@connellyReviewOccupationbasedSocial2016

(2016) - Roxanne Connelly, Vernon Gayle, Paul S. Lambert

Journal: Methodological Innovations
Link:: http://journals.sagepub.com/doi/10.1177/2059799116638003
DOI:: 10.1177/2059799116638003
Links::
Tags:: #paper #SocialClass
Cite Key:: [@connellyReviewOccupationbasedSocial2016]

Abstract

This article is a review of issues associated with measuring occupations and using occupation-based socio-economic classifications in social science research. The review is orientated towards researchers who undertake secondary analysis of large-scale micro-level social science datasets. This article begins with an outline of how to handle raw occupational information. This is followed by an introduction to the two main approaches to measuring occupations and a third lesser known but intellectually innovative approach. The three approaches are social class schemes, social stratification scales and the microclass approach. International comparisons are briefly described and a discussion of intersectionality with other key variables such as age and gender is provided. We are careful to emphasise that this article does not advocate the uncritical adoption of any one particular occupation-based socio-economic measure over and above other alternatives. Rather, we are advocating that researchers should choose from the portfolio of existing socio-economic measures in an informed and empirically defensible way, and we strongly advocate undertaking sensitivity analyses. We conclude that researchers should always use existing socio-economic measures that have agreed on and well-documented standards. We strongly advise researchers not to develop their own measures without strong justification nor to use existing measures in an un-prescribed or ad hoc manner.

Notes

“Central to this field is the recognition that the occupational structure is an important foundation for the main dimension of social stratification (Blau and Duncan, 1967: 6–7).” (Connelly et al., 2016, p. 1)

“occupations are the most powerful single indicator of levels of material reward, social standing and life chances (Parkin, 1971)” (Connelly et al., 2016, p. 1)

“The three approaches are social class schemes, social stratification scales and the microclass approach” (Connelly et al., 2016, p. 1)

“he most common justification for using occupation-based socio-economic measures is that they make reliable, parsimonious indicators of the social positions of individuals (Parkin, 1971; Rose and Pevalin, 2003)” (Connelly et al., 2016, p. 2)

“Empirical inquiries using repeated contacts data have convincingly shown that there is a high degree of income churning from year to year which makes income data unlikely to consistently represent individuals’positions in industrial economies (Jarvis and Jenkins, 1997; Jenkins and Van Kerm, 2009)” (Connelly et al., 2016, p. 2)

“Pakulski and Waters’ (1996) account of the ‘death of class’ centres on two main ideas. First, the class-based divisions peaked in industrial society and have been declining ever since. Second, although there are inequalities in modern society, these are not aligned with traditional social classe” (Connelly et al., 2016, p. 2)

“Goldthorpe and Marshall (1992) note that the concept of class which is being attacked is a concept which is never clearly defined, is most aligned to the Marxist tradition” (Connelly et al., 2016, p. 2)

“Platt (2011: 15) highlights that central concerns in contemporary class analysis include the notion of changes in the influence of class, the declining importance of class and the intersectionality of other variables such as gender and ethnicity.” (Connelly et al., 2016, p. 2)

“As Lambert (2002) demonstrates, without the use of an established protocol for coding raw occupational information (e.g. using SOC codes), it is later impossible to test for comparability between both current and future occupationbased measures.” (Connelly et al., 2016, p. 3)

“Translating raw survey data into unit group codes can be a time-consuming exercise, but the burden is greatly reduced through the use of computer-assisted and computerautomated coding procedures (Elias et al., 1993). The Computer-Aided Structured Coding Tool2 (CASCOT) is an online resource for the quick and reliable coding of occupational descriptions” (Connelly et al., 2016, p. 3)

“The British Household Panel Survey is a prime example of a large-scale social survey with well-curated occupational data that are readily available for secondary data analysis (Taylor et al., 2010).” (Connelly et al., 2016, p. 4)

“A prominent example is the United Kingdom’s long-standing ‘Registrar General’s Social Classification’9 (e.g. Szreter, 1984). There is evidence that skill-based measures are empirically very powerful, and they remain a popular choice in social research (see Elias and McKnight, 2001; Tahlin, 2007).” (Connelly et al., 2016, p. 4)

“The use of varied forms of the EGP schema is consistent with the claim that the measure is an instrument du travail rather than a definitive representation of social class groupings in the United Kingdom (Erikson and Goldthorpe, 1992: 46).” (Connelly et al., 2016, p. 5)

“The principal difference between categorical social class schemes and stratification scales is that rather than placing individuals into qualitatively distinct categories, social stratification scales place individuals at some point on a continuous or gradational one-dimensional hierarchy (Bergman and Joye, 2005).” (Connelly et al., 2016, p. 5)

“A notable example of a social stratification scale is the Cambridge Social Interaction and Stratification Scale (CAMSIS) (Prandy, 1990; Stewart et al., 1980).” (Connelly et al., 2016, p. 5)

“Two other important stratification scales are the Standard International Occupational Prestige Scale (SIOPS; see Ganzeboom and Treiman, 1996, 2003) and the International Socio-Economic Index (ISEI; Ganzeboom et al., 1992)” (Connelly et al., 2016, p. 7)

“Because they are continuous measures, all of the occupation-based social stratification scales lead to numeric values being attached to occupations, but the relative importance of a specific value is only meaningful in comparison with other occupations on the same scale” (Connelly et al., 2016, p. 7)

“Microclass approaches typically feature around 80–100 different classes. Grusky and Sørensen (1998) contend that traditional social class schemes fail to represent detailed social structures within big classes and that the social structure is not adequately represented by unidimensional hierarchical scales.” (Connelly et al., 2016, p. 8)

“Theoretically, microclasses are defined by the social and/or technical institutionalisation of occupations (e.g. plumber, baker and doctor) rather than agglomerate classes (e.g. skilled manual workers or professionals)” (Connelly et al., 2016, p. 8)

“Nevertheless, a major attraction of the microclass approach is that it facilitates the investigation of potentially important substantive differences at the detailed occupational level that may be hidden within the large categories of ‘big’ social class schemes.” (Connelly et al., 2016, p. 8)

“Gayle and Lambert, 2011; Jonsson et al., 2009” (Connelly et al., 2016, p. 8)

“Erikson et al. (2012) argue that the disaggregation of categories mean that patterns linked to microclasses cannot be clearly interpreted within the theoretical framework that is useful to a ‘big class’ measure. At a practical level, the inclusion of a large multiple category explanatory variable tends to decrease parsimony in standard statistical models” (Connelly et al., 2016, p. 8)

“Therefore, a good solution is to construct a number of measures and evaluate them through a ‘sensitivity analysis’.” (Connelly et al., 2016, p. 8)

“A suitable analytical response may be to include measures of age and gender as control variables within analysis, also investigating possible interaction effects between the age and gender measures and the occupational classification.” (Connelly et al., 2016, p. 9)

“An argument expressed by Goldthorpe et al. (1987) is that most adults reach a point of ‘occupational maturity’, around about the age of 35” (Connelly et al., 2016, p. 9)

“The ‘dominance’ approach measures all jobs in the household and typically assigns a measure based on the economically dominant occupation within the household.” (Connelly et al., 2016, p. 9)

“Another alternative is simply to incorporate the necessary individual variables that relate to occupational characteristics within the household.” (Connelly et al., 2016, p. 9)

“Finally, an enduring problem when using occupational measures is the complexity of making comparisons over time when the underlying structure of the labour market has changed. In some secondary analyses of large-scale social surveys where occupation-based measures are included as explanatory variables, this will not be an issue due to the restricted time frame of the analysis. In more specialised analyses, for example, in research on inter- and intra-generational mobility, more thought will be required regarding structural changes in the labour market. We suggest that in such analyses, specialist statistical approaches that are directed towards providing increased control to help to combat this problem should be considered (see especially Erikson and Goldthorpe, 1992).” (Connelly et al., 2016, p. 9)