]. However, even those parties that favour legislation are somewhat reticent, as this issue is controversial even among their traditional supporters. Thus, companies with this traditional pattern of parental leave are given a relatively low rank based on the register-based index interpreting policy implementation, while employees working in those companies might be very happy with the situation, interpreting it as a sign of gender equality.
This index is constructed from official registers in Sweden, a country well known for its collection of data, where registers are complete, extensive and accessible for research. Access to the same kind of linked data might be limited in other settings. Research using register data will, we believe, be better accepted if it is recognized as a first step in an analytical process aiming to better understand gendered structures in organizations.
However, the limitations of OGGI are balanced by its strengths. Not only is comparison of different organizations easy and affordable using this method, it also provides an opportunity to follow developments year by year in individual organizations.
One limitation of this study might be the selection of companies. We set out to compare those companies that were ranked as gender equal with those ranked as gender unequal using our index. We decided to invite approximately 3000 persons in total, 1500 from gender-equal companies and 1500 from gender-unequal companies. However, as few companies were ranked as gender equal on our index, we were unable to invite as many respondents from gender-equal companies. Additionally, fewer women were invited to participate, as fewer women work in the private sector. Nevertheless, the response rate for women was higher, which is normally the case in Sweden. We also had large differences in response rates between companies, from 8% to 92%. The differences were somewhat less in gender-equal companies, from 29% to 83%. We have no good explanation for the large differences. The company with the lowest response rate, 8%, had 23 employees, 21 men and two women. The sole respondent was a woman. The company with the highest response rate, 92%, employed 47 men and 11 women. At this company, 53 people answered the questionnaire. There was also one company, with a response rate of 11%, that had grown from 28 to 120 employees in four years, which might have influenced the rate of response. Our overall reflection is that we do not know in what way the large differences in response rate influenced the outcome. As the non-responders tended to be male, young and on lower incomes, those who responded might perceive their company as more equal, and might have better attitudes towards their working life and health. Hence, the discrepancy between the index and self-rated equality might be explained by these factors.
One of the aims of this study was to investigate whether a single question could provide an answer to the question of how gender equal a couple’s relationship is. The question evaluated was “How gender equal is your workplace?” One single question that is often used is self-rated health, which has been shown to be a good predictor of mortality and morbidity. We thus wanted to know whether it was possible to evaluate gender equality in the same effective way. If this single question on gender equality gave the same answer as a battery of variables, then the single question could be used. Our subsequent question was whether self-perceived gender equality gives a reliable answer concerning levels of gender equality? We do not think it does. It is not likely that individuals can appraise the entire picture when the question concerns equality at organisational level. Each person will reflect those aspects that most concern them and probably from their own perspective. On the other hand, however, no index can ever describe an individual’s situation.
Women who perceived their workplace to be quite or totally equal had higher odds of reporting good health than women who did not perceive their workplace to be equal. The results of this study therefore suggest that gender equality is beneficial for women’s health. And gender equality does not seem to be bad for men. The construction of masculinity, as described above
], may influence sickness absence patterns. In an environment where men, like women, are expected to participate in the upbringing of young children and stay at home when children are sick, it might also be more acceptable for a man to stay at home when he himself is sick (personal communication, Connell 2010). Although the changes in mortality patterns are obviously connected to shifts in smoking and drinking patterns, one could argue that steps towards increased gender equality are an underlying cause. Being a present and committed father might mean reducing mortality risks
]. This development can already be seen in men participating in their children’s upbringing
], and the long-standing life expectancy gap between men and women appears to be closing in many societies. Men take better care of their health than before and lead less risky lives, while women adopt more traditionally masculine behaviours such as smoking and alcohol consumption
To summarize, then, measuring gender equality is an important but contentious and challenging task. Nevertheless, it is necessary for policymakers, politicians, researchers and many others to monitor gender equality, and valid measures need to be developed to enable comparison and use in future policy making. In order to measure gender equality correctly, instruments must be designed, tested and validated. Our attempt in this study to compare different types of gender equality measures in relation to health is one contribution to this development. The results of our study are credible considering the contemporary health patterns and positions of the sexes in the labour market.
The OGGI can be used to monitor the development of gender equality in Swedish society and its possible impact on health. Longitudinal research at organizational level could contribute to the development of the field.