THE PSYCHOLOGICAL CONSEQUENCES OF UNEMPLOYMENT IN SOUTH AFRICA

The objective of this study was to investigate the affective experiences, attitudes to work, and job application behaviour of unemployed people. A survey design was used and samples (N = 381) were drawn from unemployed people in the North West Province. The Experiences of Unemployment Questionnaire was administered. Regarding affective experiences, being unemployed was described as very unpleasant and it was associated with boredom, loneliness, uncertainty about the future, concerns about financial matters, emptiness and conflict. When it came to the participants’ attitudes to work, the results showed that almost 96 per cent of them regarded work as important, particularly because it provides meaning. Regarding job application behaviour, the results showed that most of the participants would like to find a job within the month, and they expected to do so. Almost 78 per cent of the participants were asking people for a job at least once a week or more often. Most of them asked friends and acquaintances for employment information, but unemployed people also reported that they looked out for advertisements. People with poor education had the most negative experiences of unemployment and saw work as more important than did those with better education.


Introduction
Unemployment is a serious issue in most countries in the world, but for some countries (e.g. the United States, New Zealand, Spain and Taiwan) the problem has become severe because of the financial crisis that developed after 2007 (Wanberg, 2012). Unemployment has also been a serious problem facing South Africa (Contogiannis, 2007;Kingdon & Knight, 2004;National Planning Commission, 2011). According to Contogiannis (2007), the unemployment rate in South Africa increased from 13 per cent in 1993 to 26 per cent in 2007 (excluding discouraged workers, that is, people who had opted out of the labour market). Unemployment affects economic welfare, production, the erosion of human capital, social exclusion, crime and social instability (Dollard & Winefield, 2002;Kingdon & Knight, 2004).
The negative consequences of unemployment for well-being have been well documented (Creed & Watson, 2003;McKee-Ryan, Song, Wanberg & Kinicki, 2005). Hanisch (1999) separates the negative effects of unemployment into individual and family effects. Individual effects include physical and psychological effects. Physical effects include an increase in headaches; stomach aches; sleep problems; lack of energy; hypertension; heart disease and kidney disease. Psychological effects include increased hostility, depression, Abstract anxiety, stress, anger, fear, despair, loneliness and social isolation, and decreased self-esteem, life satisfaction, aspiration levels, concentration and personal identity. Family effects include an increase in spousal abuse, marital friction, spousal depression, family conflict, and child abuse and a decrease in family cohesion, and the well-being of children. Lucas, Clark, Georgellis and Diener (2004) showed that, although life satisfaction is moderately stable over time, unemployment affects this in the long term.
Research comparing unemployed with employed people has consistently found higher levels of psychological distress and depression, and lower levels of self-esteem among the unemployed (McKee-Ryan et al., 2005;Waters & Moore, 2001). Poor well-being in the unemployed has also been demonstrated to be largely a concern and a consequence of unemployment, and not the result of those with poorer health "drifting" into that situation (Waters & Moore, 2001). Although unemployment seems to have predominantly negative effects, factors such as the employment commitment, coping resources, cognitive appraisal and coping strategies might increase the effects of unemployment on individual well-being (McKee-Ryan et al., 2005).
While it is crucial to understand and document the psychological experiences of the unemployed in South Africa, there seem to be few studies on the topic. Van der Merwe and Greef (2003) investigated the coping mechanisms of 82 unemployed African men, while Ribton-Turner and De Bruin (2006) studied stressors and support in a group of eight unemployed adults in mid-career. Møller (1993) points out that unemployed people often "just sit around", and most of them admitted that they were bored. Boredom and despair often featured in the respondents' descriptions of their own activity schedules (Møller, 1993). Unemployment (or experiences thereof) in South Africa might be different from that in Europe. This is for the following reasons: a) In Europe, there are social security systems, while no such system currently exists in South Africa. Unemployed people in Europe still receive a reasonably good income, which cannot be said of unemployed people in South Africa. b) The South African culture is more collective than in other countries and work might be less important for individuals in this country. Masango (2005) suggests that being unemployed in South Africa might thus be less problematic .
However, based on the above description of the problem, unemployment is clearly a serious problem in the country. Furthermore, there is a lack of information on the attitudes to work, affective experiences and job application behaviour of unemployed individuals. We aim to fill the gap in the literature with results from a large-scale representative sample of unemployed people in the Potchefstroom area of the North West Province of South Africa. Information on the experiences of unemployed individuals could be used by local, provincial, and national governments to plan and implement strategies for assisting them.

Experiences, attitudes and behaviours of unemployed people
According to the International Labour Office (ILO), the unemployed comprise all those above a specific age who were without work during the reference period (i.e. they were not in paid employment or self-employment), were available for paid employment or selfemployment, and had taken active steps to seek paid employment or self-employment (ILO, 2000:429). Unemployment is a complex and multifaceted construct which involves situational (non-employment), motivational (seeking work) and medical and legal aspects (being available for work) (Paul & Moser, 2006:597). Unemployed people are therefore those who are able to work, are allowed to work, and prefer to work.
Given the above definition, unemployed people should be studied not only in terms of experiences (associated with being unemployed), but also in terms of their availability for work (employment commitment), and job application behaviour (Paul & Moser, 2006). De Witte, Hooge and Vanbelle (2010) developed a model which focuses on unemployed people's affective experiences (wellbeing), cognitions (attitudes towards work), and behaviours relating to unemployment (job application behaviours). The model by De Witte et al. (2010) was applied in Europe to aid understanding of the attitudes to work, affective experiences, and behaviours of unemployed people, but it has not yet been applied in South Africa.

Affective experiences of unemployed people
The affective experiences of unemployed people are relevant to their well-being. The latent deprivation theory (Jahoda, 1982) is the most influential for understanding the affective experiences of unemployed people (Wanberg, 2012). Jahoda argued that employment has not only the manifest function of earning a living, but also five latent functions. Deprivation of the latent functions has a negative impact on psychological well-being. These latent functions, which are associated with the satisfaction of basic human needs, include the establishment of a daily time structure, provision of regular shared experience and contacts outside the nuclear family, information on personal identity, a link with the collective purpose and the enforcement of regular activity (Jahoda, 1982). According to Jahoda (1982), adults need clear time structures and the possibility of filling their days with planned activities. The absence of such a structure results in boredom and waste of time. People also need regular shared experience and contacts outside the nuclear family. These cannot be replaced by the intensification of family life, because they provide more information and more opportunities for judgement and rational appraisal of other human beings with their unique perspectives. Information about personal identity is shaped by people's social status, which depends on the society in which they live. People tend to see themselves as others see them, but unemployed people see themselves as having no status at all, which is detrimental to their well-being. Furthermore, people need a collective purpose that transcends individual purposes, as well as the feeling of being useful, and being needed by other people. If there is no collective purpose, the result will be purposelessness and meaninglessness. Finally, regular activity is an important determinant of personal well-being. Studies have shown that unemployed people tend to report less access to the latent functions than do those employed, while those with less access generally experience lowered physical and psychological well-being (Creed & Macintyre, 2001;Creed, Muller & Machin, 2001;Evans & Haworth 1991;Haworth & Ducker 1991;McKee-Ryan et al., 2005;Paul & Batinic, 2009;Wanberg, Griffiths & Gavin, 1997). It seems that the latent functions might be partially provided in activities other than employment, such as meaningful leisure activities or attending work-related training (Creed, Hicks & Machin, 1998). Warr (1987) has developed the vitamin model, which is similar in many respects to Jahoda's (1982) model. According to Warr (1987), insufficiency or excess of nine environmental features are responsible for psychological wellbeing. The nine features, which largely mirror Jahoda's six consequences, are: opportunity for control; use of skills; interpersonal contact; external goal and task demands; variety; environmental clarity; availability of money; physical security; and valued social position. Based on Warr's (1987) approach, it would be inappropriate to suggest that unemployment is inevitably destructive or that employment is inevitably constructive: "... the impact of the transition from paid work to unemployment will be a function of changes which occur in the nine primary environmental features. In most cases, these shifts will impair mental health, but the transition can sometimes be neutral or even beneficial in its effect" (Warr, 1987:355).
According to Herbert, Drebing, Mueller and Van Ormer (2006), the effects of employment and unemployment can be analysed in terms of benefits and costs. The benefits of employment include income benefits; having structured time; contact with other people; and a sense of identity, both individual and collective (Jahoda, 1982). Potential costs associated with working include increased stress and the loss of leisure time. On the other hand, some benefits of unemployment include increased leisure time or time for other life activities. The costs of unemployment include financial stress, isolation and low self-esteem (Feather, 1990;Price, Choi & Vinokur, 2002;Winefield, Winefield, Tiggeman & Goldney, 1991).
Comparative analyses indicate that the type of job search motivation people experience is an important predictor of their unemployment experience and well-being (Vansteenkiste, Lens, De Witte & Feather, 2005). Feather and Bond (1983) found a positive association between levels of self-esteem and the structured, purposeful use of time. Further, unemployed respondents (in comparison with employed respondents) showed less engagement, less direction, and less routine in their use of time. Winefield, Tiggeman and Winefield (1992) also found that engaging in purposeful activities relieves the stress associated with both unemployment and unsatisfactory employment.

Unemployed people's attitudes towards work
Unemployed people's attitudes towards work can be studied in terms of the importance they attach to work. In the unemployment literature, this has been studied in terms of work involvement or employment commitment (Warr, Cook & Wall, 1979), the Protestant work ethic (Mirels & Garrett, 1971), and the valence of work (Feather, 1990). Warr et al. (1979:130) define employment commitment as "the extent to which a person wants to be engaged in work". Employment commitment (also referred to as work-role centrality) indicates the importance of work to an individual's sense of self (McKee-Ryan et al., 2005). Work-role centrality may stem from Protestant work ethic socialisation or from a belief that engaging in activities such as work has a lasting effect on an individual's life satisfaction (Paul & Moser, 2006;Seligman, 2002). Work gives people the opportunity of using their strengths in the service of something larger than the self, which contributes to meaning in life (Peterson, Park & Seligman, 2005;Seligman, 2002;Wrzesniewski, McCauley, Rozin & Schwartz, 1997). Paul and Moser (2006) found that the employment commitment of both employed and unemployed people was high. However, employed people showed a marginally stronger level of employment commitment. Basing their approach on consistency theory (Grawe, 2004), Paul and Moser (2008) explain the unwellbeing, psychological distress and dissatisfaction of unemployed people in terms of the incongruence between their goals and their perceived goal attainment. Individuals who are strongly committed to work will see their job search as incongruent if it is unsuccessful, with detrimental consequences. Employment commitment could mitigate or buffer the negative consequences of unemployment (Fryer & Fagin, 1993;Fryer & Payne, 1984).

Job application behaviours of unemployed people
Studies of the predictors of job search among unemployed people have often used an attitude-behaviour model (Van Hooft, Born, Teris, Van der Flier & Blonk, 2004). According to the theory of reasoned action, the immediate antecedent of job search behaviour is the intention to look for a job (Van Hooft et al., 2004). Job search intention in turn is predicted by a person's adhering to a positive or negative evaluation of job search behaviour and the perception that there is social pressure to look for a job. Furthermore, in the context of job seeking, individuals' perceived control over their actions affects their perceptions of control over job seeking behaviour, such as when people are not sure how to apply for a job or even how to write a letter of application (Van Hooft et al., 2004). Active job search is a form of problemfocused coping (McKee-Ryan et al., 2005), and is an important predictor of reemployment (Kanfer, Wanberg & Kantrowitz, 2001). McKee-Ryan et al. (2005) point out that active job seeking behaviour is often associated with decreased well-being, particularly because job seeking is often discouraging, with its associations with rejection and uncertainty. Individuals who are actively seeking a job might feel pressured to accept almost any job, including one of low quality. According to De Witte et al. (2010), reducing the value of work might become a coping strategy for decreasing the negative experiences associated with unemployment.

The effects of demographic factors
The demographic group in which unemployed people find themselves might affect their experiences, attitudes and behaviours (Kingdon & Knight, 2004). Six demographic factors were relevant to this study: gender, race, age, qualification, time unemployed and employment history.
Race might affect the experiences, attitudes and behaviours of unemployed people, especially because African, coloured and Indian people have traditionally been the victims of discrimination in South Africa. According to the National Planning Commission (2011), decades of racial discrimination have confined black people to menial labour, poor wages, poor-quality schooling and low levels of income. As pointed out by Wanberg (2012), differential skills, location and discrimination by employers affect unemployment rates, which could also impact on the experiences, attitudes and behaviours of those discriminated against.
According to Kingdon and Knight (2004), young and uneducated African people living in rural areas are the most vulnerable to unemployment. The writers point out that rural unemployment rates are higher than urban rates, which is atypical of other countries and is explained by historical policies restricting mobility. Also, many unemployed people in South Africa have never held a job. However, expanding education and skills development will reduce overall unemployment (Wanberg, 2012).
Unemployed males and females might differ regarding experiences, attitudes and behaviours for various reasons. First, males might be more inclined to experience unemployment negatively because of the sex role stereotypes that make males the breadwinners. Second, females might find it more difficult to get jobs because of gender discrimination, which could affect their affective experiences (Wanberg, 2012).
When it comes to age, older people might experience unemployment more negatively than younger people do, especially if they find it difficult to get a job (Wanberg, 2012). According to a diagnostic overview by the National Planning Commission (2011), unemployment is experienced mostly by young people (15-34 years), who are also poorly prepared for work.
De Witte et al. (2010) found that people who were unemployed in the short term experienced more psychological distress, showed more employment commitment and applied for jobs more often than those who were unemployed in the long term. They attributed these findings to an adaptation process that took place in long-term unemployed people whereby their psychological well-being increased after a while. However, these findings should be interpreted with caution, given that the unemployment contexts in South Africa and Belgium differ. For instance, unemployment in South Africa is accompanied by poverty and the absence of social grants and other forms of support from the government.
Previous employment experience is an important factor affecting the experiences, attitudes and behaviours of unemployed people. Such experience might have empowered the individual with job-specific skills and competencies, but could also have resulted in the level of knowledge and skills required to work with others in an organisational context (National Planning Commission, 2011;Wanberg, 2012).

Research objective
The objective of this study was to investigate the attitudes to work, affective experiences and job application behaviour of unemployed people in the Potchefstroom area of the North West Province.
The following hypotheses are set, based on the discussion above: Hypothesis 1: People in various race groups differ in their experiences of unemployment, the importance of work and job application behaviours.
Hypothesis 2: Males and females differ in their experiences of unemployment, the importance of work and job application behaviours.
Hypothesis 3: People with different qualifications differ in their experiences of unemployment, the importance of work and job application behaviours.
Hypothesis 4: People in different age groups differ in their experiences of unemployment, the importance of work and job application behaviours.
Hypothesis 5: People with different employment histories differ in their experiences of unemployment, the importance of work and job application behaviours.
Hypothesis 6: People who have been unemployed for a longer period differ from those who are unemployed for a shorter period in their experiences of unemployment, the importance of work and job application behaviours.

Research design
A cross-sectional survey design was used to investigate the psychological experiences of unemployed people (Howell, 2008). The research is exploratory and descriptive because very limited research has been carried out on the experiences of unemployed people in South Africa.

Participants
The total population is a stratified sample of individuals who were unemployed at the time of the study (N = 381). The participants were sampled from three areas: Potchefstroom (mainly white = 21 per cent), Promosa (Coloured = 54 per cent) and Ikageng (African = 24 per cent). The population consisted of both females and males, whether married, single, divorced, widow or widowered. The majority of the respondents fell between the ages of 17 and 30 years (54.4 per cent), with the minority (7.0 per cent) of respondents being older than 50 years. The participants' characteristics are reported in Table 1. The majority of the participants were male (53.5 per cent), preferred Afrikaans as the language of communication (65.4 per cent), and were single (70.6 per cent). A total of 57.3 per cent was between 17 and 30 years of age. Further, 24.9 per cent of the participants were occasionally employed and 20.5 per cent of them were almost always unemployed. A total of 11.3 per cent had been unemployed for longer than 7 years.

Measuring instrument
The Experiences of Unemployment Questionnaire (EUQ) was developed to measure the participants' experiences, behaviour and wellbeing for the purposes of this study. The EUQ was developed according to the model and questionnaire by De Witte et al. (2010). In line with the conceptualisation in this study, the EUQ included questions concerning unemployed people's affective experiences, attitudes towards work, and job application behaviour. The questionnaire was divided into four sections. Section 1 gathered demographic information, including gender, home language, marital status and highest qualification. The three dimensions (i.e. affective experiences, the attitude to work, and job application behaviour) were measured in the next three sections by a short-cut to the variable (i.e. a brief question summarising the issue), a list of items, and a question on the evolution of the specific concept, in order to establish whether or not it remained stable over time. Section 2 focused on the affective experiences of the unemployed person. The items were developed on the basis of Jahoda's (1982) model, and adapted according to research in the Netherlands and Belgium (De Witte et al., 2010). The participants were asked to indicate the extent to which their typical day was filled with content or activity (regular activity); their life was temporally organised and structured (time structure); they were meeting people and socialising (shared experience); they felt they were useful members of society (collective purpose); they were appreciated by other people (status); and felt confident and self-respected (personal identity). For instance, one question was: "How does it feel to be unemployed?" which should be answered on a scale ranging from 1 (very pleasant) to 5 (very unpleasant). Another question was: "How often do you experience the following?: feelings of boredom; difficulty surviving financially; uncertainty about the future; saving on personal expenditure; feeling lonely and empty; experiencing conflict; decreased self-confidence and self-esteem; how to use time; extent of social support; not feeling part of society; and feeling relaxed. A three-point scale was used, varying from 1 (often) to 3 (never). Also included was a question that tapped into the temporal dimension of unemployment ("How do you feel about being unemployed?"), on a scale ranging from 1 (feel better than before) to 4 (never felt bad).
Section 3 focused on the importance of work. The participants' desire to be in paid employment was assessed through the Employment Commitment Scale of Warr et al. (1979). Typical questions were: "How important is it to have a job?" with an answer scale ranging from 1 (very important) to 5 (very unimportant) and "How do you feel about not having a job at this moment?" with an answer scale ranging from 1 (job became more important) to 4 (job was never important).
Section 4 focuses on job application behaviour. Questions include: "When would you like to find a job?" with an answer scale ranging from 1 (within a month) to 7 (never), and "When do you expect to find a job?" with an answer scale ranging from 1 (within a month) to 7 (never). Job application behaviour was measured by presenting five different behaviours (e.g. "Search for advertisements") which had to be scored according to the frequency of behaving in this way in the recent past (scale ranging from 0 (never) to 5 (10 times or more).

Procedure
The study, which formed part of a larger project on experiences of unemployment in the North West Province, was initiated during 2005 after discussions with the executive mayor of Potchefstroom (in the North West Province). The project was planned during 2006 and funding was obtained from the National Research Foundation. During June 2006, the researchers implemented the project. Literature searches were conducted, while there were interviews and focus groups to develop the questionnaire in English. It was then translated into Afrikaans and Setswana by professional translators. A process of backtranslation ensured that the meaning of the words in the different languages was the same throughout. Thereafter, questionnaires were presented to experts to be checked for face validity and final changes were made. Three fieldworkers (who were able to speak Afrikaans, English and/or Tswana) were used to administer the questionnaires. The researchers, assisted by language practitioners, trained the fieldworkers prior to the data gathering.
Fieldwork took place during July 2006. Unemployed people were randomly selected via door-to-door selection in different areas of the town and neighbourhoods. Given the poor educational qualifications of most unemployed people, field workers conducted structured interviews with all the participants and their responses were recorded on the questionnaires. The data were captured by a computer program and checked for mistakes. Finally, the data set was prepared for statistical analysis.

Statistical analysis
The statistical analysis was carried out with the help of the SPSS program (SPSS Inc., 2009). Descriptive statistics (frequency tables) were used to analyse the data. A frequency table lists items together according to the number of times, or frequency, the items occurred. Exploratory factor analyses and Cronbach alpha coefficients were used to assess the validity and reliability of the constructs measured in this study. One-way analysis of variance (ANOVA) was used to determine the significance of differences between the scores of the demographic groups.

Descriptive statistics, factor analysis and reliability
Affective experiences of unemployed people Most of the participants (59.6 per cent) experienced unemployment as very unpleasant. Only 8.7 per cent of the participants reported that it felt pleasant or very pleasant to be unemployed. Furthermore, 72.7 per cent of the participants reported that they felt worse than before because they did not have a job. A total of 5.2 per cent of the participants reported that they never felt bad about being unemployed. Table 2 shows the experiences of unemployed people in more detail.  A relatively large number of participants experienced feelings of emptiness and lone-liness; poor social support; conflict at home; and low self-esteem and self-confidence. Interestingly, a substantial percentage of the participants felt that they were no longer part of society; never felt relaxed when unemployed; and did not know how to spend their time since being unemployed. Advantages of being unemployed included doing things they wanted to do and spending time with their families.
Unemployed people's commitment to work A total of 74.8 per cent of the participants reported that having a job became more important than before, while only 8.9 per cent reported that it was less important. Table 3 shows the evaluation of the items measuring employment commitment. Have to work to be really part of society 62.5 23.9 11.3 Do not have to work to be constructively occupied 33.1 36.7 27.8 Table 3 shows that most of the participants reported that work contributed to a meaningful life. A large percentage of the participants found it important to have work. Indeed, a large number of participants regarded work as the most important aspect of life. In fact, 70.6 per cent agreed that it was better to accept any job than to be unemployed. In addition, most of the participants reported that they could enjoy leisure time only if they had worked for it and that they had to work to be part of society.

Job application behaviour of unemployed people
A total of 62.5 per cent of unemployed people looked for work at least once a week, while only 6.6 per cent were not looking for a job at the time of the interview. Furthermore, 60 per cent of the participants were making more effort than before to find jobs, while 15.2 per cent were making less effort, and 3.7 per cent had not looked for a job at all. The job-seeking behaviours of unemployed individuals are reported in Table 4.  Table 4 shows that 47.5 per cent of unemployed individuals had inquired whether work was available 10 times or more. A total of 45.9 per cent of the participants had searched for advertisements 10 times or more.
Only 42.0 per cent of the participants had submitted applications 10 times or more. Table 5 indicates unemployed individuals' expectations as to finding a job. Within 1 year to 2 years 0.8 After 2 years 1.6 Never 3.9 Anything to find during the last month Every day go out and ask for a job 42.3 Once, twice or three times ask for a job 34.6 Once or twice a month ask for a job 15.5 Not looking for a job 6.3 Table 5 shows that 87.1 per cent of the participants would prefer to find a job within a month. Further, 75.6 per cent expected to find work within a month. 42.3 per cent of the participants asked for a job every day, with 34.6 per cent asking once, twice or three times, while 6.3 per cent were not looking for a job at that time. Only 4.2 per cent of the participants never enquired about the availability of work, while 9.2 per cent never presented themselves for work.

Factor analysis
Exploratory factor analysis was used to explore the factor structure of the EUQ. A simple principal component analysis was carried out on the 26 items of the EUQ. An analysis of the eigenvalues (> 1.00; Tabachnick & Fidell, 2007) indicated that four factors explained 45.65 per cent of the variance. The scree plot confirmed that four factors could be extracted. A principal factor analysis with a direct Oblimin rotation was then performed. The results of the principal factor analysis with loadings of variables on factors and communalities (h 2 ) are shown in Table 6. Items were retained for further analyses if their factor loadings were higher than 0.30 (Tabachnick & Fidell, 2007).  Table 6 shows that four factors were extracted. The first factor, Negative Experiences, included items such as decreased selfworth/self-esteem; loss of self-confidence; feeling disconnected from society'; feelings of emptiness; uncertainty; conflict; loneliness; and ;boredom. The second factor, Application Behaviour, included items which described behaviours triggered by unemployment, e.g. enquiring whether work was available and spontaneously presenting themselves to possible employers, submitting applications and searching for advertisements. The third factor, Importance of Work, referred to the value of work in a meaningful life and to feeling part of society. The fourth factor, Positive Experiences, included items about the benefits of unemployment in terms of social support, time for family, relaxing and leisure time. Two items, "Know how to spend time" and "Save on personal expenditure", showed low communalities, indicating that these items were not well-represented by the extracted factors.

Descriptive statistics
The descriptive statistics, alpha coefficients and Pearson correlations of the scales of the EUQ are reported in Table 7. The minimum and maximum values are included to facilitate the interpretation of the mean scores. Further, internal consistencies and correlations of the scales are reported to assist in forming reliable dimensions that could be used when comparing demographic groups. Further, the correlations between dimensions were computed to assess whether multivariate analysis of variance (MANOVA) or one-way analysis of variance (ANOVA) should be used to study differences between demographic groups. MANOVA is recommended when the scales are at least moderately related (Tabachnick & Fidell, 2007).  Table 7 shows that the alpha coefficients of the scales were acceptable, except for positive experiences, which was lower than the cut-off value of 0.70 (Nunnally & Bernstein, 1994). The Pearson correlations show a statistically significant correlation between the importance of work and negative experiences. Statistically, application behaviour is significantly and negatively related to negative experiences of unemployment. Application behaviour is related to positive experiences of unemployment. Given the small correlations between some of the scales, it was decided to use ANOVA rather than MANOVA in the subsequent analyses.
Differences between biographical groups ANOVA followed next to investigate the relationship between affective experiences, the importance of work and application behaviour of various groups, including race, gender, age, qualification, time unemployed and employment history. The results of these comparisons are reported in Table 8.  Table 8 reflects the significant effect of qualification on the negative experiences of unemployment (F (3, 317) = 4.79, p < 0.01; η 2 = 0.04). Although the extent of the effect was small (4 per cent of the variance explained), the results showed that people with a qualification of Grade 8 and lower (Mean = 1.42, SD = 0.39) experienced unemployment more negatively than did people with Grade 11 and 12 (Mean = 1.68. SD = 0.50). Qualification also had a statistically significant effect on the importance of work (F (3, 317) = 7.08, p < 0.01; η 2 = 0.06). Although the effect size was small (6 per cent of the variance explained), the results showed that people with Grade 8 and lower (Mean = 1.18. SD = 0.32) and people with Grade 9 and 10 (Mean = 1.29. SD = 0.38) found work to be more important than did people with Grade 11 and 12 (Mean = 1.40, SD = 0.40). Based on these findings, only hypothesis 3 is partially accepted. Table 8 shows that no statistically significant differences were found between the affective unemployment experiences, the importance of work and application behaviour of different race groups (hypothesis 1), males and females (hypothesis 2), different age groups (hypothesis 4), different employment history (hypothesis 5) and different periods of time unemployed (hypothesis 6). These hypotheses are rejected. The reason why few differences between affective unemployment experiences, importance of work and application behaviour demographic groups were found is that most unemployed people experience unemployment intensely negatively, they regard work as important and they reported strong application behaviour.

Discussion
The objective of this study was to investigate the attitudes to work, affective experiences and job application behaviour of unemployed people in the Potchefstroom area of the North West Province. Exploratory factor analysis resulted in four factors, i.e. negative and positive experiences of unemployment, the importance of work and job application behaviour. Qualification affected negative experiences of unemployment and the importance of work. Compared with people who had higher schooling levels, those with less schooling experienced unemployment more negatively and found work to be more important. When it came to the participants' affective experiences, the results showed that more than 80 per cent of them experienced unemployment as unpleasant. Boredom, uncer-tainty about the future, concerns about financial matters, emptiness, and conflict seemed to contribute to the unpleasantness. Regarding the participants' cognitive experiences (importance of work), the results showed that almost 96 per cent of them regarded work as important, especially because it gave meaning. The results clearly indicated that participants would like to find jobs within a month.
The results of this study showed that four factors, negative experiences of unemployment, positive experiences of unemployment, the importance of work and job application behaviour represented the constructs in this study. These factors, which correspond to the three factors reported by De Witte et al. (2010), can be defined as follows: Negative affective experiences of unemployment refer to decreased self-worth/self-esteem; loss of selfconfidence; feeling disconnected from society; and feelings of emptiness, uncertainty, conflict, loneliness and boredom. Such negative affective experiences are associated with the deprivation of latent functions, such as the establishment of a daily time structure; provision of regular shared experience and contacts outside the nuclear family; information on personal identity; a link with the collective purpose; and enforcement of regular activity (Jahoda, 1982). Studies (Creed & Macintyre, 2001;Evans & Haworth, 1991;Haworth & Ducker, 1991;McKee-Ryan et al., 2005;Paul & Batinic, 2009;Wanberg et al., 1997) showed that unemployed people reported less access to the latent functions than those who were employed, and that those with less access generally experienced lowered psychological well-being.
The negative affective experiences of unemployed people in this study correlate with the abovementioned theoretical explanation and empirical results. The results showed that unemployment was experienced as unpleasant by almost all the participants, and only a small percentage experienced it as pleasant. Negative affective experiences included boredom, finding it difficult to survive financially, uncertainty, loneliness and emptiness. Positive experiences included spending time with family and doing things that were valued (but only 40 per cent of the participants). These findings correlate with international findings on the topic (e.g. De Witte et al., 2010;Wanberg, 2012). The profile for the affective experiences of unemployment in South Africa is thus not much different from that found internationally. Jahoda (1982) maintained that time structure is an important latent benefit of work. Indeed, boredom proved to be the experience with the highest frequency in this study. A study by Møller (1993) also concluded that some respondents often or sometimes "just sit around", while most of them admitted that they were bored. Boredom and despair were often associated with the descriptions of respondents' own activity schedule (Møller, 1993). The loss of self-esteem was another issue experienced by the participants in this study. The loss of self-esteem among the unemployed is thought to be a reflection of both society and the individual, and the loss of it may be felt more intensely when one's self-definition is derived mainly from one's occupation (Møller, 1993). The study confirmed that people need to work if they are to save on personal expenditure and, according to Møller (1993), some people had a regular source of income from either employed relatives or self-employment. They stated that their household expenditure was limited to basic necessities (Møller, 1993).
The financial implication of unemployment was also a very important theme. A high percentage of the participants reported that they found it difficult to survive financially and that they had to save on personal expenditure. This was because they had no money to fulfil their basic needs. The example is illustrated by Møller (1993), who argues that the respondents described how unemployed people borrowed, begged and stole to make ends meet, while reference was made to "borrowing" from family members with the intention of returning the loan. However, it appeared that loans would effectively become gifts when the unemployed were not in a position to repay the amounts borrowed (Møller, 1993). Singer, Stacey and Ritchie (2001) have argued that, if a person is unemployed, they have little money, they cannot afford to do things and they cannot provide financial help for others.
The study also established that social support was experienced negatively. In the study by Kelvin and Jarrett (1985), it is stated that the family sees the male unemployed as 'having a problem', to which extent he is a problem to them and they may see him as 'less-than-the-man-he-was', sympathetically or disdainfully, depending on earlier issues. However, other people who are unemployed receive support from friends and seek help from them. Studies also concluded that, in general, unemployment stress is exacerbated by a low sense of social support (Linn, Sandifer & Stein, 1985). Gore (1978) found that the rural unemployed evidenced a significantly higher level of social support than did the urban unemployed. Where domestic circumstances were positive, the family could continue to provide loving support for its members, regardless of the bad times (Møller, 1993).
Participants in the current study showed that they felt worse than on the previous occasions when they were unemployed, and only a few never felt bad about it. In her study, Møller (1993) argued that the euphoria and short-lived feeling of freedom after losing one's job may be perceived as a feeling of independence and liberation from the many constraints of working life. However, when work is a central value and gives meaning to all the other aspects of life, job loss may have devastating effects (Møller, 1993). That is why most of the participants felt worse at being unemployed.
The importance of work is defined as unemployed people's cognitions of the value of work in terms of a meaningful life and the feeling of being part of society. The results of this study showed that most unemployed people indeed wanted to be engaged in work (Warr et al., 1979). Work does not just give people the opportunity of using their strengths in the service of something larger than the self (Peterson et al., 2005;Seligman, 2002;Wrzesniewski et al., 1997). A total of 70 per cent of the participants reported that they would do any job rather than be unemployed. It is seen as important to work in order to enjoy leisure time and to be part of society. Jobs have become more important than before, according to the participants. Almost all the participants regarded work as important, and most saw it as meaningful. To them, it meant that they had to work in order to find purpose in life and they were motivated by the determination to find meaning. Victor Frankl (1985), the well-known Holocaust survivormaintained that finding meaning in life is important for the individual's well-being. Work is thus regarded as an important factor in finding meaning in life.
The above findings show that most of the participants had made more effort than before to find work, which means that people search desperately for a job, regardless of what kind. Some of the participants had asked anyone they could came across whether work was available. Other studies also confirmed that people in search of a job had stated that their search would become as important as they wanted it to be (Van Hooft et al., 2004). In addition, there is the 'contact' in the shape of a third person, a relative or a mutual acquaintance, who becomes an intermediary between the job-seeker and the employer (Kelvin & Jarrett, 1985). This is true, because in the current study, people had asked their friends, as well as presenting themselves to possible employers in the effort to find a job.
Application behaviour includes efforts by unemployed people to enquire whether work is available, spontaneously presenting themselves to possible employers, submitting applications for work and searching for advertisements. When it came to job application behaviour, the results showed that most participants would have liked to find a job within the following month, and they also expected to do so in that time. Most of them had asked friends and acquaintances for employment information, but they also reported that they looked for advertisements. There are four possible interpretations of this puzzling finding. First, it is possible that participants answered the question in a socially appropriate way. Second, they might have experienced unrealistic optimism. Third, they might possibly have felt that others and society in general expected that they should find a job soon. Fourth, time might be seen from a different perspective in the African context, and may have a different meaning.
The analyses showed that qualification was the only demographic factor affecting experiences of unemployment and the importance of work. More specifically, people who had 10 years of schooling or less found work to be more important than those with 11 to 12 years of schooling. Further, people with eight years or less schooling experienced unemployment more negatively than those with more than eight years of schooling did. People with poor education were more inclined to experience decreased self-worth; loss of self-confidence; not being part of society; emptiness; uncertainty about the future; conflict at home; loneliness and boredom than were people with better education. Furthermore, people with poor education were more inclined to believe that work contributed to a meaningful life, and that it was better to accept any job at all rather than be unemployed; work is the most important aspect of life; one could enjoy leisure time only if one had worked for it; it was important to find a job; and one had to work if one was to really be part of society. Although access to and participation in education have improved since 1994, many individuals had poor educational opportunities and the quality of education is also problematic (National Planning Commission, 2011). The fact that gender, age, employment history and the length of time unemployed did not affect experiences of unemployment, the importance of work and job application behaviour indicates the pervasive effects of unemployment in the South African context. This study had various limitations. First, the study was conducted in the Potchefstroom area of the North West Province. The findings can therefore not be generalised to other towns in the province or South Africa as a whole. Second, this study made use of a cross-sectional design to study the effects of unemployment on people. Longitudinal studies are needed to study the causal effects of unemployment on individuals. Unemployed individuals with lower levels of mental health than those of the employed comparison group do not necessarily have lower levels of mental health as a consequence of the unemployment (McKee-Ryan et al., 2005). Instead, it is possible that individuals with lower levels of mental health are more likely to lose their jobs or that individuals with higher levels of mental health are more likely to find new jobs (Warr, Jackson & Banks, 1988). Therefore, while the causal relationship between unemployment and poor well-being has been documented, more research is needed to verify such relationships.

Recommendations
Based on the results of this study, three recommendations are made. First, there is an urgent need to expand the cover of social security in South Africa. The injection of cash into poor households is probably one of the most empowering ways of assisting poor people to strengthen their own coping mechanisms. Second, programs should be implemented to stimulate economic growth, job creation and skills development. Investment in infrastructure has a huge potential to redress the high unemployment and poverty levels in South Africa and also to correct the skills shortages (Thwala, 2008). Third, programs should be implemented to assist unemployed people in coping with unemployment and improving their application behaviour. Wanberg (2012) showed that poorer psychological health reduced the speed of reemployment. The government could also provide training for people who really want employment in order to enhance the skills necessary in the work situation, which would assist them in coping better with unemployment. Productive activities are lacking in the lives of the unemployed, and if they were given training they would know how to use such skills in the work environment. Skills training programs should therefore be provided so that unemployed people could also develop the idea of small businesses, probably receiving government funding.
It is recommended that future research include further investigation into the experiences of unemployed people, their job application behaviour and coping, as well as their subjective well-being in other contexts, conducted also with larger samples. Longitudinal studies could be carried out to better establish experiences of unemployment and the causal relations related to these experiences. More research is needed to determine the reliability and the validity of the Experiences of Unemployment Questionnaire using samples in South Africa.