SciELO - Scientific Electronic Library Online

 
vol.33 issue2Habilidades sociais e adaptação à universidade: Convergências e divergências dos construtosColaboração terapêutica: Estudo comparativo dois casos de insucesso terapêutico - Um caso finalizado e de um caso de desistência author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • Have no similar articlesSimilars in SciELO

Share


Análise Psicológica

Print version ISSN 0870-8231

Aná. Psicológica vol.33 no.2 Lisboa June 2015

http://dx.doi.org/10.14417/ap.922 

Attributions of causes for unemployment by unemployed workers

Pedro F. Bendassolli1, Sonia Maria Guedes Gondim2, Fellipe Coelho-Lima1

1Universidade Federal do Rio Grande do Norte, Brasil

2Universidade Federal da Bahia, Brasil

Correspondência

 

ABSTRACT

This study investigated the assignment of causes for unemployment by unemployed workers, with a view to analyzing the predictive power of sociodemographic variables for the assignment. A scale of causal attribution of unemployment, originally developed by Furnham, was applied to 376 unemployed people. After confirmatory factor analysis (CFA), the scale factors were used in a regression model containing sociodemographic variables as predictors. The CFA results support Furnham’s original three-factor model of unemployment causes (individualistic, societal, and fatalistic; χ2(100)=261.53, p<0.001; χ2/gl=2.61; CFI=0.91, TLI=0.90; RMSEA=0.06). Regression analysis identified significant prediction for only the income variable and individualistic causes factor (β=0.15, p<0.01), the income variable and societal causes factor (β=0.10, p<0.001), and the educational level variable and societal causes factor (β=-0.15, p<0.01). Societal causes presented the highest average score, which was significantly (p<0.001) different than the scores for the other two factors. The study concludes that educational level does not seem to have a linear impact on beliefs regarding the causes for unemployment, nor does gender have a significant influence on these beliefs.

Key-words: Unemployment, Attributing causes, Furnham’s scale.

 

RESUMO

Investiga-se o processo de atribuição de causas para o desemprego por trabalhadores desempregados, buscando analisar o poder preditivo de variáveis sociodemográficas sobre esse processo. Foi empregada escala de atribuição de causas do desemprego originalmente desenvolvida por Furnham e aplicada a 376 pessoas desempregadas. Após análise fatorial confirmatória, os fatores da escala foram utilizados em um modelo de regressão contendo variáveis sociodemográficas como preditoras. Os resultados da AFC dão suporte ao modelo original de três fatores de causas do desemprego (individualistas, societais e fatalistas; χ2(100)=261.53, p<0.001; χ2/gl=2.61; CFI=0.91, TLI=0.90; RMSEA=0.06). A análise de regressão identificou uma predição significativa apenas com a variável renda e o fator de causas individualistas (β=0.15, p<0.01); e as variáveis renda (β=0.10, p<0.001) e escolaridade (β=-0.15, p<0.01) e o fator de causas societais. Causas societais apresentaram as maiores médias, sendo significativamente (p<0.001) diferentes das médias dos outros dois fatores. O estudo conclui referindo-se ao fato de que o nível de escolarização não tem um impacto direto (quanto mais aumenta ou diminui um, mais aumenta ou diminui o outro) na formação de crenças sobre as causas do desemprego.

Palavras-chave: Desemprego, Atribuição de causas, Escala de Furnham.

 

The importance of understanding socioeconomic and psychological factors related to unemployment has grown throughout the world, particularly since the global financial crisis of 2007. Eurostat (February 2014) reported EU unemployment rates above 10% – for example, in Spain (25.6%), France (10.4%), Italy (13.0%), and Portugal (15.3%). Brazil has had similar unemployment rates, such as 18.7% according to DIEESE (2006), and only recently has Brazil managed to reduce and stabilize unemployment rates around 10% (Figure 1).

 

 

Two of the ways in which unemployment causes damage to a society are the permanent loss of human capital in the medium and long term, especially for the young, and the negative impact on people’s health and well-being (Choudhry, Marelli, & Signorelli, 2013). With regard to the former, the International Labor Organization special report on Trends in Global Employment 2013 stated that globally, 197 million people were unemployed in 2012, and that young people were especially prominent among the groups most affected by the financial crises (International Labour Organization [ILO], 2013). People with little work experience and less developed skills, as is the case with young people, offer a lower human capital than adult workers and are thus more vulnerable. With regard to the negative impacts of unemployment on health and well-being, Boyce, Wood, Daly, and Sedikides (in press) conducted a longitudinal study of employment, following the occupational history of more than 6,000 people, and they found strong evidence that the unemployment experience is associated with personality changes, with repercussions for ways of thinking, feeling, and behaving.

While acknowledging that the problems of unemployment must be addressed by studies and by economic, political, and social actions that are complex at the macro level (e.g., Blustein, Medvide, & Wan, 2012; Kapuvári, 2011), we believe that analyses of personal beliefs regarding the causes of unemployment can ultimately contribute to the development of public policies on psychosocial support for the unemployed.

According to Feather (1985), attributions of causes for events are strongly based on people’s values and attitudes. People who are conservative tend to attribute internal and individualistic causes to the unemployment condition more than societal ones. Public policies focused on actions at the macroeconomic level should recognize that socialization processes interfere with attribution processes and that they impact motivational levels and strategies for dealing with unemployment. Thus, studies on causal attributions made by unemployed workers with regard to unemployment are of great importance.

One of the basic premises of common sense psychology is that people act as naive scientists and try to explain events in the world, giving them meaning, which leads to a greater sense of control over the events (e.g., Crittiden, 1983; Martinko, Douglas, & Harvey, 2006). The classical authors who have studied the process of psychological attribution (e.g., Heider, 1958; Kelley, 1973; Weiner, 1986) have suggested that the following are important ingredients for understanding this phenomenon: (i) the person making the attribution of cause (the actor of the action or its observer), (ii) the source of the cause (internal or external to the person making the attribution), (iii) the level of control over the cause (modifiable or not), and (iv) the stability of the cause over time (mutable or immutable).

Empirical studies over the years have indicated that individual variables are correlated with an individualistic style of assigning causes to successes or failures in one’s personal life, for example, gender (e.g., Campbell & Henry, 1999) and motivational values profile (e.g., Gondim, Álvaro-Estramiana, Luna, Oliveira, & Souza, 2010). In the case of unemployment, which is generally considered a failure event, empirical evidence suggests that one’s attributional style (with respect to its causes) can have a negative impact on self-esteem and psychological well-being (e.g., Álvaro-Estramiana, Gondim, Garrido-Luque, Luna, & Dessen, 2012; Guimarães, Demazière, & Sugita, 2010; Jahoda, 1979; Moyano-Díaz, Gutiérrez-Peña, Zúñiga-Cabrera, & Cornejo, 2013).

One of the first researchers to study and create a questionnaire for measuring attributions of unemployment causes was Furnham (1982). The findings of his study, which was conducted in the UK, corroborated the role of the actor or observer in causal attributions for unemployment. He found that employees tend to blame the individual for unemployment more than do unemployed people, who attribute unemployment to external causes (society, economy, and technology). Another conclusion of Furnham’s study was that although a high educational level increases an individual’s chances of attributing unemployment to external causes – since we can presume that in this case the individual has acquired a more critical stance on the society, thanks to the education – the attributor’s role as observer (employee) or actor (unemployed) influences whether the attribution identifies internal or external causes. Observers tend to attribute unemployment to individualistic causes rather than societal factors – an “attribution bias” phenomenon that has been pointed out by Jones and Davis (1965).

Furnham (1988) and Furnham and Rawles (1996) proposed that there are three main types of reasons given for unemployment: (a) individualistic reasons related to dispositional and personal factors (internal causes), (b) societal reasons involving policies and organizations (external and structural), and (c) fatalistic reasons related to technological developments whose inevitability brings about negative consequences. Gondim et al. (2010) subsequently attempted to validate Furnham’s (1988) model in Brazil. To this end, they used a sample of employed and unemployed individuals from the northeast region of the country, and they analyzed the obtained data with exploratory analytical processes. Their analysis identified several problems in adapting the measure to the Brazilian reality, as will be described below. Thus, the primary objective of the present study is to conduct a confirmatory factor analysis (CFA) of Furnham’s model with the aim of contributing to the cultural adaptation of this instrument and its psychometric development.

Additionally, this study has a second objective, which is to analyze the explanatory power of some sociodemographic variables in attributions of fatalistic, societal, and individualistic causes, with a view to testing some suggestions that have been made in the literature in this regard. For example, studies have suggested that men and women react differently to job loss (e.g., Burke, 1985) and that women tend to make more external attributions of causes for job loss than do men (Feather & O’Brien, 1986). This can happen because women do not see their work as a core value or because their earnings would be secondary for the family income. However, a study by Leana and Feldman (1991) concluded that men and women employ different strategies to deal with job loss and that this is probably because of their socialization processes. Men engage in more active job search strategies, while women tend to seek social support. Additionally, Wooten and Valentin (2008) found gender differences in causal attributions concerning job loss (women attribute it to more external causes) and in emotional reactions to it (women become more depressed, while men express more anger). But these authors concluded that these gender differences practically disappear when age, ethnicity, marital status, education, employment status, and income were controlled for. It is this theoretical context that situates the second objective of the present study.

 

Method

 

Participant

Three hundred and seventy-six (376) unemployed people living in a northeast Brazilian municipality and attending local units of the National System of Labor Intermediation (SINE, in Portuguese) – which is linked to the Brazilian Federal Government’s Ministry of Labor – participated in the accidental sampling. SINE aims to integrate policies on the routing of workers for available job positions. Therefore, it acts as a labor broker, connecting potential employees to potential employers (as a kind of public employment agency). In addition, its mission is to support the generation of employment and income, professional training, and access to unemployment insurance – which is paid to an employee, provisionally, when the employee loses his or her job due to dismissal and is seeking a new job. SINE was established by Brazilian Law (Law, nº 8.019/1990).

The sample was evenly distributed between the two sexes (50.7% men), with a predominance of people with an incomplete higher education (53.1%), followed by those with a complete secondary school education (13.3%), a complete higher education (10.3%), a complete postgraduate education (8.5%), a complete basic education (8.0%), and an incomplete secondary education (6.4%). More than half of the workers were youth between the ages of 15 and 29 years (55.8%), with a majority of the youths between 18 and 24 years of age (30.8%). For most (86.7%), personal income from their last jobs did not exceed twice the minimum wage (at the time R$678.00 monthly, about €250.00), with nine people (2.4%) earning more than five times the minimum wage. No participant had been unemployed for more than six months, and more than half (59.8%) had held only one job in the past two years, suggesting a reduced turnover among these workers. Only those who had not entered into any formal or informal employment participated in the survey.

 

Instrument

The instrument used in this study evolved from Furnham’s (1988) model, which was described in the introduction. In an attempt to enhance Furnham’s scale of causal attributions for unemployment and also to find evidence of cross-cultural validation of the three factorial dimensions of Furnham’s model, Mylonas (2007) developed an instrument of 44 items that included 20 items from the original scale and applied it to a sample of 250 people. The results offered support for a triarchic model with 19 items constituting the three dimensions (7 items for individualistic causes, 6 for societal causes, and 6 for fatalistic causes), with only 8 items coming from the original scale. Mylonas and Furnham (2014) subsequently confirmed the new measure in a cross-cultural study of samples from eight countries.

Gondim et al. (2010) adapted Mylonas and Furnham’s questionnaire to Portuguese and attempted to confirm the triarchic solution for Brazil – but they were not able to do so, in part because Brazilians ascribe unemployment to multiple causes. The authors used an exploratory factor analysis, and this revealed three factors (KMO=0.68), each of which merged individualistic, societal, and fatalistic causes, with factor loadings ranging from 0.35 to 0.68 for the first factor (8 items, Cronbach’s alpha 0.65), 0.36 to 0.62 for the second factor (8 items, Cronbach’s alpha 0.75) and 0.58 to 0.77 for the third factor (3 items, Cronbach’s alpha 0.65). The first two factors jointly explain 13% of the variance, and the third explains 9%. According to Gondim et al., this mix of items reveals some measurement problems when applied to the Brazilian sample.

The present study uses Gondim et al. (2010)’s adaption of Mylonas and Furnham’s questionnaire for assigning causes for unemployment. Each item is a judgment on the cause for unemployment: individualistic (e.g., “Lack of effort and laziness among unemployed people”), societal (e.g., “The educational system does not correspond to the current job market”), or fatalistic (e.g., “Demographic and population changes”). To answer the items, participants used a 4-point Likert scale ranging from 1 (not at all related to unemployment) to 4 (extremely related to unemployment). A section was added to the questionnaire to collect demographic information from the participants, including gender, number of formal jobs in the last two years, age, education, and income.

 

Data collection procedures. The instrument was applied presentially in an assisted form to unemployed workers who were using four local units of SINE in 2013-2014. After obtaining authorization from the Office of the Secretary of State, the applicators’ visits to the SINE units commenced. These visits occurred two to three times per week in intercalated shifts, during the periods from November 2013 to February 2014 and April 2014 to May 2014.

The unemployed workers who participated in this research were approached while they were waiting to be served at the SINE agency. The researchers presented the objective of the research and the ethical aspects involved in their participation and then, with the consent of the worker, they applied the instrument on the attribution of causes. The total time for application of the questionnaire ranged between 30 and 60 min. In the case of respondents who had a lower educational level, we opted to base the application on a strategy of presenting each item with cards of different colors and shades to differentiate the intensity of the response, from the brightest (not at all related to unemployment) to the darkest (extremely related to unemployment).

 

Data analysis procedures. First, we examined the database to identify incomplete questionnaires. We found that there were very few of these (less than 1%), and we decided to exclude them (listwise). There were no univariate extreme cases (based on z-scores). However, we excluded eight multivariable extreme cases (based on the Mahalanobis distance, p<0.01). The asymmetry indices of the variables ranged from |0.02| to |0.98| and had a kurtosis of |0.02| to |1.30|.

Given the two objectives of this study, we submitted the data to two types of analysis. The first was a confirmatory factor analysis (CFA). Considering that the variables were measured at ordinal levels (four classes), and to minimize the effects of the distribution of the data (for example, a possible nonnormal distribution), we used the robust maximum likelihood extraction method, with Satorra-Bentler correction. We conducted the analysis with a polychoric correlation matrix. The indices we used to assess the goodness of fit were the difference in the chi-square for degrees of freedom (χ2/gl), whose expected value is less than 2 (Arbuckle, 2008), and the comparative fit index (CFI) and Tucker-Lewis index (TLI), whose values are both expected to be greater than or equal to 0.90 (Bollen, 1989; Byrne, 2010; Kline, 2011). We also used the root mean square error of approximation (RMSEA), whose expected value is less than 0.06 (Brown, 2006), and we assessed the degree of reliability of the factors using Cronbach’s alpha. Finally, we used LISREL software (v. 8.80) for the analysis.

After validation of the model, we conducted the second analysis, related to the second purpose of this article: we took the three belief factors concerning the causes for unemployment as dependent variables and inserted them into a regression model to determine the effect of the sociodemographic (independent) variables on these factors. First, we reclassified the initial six levels of education in order to facilitate the analysis and also in view of theoretical considerations that we address in the discussion section. Then we converted schooling (which is categorical) into three dummy variables. Because of the exploratory nature of our investigation of the relationship between these variables and the attributions of causality, as described in the introduction, we carried out an inspection of the correlation matrix. Only sociodemographic variables that correlated significantly (p<0.05) with some factor of causality attribution were included in the model. We conducted the regression analysis (linear multivariate) itself with SPSS software (v. 20), using the stepwise (exploratory) method.

 

Results

Initially, the CFA resulted in a model that was poorly fitted to the data. Using the 19 original questions, provided to saturate the three factors, we obtained the following indicators: χ2(149)=490.53, p<0.001; χ2/gl=3.29; CFI=0.84, TLI=0.84; RMSEA=0.07 (IC of 90%=0.07, 0.08), P[RMSEA≤0.05]=0.001. Based on this initial model, we proceeded to respecify some parameters.

First, we retained in the model only those items whose factorial weight was greater than or equal to 0.30. Based on this criterion, three items, all of which were causal factors of unemployment, were excluded: “Inability of unemployed people to adapt to new conditions” (individualistic cause), “Production facilities and enterprises have been relocated to other areas or even to other countries” (societal cause), and “The introduction of widespread automation” (fatalistic cause).

Then we considered suggestions for modification (modification indices) that were given by the software itself – taking into account, however, their consistency with the theoretical basis of reference. We therefore decided to allow the errors of two items to co-vary with each other, both factors being societal causes: “The educational system does not correspond to the current job market” and “Unemployed people do not qualify for contemporary market needs”, as the semantic formulations for these two items are close to each other (at least in the Portuguese version of the questionnaire).

As a result of these respecifications, the new model, consisting of 16 items, presented better fit indices: χ2(100)=261.53, p<0.001; χ2/df=2.61; CFI=0.91, TLI=0.90; RMSEA=0.06 (90% CI=0.05, 0.07), P[RMSEA≤0.05]=0.004. Concerning the reliability of the three factors, Cronbach’s alpha was 0.68 for individualistic causes, 0.61 for societal causes, and 0.55 for fatalistic causes. These rates are slightly lower than those found in the previous study in Brazil using the same instrument (Gondim et al., 2010). The weight factor for each item is shown in Figure 2, as well as the correlations between factors (all significant, p<0.001).

 

 

Turning to the results of the regression analysis, inspection of the correlation matrix composed of the sociodemographic variables reveals that only two variables were correlated (although weakly) with two factors of unemployment causes: individualistic causes and income (r=0.14, p<0.01), societal causes and income (r=0.12, p<0.01), and societal causes and education (incomplete higher education, r=-0.16, p<0.01; complete secondary education, r=0.13, p<0.01; incomplete secondary education, r=0.08, p<0.04). We then included these variables in the regression models, leading to the indicators shown in Table 1.

 

 

As can be seen in Table 1, the predictive power of only two of the sociodemographic variables used in the analysis is significant, albeit very low – they account for only 2% of the observed variability in the individualistic causes factor and 3.7% of the variability in the societal causes factor (Model 2: F(1,358)=3.88, p<0.05). Note that an increase in income slightly increases the attribution of individualistic causes (β=0.15, p<0.01) and societal causes (β=0.10, p<0.001). Turning to educational level, we noted a negative effect on the scores of societal causes when going from participants with a basic education (baseline group) to those with an incomplete higher education (β=-0.15, p<0.01). That is, participants with a higher level of schooling (in this case, incomplete higher education) tended to ascribe lower scores to societal causes for unemployment.

Finally, Table 2 shows the means of each of the three factors of attribution of causality for unemployment and their differences. As can be seen, the social causes factor exhibits significantly higher scores than the individualistic and fatalistic causes for unemployment.

 

 

 

Discussion

The objective of this paper was twofold: to confirm the structure of Furnham’s model of attribution of causes for unemployment and to investigate the predictive role of sociodemographic variables for the components of this model.

With regard to the first objective, although the model was confirmed as a whole (structural confirmation, since we found the same number of factors and the same pattern of saturation of the items in the factors), some aspects of our findings deserve to be mentioned. Firstly, not all items were confirmed. Only 16 out of the 19 items of the original version were retained based on the weight factors for the items (λ~0.30). This suggests that the deleted items may not be good representatives of latent variables for the Brazilian sample.

Secondly, the introduction of a trajectory between two items that are a part of the societal cause factor suggests that there may be a problem of semantic overlap between these (the items are supposed to evaluate different facets of the latent variable, but the proximity of the writing leads to a similar interpretation).

Thirdly, inspection of the model residues revealed that several other items (at least three with an high impact on respecification of the model) have their errors correlated, which suggests that a fraction of the behavior of these items that is not explained by the respective factors can be explained by other factors not provided for in the model. This strengthens the argument that unemployment has multiple causes, and this should be considered in addition to the three dimensions contained in the Furnham (1982) model. In their cross-cultural research, Álvaro-Estramiana et al. (2012) compared Brazilian and Spanish samples and found that Brazilians attribute unemployment to multiple causes, unlike Spaniards, who follow tendencies similar to samples from other countries (see also Mylonas et al., 2010, 2014).

Also in this connection, looking at the scale items that were excluded in this study, the possible influence of context can be perceived. Two of these items (“Production facilities and enterprises have been relocated to other areas or even to other countries” and “The introduction of widespread automation”) refer to elements that are not found in the local reality of the current sample of participants. Unlike the reality for which this instrument was originally designed and implemented (European countries), the context of insertion for the participants in the present study is newly industrialized and lies primarily in the services sector.

Concerning the second objective, in which we tested the weight of sociodemographic variables in the attributions of causes for unemployment, the results suggest that there are no differences in the variance of responses to attributions in terms of age or gender. In the sample for this study, the only connections that were found were between education and income and attributions of cause. These results seem to corroborate Wooten and Valentin’s (2008) findings that the impact of certain sociodemographic variables on causal attributions for unemployment is spurious – for example, the gender variable that has been extensively studied in the literature on unemployment (e.g., Burker, 1985; Feather & O’Brien, 1986; Leana & Feldman, 1991).

The positive association found between attributions of individualistic causes and income may be explained by attribution theory from social psychology (Heider, 1958; Kelley, 1973; Weiner, 1986). This theory suggests that a person who is successful tends to ascribe merit for this to him- or herself, but an unsuccessful person will ascribe responsibility for failure to contextual factors, which strengthens the assumption that one’s success depends on oneself while failure (unemployment) depends on a lack of effort and dedication on the part of others. This reinforces the individualistic culture that believes that unemployment can be combated through entrepreneurial attitudes. This way of thinking has been linked to a stratum of the population in ascendancy, who tend to adhere to liberal values with a more individualistic emphasis (Estanque, 2003; Pochmann, 2012), as well as conservatives (Feather, 1985).

An understanding of the predictive behavior of the educational level variable must take into account the characteristics of the sample used and the method employed. In the regression models we tested, the only portions of the sample whose behavior influenced the dependent variable (in this case, the societal causes factor) were participants with a basic education and those with an incomplete higher education (the latter comprised about 50% of the sample).

We have no evidence to believe that this finding can be generalized. For this reason, further studies should test the relationship between different schooling levels and attributions of causality. One question we can formulate based on our finding concerning the role of schooling in the attributional styles is: Why didn’t the intermediate level of schooling (in our case, secondary education) affect the mechanisms of causal attributions? Could this result be related to our sample characteristics? In any case, it would be interesting, for example, to control new samples in terms of schooling (there is an imbalance in the present sample since 50% of the participants were at the incomplete higher education level) to investigate the assumption that the transition between different levels of schooling could indeed influence the breadth of personal backgrounds and consequently affect the patterns of attribution of causes of unemployment.

A finding that is worth further exploration using different set of samples is that societal causes for unemployment had a comparatively higher overall score (M=2.90, SD=0.58; Table 2) than did individualistic and fatalistic causes, and this difference was statistically significant in relation to the other averages (see Table 2). A plausible explanation is that the government is viewed as responsible for the unemployment situation, since it controls the educational system. Another possible explanation is that the participants in some way understand that unemployment is a temporary situation that can be surmounted provided that conditions for better training are met. This result may be associated with the belief in the importance of training, as advocated by the theory of human capital, according to which the level of training or qualification for work has a positive impact on the achievement of better jobs, wages, and employability (Saul, 2004; Vatin, 1999). However, the theory of human capital has recently been criticized (e.g., Adamson, 2009; Bröckling, 2011; Frigotto, 2010, 2011) and has also been challenged by what Chroudhly et al. (2013) noted concerning the loss of human capital that is associated with high rates of unemployment, especially among young people, who supposedly are more informed and ready for the demands of work.

In conclusion, it is worth pointing out that the results presented in this paper confirm that an instrument for measuring attributions of causality based on the separation of causes into individualistic, fatalistic, and societal is problematic – at least with respect to the Brazilian reality. Additionally, and this may have something to do with the previous point, there was no clear evidence of the role of sociodemographic variables in understanding how people attribute causes for unemployment.

The results of this study have generated several issues that deserve further study. For example, they indicate that the influence of schooling on the assignment of causes for unemployment is not evident. However, as mentioned above, this study’s understanding of the role of educational level could have been hampered by the sample characteristics and methodological decisions adopted. Concerning the former, the sample is composed mainly of young people (55% below the age of 29 years), and is characterized by short periods of unemployment (88% were unemployed for two months or less). Even the relationships found between income and individualistic and societal causal attributions needs some qualification, because the income variability within the sample is small, considering that 86% of the participants earned no more than twice the minimum monthly wage. Again, this limitation indicates the need for further research on this important issue.

 

References

Álvaro-Estramiana, J. L., Gondim, S. M. G. G., Garrido-Duque, A., Luna, A. F., & Dessen, M. (2012). Desempleo y bienestar psicológico en Brasil y España: Un estudio comparativo. Revista Psicologia: Organizações e Trabalho, 12, 1-12.         [ Links ]

Adamson, M. (2009). The human capital strategy. Ephemera: Theory & Politics in Organization, 9, 271-284.         [ Links ]

Arbuckle, J. L. (2008). AMOS 17 users’ guide. Chicago: SPSS.

Blustein, D. L., Medvide, M. B., & Wan, C. M. (2012). A critical perspective of contemporary unemployment policy and practices. Journal of Career Development, 39, 341-356.         [ Links ]

Bollen, K. A. (1989). Structural equations with latent variables. New York: John Willey.         [ Links ]

Boyce, C. J., Wood, A. M., Daly, M., & Sedikides, C. (in press). Personality change following unemployment. Journal of Applied Psychology.         [ Links ]

Bröckling, U. (2011). Human economy, human capital. A critique of biopolitical economy. In U. Bröckling, S. Krasmann, & T. Lemke (Orgs.), Governmentality: Current issues and future challenges (pp. 247-268). New York: Routledge.         [ Links ]

Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York: The Guilford Press.         [ Links ]

Burke, R. (1985). Comparison of experiences of men and women following a plant shutdown. Psychogical Reports, 57, 59-66.         [ Links ]

Byrne, B. M. (2010). Structural equation modeling with AMOS. New York: Routledge.         [ Links ]

Campbell, C. R., & Henry, J. W. (1999). Gender differences in self-attributions: Relationship of gender to attributional consistency, style, and expectations for performance in a college course. Sex Roles, 41, 95-104.         [ Links ]

Choudhry, M., Marelli, E., & Signorelli, M. (2013). Youth and total unemployment rate: The impact of policies and institutions. Rivista Internazionale di Scienze Sociali, 121, 63-86.         [ Links ]

Crittiden, K. S. (1983). Sociological aspects of attribution. Annual Review of Sociology, 9, 425-446.         [ Links ]

Departamento Intersindical de Estatística e Estudos Socioeconômicos – DIEESE. (2006). Pesquisa de emprego e desemprego: Mercado de trabalho metropolitano. São Paulo, SP: DIEESE.

Estanque, E. (2003). O efeito classe média: Desigualdades e oportunidades no limiar do século XXI. In M. V. Cabral, J. Vala, & A. Freire (Orgs.), Atitudes sociais dos portugueses: Desigualdades sociais e percepções de justiça (pp. 69-105). Lisboa: Imprensa de Ciências Sociais.         [ Links ]

Feather, N. T. (1985). Attitudes, values, and attributions: Explanations of unemployment. Journal of Personality and Social Psychology, 48, 876-889.         [ Links ]

Feather, N. T., & O’Brien, G. E. (1986). A longitudinal analysis of the effects of different patterns of employment and unemployment on school leavers. British Journal of Psychology, 77, 459-479.

Frigotto, G. (2010). A produtividade da escola improdutiva (9ª ed.). São Paulo: Cortez.         [ Links ]

Frigotto, G. (2011). Novos fetiches da pseudoteoria do capital humano no contexto do capitalismo tardio. In J. De Andrade & L. G. De Paiva (Orgs.), As políticas públicas para a educação no Brasil contemporâneo (pp. 18-35). Juiz de Fora: Editora da Universidade Federal de Juiz de Fora.         [ Links ]

Furnham, A. (1982). Explanations for unemployment in Britain. European Journal of Social Psychology, 12, 335-352.         [ Links ]

Furnham, A. (1988). Lay theories: Everyday understanding of problems in the social sciences. Oxford: Pergamon Press.         [ Links ]

Furnham, A., & Rawles, R. (1996). Job search strategies, attitudes to school and attribution about unemployment. Journal of Adolescence, 19, 355-369.         [ Links ]

Gondim, S. M. G., Álvaro, J. L., Luna, A. F., Oliveira, T., & Souza, G. C. (2010). Atribuições de causas do desemprego e valores pessoais. Estudos de Psicologia (UFRN), 15, 309-317.         [ Links ]

Guimarães, N., Demazière, D., & Sugita, K. (2010). O desemprego como experiência biográfica: São Paulo, Paris, Tóquio. In N. Guimarães (Org.), Desemprego, uma construção social: São Paulo, Paris e Tóquio (pp. 87-201). Belo Horizonte: Argvmentvm.         [ Links ]

Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley and Sons.         [ Links ]

Holmqvist, M. (2009). Medicalization of unemployment: Individualizing social issues as personal problems in the Swedish welfare state. Work Employment & Society, 23, 405-421.         [ Links ]

International Labour Organization (ILO). (2013). Tendencias mundiales del empleo 2013. Resumen Ejecutivo. Disponível em http://www.oit.org.br/sites/default/files/topic/gender/doc/tendenciasmundiaisdempregoresumoespanhol_974.pdf         [ Links ]

Jahoda, M. (1979). The impact of unemployment in the 1930’s and the 1970’s. Bulletin of British Psychological Society, 32, 309-314.

Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process in person perception. Advances in Experimental Social Psychology, 2, 219-266.         [ Links ]

Kapuvári, V. (2011). Psychological effects of economic recession and unemployment. European Journal of Mental Health, 6, 83-93.         [ Links ]

Kelley, H. H. (1973). The processes of causal attribution process. American Psychologist, 28, 107-128.         [ Links ]

Kline, R. B. (2011). Principles and practice of structural equation modelling. New York: The Guilford Press.         [ Links ]

Leana, C. R., & Feldman, D. C. (1991). Gender differences in responses to unemployment. Journal of Vocational Behavior, 38, 65-77.         [ Links ]

Law nº 8.019, de 11 de abril de 1990. Altera a legislação do Fundo de Amparo ao Trabalhador (FAT), e dá outras providências. Brasília. Disponível em http://www.planalto.gov.br/ccivil_03/leis/L8019.htm

Martinko, M. J., Douglas, S. C., & Harvey, P. (2006). Attribution theory in industrial and organizational psychology: A review. International Review of Industrial and Organizational Psychology, 21, 127-188.         [ Links ]

Moyano-Díaz, E., Gutiérrez Peña, D., Zúñiga Cabrera, K., & Cornejo, F. A. (2013). Employees and unemployed, causal attributions and subjective well-being. Psicologia & Sociedade, 25, 440-450.         [ Links ]

Mylonas, K. (2007). Psychometric adaptation of the Furnham “Reasons of Unemployment” questionnaire for the Greek population. Paper presented at the 9th European Conference on Psychological Assessment, Thessaloniki, Greece.

Mylonas, K., Furnham, A., & country collaborators. (2014). Bias in terms of culture and a method for reducing it: An eight-country explanations of unemployment scale study. Educational and Psychological Measurement, 74, 77-96.         [ Links ]

Mylonas, K., Furnham, A., Konstantinidis, E., Papazoglou, S., Divale, W., Leblebici, C., . . . & Boski, P. (2010). The explanations of unemployment scale: An eight-country study on factor equivalence.         [ Links ] Comunicação apresentada no XX Congress of the International Association for Cross Cultural Psychology, Melbourne.

Pochmann, M. (2012). Nova classe média?. O trabalho na base da pirâmide social brasileira. São Paulo: Boitempo.         [ Links ]

Saul, R. P. (2004). As raízes renegadas da teoria do capital humano. Sociologias, 6, 230-273.         [ Links ]

Vatin, F. (1999). Le travail, sciences et société. Bruxelles: Ed. de l’Université de Bruxelles.

Weiner, B. (1986). An attribution theory of motivation and emotion. New York: Spring-Verlag.         [ Links ]

Wooten, K. C., & Valenti, A. M. (2008). The questionable impact of gender on job loss reactions: Implications for employment counseling. Journal of Employment Counseling, 45, 146-155.         [ Links ]

 

CORRESPONDÊNCIA

A correspondência relativa a este artigo deverá ser enviada para: Pedro F. Bendassolli, Departamento de Psicologia, Universidade Federal do Rio Grande do Norte - Campus Universitário Central, Caixa Postal 1524, Campus Universitário Lagoa Nova, CEP 59078-900, Natal/RN, Brasil. E-mail: pbendassolli@gmail.com

 

Agradecemos o apoio da Secretaria do Estado de Trabalho, Habitação e Assistência Social do Rio Grande do Norte.

 

Submissão: 01/07/2014 Aceitação: 24/02/2015

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License