Can they work it out and do they get any satisfaction? Young Swedish physically disabled men's and women's work involvement and job satisfaction

In this article, results are reported from a study of what value young (20-35 years) disabled men and women ascribe to a job and the job satisfaction they have. Data for the study were collected vi ...


Introduction
This article reports a study on how physically disabled young women and men value work and what job satisfaction they have. As an introduction a short summary of the research area is given. The introduction shortly discusses the employment situation of disabled persons, including gender differences in the labour market. The introduction also discusses the character of barriers that disabled people are confronted with when trying to gain employment and the importance of understanding how individuals value work and experience job satisfaction.
In discussions about work and its value, it is often emphasized that people achieve better health by working. Jahoda's (1979) functionalistic approach is often used to explain the importance of a job. It is assumed that, apart from the purely economic gains, a job contributes towards satisfying several psychological needs such as the need for structure in everyday life, the need for social contacts outside the family and the need for status and identity (see for example Hammarströ m 1996;Clasen, Gould, and Vincent 1998;Barnes, Mercer, and Shakespeare 1999;Szebehely, Fritzell, and Lundberg 2001). People with disabilities have the same needs to satisfy in this respect as other groups in society. However, a disability has been shown to contribute to discrimination on the labour market (Barnes, Mercer, and Shakespeare 1999). Statistics from Sweden demonstrate, for example, that people with a disability have a lower level of employment (65%) than persons without disabilities (77%). Persons with a disability are unemployed to a higher degree than other groups and work part time more often than the population as a whole (Statistiska Centralbyrån 2003a). In this respect, conditions in Sweden and internationally are similar. Research also demonstrates that individuals with a disability are unemployed in the long term and granted an early retirement pension more often than individuals without disabilities (Barnes, Mercer, and Shakespeare 1999;Bricout and Bentley 2000;Jolly 2000;Reynolds, Nicholls, and Alferoff 2001;Barnes and Mercer 2005).
When it comes to conditions for the two sexes, research indicates that the situation for women with a physical disability is more difficult than it is for men (Jolly 2000). Research shows that women with a disability are employed outside the home to a lower degree than men and that they also have lower incomes than men (Jolly 2000). This latter finding can presumably be explained by the fact that women more often are working in relatively low-paid jobs, for example different types of service jobs, and that they more often work part time (Jolly 2000). Research also shows that women with a disability more often telecommute from the home than do men (Jolly 2000). This seems to have negative consequences since the jobs women as a group perform from the home are, generally speaking, associated with relatively low pay and the lack of colleagues tends to lead to isolation (Jolly 2000).
The barriers to the labour market, which persons with a disability are confronted with, are of two kinds. The first of these are individual-related, i.e., factors that can be related to the individual him/herself in the form of the degree of disability, individual educational level, own attitudes and so on. The second category is made up of societal factors, i.e., factors that can be related to both social and physical surroundings in the form of a poorly adapted physical environment, employers' negative attitudes (Thornton and Lunt 1997;Barnes, Mercer, and Shakespeare 1999;Bricout and Bentley 2000;Reynolds, Nicholls, and Alferoff 2001), poor employment prospects (Gonäs 1991) and demands on the employee's flexibility and education (Arnell Gustavsson 1996).
It could be claimed that it is equally important to study barriers of a subjective or experienced nature as it is to understand objective barriers that exist for people with physical disabilities. This is because a complete understanding of individuals' actual life careers cannot be acquired without knowledge of both the above-mentioned dimensions (cf. Layder 1993).
When it comes to disabled persons' relation to paid work, the reason for their lower level of employment must be sought in the interactions between individual (biological as well as psychological and socio-psychological) factors and environmental factors rather than in isolated events and factors. This approach is in line with the life-course view developed by Sampson and Laub (2005) in another field, namely, the development of crime. According to the authors, the development of an individual's career in a certain area should be seen as the 'constant interaction between individuals and environment as coupled with random developmental noise and a purposeful human agency' (12).
How physically disabled people value work and what job satisfaction they have are two important factors of an individual-related nature, which determine their labour market careers. This is because both these factors affect the individual's ability to get and keep a job. Research shows, namely, that persons who attribute high value to work look for work to a higher degree (Å berg 1997;Bolinder 1999;Julkunen 2002). This applies to both the formal search methods (answering job ads in newspapers) and informal search methods (contacting friends and acquaintances and taking other initiatives), which the individual can utilize. Studies also demonstrate that high motivation to work increases the individual's prospects of being reemployed (Å berg 1997;Bolinder 1999;Julkunen 2002). Similarly, there is evidence that job satisfaction has a direct impact on the employee's performance in the workplace since employees with high job satisfaction perform better than persons with low job satisfaction (Furåker 2003). Low job satisfaction also leads more often to sickness absence and less chance of returning to work after periods of sickness absence (Hö ög 1985;Furåker 2003).
In order to shed further light on individual-related factors that are of importance to the labour market situation facing young, physically disabled persons, the aim of the study is to investigate if there are any differences between physically disabled young women and men concerning value of work and experience of job satisfaction.

Method Respondents
The study includes persons living in Sweden and aged 20Á35 years, the age group most active in the labour market. Swedish studies namely show that physically disabled persons younger than 20 years often are still in school while individuals older than 31 are often on early retirement pension and that participation in higher education in this group is rather low, only 18% (see for example Larsson 1997;Statistiska Centralbyrån 1997).
The respondents were selected from the National Labour Market Board's (AMS) register. They were selected on the basis of having satisfied the following criteria at the time of the study: (1) That they were classified as physically disabled in the official register, i.e., they suffered from impairment in their neck, shoulders, back, legs, arms or hands (see also Statistiska Centralbyrån 2003b). In Sweden, the degree of disability is defined and categorized partly on the basis of a medical evaluation of the degree of difficulty in coordinating movements and partly on the basis of a report on the individual's working capacity. The latter is determined by AMS after talking with the individual using self-evaluations and assessment of work performance; (2) That they were aged between 20 and 35 years; (3) That they were the subject to some form of labour market policy measure initiated by AMS. The employment situation of the respondents at the time of the study is accounted for in Table 1. Research procedure and non-response rate Postal questionnaires were mailed to respondents during the first half of 2003. A total of 1690 questionnaires were mailed and 706 answers were received, i.e., a response frequency of 42%. It is known that questionnaire studies conducted in Sweden among persons with a functional impairment are, generally speaking, associated with a non-response rate of about the same size as in the present study (see for example Larsson 1997); nevertheless, it is important to reflect on how this may have affected the validity of the study. Conceivable reasons for the large non-response rate could be that some of the persons asked did not regard themselves as having a physical disability and thus felt that the study was irrelevant to them. Another reason could be that some of the addresses received from the National Labour Market Board's register were incorrect and that the persons receiving the questionnaire thus felt (quite rightly) that they did not belong to the category described. Yet another reason could be that some of the respondents felt that their address had been given out in an incorrect way and thus were unwilling to answer. These assumptions were confirmed by the fact that four selected respondents phoned the research leader, claiming that they did not have any physical disability. Others got in touch and expressed dismay at their address having been given out by the authority. Yet another possible reason for the non-response, proposed by Barron (1997), could be that the respondents were quite simply tired of answering questions about their private life, which they felt were asked by representatives of the authorities.
It would have been possible to check for non-response bias by contacting the recipients of the questionnaire by phone (Ejlertsson 1996). However, it was decided not to use this procedure as the respondents were considered to be a sensitive group, who could be resentful if the researchers made contact once more. Instead, an analysis was made of register data and the questionnaires received. This analysis shows that there were no large differences in the average age of the persons in the register data and the respondents (average was 29 years and 30 years). However, the non-response analysis shows that men were consistently underrepresented among respondents in all groups. Thirty-five percent of the respondents are men, as against 43% of all. From a gender perspective, there is thus reason for a degree of caution when interpreting the results.

Research instrument
The concept value of work refers to the value an employee attributes to paid work. This concept is often measured in both economic and psychosocial terms (see for example Nordenmark 1999;Julkunen 2002). In the present study, the concept value of work has been measured with the help of the Work Involvement Scale (WIS), designed by Warr, Cook and Wall (1979). WIS measures 'the degree to which a person wants to be engaged in work ' (1979, 133). The instrument measures the respondents' work attitudes and aspects of psychological wellbeing in work and consists of six questions. Five of the six questions from the original WIS were used in this study. These questions were translated into Swedish by Isaksson and Svedberg (1989) in a previous study. The five questions used were: (1) Even if I won a great deal of money on the pools I would continue to work somewhere; (2) I should hate to be on the dole; (3) I would soon get very bored if I had no work to do; (4) The most important things that happen to me involve work; (5) If unemployment benefits were really high I would still prefer work.
The sixth question in WIS, Having a job is very important for me, was replaced with the question I only want to have a job for the sake of the salary. This change was made in order to capture the economic importance of work to the individual. Another reason for the change was a desire to be able to compare with the answers given in an earlier interview study made by the one of the researchers (see Dag 2004).
Several studies in the field of working life confirm the scientific importance of measuring both the above-mentioned areas (see for example Goldthorpe et al. 1968;Korpi 1978;Priebe et al. 1998;Nordenmark 1999;Julkunen 2002).
The answers concerning the value of work have been obtained with the help of a 'verbal descriptive scale' consisting of four answering alternatives, which are as follows: 0 0Disagree completely; 1Á20Agree to some extent; 3Á40Agree in part; and alternative 5Á60Agree completely.
The concept job satisfaction has been used since the first decades of the twentieth century (Wright 2006). Questions about job satisfaction are usually in the form of what degree the employee is satisfied or, alternatively dissatisfied, with different aspects or dimensions of his/her work. Job satisfaction can also be described as an emotional state linked to work (Smith, Kendall, and Hulin 1969;Warr, Cook and Wall 1979;Butler 1990;Arches 1991;Kaufmann and Kaufmann 2005). This subjective experience of job satisfaction can be affected by the person's experience of, demands on and expectations of his work (Månsson 2006). Questions about job satisfaction often concern areas such as salary, promotion, colleagues, the boss and the work itself (cf. Smith, Kendall, and Hulin 1969).
In the present study, the respondents' job satisfaction is based on their subjective experiences of their present or most recent employment and includes negative and positive valuations of the respondents' own work environment (Smith, Kendall, and Hulin 1969;Warr, Cook and Wall 1979;Trygged 1996). The concept job satisfaction has been measured with the help of six questions concerning the following areas: (1) the physical conditions in the workplace; (2) colleagues; (3) acknowledgement of work well done; (4) the attitude of the employee's immediate superior; (5) his/her salary; and finally (6) practical help received in the workplace.
Different studies of the life situation of physically disabled persons in the labour market have emphasized the significance of all six questions areas. One example that can be mentioned is the physical conditions in the workplace, but also practical assistance in the workplace is of importance for the opportunities of a physically disabled person to get and keep his/her job (Russell 1998;Barnes, Mercer, and Shakespeare 1999;Barron 1997;Dag 2006). Questions about colleagues and immediate superiors were included because these aspects have been shown to be important for the individual's job satisfaction (Hatton et al. 1999). Similarly, the authors find that acknowledgement of work performance is important for job satisfaction. The question about salary was included since research shows that physically disabled persons in general earn less than others (Russell 1998;Barnes, Mercer, and Shakespeare 1999).
The answers concerning job satisfaction have been obtained with the help of a 'verbal five step descriptive scale' (1) I am very satisfied; alternative (2) I am fairly satisfied; alternative (3) I am neither satisfied nor dissatisfied; alternative (4) I am rather dissatisfied; and finally, alternative (5) I am very dissatisfied.

Analysis
The statistical analyses have been performed with the help of frequency and multivariate calculations. The multivariate analysis was performed with logistic regression analysis which is a non-parametrical test recommended when analyzing ordered categorical data (Svensson 2001). Spearman's rank correlation coefficient (r s ) has been used as the correlation measurement. Spearman's rank correlation coefficient is a useful method for calculating the correlation between two variables of ordinal scale level. The Mann-Whitney test, which is also useful for data of a nonparametric nature, was used to examine whether two samples of the data material differed from each other (Brace, Kemp, and Snelgar 2003).

Results
The employment situation of the respondents at the time of the study is accounted for in Table 1. Table 1 shows that 28% of the respondents were unemployed, 14% had an employment limited in time and 14% had permanent employment. The largest proportion of the respondents were subjected to some sort of labour market policy measure. The table also shows that 7% were students. Table 2 shows that the greatest proportion of the respondents (55%) were married or cohabitant, 25% were single persons without children, 12% were single parents. Eight percent of the respondents were living with their parents.

Value of work
The results (not displayed in the table) show that regardless of gender and employment situation 90% of all the respondents partly or fully agree that they would work even if they won a large sum of money and that only 38% want to have a job for the sake of the salary. Ninety percent partly or fully agree that they hate being unemployed. Eighty-eight percent state that they get bored if they do not have a job and 83% say that a job is one of the most important events in their lives. An overwhelming 94% said that they would prefer to work even if unemployment benefits were generous. Results (also not displayed in the table) show that, regardless of gender, 90% of those unemployed, employed (91%) and participating in labour market interventions (91%) partly or fully agree that they would work even if they won a large sum of money. A relatively high percentage of the respondents who were studying (71%) responded in the same way. Answers concerning value of work, based on employment situation and gender, are accounted for in detail in Table 3. Table 3 shows that both men and women in general attribute a great psychosocial value to work. A slightly larger proportion of the women (91%) and the men (88%) partly or fully agree that they would work even if they won a large sum of money. Ninety-one percent of the women, and 87% of the men, also state that they hate or would hate being unemployed. Table 3 also shows that more women (85%) than men (79%) say that a job is one of the most important events in their lives (p00.05). In addition, more women (96%) than men (92%) state that they would prefer to work even if employment benefits were generous (p 00.05). The impression that women assign greater psychosocial value to work than men do is supported by the fact that fewer women (34%) than men (46%) partly or fully agree that they want a job solely for the money it pays (p00.00).
No large differences are found between respondents of the sexes and how they value work if their whole employment situation is taken into account. Table 3 shows that that respondents generally speaking attribute a high value to paid work regardless of whether they are unemployed, participate in labour market interventions, are employed or are studying.
In Table 3 it can also be seen that both women and men, regardless of employment situation, value paid work highly. The table shows that women assign higher values to all the statements about the value of work than do men in all employment situations with the exception of the statement that they want a job for the sake of the money it pays. In the case of the latter statement, the results show instead that more men than women, regardless of employment situation, said that they want a job for the sake of the money it pays. Additionally, fewer unemployed women (34%) than unemployed men (45%) say that they want to work for sake of the money. Fewer women participating in a labour market intervention (38%) than men in the same situation (44%) state that they want to work solely for the sake of the money. It is also noteworthy that the differences between men and women who are employed or are studying are relatively large. Here, the results show that far fewer women who have some form of employment (29%) and women who are studying (29%) state that they want to work for the money they are paid than do men with a job (50%) (p0B0.05) and men who are studying (58%) (p00.05).
As regards the relationship between the respondents' marital status and how they value work, the results show that the women, regardless of civil status, value a job somewhat more than men do (see Table 4). Table 4 shows that a high proportion of single women with children (94%) and men in the same situation (79%) state that they would want to work even if they won a large sum of money. Similarly, a high proportion of single women without children (96%) and men without children (92%) say that they would want to work even if they won a large sum of money. The data show similar results for married/co-habiting men and women, i.e., 89% of women and 86% of men state that they would want to work even if they won a large sum of money. We also see that fewer women than men in all the categories want to work for the sake of the salary paid. For example, far fewer women living alone with children (23%) than men in the same situation Scandinavian Journal of Disability Research 293 Notes: Significance levels *00.05, ** 0B0.05 ***00,00.
(47%) say that they want to work for the sake of the salary paid (p00.05). The results also show that fewer married/co-habiting women (36%) than married/cohabiting men (43%) state that they only want to work for the sake of the salary paid (p 00.00). Table 5 shows that the respondents for the most part express high job satisfaction. Seventy-nine percent state that they are satisfied with the physical conditions in the workplace or where they are studying and 95% also say that they are satisfied with their workmates or fellow students. Table 5 shows that 85% are satisfied with their work or study acknowledgement and 84% are satisfied with their immediate superior or teacher. However, this pattern changes when it comes to pay structure. This question rates the least satisfaction with only 42% of the respondents stating that they are satisfied. From Table 5 it can be seen that the respondents, independently of their marital status or employment situation, have high job satisfaction. However, independently of their marital status, the employed are satisfied with the pay structure to a higher degree than the other groups (unemployed, employment measures and studying). Table 5 shows that the employed, independent of marital status, are satisfied with the acknowledgement they get for the work they do to a higher degree than the other groups.

Job satisfaction
No significant differences between the sexes were found in the questions about job satisfaction.

Value of work in relation to job satisfaction and other conditions
A multivariate analysis performed with logistic regression analysis shows that gender (p 00.048), marital status (p 00.041), Acknowledgment for work performance (p 0 0.039) and pay structure (p 00.008) have significant effects in the respondents value of the psychosocial dimension of work (see Table 6). According to Table 6, the probability for attributing a high psychosocial value to work is higher among women than men (odds ratio 1.52). The results also show that respondents that are satisfied with the pay structure (odds ratio 1.25) and respondents who are satisfied with acknowledgment for work performance (odds ratio 1.19) attribute a higher psychosocial value to work compared to respondents that are dissatisfied with the acknowledgment they get for the work they do and respondents that are dissatisfied with their salary. Further, the results show that single persons with or without children attribute a lower psychosocial value to work than those who are cohabiting or married (odds ratio 0.81). Table 7 shows that men with high job satisfaction assign a higher psychosocial value to work than do men with low job satisfaction. The difference in valuation of work between women with high job satisfaction and women with low job satisfaction is small. Men who are satisfied with the acknowledgement they are given for work performance (85%) state more often than men who are dissatisfied (56%) that work is one of the most important aspects of their lives. The corresponding figures for the women are 87% (satisfied) and 81% (dissatisfied) (r s 0.282, p 0.016). Further, men who are satisfied with their salary (90%) state more often than the men who are dissatisfied (76%) that work is one of the most important events in their lives. The corresponding figures for the women are 90% (satisfied) and 82% (dissatisfied) (r s 0.195 p 0.001).
A cross-tabulation of the respondents' satisfaction with physical conditions in the workplace and their value of work (as regards to how they agreed to the statement: I only want to have a job for the sake of the salary) is shown in Table 8.
From Table 8 it can be seen that 43% of the men who are satisfied with the physical conditions in the workplace state that they work only for the money they are paid compared with 69% of the men who are dissatisfied with the physical conditions. The corresponding figures for the women are 31% (satisfied) and 37% (dissatisfied) (r s (.347, p 0.000).

Discussion
Because of the large non-response rate in the study the results should be interpreted with caution. With this in mind, an attempt is made below to summarize and discuss the main results of the study.
The results show that the respondents attributed a high psychosocial value to work and that job satisfaction was high. It can also be seen from the results that, regardless of employment situation, the respondents valued paid work highly.
The results also show that men tended to attribute a higher value to the economic aspects of paid work than did women and that women valued the psychosocial aspects more highly than did men. This difference remained even when the respondents' employment and marital status were taken into account.
A greater part of the respondents were very satisfied with their jobs and those respondents valued their work more highly than the other respondents. An important difference between female and male respondents was, however, that which documented job satisfaction. Respondents who attributed a high and a low value respectively to their work were much less distinct among the women than among the men.
An interesting question is whether the results for the physically disabled as regards value of work and job satisfaction differ from earlier studies of persons with and without physical disabilities in Sweden and other Nordic countries. Such a comparison shows that the respondents' answers do not make them unique compared with other groups or women and men in general. Similar results have been found in studies of other groups, e.g., unemployed single men in need of social welfare benefits (Isaksson and Svedberg 1989) and unemployed young women and men (Malmberg-Heimonen and Julkunen 2002). Isaksson and Svedberg (1989) interviewed 86 men between 20 to 35 years of age. These respondents also valued work highly. On a scale from one to seven they scored 4.39. The study conducted by Malmberg-Heimonen and Julkunen (2002) included data from Denmark, Finland, Iceland, Norway, Scotland and Sweden. Eight thousand, six hundred and fifty-four unemployed young women and men between 18Á24 years participated. The results show that the respondents attributed a high psychosocial value to work. On the same scale as used in the study conducted by Isaksson and Svedberg (1989) the unemployed young women in Finland (4.71) and Sweden (5) attributed a higher value than men in the same countries (Finland: 4.58 and Sweden: 4.72). Unemployed young women also reported more negative feelings concerning unemployment than the young men in these countries. The young single women attributed a higher value to work than the young single men (the results were significant for Finland, Iceland and Sweden but no figures are presented in the research report). In all countries, women with children had less negative experiences of unemployment than both men and women who were single.
Our respondents' answers based on gender and age can also be compared with a study of employment commitment and psychological well-being among unemployed men and women in Sweden, as reported by Nordenmark (1995Nordenmark ( , 1999. This research project was based on a random sample of 3500 unemployed from the so-called 'HÄ NDEL' register (response rate 74%). This study showed that men and women in Sweden of all age categories valued paid work highly (Nordenmark 1995). Ninety-three percent of all men and women stated that it is very important to have a job, 81% said that they hate being unemployed and 77% reported that paid work is one of the most important things in their lives. In another study from the same research project (Nordenmark 1999) the value of work in different age groups was reported. These results showed that the younger the respondents the higher psychosocial value they attributed to work. On a scale from one to 20 the age group from 18Á25 years scored highest (16.33) while the age groups from 51Á60 years scored the lowest (14.08).
Research shows that answers given by physically disabled men and women in some respects differ from the answers given by non-disabled men and women in other studies (cf. Bielby 1992;Nordenmark 1999;Kulik 2000). These studies show that men attribute a higher psychosocial value to work than do women. The present study has shown that female respondents valued the importance of work highly (regardless of how satisfied they were with the conditions in their workplace). Furthermore, the female respondents, regardless of family type, valued work more highly than did the men.
In sum findings on job satisfaction of disabled women and men show that disabled women attribute a higher psychosocial value to work while disabled men ascribe a higher economical value. This can be compared with findings showing that non-disabled men attribute a higher psychosocial value to work and that nondisabled women ascribe a higher economical value to work.
One might ask how these differences between women with a physical disability and other women can be explained. A possible reason for the differences could, quite simply be attributed to the participating women's worse position in the labour market compared with that of other women. Another explanation could be that women with a physical disability (irrespective of civil status) regard work as a potentially more important source of autonomy than do non-disabled women or men in the same situation, since women with a physical disability are more dependent on other people in their day-to-day life and that paid work reduces this dependence. This is confirmed by other research conducted among physically disabled people, which shows that young men, compared with young women, are provided with opportunities to be less dependent on other people's help (Barron 1996). This is also supported by research showing that young women with a disability are opposed to such preconceived notions, which are often attributed to them, that it is suitable for women to be more dependent on assistance from society than it is for men in the same situation (Barron 1997). Another possible reason for the difference in values is that the female respondents, to a higher degree than men, expect paid work to be an important way to break social isolation and acquire new or retain established social relationships. This hypothesis can be related to research showing that women with a disability have a weaker social network than men with a disability do (see for example Dag 2004;Statistiska Centralbyrån 2003a). Research shows that men with a disability compared with women with a disability identify more strongly with individuals of the same sex who are not disabled (see for example Barron 1997). If this is linked to the differences in values found among the respondents the explanation could be that disabled men, compared with women, have an identity more strongly tied to what is considered to be 'normal' (for their gender) and that this link to the 'normal' (in the sense of nondisabled) results in their dependence on paid work from a psychosocial perspective, as a confirmation of their 'normality', being less apparent.
Previous studies of unemployment from Sweden have suggested that some groups of unemployed persons have adapted to the life of being unemployed and have low levels of work commitment (Svenning 1993). On the basis of such findings, it has been proposed that attitudinal factors (i.e., individual-related factors) are important when trying to understand why certain groups remain unemployed longer than others (Engbersen 1989). One such factor that has been emphasized is the will of the groups studied to get a job (for Swedish studies, see for example Å berg 1997;Bolinder 1999).
The present study shows that the disabled respondents do not value paid work less than does the population in general and that, as mentioned in the introduction, there is a connection between high motivation to work and actively looking for work. Accordingly, it is unlikely that the lower level of employment (28%) in the group studied would be due to them being less active than other groups are in attempting to enter the labour market. As the respondents also report relatively good job satisfaction, the culture of employment hypothesis would seem to be an even less likely explanation. The fact that the physically disabled as a group do not seem to be particularly willing to accept an early retirement pension as an alternative to paid work even though they are offered this option, does not support the hypothesis of a culture of unemployment (Dag 2004). Taken as a whole, the results indicate that societal factors like insufficiently adapted workplaces, employers' negative attitudes, and the interplay between these factors and the individual's physical status constitute the most important types of barriers for the group of young people studied.
Some researchers claim that the limited presence of people with physical disabilities in the labour market indicates that they are socially excluded from their civil rights since paid work is of such central importance in people's lives (cf. Abberley 1996Abberley , 1997. Bearing in mind that a comprehensive body of research also shows that the unemployed have a lower level of wellbeing than the employed and that persons who are unemployed for long periods of time have the lowest level of wellbeing (see e.g. Julkunen and Malmberg-Heimonen 1998;Warr, Jackson, and Banks 1998), the greater difficulties in getting a job for the physically disabled, must be regarded as a 'double handicap'. Given their high motivation to work, it is even more obvious that improved opportunities for work for disabled persons should be a priority for social policy.
In view of the differences in the valuation of the psychosocial aspects of a job for men and women shown in the present study, and the differences in needs that may exist between men and women with physical disabilities, there is reason to pay special attention to women's need for paid work and to implement special measures to meet this need.