Gender segregation in the labour market and ...

4 downloads 0 Views 85KB Size Report
Keywords: Labour market segregation; Work postures; Gender. 1. ... (md), minimum to maximum values (range), and Mann-Whitney test statistics (P) used to ...
Gender segregation in the labour market and exposure to awkward work postures Ola LEIJON1,2, Eva BERNMARK1,2, Lena KARLQVIST1,3, Annika HÄRENSTAM1,3

Address: 1 Department of Public Health Sciences, Division of Occupational Medicine, Karolinska Institutet, Norrbacka, SE-171 76 Stockholm, Sweden 2 Department of Occupational and Environmental Health, Stockholm County Council, Norrbacka, SE-171 76 Stockholm, Sweden 3 National Institute for Working Life, SE-113 91 Stockholm, Sweden Topic classification: Exposure; Ergonomics; Prevention Keywords: Labour market segregation; Work postures; Gender 1. Introduction

Gender segregation in the labour market may entail occupational and gender differences in social status, salary, position, and advancement possibilities. Moreover, segregation of women and men into different jobs is often cited as one of the most plausible explanations for gender differences in exposures and, hence, gender differences in work-related musculoskeletal disorders. However, few studies have focused on the distribution of exposure to physical risk factors such as awkward work postures, in relation to gender segregation in the labour market. Present study focuses on differences and similarities in exposure to sitting/standing and awkward arm and trunk postures between workers in female-dominated, male-dominated or gender-mixed occupations (horizontal segregation) and between workers with low or high influence at work (vertical segregation). Interaction effects between horizontal segregation, vertical segregation and gender in relation to exposure to awkward work postures are also investigated. 2. Methods

The data were collected in five Swedish counties between 1995 and 1997 (Härenstam et al, 2003). A sample of 78 matched pairs (one woman and one man) representing 83 different job titles, were selected for this study. Each pair was employed in the same type of work, i.e. working with people, data or things (Kohn and Schooler, 1983), and had the same qualification level according to the socio-economic classification, categorised into low, medium or high level. Gender distribution in the occupation was used as an indicator variable of horizontal segregation in the labour market. Each worker was classified into one of the three groups: female-dominated job (>70% women holding the job title), male-dominated job (>70% men holding the job title) and gender-mixed job (30-70% of both women and men holding the job title). Influence at work was used as an indicator variable for vertical segregation in the labour market. Data on level of influence was gathered by a job analysis where each subject was followed for usually one typical workday by one of four well-trained observers (Waldenström et al., 2003). The data was dichotomised into low or high influence at work. Direct, full workday, measurements of sitting/standing, arm and trunk postures were taken, for each of the 156 subjects, with two different technical instruments. Both duration of body segment postures and frequency of movements were recorded. Associations between gender segregation in the labour market and exposure were analysed in the context of gender order theory. 3. Results

Subjects in female-dominated occupations experienced shorter duration of sitting work posture

compared to subjects in male-dominated or gender-mixed occupation (Jonckheere-Terpstra P = 0.011). There were no statistically significant differences between the groups concerning arm or trunk postures and movements. Subjects with a low level of influence experienced higher duration and frequencies of awkward work postures and movements compared to those with high influence at work (Table 1). Table 1. Vertical segregation and work postures/movements for subjects in the matched pairs. Median values (md), minimum to maximum values (range), and Mann-Whitney test statistics (P) used to compare the two groups with low or high influence at work.

duration of sitting work posture (%) posture changes sit-stand per hour (freq.) duration of arm elevation >60º dominant arm (%) non-dominant arm (%) arm elevations per hour exceeding 60º dominant arm (freq.) non-dominant arm (freq.) duration of trunk bend >40º (%) trunk bends per hour exceeding 40º (freq.)

Low influence at work md range 38 1–84 10 2–33

High influence at work md range 62 7–92 11 1–31

P 0.001 0.188

3 3

1–15 1–17

2 2

0–15 0–14

0.003 0.003

114 86 7 111

19–557 23–344 1–35 31–369

58 40 6 70

11–356 9–269 0–24 16–268

40o. The three interaction effects had the same direction: higher exposure for subjects with low level of influence compared to those with high influence and higher exposure for women compared to men (Two-way analyses of variance, ANOVAs, P = 0.019–0.035). 4. Discussion and conclusions In summary, the analysis of gender segregation showed that vertical segregation in the labour market, as measured in present study, appears to be the strongest determinant for exposure differences to awkward work postures. Furthermore, interaction effects indicated gender differences to disadvantage for women. This finding may imply that influence at work does not account for all relevant aspects that might explain exposure differences in relation to vertical segregation. Finally, the lack of exposure differences due to horizontal segregation indicates that no general recommendations can be made, either for women and men or girls and boys, when considering exposure to physical load (awkward work postures) and choice of future occupation. 5. References

Härenstam A, et al. 2003. Patterns of work and living conditions: a holistic, multivariate approach to occupational health studies. Work & Stress 17:73-92. Kohn ML, Schooler C. 1983. Work and personality: an inquiry into the impact of social stratification. Norwood, NJ: Ablex, 389 p. Waldenström K, et al. 2003. Does psychological distress influence reporting on demands and control at work? Occup Environ Med 60:887-891.