Problematizing gender, work and health: the relationship between gender, occupational grade, working conditions and minor morbidity in full-time bank employees

https://doi.org/10.1016/S0277-9536(98)00287-1Get rights and content

Abstract

It is commonly asserted that while women have longer life expectancy than men, they have higher rates of morbidity, particularly for minor and psychological conditions. However, most research on gender and health has taken only limited account of the gendered distribution of social roles. Here we investigate gender differences in morbidity whilst controlling, as far as possible, for one major role, namely participation in paid employment. There is substantial segregation of the labour market by gender; men and women typically work different hours in different occupations which involve varying conditions and differing rewards and costs. Here, we examine men and women working full-time for the same employer. This paper reports on a postal survey of employees (1112 men and 1064 women) of a large British bank. It addresses three main questions: do gender differences in minor morbidity remain if we compare men and women who are employed in similar circumstances (same industry and employer)? What is the relative importance of gender, grade of employment within the organisation, perceived working conditions and orientation to gender roles for minor morbidity? Finally, are these factors related to health differentially for men and women? There were statistically significant gender differences amongst these full-time employees in recent experience of malaise symptoms, but not in physical symptoms or GHQ scores. Controlling for other factors did not reduce the gender differences in malaise scores and produced a weak, but significant, gender difference in GHQ scores. However, gender explained only a small proportion of variance, particularly in comparison with working conditions. Generally similar relationships between experience of work and occupational grade and morbidity were observed for men and women. Throughout the paper, we attempt to problematize gender, recognising that there are similarities between women and men and diversity amongst women and amongst men. However, we conclude that the gendered nature of much of adult life, including paid work, continues to shape the experiences and health of men and women at the end of the twentieth century.

Introduction

It is commonly observed that women live longer than men, but that they have higher rates of morbidity and health service use than do men, particularly for minor and psychological conditions (see other papers in this issue; also Nathanson, 1975; Waldron, 1976; Gove, 1984; Verbrugge, 1985, Verbrugge, 1989; Verbrugge and Wingard, 1987; Miles, 1991). Suggested explanations for these differences include biological factors, reporting differences, different access to health care, and acquired risks relating to different exposures or behaviours (Verbrugge, 1985). This study focuses on gender differences in minor morbidity. It was designed to try to control for some of the `acquired risks' that are differentially distributed among men and women because they tend to occupy different social positions in contemporary British society (and indeed in all societies).

Given the pervasiveness of a gendered division both of labour and of rewards and costs associated with that labour, the distribution of occupational and family roles and of material and non material resources differs between men and women. Most studies of gender and health describe men and women in the aggregate. Since these roles and resources might themselves be directly or indirectly health promoting or health damaging, it may be their differential allocation among women and men, rather than other aspects of their gender (such as biological vulnerability, tendency to take risks or reporting behaviour) that generate different morbidity rates for males and females. Thus, aggregate comparisons may not be comparing like with like.

In order to explore the importance of this explanation for observed differences in morbidity between men and women, it would be necessary to control for as many role occupancy and socioeconomic conditions as possible and then to observe whether differences between men and women remain, increase or decrease. However, it is important to bear in mind that gender is one of the first attributes ascribed to an individual in all social encounters, usually from the moment of birth (in the absence of the revelation of pre-natal test results), throughout life and even in reconstructions of people's lives following death. As Lorber and Farrell (1991)have pointed out,

The reason for gender categories and the constant construction and reconstruction of differences between them is that gender is an integral part of any social group's structure of domination and subordination and division of labour in the family and the economy. As a major social status (if not the major social status) gender shapes the individual's opportunities for education, work, family, sexuality, reproduction, authority and the chance to make an impact on the production of culture and knowledge... Gender is built into the social order... The major social institutions of control — law, medicine, religion, politics — treat men and women differently (1991: 1–2. Emphasis in original)

Whilst acknowledging the embeddedness of gender in all aspects of our lives and the impossibility of separating `gender' from cumulative experiences of work and other major social positions (such as family roles) in any individual's biography, we think that recent theoretical perspectives and changes in the labour market lend themselves to an examination of gender and health in men and women in similar employment situations.

What we have attempted in this paper is to look at gender differences from a particular perspective by giving salience to the work `role'. Thus, we have tried to look at gender differences in morbidity whilst controlling, as far as possible, for one major role, namely participation in the formal labour market. We address 3 main questions: (1) If we compare men and women working full-time in similar jobs for the same organisation, what is the extent, direction and magnitude of gender differences in minor morbidity? (2) How important is the grade of employment within the organisation, the working conditions experienced by respondents and orientation to gender roles for these measures of minor morbidity? (3) Do these factors appear to be related to health differentially for men and women? In addressing these questions, we try to compare `like with like' as much as possible both through the selection of our respondents and by controlling for age and other personal circumstances such as marital and parental status.

The term `gender' is commonly used to signal the importance of the social and cultural aspects of being a woman or a man in a particular place and time. However, having declared an interest in men and women as socially constructed, as well as biological, beings, most theorists then go on to compare groups of (biologically defined) men with groups of (biologically defined) women unproblematically. As Alvesson and Due Billing (1997): p. 25) suggest: “It is very easy to classify people according to their (biological) sex, but defining the meaning and significance of this and finding out when, how and why men and women are treated differently can become a difficult task”.

The distinction between sex and gender was popularised by Oakley in the 1970s; in this formulation sex was used to refer to biological differences between males and females, assumed to be universal and unchanging, while gender was used to refer to culturally constructed notions of masculinity and femininity, argued to be highly variable (Oakley, 1972). Although this conceptualization has been useful theoretically and powerful politically, it is increasingly recognised as problematic. The distinction between the two has again become blurred in its usage (for instance those interested in comparing women and men's health have sometimes used `gender differences' and `sex differences' interchangeably and sometimes implicitly or explicitly contrasted gender with sex differences), perhaps because “much research in the sociology of health and illness now uses feminist theory only tacitly”, as Annandale and Clark (1996), p. 17) have contended. Also the conception of the biologically `given' nature of sex has been shown to be more complex and the processes and bases on which sex grouping is assigned and maintained problematic (Birke and Vines, 1987). Not only is there constant interaction between the biological and the social, but biological `facts' are socially constructed, while cultural views affect which `facts' are selected and how much importance they are given (Connell, 1985, Morgan, 1986). As Barrett has remarked, “contemporary Western feminism, confident for several years of its `sex–gender' distinction, has found, these various categories radically undermined by the new `deconstructive' emphasis on fluidity and contingency” (Barrett, quoted in Annandale, 1998).

However, despite this debate, a widespread assumption of `difference' between men and women has continued. For example, a highly prevalent (although often implicit) assumption in medical sociology and epidemiology is that whether one is male or female is one of the main determinants of one's health. Its pervasiveness is demonstrated by the way morbidity or mortality rates are almost always presented separately for males and females or controlled for age and gender, although it is seldom made explicit whether this is premised on an underlying biological or social rationale. Differences are seldom noted or discussed. This separate presentation then further reinforces assumptions that differences in health and longevity between men and women are substantial and universal (Macintyre, 1993).

Clarke (1983): p. 62) has argued that the data are much more complex than are commonly described: “Despite the plethora of studies on sex and illness, one would have an exceedingly difficult task should one want to describe the differences in the morbidity experience of men and women” and more recently, a number of authors have questioned whether there is a (necessary or inevitable) female excess in morbidity at all. Using cross-sectional interview data from 523 Canadian men and women, Kandrack et al. (1991)found no significant gender differences for self-rated health, preventative health beliefs or sickness absence and gender explained no more than 3% of the variance in bed days, recency of last medical check-up and number of visits to the doctor in the last year. Similarly, Macintyre et al. (1996): p. 617) were struck by the lack of a female excess of ill health and by the “complexity and subtlety of the pattern of gender differences across different measures of health and across the life course” when they examined data from 2 British studies. They found little or no gender difference for those who reported their health to be `fair' or `poor' or in those reporting any longstanding or limiting longstanding illness in either study. While there was a female excess in reporting malaise symptoms in both studies at all ages, this was much less common for physical symptoms. They concluded that a “female excess is only consistently found across the life span for more psychological manifestations of distress and is far less apparent or reversed, for a number of physical symptoms and conditions” (Macintyre et al., 1996, p. 621). Kandrack et al. (1991)concluded that non statistically significant gender differences are less likely to be reported than significant findings, so that difference, rather than similarity, has been emphasised in the literature. As we have argued before, the sociology of gender and health inherits a legacy of polarising men and women, male and female, masculine and feminine; “the dichotomisation of gender within social science and the observation that health status does indeed vary by sex, together predispose researchers to focus on the differences between men and women” (Annandale and Hunt, 1990: p. 25).

Feminist post-modernists have been some of the most vocal critics of those who have ignored similarities between men and women and diversity within men and women. They argue that categories such as `women' and `men' gloss over differences between women and between men, such as class, ethnicity, age and sexuality. Thus, a feminist position is not possible as, “none of us can speak for `woman' because no such person exists” (Flax, 1990: p. 56). Similarly, Connell (1985)has argued that conceptualising men as a homogenous group is not helpful.

Annandale and Clark (1996), see also Annandale, 1998) have used this viewpoint to challenge the concept of gender as it has recently been used in medical sociology. They argue that the discursive categories of `men' and `women' can be unhelpful when attempting to understand health, as this can lead to the valorising of gender differences. Men's `good' health is often seen as a backdrop against which to construct women's `poor' health, leaving men's health as `invisible' while women's health is problematized. “Ironically, it is almost as if women cannot be well any more (and... men cannot be ill)” (Annandale and Clark, 1996: p. 29).

A number of feminists have expressed their unease with postmodernism. While Nicholson (1990)is critical of the legacy of the Enlightenment which presents (white, middle-class male) scholarship as a value-free, `view from nowhere', she questions the wisdom of a move to the relativistic `view from everywhere' that some postmodernists favour. Similarly, Di Stefano (1990): p. 77) is wary of moving to a `perplexing plurality of difference' and Jackson (1992)argues that post-modernism is potentially dangerous for feminism as it may undermine feminist politics; while acknowledging the problems of using `women' as an unitary category, to discard it denies the existence of women as a political group and the material realities which constrain women. “Women are being deconstructed out of existence and `gender' is replacing women as the starting point of feminist analysis” (Jackson, 1992: 31).

In the analysis which follows, we compare minor morbidity amongst men and women, as we believe that these socially constructed categories persist in having profound significance at the end of the twentieth century. We explicitly do not want to suggest that the categories `women' and `men' cannot be used. However, as we are concerned too with the complexities of gender, we have also attempted to explore some of the diversity within as well as between groups of men and women. Throughout this paper, we have chosen to use the term `gender' (through the lack of any better term) to distinguish between men as a group and women as a group. In addition, we have included a group of variables we have called `orientation to gender roles', to try and give empirical recognition to the fact that attributes such as `masculinity' or `femininity' are not necessarily coincident with gender (see Annandale and Hunt, 1990).

A number of studies of both white collar and blue collar workers have shown that various job related factors (such as work-related self esteem, work pressure, job satisfaction, physical working environment, degree of variety and challenge in work, decision latitude) are extremely powerful predictors of physical and psychological morbidity (Hurrell and Smith, 1981; Lowe and Northcott, 1988; Loscocco and Spitze, 1990; Lindstrom, 1991; Cahill and Landsbergis, 1996; Stansfeld et al., 1997, Walters et al., 1997). Several of these have shown that these features of the work environment may explain more variance in morbidity than whether the subject is male or female (Hurrell and Smith, 1981; Lowe and Northcott, 1988; Lindstrom, 1991; Cahill and Landsbergis, 1996).

Fewer women that men work full-time in Britain; in 1996, only 37% of women aged between 16 and 59 were in full-time employment compared to 70% of men aged between 6 and 64 (ONS, 1997). Although there has been a substantial increase in the number of women in traditionally male occupations, occupational gender segregation is one of the most enduring features of the labour market (Reskin and Roos, 1990; Williams, 1993). Horizontal and vertical occupational segregation of the labour market contributes to low-paying jobs with little likelihood of advancement for many women. Women earn on average two-thirds of men's earnings, occupy less prestigious jobs and are often clustered at the lower end of the status scale in each category where they may be subject to time-pressures and work allowing little autonomy (Sorensen et al., 1985; Hibbard and Pope, 1987; Lennon-Clare, 1987; Kauppinen-Toropainen et al., 1988; Arber and Gilbert, 1992; Hunt and Emslie, 1998). Jobs are gender-typed as either `male' or `female', although the characterisation of a particular job may change over time (see Alvesson and Due Billing, 1997for examples). Hunt and Emslie (1998)classified 670 respondents in paid employment in a random community sample of 40-yr olds living in the West of Scotland according to the `gender ratio' of each occupation (see Kanter, 1977). They found that 79% of women worked in `male minority' occupations (between 61% and 80% women), while 47% of men worked in `female token' occupations (up to 20% women) and 30% of men worked in `female minority' occupations (between 21% and 40% women). In Britain, most women are concentrated in a small number of sectors and occupations such as in clerical, service, sales and factory work and the so-called `semi-professions' of nursing, librarianship, teaching and social work. Legge (1987)has argued that occupations perceived as peripheral are more likely to have women in senior grades, while those seen as important to strategic decision making are unlikely to place women in positions of power. Indeed, women have often only gained entry to an occupation because it has become less attractive to men, due to decreased rewards, poorer working conditions or the increase of duties traditionally seen as `female'.

Much previous work on gender, roles and health has not treated men and women symmetrically. Whereas men's health has been evaluated primarily in terms of their `traditional' role as breadwinners, the relationship between women and work has often been conceptualised as secondary to that of their role in the household. Thus paid work may be seen as an additional role for women, instead of being understood as an indicator of status, income and class position as it is for men (Arber, 1991; Hunt and Annandale, 1993). The implications for health of the actual content of the job and the working conditions under which it is performed tend to be less frequently studied for women than for men. Conversely, very few studies examine men's, as compared to women's, domestic and family responsibilities and their implications for health.

Feldberg and Glenn (1979)summarised the tendency to take a gender segregated approach almost 20 yr ago, coining the expression `job model' for men and `gender model' for women. The job model uses the paid work people do to explain employees' health or their behaviour both at work and away from work. In the gender model, characteristics of the job itself are ignored in favour of personal characteristics and family circumstances.

This separation into job and gender models implies that the relationship between paid work or domestic roles and health differs between men and women, ie that men's and women's health have different social determinants. This implication has rarely been examined systematically (although see Hunt and Annandale, 1993; Popay et al., 1993; Macintyre and Hunt, 1997). One aim of this paper is thus to examine whether there are significant differences between men and women in the associations between measures of minor recent morbidity and the predictors examined, namely occupational grade, perceived working conditions and orientation to gender roles.

In summary, the questions we seek to address here are: first, do gender differences in health remain when we compare men and women who are employed in similar circumstances (full-time for the same single organisation); secondly, what is the relative importance of gender (being a man or a woman), occupational grade, perceived working conditions and orientation towards gender roles in explaining minor morbidity amongst these bank employees and thirdly, is there any evidence that the relationships between these `predictors' and health are different for men and women?

Section snippets

Sample

The study was designed specifically to study gender differences in morbidity after controlling for occupational participation, employment sector and employer. We sought a sample of men and women working full-time (35 h or more per week) in similar jobs within the same organisation. The service sector has expanded hugely over the last 30 yr with the transformation from a manual to a service-producing economy in Britain and the majority of female workers throughout the European Union work in the

Results

Results for the bivariate analysis are given in Table 1. Women reported a significantly larger mean number of malaise symptoms than men (1.5 vs. 1.3 respectively). However, the number of physical symptoms reported and GHQ scores, were very similar for women and men.

We then went on to compare the health of men and women after controlling for a wide range of variables, focusing particularly on occupational grade, perceived working conditions and orientation to gender roles. The results of a

Discussion

This paper aimed to make a specific contribution to debates about gender differences in health. Many previous studies of gender and health have used general population samples which, given the strongly gendered nature of contemporary society, means that comparisons between men and women may be confounded with comparisons between other axes of social differentiation such as employment and domestic roles and access to material and non material resources. The study sought, as far as possible, to

Conclusion

The results in this study lend more support to the `job' model, which privileges the characteristics of paid employment in explanations of health and working conditions, than the `gender' model, which focuses on family and personal characteristics (Feldberg and Glenn, 1979). According to our analysis, it is not whether employees are male or female, whether they are married or unmarried or whether or not they are parents which is most predictive of minor morbidity, but how they experience their

Acknowledgements

We would like to thank the Bank personnel staff and participants for their help in administering and participating in this research. We would like to thank Geoff Der for his statistical help and Pat Fisher, Lindsay Macaulay and Margaret Reilly for their clerical assistance. Particular thanks are due to Ellen Annandale and an anonymous referee, for their perceptive criticisms of an earlier version of this paper. The Medical Research Council funded this study and the authors.

References (85)

  • I Waldron

    Why do women live longer than men?

    Soc. Sci. Med.

    (1976)
  • S Wyke et al.

    Competing explanations for associations between marital status and health

    Soc. Sci. Med.

    (1992)
  • Acker, J., 1991. Hierarchies, jobs, bodies: a theory of gendered organizations. In: Lorber, J., Farrell, S.A. (Eds.),...
  • P.K Adelmann

    Occupational complexity, control, and personal income: their relation to psychological well-being in men and women

    J. Appl. Psychol.

    (1987)
  • Alvesson, M., Due Billing, Y., 1997. Understanding Gender and Organizations. Sage Publications,...
  • Annandale, E., 1998. The Sociology of Health and Medicine. Polity Press,...
  • E Annandale et al.

    What is gender? Feminist theory and the sociology of human reproduction

    Sociol. Health Illness

    (1996)
  • E Annandale et al.

    Masculinity, femininity and sex: an exploration of their relative contribution to explaining gender differences in health

    Sociol. Health Illness

    (1990)
  • Arber, S., Gilbert, N., 1992. Re-Assessing women's working lives: an introductory essay. In Women and Working Lives....
  • M.H Banks et al.

    The use of the general health questionnaire as an indicator of mental health in occupational studies

    J. Occup. Psychol.

    (1980)
  • Bartley, M., Popay, J., Plewis, I., 1992. Domestic conditions, paid employment and women’s experience of ill-health....
  • Bem-Lipsitz, S., 1981. Bem Sex-role Inventory Professional Manual. Consulting Psychologists Press, Palo Alto,...
  • C.E Bird et al.

    Gender, time use, and health

    J. Health Soc. Behav.

    (1991)
  • Blaxter, M., 1990. Health and Lifestyles. Routledge,...
  • Bordo, S., 1990. Feminism, postmodernism, and gender-scepticism. In Feminism/Postmodernism. Routledge, New York, pp....
  • J Cahill et al.

    Job strain among post office mailhandlers

    Int. J. Health Serv.

    (1996)
  • J.N Clarke

    Sexism, feminism and medicalism: a decade review of literature on gender and illness

    Sociol. Health Illness

    (1983)
  • Cockburn, C., 1990. Men’s power in organizations: ‘equal opportunities’ intervenes. In: Men, Masculinities and Social...
  • R.W Connell

    Theorising gender

    Sociology

    (1985)
  • Cox, B.D.M., Blaxter, A.L.J., Buckle, et al., 1987. The Health and Lifestyle Survey. Health Promotion Research Trust,...
  • Di Stefano, C., 1990. Dilemmas of difference: feminism, modernity, and postmodernism. In: Nicholson, L.J.,...
  • Emslie, C., 1997. Banking On Good Health? Gender Differences in Minor Morbidity Amongst Men and Women Working Full-Time...
  • Emslie, C., Hunt, K., Macintyre, S., 1997. Gender differences in minor morbidity amongst men and women working full...
  • R.L Feldberg et al.

    Male and female: job versus gender models in the sociology of work

    Soc. Problems

    (1979)
  • Flax, J., 1990. Postmodernism and gender relations in feminist theory. In: Feminism/Postmodernism. Routledge, New York,...
  • Goldberg, D., 1972. The Detection of Psychiatric Illness by Questionnaire. Oxford University Press,...
  • W.R Gove

    Gender differences in mental and physical illness: the effects of fixed roles and nurturant roles

    Soc. Sci. Med.

    (1984)
  • Halford, S., Savage, M., Witz, A., 1997. Gender, careers and organisations. Current developments in banking, nursing...
  • V.S Helgeson

    The role of masculinity in a prognostic predictor of heart attack severity

    Sex Roles

    (1990)
  • J.H Hibbard et al.

    Employment characteristics and health status among men and women

    Women and Health

    (1987)
  • K Hunt et al.

    Just the job? Are the health effects of domestic and paid work gender-specific?

    Sociol. Health Illness

    (1993)
  • Hunt, K., Emslie, C., 1998. Men's work, women's work? Is there an effect of taking account of occupational sex ratios...
  • Cited by (0)

    View full text