Profiling the unemployed from selected communities in South Africa based on their experiences,commitment to employment,and job search behaviour

Abstract

Unemployment has negative consequences for individuals’ psychological well-being. Consequently, interventions should be designed and implemented to alleviate the psychological burden of unemployment. The design of these interventions should, however, be approached with care, as ‘the unemployed’ may not be a homogeneous group. The aim of the study was to determine whether the four already identified (the optimists, the desperate, the discouraged, and the adapted) South African unemployment profiles could be replicated in other unemployed communities in South Africa. The study also aimed to examine the associations between these profiles and negative emotions and basic psychological need frustration. To establish the replicability of the types, a multiphased sampling design was followed to recruit 867 unemployed people residing in Boipatong and Orange Farm in the Gauteng Province in South Africa. Through latent profile analysis, the study replicated the four profiles: the optimists, the desperate, the discouraged, and the adapted. The profiles were differentially associated with negative emotions and psychological need frustration, further attesting to the validity of the profiles. The results of the study can be applied towards creating tailored interventions for the different types of unemployed people from South African communities to enhance the efficacy of these interventions.

Keywords

Attitudes behaviour experiences latent profile analysis psychosocial typology unemployment

South Africa’s unemployment rate ranges between 30% and 40%, depending on whether the narrow or expanded definition is used (Statistics South Africa [Stats SA], 2020). Unemployment is a challenging issue and is not only a significant economic and societal burden (e.g., criminality, family breakdown, poverty, and substance abuse) on countries, but also a psychological burden on most unemployed individuals (Bernstein, 2019; Brand, 2015). Despite the high unemployment rate and the associated detrimental psychological consequences, few psychologically oriented unemployment interventions exist within the South African context (Paver et al., 2019). We should, however, caution against blindly developing and implementing such interventions because ‘the unemployed’ may not be a homogeneous group (Knopf, 2013), and meta-analytic evidence shows that interventions tailored to the specific groups are more effective than one-size-fits-all approaches (Liu et al., 2014).

De Witte (1992) developed a typology to categorize different types of unemployed based on their experiences, their attitudes towards employment, and their job search intensity. Based on these three dimensions, five types of unemployed were identified: the optimist, desperate, discouraged, adapted, and withdrawn. The typology was replicated in Belgium among a sample of female unemployed (De Witte & Wets, 1996). Within the South African context, research also confirmed that ‘the unemployed’ consisted of homogeneous subgroups, but only four of the five types were identified (Van der Vaart et al., 2018). This echoes the findings of De Witte and Hooge (1995) among a sample of short-term unemployed.

Despite the value-add of previous research using this typology, three notable gaps exist. First, research was mainly conducted in a more affluent country with unemployment benefits. Such benefits may account for differences in emerging profiles (Van der Vaart et al., 2018). Second, only one study was conducted in South Africa, and the sample was not a true reflection of the unemployed population. The sample consisted mainly of Afrikaans-speaking people (65.4%), who can reasonably be assumed to have been either White or coloured. Stats SA (2020) reports that 80.85% of South Africa’s working population consists of Black Africans, with the unemployment rate being the highest among them. Finally, none of the previous studies explored the differences between these types based on their negative emotions and the frustration of their basic psychological needs. It is best practice in person-centred research to include covariates to strengthen the construct validity of the profiles and to enhance their utility for practice (Morin et al., 2020).

The current study aimed to fill these gaps by determining whether the four already identified unemployment profiles could be replicated and by examining the associations between the profiles and negative emotions and psychological need frustration. Studies have reported lower levels of psychological and physical health for the unemployed (compared with the employed), which could be attributed to the negative emotions (Wanberg, 2012) and the frustration of their basic psychological needs experienced as a result of unemployment (Vansteenkiste & Van den Broeck, 2018).

Dimensions of the unemployment experience

Stats SA (2020) focuses on unemployed people’s job-seeking behaviours and identifies two types of jobseekers: seekers versus non-seekers (or discouraged jobseekers). However, individuals’ experience of unemployment is a multifaceted construct and necessitates its study from a multidimensional perspective. Three dimensions are of importance from a psychological perspective: the unemployed’s experiences, their commitment to employment, and their job search intensity (De Witte et al., 2012; Vleugels et al., 2013). First, experiences relate to the unemployed’s deprivation of the opportunity to (a) set mutual goals, (b) have contact with others, (c) structure their time, (d) engage in regular activities, and (e) gain social status (Jahoda, 1982). Some unemployed may also have positive experiences (Van der Vaart et al., 2018). Second, ‘the extent to which a person wants to be engaged in paid work’ (Warr et al., 1979, p. 130) reflects their commitment to employment (De Witte et al., 2012). Third, job search intensity refers to the frequency with which a person engages in various job search activities (Wanberg, 2012).

Not only are these dimensions important, but they also interact with one another. For example, an unemployed person might search for a job more frequently, but this could lead to more negative experiences. Similarly, individuals might be committed to employment, which would enhance not only their search behaviour but also their negative experiences (De Witte, 1992).

Based on these dimensions, four types of unemployed were identified within the South African context. The optimists experienced their situation as somewhat negatively (but also as somewhat positively), and they were somewhat committed to employment, but were searching quite intensely for work. The desperate experienced unemployment extremely negatively, were committed to employment, and were likewise searching quite intensely for work. The discouraged were also negative, but they reported some positive experiences. Although work was important, there was little engagement in job search activities. The adapted had more positive than negative experiences, were less committed to work, and showed little interest in job search activities (Van der Vaart et al., 2018). The withdrawn were characterized by few negative experiences, low levels of employment commitment, and very low levels of job search behaviour (De Witte & Wets, 1996). The withdrawn were identified in Belgium and their presence might be attributable to the existence of an unemployment grant, which allows them to temporarily withdraw from the labour market (Van der Vaart et al., 2018). Therefore, the following is hypothesized:

Hypothesis 1: Four unemployed types (i.e., the optimist, desperate, discouraged, and adapted) exist, differing in their experiences, their commitment to employment, and their job search behaviour.

We expected these profiles to differ in their negative emotions and basic psychological need frustration. According to the appraisal theory (Arnold, 1960; Lazarus, 1966), emotions are adaptive responses that reflect judgements of environmental characteristics. Emotions have significant implications for one’s well-being and rational thinking (Moors et al., 2013). In line with the appraisal theory, the study was guided by a list of discrete emotions typically experienced in a workplace setting, that is, anger, anxiety, guilt, shame, envy, and jealousy (see Lazarus & Cohen-Charash, 2001).

The optimists and adapted might experience fewer negative emotions, as they were less involved in searching for a job (with less opportunity for rejection), attached less value to employment (with less dissonance between their values and their employment status) (Festinger, 1957), and reported fewer negative experiences. The adapted might also have adapted to unemployment psychologically (Van der Vaart et al., 2018). The opposite would hold for the discouraged and desperate. The desperate (with their higher levels of job search, commitment to employment, and negative experiences) would experience the most negative emotions. Therefore, the following is hypothesized:

Hypothesis 2: The optimists and the adapted will report the lowest levels of negative emotions (2a), followed by the discouraged (2b), with the desperate reporting the highest levels of negative emotions (2c).

According to the self-determination theory (SDT), people have three basic psychological needs: the need to feel that (a) their actions align with their wishes (i.e., autonomy), (b) they are competent and able to master their environment (i.e., competence), and (c) they belong and are cared for (i.e., relatedness). If satisfied, individuals experience optimal psychological functioning. However, need satisfaction is only one side of the coin, as these needs can be frustrated. The consequences of need frustration are more negative than mere low need satisfaction (Vansteenkiste et al., 2020). Unfortunately, unemployed people have a higher inclination towards need frustration than (low) need satisfaction (Vansteenkiste & Van den Broeck, 2018).

We expected a similar pattern for need frustration as for negative emotions. The optimists and adapted, receiving fewer instructions (on how and when to search for jobs) from others, experiencing fewer rejections, and placing less pressure on themselves (because of the value attached to employment), might feel less frustrated in their needs. The opposite would hold for the discouraged and desperate because of their intense job search, the high value they placed on employment, and their negative experiences of unemployment. Continuous unsuccessful attempts to find a job might lead to feelings of incompetence and isolation, especially if accompanied by pressure from others (Vansteenkiste & Van den Broeck, 2018). Therefore, the following is hypothesized:

Hypothesis 3: The optimists and the adapted will report the lowest levels of psychological need frustration (3a), followed by the discouraged (3b), with the desperate reporting the highest levels of psychological need frustration (3c).

Method

Participants

The sample consisted of 867 participants from Boipatong (54.20%) and Orange Farm (45.80%). Slightly more females (54.30%) than males (45.70%) participated. Almost all participants (99.20%) were Black Africans, and most spoke either Sesotho (46.80%) or isiZulu (28.50%). The majority reported not having completed (58.70%) or only having completed (38.70%) secondary education. On average, they were 32 years old (SD = 10.46), and most had been unemployed for more than 2 years (64.90%). Several of them were single (79.30%). More than half were living with parents, grandparents, or other family (50.50%), with no other income in the household. Only 22.50% received grants themselves. The sample was mostly characteristic of the unemployed in South Africa: the unemployment rate is higher among Black African females, individuals between the ages of 15 and 24, and those with a qualification lower than secondary school level (Stats SA, 2020).

Instruments

A biographical questionnaire was used to measure characteristics that are commonly measured within an unemployment context: gender, age, educational level, marital status, living situation, township, unemployment duration, employment history, social assistance (self or others) or another form of income earned by others in the household, and the number of individuals financially dependent on the unemployed participant.

The Experiences of Unemployment Questionnaire (EUQ; De Witte et al., 2010) was utilized to measure participants’ affective experiences, attitudes towards employment, and job search behaviour. Questions tapping into negative and positive experiences consisted of 10 and 6 items, respectively. Participants had to rate their negative (e.g., ‘I feel bored’) and positive (e.g., ‘I have more time to spend with my family members’) experiences on a 3-point frequency scale ranging from 1 (never) to 3 (often), which was recoded to 0 (never) to 2 (often) for analysis. The decision to use a 3-point frequency scale was to avoid confusion brought about by too many options (Flaskerud, 2012). The Negative Affective Experiences subscale was proven to be reliable (α = .91) and valid, as was the Positive Affect Experiences subscale (α = .80) (Van der Vaart et al., 2018).

The importance of work was measured by seven items based on the Employment Commitment Scale of Warr et al. (1979). Participants had to indicate to what extent they agreed with a range of statements (e.g., ‘It is better to accept any job than to be unemployed’) on a scale ranging from 1 (disagree) to 3 (agree). Previous research in the South African context supported the reliability and the validity of the scale (α = .73; .90) (De Witte et al., 2012; Van der Vaart et al., 2018). Job search behaviour was measured by asking how many times participants had performed any of the five different behaviours (e.g., ‘Asked friends, family or acquaintances if they were aware of any work’), reflected on a frequency scale ranging from 0 (never) to 4 (10 times or more). This scale was found to be reliable and valid in a South African context (α = .78; .91) (De Witte et al., 2012; Van der Vaart et al., 2018).

A self-constructed questionnaire, which was based on a comprehensive review of the literature, existing questionnaires (such as the Job Affective Scale [JAS] and the Job-Related Affective Well-Being Scale [JAWS]; Burke et al., 1989; Van Katwyk et al., 2000), and feedback from the unemployment advisory board, was used to measure discrete negative emotions that are relevant in this context. Validity information (supporting the unidimensionality and the construct [convergent and concurrent] validity of the scale) is reported in Tables S1, S2, and S3 of the online supplements (https://osf.io/ecrm8/?view_only=75661f0c2ad746359e9d006a3960243f). Participants had to rate their negative (e.g., ‘anger’, ‘fear’, and ‘guilt’) emotions on a 3-point frequency scale ranging from 1 (never) to 3 (often).

The Basic Psychological Need Satisfaction and Frustration Scale (BPNSFS; Chen et al., 2015) was used to measure basic psychological need frustration. The Need Frustration subscale contained 10 items, 3 items for each of the needs, except relatedness, which was measured through 4 items. Respondents had to indicate to what extent they agreed with a range of statements reflecting autonomy frustration (e.g., ‘I feel pressured to do too many things’), competence frustration (e.g., ‘I have serious doubts about whether I can do things well’), and relatedness frustration (e.g., ‘I have the impression that people I spend time with dislike me’) on a 3-point Likert-type scale ranging from 1 (disagree) to 3 (agree). Chen et al. (2015) reported reliability coefficients ranging between .64 and .89.

Procedure

Participants were recruited from two informal settlements in the Gauteng province of South Africa. Convenience sampling (i.e., recruiting unemployed participants roaming the streets, and door-to-door) and volunteer sampling (i.e., advertisements in community newspapers and on community radio stations) were used to obtain a heterogeneous sample of unemployed people who fitted the inclusion criteria in line with the expanded definition of unemployment (Stats SA, 2020). The fieldworkers (unemployed volunteers from communities other than those sampled) assisted research participants by means of structured interviews in completing the translated (e.g., into isiZulu and Sesotho) questionnaires. A back-translation judgemental design was employed to determine the equivalence of the translated questionnaires (see De Kock et al., 2013).

Ethical considerations

The Economic and Management Sciences Research Ethics Committee of the North-West University granted ethics approval for the study (NWU-00635-20-A4). A letter of information, adhering to strict ethical requirements, and a consent letter requesting voluntary participation were handed out to participants prior to the structured interviews.

Data analysis

Mplus 8.4 (L. K. Muthén & Muthén, 1998–2019) was used for data analysis. The mean- and variance-adjusted weighted least squares (WLSMV) estimator was used to perform confirmatory factor analysis (CFA). We used indices to assess fit based on recommendations by Kline (2016) and Wang and Wang (2020): comparative fit index (CFI) and Tucker–Lewis index (TLI) values higher than .90, root mean square error of approximation (RMSEA) values lower than .08, and standardized root mean square residual (SRMR) lower than .10. The reliabilities of the scales were calculated using the ordinal version of the Cronbach’s alpha reliability coefficient (Gadermann et al., 2012). Latent profile analysis (LPA) was performed on the factor scores saved from preliminary measurement models, a practice which has become more common in recent applications of mixture models (Meyer & Morin, 2016). LPA is a model-based approach to cluster individuals or cases into groups (latent profiles) based on their responses to observed continuous variables (B. Muthén & Muthén, 2000). Being model-based, formal statistical procedures were used to determine the optimal number of profiles: lower Bayesian information criterion (BIC) values and a non-significant bootstrap likelihood ratio test (BLRT) test (Nylund et al., 2007).

Finally, we used the Bolck-Croon-Hagenaars (BCH) method (see Bakk & Vermunt, 2016, for a detailed description) to compare the levels of negative emotions and basic psychological need frustration across the profiles.

Results

CFA for the profile variables

CFA models were estimated to establish the factor structure for the typology (profile) variables and the correlates. The CFA models for the typology variables were specified without the correlates to retain independence between the profiles and the correlates. The four latent typology variables (i.e., negative experiences, positive experiences, commitment, and job search with their respective observed indicators or items) were specified as separate, but related, factors. Based on the fit statistics, the hypothesized model (Model 1) did not yield acceptable fit to the data. Model development was performed by deleting one item from the negative experiences scale (‘I must save on my personal expenditure’) and one item from the importance of work scale (‘People do not have to work as such to be constructively occupied’) because of inadequate factor loadings (.26 and −.07). The errors of two negative experience items (‘My self-worth has decreased’ and ‘I have lost my self-confidence’) and two job search items (‘Searched for advertisements in newspapers and weeklies’ and ‘Searched for advertisements on the internet [e.g., job or organisational websites] or social media [e.g., Facebook, LinkedIn]’) were allowed to correlate due to high modification indices (81.92 and 62.61). The revised model (Model 1a) had a better fit according to most of the indices (see Table 1).

Table 1.

Measurement models for the psychosocial variables (N = 867).

Model	χ²	p value	df	TLI	CFI	RMSEA	SRMR
Model 1	1305.91	.00	399	.85	.87	.05 [.05, .05]	.08
Model 1a	974.43	.00	342	.89	.90	.05 [.04, .05]	.07

χ2: chi-square; df: degrees of freedom; TLI: Tucker–Lewis index; CFI: comparative fit index; RMSEA: root mean square error of approximation; SRMR: standardized root mean square residual.

LPA

The optimal number of classes was determined by estimating five models with an increasing number of latent classes and comparing the fit statistics of these models. Table 2 shows the fit indices for the models.

Table 2.

Comparison of LPA models.

Model	Log likelihood	#fp	Scaling	AIC	BIC	ABIC	Entropy	LMR LR testp	ALMR LR testp	BLRTp
1-profile LPA	−2659.52	8	0.92	5335.03	5373.15	5347.75	–	–	–	–
2-profile LPA	−2460.32	13	1.05	4946.64	5008.58	4967.30	.68	.00	.00	.00
3-profile LPA	−2397.37	18	1.36	4830.74	4916.51	4859.34	.69	.09	.09	.00
4-profile LPA	−2351.86	23	1.14	4749.72	4859.32	4786.28	.77	.00	.00	.00
5-profile LPA	−2317.07	28	1.10	4690.14	4823.56	4734.64	.78	.00	.00	.00

LPA: latent profile analysis; AIC: Akaike information criterion; BIC: Bayesian information criterion; ABIC: Adjusted Bayesian information criterion; LMR LR: Lo-Mendell-Rubin likelihood ratio test; ALMR LR: Adjusted Lo-Mendell-Rubin likelihood ratio test; BLRT: Bootstrap likelihood ratio test.

For all models, the BIC values failed to reach a minimum, and the BLRT remained statistically significant (p < .05). Since these tests are all tests of statistical significance, determining the optimal number of profiles can be influenced by the sample size (Marsh et al., 2009) and it is likely that the indicators may improve without reaching a minimum (Howard et al., 2016). An examination of the plots indicated that adding a fourth profile resulted in the addition of a qualitatively and quantitatively different profile, compared with the model with three profiles, but adding a fifth profile resulted in a reduction of the number of desperate unemployed to form another profile that one would also label ‘discouraged’. Hence, the five-profile solution has two discouraged groups and does not add anything theoretically meaningful. In addition, the fit indices are only slightly better for the fifth profile. Therefore, the more parsimonious four-profile model was preferred. The solution was theoretically interpretable and yielded adequate profiling; the posterior class membership probabilities were all above .87, and the entropy value (.77) was also quite high (Clark, 2010).

As displayed in Table 3 and Figure 1, four profiles could, thus, be identified, differing on the typology variables. Therefore, Hypothesis 1 stating that four unemployed types exist was accepted.

Table 3.

Mean-level differences for the types of unemployed.

	Optimists	Desperate	Discouraged	Adapted
Psychosocial variables
Negative experiences	−.32^c (1.07)	.70^a (1.88)	.05^b (1.40)	−.69^d (0.77)
Positive experiences	−.03^b (0.94)	.04^ab (1.05)	−.01^b (0.96)	.12^a (1.06)
Commitment to employment	−.43^d (2.58)	.65^a (3.00)	.20^b (2.90)	−.15^c (2.06)
Application behaviour	−.06^b (2.08)	.12^a (2.14)	.05^a (2.20)	−.30^c (1.75)
Correlates
Negative emotions	−.24^c (2.03)	.60^a (2.64)	.03^b (2.27)	−.51^d (1.79)
Autonomy frustration	−.10^c (1.99)	.23^a (2.36)	−.00^b (2.16)	−.12^c (1.99)
Competence frustration	−.10^c (1.79)	.24^a (2.02)	.03^b (1.90)	−.14^c (1.77)
Relatedness frustration	−.19^c (1.83)	.47^a (2.37)	−.01^b (2.01)	−.21^c (1.82)

Indicators estimated from scaled scores are indicated in parentheses. Within rows, means with different letters are significantly different from each other.

Figure 1.

Psychosocial profiles.

Profile 1 (optimists) showed an equally moderate score on negative and positive experiences, considered work important (but less so than the desperate and the discouraged), and showed moderate application behaviour. The optimists constituted 34.60% of the sample. Profile 2 (desperate) scored high on negative experiences (outweighing the moderate positive experiences) and considered work to be very important (more so than the other profiles), but showed moderate application behaviour (although more than the optimists and the adapted). The desperate constituted 19.30% of the sample. Profile 3 (discouraged) scored moderately high to moderate on negative and positive experiences and considered work important (more so than the optimists and the adapted), but showed moderate application behaviour. The discouraged constituted 37.90% of the sample. Profile 4 (adapted) scored low on negative experiences and moderate on positive experiences, attached little importance to work, and showed moderately low application behaviour. The adapted constituted 8.30% of the sample.

Statistically significant associations were detected between the different profiles and certain demographic variables (age, educational level, unemployment duration, number of dependants, and grants), but these associations were, at best, only small effects.

Negative emotions and need frustration as correlates of profile membership

Two competing models were specified for the correlates. In the first model, negative emotions and psychological need frustration were both specified as unidimensional. This model fitted the data poorly (χ² = 1070.85, df = 118, p < .001; RMSEA = .10 [.09, .10]; CFI = .86; TLI = .84; SRMR = .07). In the second model, negative emotions were specified as unidimensional, whereas psychological need frustration was modelled as three dimensions: autonomy, competence, and relatedness frustration. Model 2 had a good fit (χ² = 420.79, df = 113, p < .001; RMSEA = .06 [.05, .06]; CFI = .96; TLI = .95; SRMR = .05). We had no theoretical rationale to expect differences between the types of unemployed and the different negative emotions and the different dimensions of psychological need frustration. Therefore, they were specified as unidimensional. Fit statistics, however, showed that psychological need frustration had to be treated as multidimensional. Descriptive statistics, and reliability and correlation coefficients are reported in Table S3 of the online supplements. The internal consistency of the scales is supported by reliability coefficients equal to or higher than .70.

Mean-level factor score differences were examined between the four types of unemployed regarding negative emotions and psychological need frustration. As displayed in Table 3, significant differences can be reported between the different types of unemployed.

The desperate reported the most negative emotions, followed by the discouraged and the optimists, with the adapted reporting the fewest negative emotions. Therefore, Hypotheses 2a to 2c, stating that the optimists and the adapted will report the lowest levels of negative emotions (2a), followed by the discouraged (2b), with the desperate reporting the highest levels of negative emotions, were accepted. Post hoc analyses were performed in which the comparisons were made on the level of the separate emotions. The same pattern emerged for most of the emotions, with the exception of shame (the optimists experienced less shame than the adapted) and jealousy (the optimists and the adapted experienced equal amounts of jealousy). The desperate reported that they experienced the most psychological need frustration, followed by the discouraged. The optimists and the adapted experienced equal amounts of psychological need frustration, but less so than the other two profiles. Therefore, Hypotheses 3a to 3c, stating that the optimists and the adapted will report the lowest levels of psychological need frustration (3a), followed by the discouraged (3b), with the desperate reporting the highest levels of psychological need frustration (3c), were accepted. The results also showed that the different types of unemployed followed the same pattern for all three needs (desperate > discouraged > adapted = optimists).

Discussion

The first aim of this study was to establish whether the four unemployment profiles could be replicated. In line with our expectations, the four types were replicated, differing in their experiences, their commitment to employment, and their job search behaviour. Consequently, ‘the unemployed’ was not a homogeneous, but rather a heterogeneous group, consisting of several homogeneous subgroups. The profile composition was in line with previous research (see De Witte & Hooge, 1995; Van der Vaart et al., 2018). However, the prevalence of the respective profiles bore noteworthy differences from previous research. For example, roughly three times more optimists were identified in this sample than in the one by Van der Vaart et al. (2018), and roughly a quarter more than in the study by De Witte (1992). Also, roughly a third fewer desperate were identified compared with the previous South African study. The number of discouraged remained very similar in comparison with the previous South African study by Van der Vaart et al. (2018), but was almost three times higher than the initial research by De Witte (1992). The adapted decreased by almost two-thirds from the prevalence found by previous studies (De Witte, 1992; Van der Vaart et al., 2018).

These differences could be ascribed to five aspects: legislation, beliefs, family arrangements, structural challenges associated with job search, and high levels of poverty. In South Africa, legislation, such as the Employment Equity Act and the Broad-Based Black Economic Empowerment Act, makes provision for redressing past imbalances through preferential treatment of Black South Africans when appointing employees (Horwitz & Jain, 2011). In a recent study, Du Toit et al. (2018) found that the unemployed believed in the power of external forces to secure employment. Both legislation and beliefs might account for the higher prevalence of optimists and the lower prevalence of the desperate among this sample as it brings hope. Furthermore, multigenerational and extended family living arrangements (common in South Africa) generally provide a buffer against the negative effects and help to absorb the impact of unemployment (Barnett, 2008), which would also account for fewer desperate unemployed. Even though South Africa’s unemployed are extremely poverty stricken, Fourie (2012) highlights desperate individuals’ constant battle in trying to overcome various barriers (e.g., the cost of searching for a job and the local unemployment rate), ultimately leading to discouraged jobseekers. Poverty, especially among disadvantaged communities, might also result in fewer adapted unemployed, as they do not have the luxury to withdraw from the labour market (Van der Vaart et al., 2018).

The second aim of this study was to examine the associations between these profiles and negative emotions and psychological need frustration. Examining these associations further attested to the validity of these profiles and provided valuable insights for intervention development. The optimists and adapted experienced fewer negative emotions and less psychological need frustration compared with the desperate and discouraged. This result could be the consequence of three aspects. First, the optimists and the adapted were less involved in job search behaviour and are, therefore, less exposed to the challenges and rejection accompanying a job search. Second, they were less committed to employment (the adapted even more so than the optimists), which further lowered their negative emotions and psychological need frustration. In addition, the adapted might be engaged in a ‘psychological adaptation process’ in which they had adapted to their role as unemployed (De Witte et al., 2010). In the third place, the optimists were more employable as they are younger, educated, and short-term unemployed (Van der Vaart et al., 2018) compared with the other groups, resulting in more positive outcomes. In contrast, the results indicated that the desperate and discouraged experienced more negative emotions and psychological need frustration. This could be attributed to their high levels of employment commitment, their continuous involvement in job search behaviour and subsequent rejection, the constant need to overcome various obstacles and challenges (Fourie, 2015), and their higher levels of employment commitment.

Having concluded that ‘the unemployed’ is not a homogeneous group, it is important that these groups are optimally supported according to their differential experiences. Since the optimists and adapted have fewer negative emotions and their needs are less frustrated, interventions that optimize these aspects for these two groups are ideal. For them, economic interventions (e.g., job creation) may be more beneficial. Psychological interventions could be more preventive (e.g., problem-solving skills to enable them to deal with challenges in the job-seeking process). The desperate and discouraged require remedial psychological interventions. Interventions should be targeted at lowering their negative emotions and their psychological need frustration (Vansteenkiste & Van den Broeck, 2018). For example, psychological needs could be satisfied, by stakeholders who are dealing with the unemployed, in the following ways: sharing decision-making (autonomy support), providing uplifting feedback (competence support), and showing interest (relatedness support). Ultimately, by implementing tailored interventions, the effectiveness of interventions will be ensured (Koopman et al., 2017). From a policy perspective, stakeholders could consider prioritizing interventions for the desperate and discouraged as they are the two most prevalent groups and psychologically, they suffer the most. Furthermore, policymakers could focus on preventing the optimist from developing into one of the other four groups as it has economic and/or psychological consequences.

The study made use of a cross-sectional design, which limited any claims about causality. The study was only conducted in two communities and only included one ethnic group. Although the sample is more representative than that of Van der Vaart et al. (2018), these two limitations might limit the generalizability of the findings. Due to the lower levels of literacy of typical unemployed people, understanding of the content of the various questionnaires could have proven more difficult. Future studies could employ a longitudinal design, and they could also replicate the study in more communities to strengthen, validate, and ensure meaningful interpretation of the different types of unemployed people within a non-Western context. A longitudinal study would additionally enable researchers to explore the transitioning of unemployed individuals between different profiles as well as to detect changes in the levels of psychological need satisfaction and identified variables because of (changes in) profile membership. Furthermore, future research could investigate positive emotions and psychological need satisfaction to provide a comprehensive overview of the psychological functioning of the unemployed.

Conclusion

The study confirmed the existence of four types of unemployed people in South African communities; in other words, ‘the unemployed’ are similar in that they are unemployed, but different in that they experience unemployment differently. The types also differ on other psychological correlates, which further supports interventions tailored to their needs.

Supplemental Material

sj-pdf-1-sap-10.1177_0081246320978969 – Supplemental material for Profiling the unemployed from selected communities in South Africa based on their experiences, commitment to employment, and job search behaviour

Supplemental material, sj-pdf-1-sap-10.1177_0081246320978969 for Profiling the unemployed from selected communities in South Africa based on their experiences, commitment to employment, and job search behaviour by Ivan Putter, Leoni van der Vaart, Hans De Witte, Sebastiaan Rothmann and Anja Van den Broeck in South African Journal of Psychology

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors gratefully acknowledge the support of the Flemish Interuniversity Council—University Development Cooperation (VLIR-UOS) for awarding funding that enabled the research and authorship of this article.

ORCID iDs

Ivan Putter

Leoni van der Vaart

Hans De Witte

Sebastiaan Rothmann

Supplemental material

Supplemental material for this article is available online.

References

Arnold

M. B.

(1960). Emotion and personality. Columbia University Press.

Bakk

Vermunt

J. K.

(2016). Robustness of stepwise latent class modeling with continuous distal outcomes. Structural Modeling: A Multidisciplinary Journal, 23(1), 20–31.

Barnett

M. A.

(2008). Economic disadvantage in complex family systems: Expansion of family stress models. Clinical Child and Family Psychology Review, 11(3), 145–161.

Bernstein

(2019, June 25). How to put the long-term unemployed back to work. The Washington Post. https://www.washingtonpost.com/outlook/2019/06/25/how-put-long-term-unemployed-back-work/

Brand

J. E.

(2015). The far-reaching impact of job loss and unemployment. Annual Review of Sociology, 41, 359–375.

Burke

M. J.

Brief

A. P.

George

J. M.

Roberson

Webster

(1989). Measuring affect at work: Confirmatory analyses of competing mood structures with conceptual linkage to cortical regulatory systems. Journal of Personality and Social Psychology, 57(6), 1091–1102.

Chen

Vansteenkiste

Beyers

Boone

Deci

E. L.

Van der Kaap-Deeder

Duriez

Lens

Matos

Mouratidis

Ryan

R. M.

Sheldon

Soenens

Van Petegem

Verstuyf

(2015). Basic psychological need satisfaction, need frustration, and need strength across four cultures. Motivation and Emotion, 39(2), 216–236.

Clark

S. L.

(2010). Mixture modeling with behavioral data (UMI No. 3405665) [Doctoral dissertation, University of California]. ProQuest Dissertations and Theses Global.

De Kock

Kanjee

Foxcroft

. (2013). Cross-cultural test adaptation, translation, and tests in multiple languages. In Foxcroft

Roodt

(Eds.), Introduction to psychological assessment in the South African context (4th ed., pp. 83–106). Oxford University Press.

10.

De Witte

. (1992). Tussen optimisten en teruggetrokken: Een empirisch onderzoek naar het psychosociaal profiel van langdurig werklozen en deelnemers aan de Weer-Werkactie in Vlaanderen [Between optimism and withdrawn: An empirical investigation into the psychosocial profile of the long-term unemployed]. Hoger Instituut voor de Arbeid, KU Leuven.

11.

De Witte

Hooge

. (1995). Een vergelijking van typologieen van kort- en langdurig weklozen: wetenschappelijke en beleidsmatige implicaties [A comparison between the short- and long-term unemployed: Scientific and policy implications]. In Van der Hallen

(Ed.), De Arbeidsmarktonderzoekersdag (pp. 126–142). Hoger Instituut voor de Arbeid, KU Leuven.

12.

De Witte

Hooge

Vanbelle

. (2010). Do the long-term unemployed adapt to unemployment? Romanian Journal of Applied Psychology, 12, 8–14.

13.

De Witte

Rothmann

Jackson

. (2012). The psychological consequences of unemployment in South Africa. The South African Journal of Economic and Management Sciences, 15(3), 235–252.

14.

De Witte

Wets

. (1996). On the heterogeneity of the experience of long-term unemployment among young women. In De Goede

M. P. M.

De Klaver

P. M.

Van Ophem

J. A. C.

Verhaar

C. H. A.

De Vries

(Eds.), Youth: Unemployment, identity and policy (pp. 65–85). Ashgate.

15.

Du Toit

De Witte

Rothmann

Van den Broeck

. (2018). Unemployment experiences in context: A phenomenological study in two townships in South Africa. Journal of Psychology in Africa, 28(2), 122–127.

16.

Festinger

(1957). A theory of cognitive dissonance. Stanford University Press.

17.

Flaskerud

(2012). Cultural bias and Likert-type scales revisited. Issues in Mental Health Nursing, 33(2), 130–132.

18.

Fourie

C. v. N.

(2012). The South African unemployment debate: Three worlds, three discourses? http://www.saldru.uct.ac.za/documentation/redi-papers/321-the-south-african-unemployment-debate-three-worlds-three-discourse-revised/file

19.

Fourie

C. v. N.

(2015). The South African unemployment debate: Three words, three discourses? https://www.researchgate.net/publication/275715297_The_South_African_unemployment_debate_Three_words_three_discourses

20.

Gadermann

A. M.

Guhn

Zumbo

B. D.

(2012). Estimating ordinal reliability for Likert-type and ordinal item response data: A conceptual, empirical, and practical guide. Practical Assessment, Research, & Evaluation, 17(3), 1–13.

21.

Horwitz

F. M.

Jain

(2011). An assessment of employment equity and Broad Based Black Economic Empowerment developments in South Africa. Equality, Diversity and Inclusion, 30(4), 297–317.

22.

Howard

Gagné

Morin

A. J. S.

Van den Broeck

(2016). Motivation profiles at work: A self-determination theory approach. Journal of Vocational Behavior, 95–96, 74–89.

23.

Jahoda

(1982). Employment and unemployment: A social-psychological analysis. Cambridge University Press.

24.

Kline

R. B.

(2016). Principles and practice of structural equation modeling (4th ed.). Guilford Press.

25.

Knopf

(2013). Typologies of homelessness: Moving beyond a homogeneous perspective. https://pdfs.semanticscholar.org/73e6/e8d1f32bcf6857e36729c78a08e5ce35966b.pdf?_ga=2.139630705.373767016.1581959406-22165168.1568654019

26.

Koopman

M. Y.

Pieterse

M. E.

Bohlmeijer

E. T.

Drossaert

C. H. C.

(2017). Mental health promoting interventions for the unemployed: A systematic review of applied techniques and effectiveness. International Journal of Mental Health Promotion, 19(4), 202–223.

27.

Lazarus

R. S.

(1966). Psychological stress and the coping process. McGraw-Hill.

28.

Lazarus

R. S.

Cohen-Charash

(2001). Discrete emotions in organizational life. In Payne

R. L.

Cooper

G. L.

(Eds.), Emotions at work: Theory, research and applications for management (pp. 45–81). John Wiley & Sons.

29.

Liu

Hang

J. L.

Wang

(2014). Effectiveness of job search interventions: A meta-analytic review. Psychological Bulletin, 140(4), 1009–1041.

30.

Marsh

H. W.

Lüdtke

Trautwein

Morin

(2009). Classical latent profile analysis of academic self-concept dimensions: Synergy of person- and variable-centered approaches to theoretical models of self-concept. Structural Equation Modeling, 16(2), 191–225.

31.

Meyer

J. P.

Morin

A. J. S.

(2016). A person-centered approach to commitment research: Theory, research, and methodology. Journal of Organizational Behavior, 37(4), 584–612.

32.

Moors

Ellsworth

Scherer

K. R.

Frijda

N. H.

(2013). Appraisal theories of emotion: State of the art and future development. Emotion Review, 5(2), 119–124.

33.

Morin

A. J. S.

McLarnon

M. J. W.

Litalien

(2020). Mixture modeling for organizational behavior research. In Griep

Hansen

S. D.

(Eds.), Handbook on the temporal dynamics of organizational behavior (pp. 351–379). Edward Elgar.

34.

Muthén

L. K.

(2000). Integrating person-centered and variable-centered analyses: Growth mixture modeling with latent trajectory classes. Alcoholism: Clinical and Experimental Research, 24(6), 882–891.

35.

Muthén

L. K.

Muthén

B. O.

(1998–2019). Mplus user’s guide (8th ed.).

36.

Nylund

K. L.

Asparouhov

Muthén

B. O.

(2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal, 14, 535–569.

37.

Paver

Rothmann

Van den Broeck

De Witte

(2019). Labour market interventions to assist the unemployed in two townships in South Africa. South African Journal of Industrial Psychology, 45(0), a1596.

38.

Statistics South Africa. (2020). Quarterly labour force survey: Quarter 1, 2020. http://www.statssa.gov.za/publications/P0211/P02111stQuarter2020.pdf

39.

Van Der Vaart

De Witte

Van den Broeck

Rothmann

. (2018). A psychosocial typology of the unemployed in South Africa. South African Journal of Psychology, 48(2), 179–192.

40.

Van Katwyk

P. T.

Fox

Spector

P. E.

Kelloway

E. K

. (2000). Using the Job-Related Affective Well-Being Scale (JAWS) to investigate affective responses to work stressors. Journal of Occupational Health Psychology, 5(2), 219–230.

41.

Vansteenkiste

Ryan

M. R.

(2013). On psychological growth and vulnerability: Basic psychological need satisfaction and need frustration as a unifying principle. Journal of Psychotherapy Integration, 23(3), 263–280.

42.

Vansteenkiste

Ryan

R. M.

Soenens

(2020). Basic psychological need theory: Advancements, critical themes, and future directions. Motivation and Emotion, 44, 1–31.

43.

Vansteenkiste

Van den Broeck

(2018). Understanding the motivational dynamics among unemployed individuals: Refreshing insights from the self-determination theory perspective. In Klehe

van Hooft

E. A. J.

(Eds.), The Oxford handbook of job loss and job search (pp. 1–23). Oxford University Press.

44.

Vleugels

Rothmann

Griep

De Witte

(2013). Does financial hardship explain differences between Belgian and South African unemployed regarding experiences of unemployment, employment commitment and job search behaviour? Psychologica Belgica, 53(2), 75–95.

45.

Wanberg

C. R.

(2012). The individual experience of unemployment. Annual Review of Psychology, 63, 369–396.

46.

Wang

(2020). Structural equation modeling (2nd ed.). John Wiley & Sons.

47.

Warr

P. B.

Cook

J. D.

Wall

T. D.

(1979). Scales for the measurement of work attitudes and aspects of psychological well-being. Journal of Occupational Psychology, 52(2), 129–148.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB