The role of income, life-style deprivation and financial strain in mediating the impact of unemployment on psychological distress: evidence from the Republic of Ireland.
Available evidence suggests that income affects psychological well-being indirectly via subjectively appraised financial strain.
This paper presents an analysis of the relationships between measures of household income and resources, subjectively appraised financial strain and psychological distress drawing on a nationally representative sample of 3294 households in the Republic of Ireland.
Our analysis demonstrates that objectively assessed exclusion from customary life-styles, involving deprivation of socially defined necessities, is associated with increased psychological distress.
A compelling case can be made that, with hindsight, the most striking feature of the current literature on the consequences of unemployment will be seen to be its remarkable lack of emphasis on poverty. No originality is claimed in making this observation. Kelvin & Jarret (1985) note that, while there is widespread recognition that unemployment brings both economic and psychological problems, consideration of the relationship between the problems is remarkably rare. Similarly Fryer (1988, 1990), in a number of recent publications, has drawn attention to the fact that, while financial hardship is repeatedly mentioned in the literature, it is not allocated a central role.
The neglect of poverty is made more surprising by the fact that the seminal 1930s field study of Marienthal, an Austrian village stricken by mass unemployment, showed that there was a clear connection between a family's attitude and its economic situation (Jahoda, Lazarsfeld & Zeisel, 1933/1972). Subsequently, however, the classic studies were interpreted in a particularly one-sided fashion. Jahoda's (1982) own subsequent explanatory emphasis was on the latent, non-economic, consequences of unemployment; involving the loss of social contact, activity, status, purposefulness and time structure. Others followed Jahoda in making psychological, rather than material deprivation theoretically central (Warr, 1987). As a consequence, despite the evidence that unemployment in the 1980s still involved the manifest consequence of a substantial reduction in income, sight has been lost of the impact of declining resources on agency and coping capacity (Fryer, 1986; Fryer & Payne, 1986; Heady & Smith, 1989).
One factor contributing to this partiality has been the tendency to view unemployment from a life-event perspective and to label it as an acute stressor. More recently, however, a number of authors have emphasized the need to take into account the process through which events adversely restructure social and economic conditions of life (Mirowsky & Ross, 1989; Pearlin, 1990). Chronic stress arises from the dogged -- slow to change -- problems of daily life. The two types of stress converge when life-events, such as unemployment, lead to hardship in basic enduring economic and social circumstances (Pearlin et al., 1981; Ross & Huber, 1985).
The evidence available, from the limited number of empirical studies that have used multivariate analysis, when assessing the association between income and psychological health, indicates that the expected pattern of association between financial strain and psychological health during unemployment is found when using subjective measures of the former, but not when employing objective measures (Ullah, 1990).
A second distinction is that between direct and indirect effects of income. Some explanations, Ullah (1990) notes, have emphasized the indirect effect of income on psychological health through its impact on social and leisure activities. Others, he suggests, point to the possibility of direct effects on aspects of well-being, such as one's sense of value and status, although none offer an explanation of how income might be directly related to mental health. Whatever the merit of the statistical distinction between the direct and indirect impact of income, for neither psychologists nor sociologists can income per se provide an adequate explanation of psychological health in the absence of an understanding of the psychological processes involved, and their social context.
Even those studies which have examined the impact of economic hardship, in general, have failed to draw on the wider psychological and sociological literature relating to poverty or relative deprivation. As a consequence, there exists a great deal of confusion regarding the manner in which the relationships between unemployment, income, and financial hardship ought to be conceptualized. We have substantial evidence available to us from other areas of research, such as job satisfaction and equity evaluation, that attempts to explain complex subjective phenomena on the basis of variables such as income without attention to the interpretative processes involved provide a very poor return on effort expended (Adams, 1963; Goldthorpe, Lockwood, Bechhofer & Platt, 1968; Locke, 1969). In the absence of an understanding of how objective deprivation is related to perceptions of financial strain, it is impossible to be sure that the direction of causality, between the latter and mental health, is not the opposite to that hypothesized.
Income, life-style deprivation and the appropriate unit of analysis
In this study we operate with a relative, rather than an absolute, definition of poverty. The most influential formulation of such a concept is in terms of exclusion from ordinary living patterns, customs and activities arising from a lack of resources (Townsend, 1979). Such a perspective directs attention to the household or family unit level of analysis. In terms of income, it suggests that we should follow the practice of focusing on household disposable income, suitably adjusted for variations in household size and composition.
Even where income is measured with as much precision as possible, and at the appropriate level, it may still be argued that it does not provide an adequate measure of poverty, defined as low consumption; some of those not on low income suffer deprivation in consumption and far from all members of low-income groups experience such deprivation (Callan, Nolan & Whelan, 1993; Ringen, 1987, 1988). In discussions of the criteria which should be applied to items making up a deprivation index, there are two recurring themes. The first is necessity. Mack & Lansley (1985, p. 39) tackle this issue by explicitly taking into account the prescriptions of the community. They thus define poverty in terms of enforced lack of socially perceived necessities.
The second theme, although frequently less explicit, is perhaps even more important. It centres on the relationship of specific deprivations to other aspects of deprivation and, by implication, to resources. Mack & Lansley (1985, p. 41), while terming the enforced lack of any socially perceived necessity a deprivation, conclude that such deprivation will be termed poverty only when it affects a person's way of life. They assume that 'poverty is a situation where such deprivation has a multiple impact on household's way of life' (Mack & Lansley, 1985, p. 171). Similarly, Ringen (1987, p. 161) takes the criterion of exclusion from one's society to involve a standard of living which is characterized as a state of general deprivation.
The foregoing brings out the need for a systematic analysis of the dimensions of life-style deprivation. Such analysis has been almost entirely absent in the poverty literature. In our subsequent analysis we will attempt to distinguish theoretically relevant dimensions of deprivation. Our first step will involve identifying a set of items or activities, widely regarded as necessities, the enforced absence of which accords with our notion of poverty. Such a set of items we will term primary deprivation. We would wish to distinguish this from enforced absence of possessions/activities which are not socially sanctioned as necessities, which we might describe as secondary deprivation. In addition, it is reasonable to expect that items might also be distinguished from each other by the degree to which their possession is related to current resources. In particular, the range of housing items which are included in this study are such that their relationship to current income is likely to be modest because of the influence of public housing, stage of the life-cycle, and the likelihood that some of the households, which are currently experiencing poverty, may have acquired these items some time previously.
Our theoretical approach implies that, while our objective is to explain the psychological distress of individuals, income and life-style deprivation are measured at the household level. This approach has the substantial advantages that it allows us to build on the insights provided by the classic studies into the relationship between economic deterioration and family experience. This is particularly important because, as McGhee & Fryer (1989, p. 239) stress, while psychologists have made considerable progress in describing the health consequences of unemployment, this has not been matched by advances which locate the experience of unemployment in the broader context of the family.
In this study we seek to examine the relationship between income, life-style deprivation, perceived financial strain and psychological distress. A particular feature of the study is our desire to establish the consequences of unemployment not only for the individual involved but also for their spouses.
The Survey of Poverty, Income Distribution and Usage of State Services carried out by the Economic and Social Research Institute, Dublin, in 1987, provides the database for our analysis. A detailed description of this survey is provided in Callan, Hannan, Nolan, Whelan & Creighton (1989). The survey was designed to provide a national sample from the population of the Republic of Ireland resident in private households. Thus those living in institutions -- hospitals, hotels, prisons, etc. -- did not form part of the population. The sampling frame was the Register of Electors. Sampling was performed using the RANSAM programme developed at the ESRI (Whelan, 1979). This implements a multi-stage random sample incorporating both stratification and clustering and giving each individual on the Register an equal probability of being selected.
The response rate was 64.3 per cent of households. This figure can be compared with that for the Household Budget Survey in 1980 of 56 per cent. Non-response introduces bias into the resulting sample only if it is non-random. A reweighting scheme, to correct for identified biases in the selection of households, was developed based on the 1986 Labour Force Survey. The reweighting was implemented on the basis of (i) household location; (ii) occupation of the head of household; (iii) number of adults in the household; and (iv) age of the household head. Since information was collected at both household and individual levels, a second-stage reweighting was undertaken at the individual level to allow for non-response within households.
The survey which involved fully structured interviews was conducted by personal visits by members of the ESRI's panel of trained interviewers. A wide range of information was gathered through a number of questionnaires, including the following:
(i) For each household one questionnaire covered household characteristics and composition;
(ii) For each individual aged 15 or over and not in full-time education an individual questionnaire dealt with the respondent's income, labour force status, level of psychological distress, etc.;
(iii) Where a full questionnaire could not be completed an abbreviated one, focusing on key information required for estimation of household income, and for reweighting purposes, was administered.
The full range of individual information is available for 6764 individuals. In Table 1 we set out the distribution of the sample across key socio-demographic characteristics. At various points in the analysis we focus on different levels and subsections of the sample. Thus in constructing measures of income and life-style deprivation we operate at the household level. In assessing the impact of unemployment on psychological distress we draw on the individual level data, and restrict our attention to those in the labour force who number 3833.
Psychological distress. Psychological well-being was measured using the 12-item version of the General Health Questionnaire (Goldberg, 1972, 1978) and the GHQ scoring procedure. In order to make it possible for the GHQ to be administered by the interviewers it was necessary to introduce some changes to the combinations of items and answer formats. The procedure adopted was intended to avoid grouping of 'positive' or 'negative' items or the need for repeated changes of response format. The approach taken was to divide the items into two groups of six, each of which was allocated to one of the two possible response formats. The alpha coefficient for the 12-item scale was found to be .82.
One possibility that we need to consider is whether the changes that we have made in the item-response format combinations may have influenced the distribution of GHQ score. The evidence provided by Goldberg (1972) indicates that the level of 'pathological' response is marginally higher for the items where the format was changed. Our own results show the unchanged items are marginally more likely to provide higher numbers in the 'pathological' categories. However, the differences are extremely slight and highly unlikely to be of substantive significance. The split-half correlation coefficient between the two sets of items is .73 (Whelan, Hannon & Creighton, 1991).
Table 1. Distribution of the sample of individuals who completed full questionnaires by selected sociodemographic characteristics Sex Labour force status Marital status Male 48.6 Employees 36.4 Married 64.0 Female 51.4 Self-employed 10.0 Single 25.7 Unemployed 10.3 Other 0.3 Ill/disabled 3.1 Retired 9.0 In home duties 31.3
Unemployment. The concept of unemployment adopted in this study, like that in the EC Labour Force Survey, is dependent upon the respondents' evaluation of their own employment status. This contrasts with the Live Register definition which is primarily determined by the criteria relating to entitlement to financial provision. The former approach may include as unemployed those seeking their first job and women in home duties seeking work. It may exclude such registrants who opt for the categories of 'home duties' or 'retired' when asked to describe their current status.
Income. Detailed information was collected on income for each member of the household on a variety of income sources. This information was combined to arrive at an estimate of household income. Following the standard practice of the UK Family Expenditure Survey the focus was on current weekly disposable income, since this is most directly relevant to the command over resources which determines a family's standard of living.
In comparing income levels across households, some adjustment is required to take differences in needs, arising from variation in household size and composition, into account. This is done by the application of adult equivalence scales designed to convert each household to a common basis. In this case each additional adult in a household has been given a weight of .67 and each child a weight of .33.
Although the information available to us allows for income to be treated as a continuous variable, we have chosen to operate in our analysis with an aggregated decile measure. The reason for this is that a number of major surveys of poverty have found extremely weak correlations between income measured continuously and individual life-style items. In the ESRI survey the average correlation was .11 which is almost identical to that reported by Townsend (1979) and Mack & Lansley (1985). These findings suggest that the correlations are being attenuated by measurement error. One simple, but surprisingly effective solution, is to divide the data into intervals. When income is grouped by decile the average correlation is doubled. This approach is actually an instance of the instrumental variable method developed by Bartlett (1949) and could be regarded as a variation on more recent applications of this approach (Fairbrother, 1991).
Our final measure of income is, therefore, household adult equivalent disposable income aggregated to decile level.
Life-style deprivation. The choice of life-style deprivation items to be included in the study was influenced by the range of indicators employed in other major studies of poverty (Mack & Lansley, 1985; Townsend, 1979). The 24 items on which our analysis is based, which are set out in full in the Appendix, are made up of 17 of the Mack & Lansley (1985) pool of items together with seven additional items.
Financial strain. In attempting to measure perceptions of financial strain, 'heads of households' and 'household managers' were asked, taking into consideration the household's total income, if the household was able to 'make ends meet'. The choice of head of household was determined by the respondents. The household manager was defined as 'the person who buys most of the groceries'. Influenced by results regarding the relationship between life-style deprivation and psychological distress which will be presented later, attention was concentrated on the contrast between those respondents experiencing a 'great difficulty' and all other items. Individuals were allocated to the former category if either the head of household or the household manager responded in this fashion.
As we have already noted, previous work assumes a single dimension of deprivation. On the basis of the distinctions we have made earlier between necessities and non-necessities, and variation in the strength of the association between current resources and life-style items, we hypothesized the three dimensions of deprivation outlined earlier, i.e. primary, secondary and housing deprivation.
In order to pursue this hypothesis we make use of factor analysis. Since there are a number of problems involved in the application of conventional factor analysis procedure to dichotomous items we have made use of Muthen's (1978) Generalized Least Squares procedures as incorporated in the computer program Liscomp (Mislevy, 1986; Muthen, 1988).
For the 20 items for which we have data on absence and enforced absence we have concentrated exclusively on the latter. Together with the additional four items, they were entered into a factor analysis specifying a three-factor solution. The results of this analysis are set out in Table 2 and provide an unambiguous picture.
TABULAR DATA OMITTED
The items which load on the first factor we have labelled primary lifestyle deprivation. In order to develop a measure of primary deprivation, which can plausibly be interpreted as the life-style component of poverty, we have avoided including items in the index which, while having their highest loading on this factor, have loadings on other factors which are not a great deal lower. The remaining items refer to lack of basic food, clothing, heating, etc. and the experience of persistent debt difficulties arising from dealing with routine expenses. The consumption items all relate to current consumption. The items making up the dimension have relatively low levels of non-possession and high levels of socially defined necessity; the scale score runs from 'zero' to 'eight' and 32 per cent of respondents reside in households where one, or more, of these deprivations is experienced.
The second dimension -- secondary life-style deprivation -- includes items which refer to exclusion from, basically, middle-class, or comfortable working-class, life-style patterns, including holidays, leisure activities and consumer durables with significant current expenditure costs associated with them. Here we find much higher levels of non-possession and lower levels of socially defined necessity. The scale runs from 'zero' to 'nine', and three in four households suffer an enforced lack of at least one item.
The final factor relates to the housing deprivation factor. This factor includes characteristics relating to housing quality and durable household goods. The scale score runs from 'zero' to 'seven', and one in eight households suffer an enforced lack of one, or more, items. The alpha reliability coefficients for the scales are as follows:
(i) primary deprivation .70
(ii) secondary deprivation .76
(iii) housing and household capital deprivation .70
If one fails to specify the number of factors expected, the distinctions between these dimensions remain clear. However, further fragmentation occurs within each set of items culminating in a seven-factor solution. The reliability coefficients, for scales based on such a solution, would necessarily be lower.
Table 3 shows a correlation matrix of key variables. Unemployment is significantly correlated with GHQ score (r = .32). The contrast here is between those unemployed and those at work. The income measures display a correlation of -.21 with the GHQ score. Both primary and secondary life-style deprivation and financial strain have slightly higher correlations. Unemployment is significantly associated with income, each of the dimensions of life-style deprivation, and financial strain. The correlation range from -.42 in the case of income to .11 in the case of housing deprivation. Before turning our attention to the manner in which these variables mediate the impact of unemployment on psychological distress it will be useful to examine the degree of intercorrelation between these variables.
Income has its highest correlation with secondary deprivation (r = .49) and the lowest with housing deprivation (r= -.22), while primary deprivation occupies an intermediate position (r = -.39). This pattern of results seems reasonable on theoretical grounds. Housing and household capital items are accumulated over a period of time, and we might expect life-cycle and location factors to be at least as important as income. With regard to primary deprivation, precisely because of the extremes of deprivation being tapped, we would expect not only that people would draw on savings, or other accumulated resources, to provide such items but also that they would be extremely likely to make use of available sources of social support and, indeed, to do so with some measure of success.
Table 3. Correlation matrix of key variables for individual in the labour force 1 2 3 4 5 6 1 GHQ 2 Unemployment .32 3 Income -.21 -.42 4 Primary deprivation .29 .35 -.39 5 Secondary deprivation .23 .34 -.49 .53 6 Housing deprivation .08 .11 -.22 .31 .29 7 Financial strain .24 .30 -.43 .45 .49 .18 * All correlations are significant at least at the .01 level.
Despite the qualifications it is necessary to make regarding the use of income as a variable, our measure of household equivalent income does display a much stronger relationship to financial strain (-.43) than that reported by Ullah (1990) using individual income data (-.24). Both primary and secondary deprivation actually have marginally higher correlations. The multiple correlation of financial strain with income and life-style deprivation is .57. This finding helps to account for the fact that, unlike other studies, our measures of objective deprivation, including income, are as strongly related to GHQ as is financial strain. It would appear that significant advantages are gained by measuring resources and deprivation at household level.
While all of the life-style deprivation dimensions are related to financial strain they are, in turn, strongly related to each other. Before proceeding to a multivariate analysis, aimed at understanding how the impact of unemployment is mediated, it is necessary to decide how to treat such variables. It seems most sensible to think in terms of cumulative deprivation. The question arises, for instance, of whether secondary deprivation has any effect when it occurs in the absence of primary deprivation?
In order to pursue this line of inquiry, we start by assigning causal priority in their influence on psychological health to life-style deprivation as follows:
(i) primary life-style deprivation
(ii) secondary life-style deprivation
(iii) housing deprivation
Considerations of both parsimony and meaning dictate that primary deprivation should have priority. Secondary deprivation takes precedence over housing deprivation because of the evidence that the latter is much more weakly related to current income, although this finding may not generalize to societies with a different balance of public and private sector housing. In effect, the procedure we adopt is one of semi-partial correlation, where each of the life-style variables has its relationship to the causally prior dimensions taken into account. What remains is a residualized variable which is independent of all prior variables. When this procedure is adopted, housing deprivation ceases to have any significant impact on the other key variables, and it has been dropped from the remainder of the analysis.
To assess the independent contribution of our key variables on GHQ score, multiple regression analyses were performed with forced stepwise inclusion of predictor variables. In Table 4 the beta weights of a series of equations are shown. We start by assessing the impact of unemployment on GHQ score and proceed to assess the impact of inclusion of, in turn, (i) income; (ii) primary and secondary deprivation; and (iii) financial strain. In each of the equations we also control for the effects of physical illness, age, sex and marital status although to simplify the presentation we have not reported the coefficients.
TABULAR DATA OMITTED
The addition of income to the analysis in equation 2 reduces the beta weight for unemployment from .31 to .27. Entering the life-style deprivation dimensions, as is clear from equation 3, further reduces the unemployment coefficient to .23 and income becomes insignificant. This result is particularly important. Earlier it was stressed that the impact of income is necessarily indirect, in the sense that, ultimately, we must provide an account of the nature of the intervening psychological pressures. In fact, there is no need to go beyond the association between income and life-style deprivation in order to explain the original correlation between income and psychological distress. Primary deprivation, together with unemployment, is the most significant influence on GHQ score. Finally, in equation 4, with the introduction of the financial strain variable, the unemployment coefficient drops to .22, and the life-style coefficients are somewhat reduced, while the financial strain variable has a statistically significant effect. No evidence was found for the existence of any significant interaction between income/life-style deprivation and unemployment. Low income and enforced deprivation of basic commodities, where they occur, have adverse mental health consequences for those at work, just as much as for the unemployed. The latter, of course, must cope with additional problems.
Overall the results show that life-style deprivation and, in particular, primary deprivation play a significant role in mediating the impact of unemployment. The effect of income can be accounted for entirely by its relationship to life-style deprivation. Financial strain has an independent effect but its role in explaining psychological distress is modest, once we have taken the impact of objective deprivation into account. Finally, while the role of deprivation in mediating unemployment emerges unambiguously, it must be stressed that unemployment continues to have a substantial independent effect.
The independent role of unemployment in this case provides a striking contrast to the situation we find when we look at the impact of husbands' unemployment on the GHQ scores of married women. Our analysis is based on 1703 married women in our sample. The set of equations presented in Table 5 are identical to those presented in Table 4, except that it is now husband's unemployment on which we focus. The unemployment of a husband clearly has a significant effect on a woman's level of psychological distress, as the beta coefficient of .18 in equation 1 shows, although the effect is a good deal less than in the case of personal unemployment. Similarly, the introduction of income in equation 2 reduces the size of the unemployment coefficient to .12. As in the case of unemployment, the effect of income is entirely accounted for by its relationship to life-style deprivation, as can be seen from equation 3. What is particularly notable though is the finding contained in equation 4 that, for married women, their husbands' unemployment ceases to have an effect on their GHQ scores when we control for life-style deprivation. This result again illustrates the consequences of neglecting the role of poverty. For married women it is economic deprivation, associated with their husbands' unemployment, which is the crucial factor affecting their psychological health. However, deprivation is not an unusual phenomenon among such women. Close to two-thirds of women with unemployed husbands reside in households which suffer an enforced lack of at least one primary life-style item, compared to less than a quarter of women whose husbands are at work. Finally, financial strain once again has an independent effect even when we control for other factors.
TABULAR DATA OMITTED
The starting point of the paper was the neglect of the role of poverty in explaining the impact of unemployment on psychological health. The case was argued that the conclusion that subjective measures of financial strain, but not objective measures, are related to psychological health, does not provide a satisfactory basis for understanding the manner in which economic deprivation mediates the impact of unemployment. Furthermore, it was argued that understanding the impact of income requires that the process through which it operates is specified. When income is measured at a household level we do in fact find that income bears a clear relationship to GHQ scores. We also find that this effect operates indirectly through life-style deprivation. Financial strain continues to have an independent effect but the effect is modest enough not to detract from the significance of our findings regarding the impact of objective economic deprivation.
We have no desire to replace a one-sided emphasis on the latent functions of employment with claims for the exclusive importance of material deprivation. Unemployment continues to have a substantial and damaging effect on the psychological health of the unemployed individual even when we control for income, life-style deprivation and financial strain. Employment does indeed provide more than money.
Analysis at the household level, however, reinforces the view that poverty is a crucial mediating factor. A focus on the household helps to avoid the danger of removing the social context which makes the experience of individuals meaningful (Fryer, 1988). A study of households also allows us to pursue the frequently proffered suggestion that recognition should be paid to the impact of unemployment on the family rather than just the individual (McGhee & Fryer, 1989; McKee & Bell, 1985). While the risk of poverty which is associated with unemployment is one of the important ways in which job loss is translated in psychological distress, unemployment also involves exclusion from a range of experiences and associated psychological benefits, and exposure to the potentially stressful demands of the new role of being unemployed (Jahoda, 1982; Warr, 1987). For the wives of unemployed men the situation is rather different. While the husbands' altered role can clearly have implications for their pattern of activities, any alterations in their own roles are likely to be modest in comparison to those to which their husbands must accommodate. Our findings indicate that where a husband's unemployment does not lead to economic deprivation it does not appear to have any impact on the wife's level of psychological distress.
The result does not necessarily imply that a wife's response to her husband's unemployment takes an entirely economic form. However, whatever the emotional aspects of her response are, they do not seem to involve a heightened probability of psychological distress. Our finding with regard to the impact of economic deprivation is consistent with a body of research which has argued that male unemployment may carry a heavy managerial role for women who are forced to live on their wits (Pahl, 1980, 1983; McGhee & Fryer, 1989; McKee & Bell, 1986).
The results we have presented show the importance of going beyond income per se and taking a broader view of deprivation and resources. This is illustrated, with particular force, by the impact of primary deprivation on psychological health. It is deprivation, of this rather basic sort, which involves the enforced absence of socially defined necessities, such as food, clothes and heating, which has the most striking effect. Thus, acute and chronic stress converge to produce an impact on psychological health which stems from hardship in basic enduring economic circumstances, and the experience of what has been described as 'economic brinkmanship' (Pearlin et al., 1981); in which resources are persistently insufficient to permit the development of adequate coping strategies.
The findings, presented here, clearly demonstrate the role of poverty in mediating the impact of unemployment not only for the individuals involved but also for the members of their families. The original Marienthal conclusion, that the deterioration in mental health among members of households affected by unemployment is intimately related to the erosion of economic resources, appears to be no less true today.
I would like to acknowledge the debt I owe to all of my ESRI colleagues who participated in the Survey of Income Distribution, Poverty and Usage of State Services. More particularly, I am grateful to Damian Hannan and Sean Creighton for the significant contributions they made to the work on life-style and psychological distress reported in the paper. The paper also benefited from helpful comments from Denis Conniffe, David Fryer, Brian Nolan, Sean O'Riain and the referees.
Adams, J. (1963). Towards an understanding of inequity. Journal of Abnormal and Social Psychology, 67, 422-436.
Bartlett, M.S. (1949). Fitting a straight line when both variables are subject to error. Biometrika, 5, 207-212.
Callan, T., Hannan, D., Nolan, B., Whelan, B. & Creighton, S. (1989). Poverty and the social welfare system in Ireland. ESRI, Paper No. 146.
Callan, T., Nolan, B. & Whelan, C. T. (1993). Resources, deprivation and the measurement of poverty. Journal of Social Policy, 22.
Fairbrother, R. W. (1991). A general estimation procedure based on group means. Computational Statistics and Data Analysis, II, 252-264.
Fryer, D. (1986). Employment deprivation and personal agency during unemployment. Social Behaviour, 1, 3-23.
Fryer, D. (1988). The experience of unemployment in social context. In S. Fisher & J. Reason (eds), Handbook of Life Stress, Cognition and Health. Chichester: Wiley.
Fryer, D. (1990). The mental health costs of unemployment: Toward a social psychological concept of poverty. British Journal of Clinical and Social Psychiatry, 7(4), 164-175.
Fryer, D. & Payne, R. (1986). Being unemployed: A review of the literature on the psychological experience of unemployment. In C. L. Cooper & I. Robertson (Eds), International Review of Industrial and Organisational Psychology. Chichester: Wiley.
Goldberg, D. P. (1972). The Detection of Psychiatric Illness by Questionnaire. Oxford: Oxford University Press.
Goldberg, D. P. (1978). Manual for the General Health Questionnaire. Windsor, Berks: National Foundation for Educational Research.
Goldthorpe, J., Lockwood, D., Bechhofer, F. & Platt, J. (1986). The Affluent Worker: Industrial Attitudes and Behaviour. Cambridge: Cambridge University Press.
Heady, P. & Smith, M. (1989). Survey Standards during Unemployment, vol. I. London: HMSO.
Jahoda, M. (1982). Employment and Unemployment: A Socio-Psychological Analysis. Cambridge: Cambridge University Press.
Jahoda, M., Lazarsfeld, P.F. & Zeisel, H. (1933/1972). Marienthal: The Sociography of an Unemployed Community. New York: Aldine-Atherton.
Kelvin, P. & Jarrett, J. E. (1985). Unemployment: Its Social Psychological Effects. Cambridge: Cambridge University Press.
Locke, L. A. (1969). What is job satisfaction? Organizational Behaviour and Human Performance, 4, 3-23.
Mack, J. & Lansley, S. (1985). Poor Britain. London: George Allen & Unwin.
McGhee, J. & Fryer, D. (1989). Unemployment, income and the family: An action research approach. Social Behaviour, 4, 237-252.
McKee, L. & Bell, C. (1985). Marital and family relations in times of male unemployment. In B. Roberts, R. Finnegan & D. Gallie (Eds), New Approaches to Economic Life. Manchester: Manchester University Press.
Mirowsky, J. & Ross, C. (1989). Social Causes of Psychological Distress. New York: Aldine de Gruyter.
Mislevy, R.J. (1986). Recent developments in the factor analysis of categorical variables. Journal of Educational Statistics, 11, 3-31.
Muthen, B. (1978). Contributions to factor analysis of dichotomous variables. Psychometrika, 43, 551-560.
Muthen, B. (1988). Liscomp (Computer Programme) 2nd ed. Chicago: Scientific Software.
Pahl, J. (1980). Patterns of money management within marriage. Journal of Social Policy, 9, 313-335.
Pahl, J. (1983). The allocation of money and the structuring of inequality within marriage. Sociological Review, 31, 235-262.
Pearlin, L., Menaghan, E. Lieberman, M. & Mullan, J. T. (1981). The stress process. Journal of Health and Social Behaviour, 22, 337-351.
Pearlin, L. (1990). The sociological study of stress. Journal of Health and Social Behaviour, 30, 241-256.
Ringen, S. (1988). Direct and indirect measures of poverty. Journal of Social Policy, 17, 351-366.
Ringen, S. (1988). The Possibility of Politics. Oxford: Oxford University Press.
Ross, C. E. & Huber, J. (1985). Hardship and depression. Journal of Health and Social Behaviour, 26, 312-327.
Townsend, P. (1979). Poverty in the United Kingdom. Harmondsworth: Penguin.
Ullah, P. (1990). The association between income, financial strain and psychological well-being among unemployed youths. Journal of Occupational Psychology, 63, 317-330.
Warr, P. B. (1987). Work, Unemployment and Mental Health. Oxford: Clarendon Press.
Whelan, B. J. (1979). RANSAM: A random sample design for Ireland. The Economic and Social Review, 10(2), 169-174.
Whelan, C. T., Hannan, D. & Creighton, S. (1991). Unemployment, Poverty and Psychological Distress. Dublin: The Economic and Social Research Institute.
Life-style deprivation items
Here is a list of some things a person (a household) might be able to have or to do.
(a) Could you tell me which ones you believe are necessities, that is things that every household (a person) should be able to have and that nobody should have to do without.
(b) Looking again at the list, could you tell me which of the things listed do you not have or cannot avail of.
(c) Of the things you don't now have, which ones would you like to have but must do without because of lack of money?
2. Washing machine.
5. Colour television.
6. A week's annual holiday away from home (not staying with relations).
7. A dry, damp-free dwelling.
8. Heating for the living room when it is cold.
9. Central heating in the house.
10. An indoor toilet in the dwelling (not shared with other households).
11. Bath or shower (not shared with other households).
12. A meal with meat, chicken or fish every second day.
13. A warm, waterproof overcoat.
14. Two pairs of strong shoes.
15. To be able to save some of one's income regularly.
16. A daily newspaper.
17. A roast meat joint or its equivalent once a week.
18. A hobby or leisure activity.
19. New, not second-hand, clothes.
20. Presents for friends or family once a year.
21. There is a day during the previous week when the household manager, i.e. the person who buys most of the groceries for the household, did not have a substantial meal at all -- from getting up to going to bed.
22. The household manager has to go without heating during the last year through lack of money, i.e. having to go without a fire on a cold day, or to go to bed early to keep warm or light the fire late because of lack of coal or fuel.
23. Head of household has not had an afternoon or evening out in the last fortnight because of lack of resources.
24. The household is:
(a) Currently in arrears on rent, mortgage, electricity or gas; or
(b) Has had to go into debt in the last 12 months to meet ordinary living expenses such as rent or food, Christmas or back-to-school expenses; or
(c) Has had to sell or pawn something worth |pounds~50 or more to meet ordinary living expenses; or
(d) Has received assistance from a private charity in the last year.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Special Issue: Marienthal and Beyond: 20th Century Research on Unemployment and Mental Health|
|Author:||Whelan, Christopher T.|
|Publication:||Journal of Occupational and Organizational Psychology|
|Date:||Dec 1, 1992|
|Previous Article:||Expectancy-value theory and unemployment effects.|
|Next Article:||Towards a social psychology of the labour market: or why we need to understand the labour market before we can understand unemployment.|