Occupational status

 Summary prepared by Sarah Burgard and Judith Stewart in collaboration with Joseph Schwartz. Last revised July, 2003.

 Chapter Contents:
a. Background
b. Measurement approaches
c. References

 Background

Occupational status is one component of socioeconomic status (SES), summarizing the power, income and educational requirements associated with various positions in the occupational structure. Occupational status has several advantages over the other major indicators of SES, which are most commonly educational attainment and personal or family income. First, occupational status reflects the outcome of educational attainment, provides information about the skills and credentials required to obtain a job, and the associated monetary and other rewards. For example, professionals are differentiated from manual workers by selection on educational attainment that influences patterns of remuneration. Occupational status is also likely to be a better indicator of income over the long term than is income information collected at any single point in time, because in the short-term, income can be quite volatile (Williams and Collins 1995). Finally, occupational status is a promising measure of social position that can provide information about job characteristics, such as environmental and working conditions, decision-making latitude, and psychological demands of the job.

Occupational status is hypothesized to be related to health because (1) it positions individuals within the social structure, which defines access to resources and constraints that can have implications for health and mortality (Mare 1990; Moore and Hayward 1990), and (2) each particular job has its own set of demands and rewards that can influence health, such as physically hazardous or psychologically stressful working conditions (House et al. 1980; Karasek et al. 1981), as well as effects of the job on lifestyle factors including drinking, smoking, and obesity (Sorenson et al. 1985; House et al. 1986). Income and prestige gained from an occupation influence health-related behaviors, choice of community setting and social networks, as well as providing the funds to purchase medical care, healthy foods, and a safe living environment. Members of different occupational groups also vary in risk factor development and health behaviors because selection criteria for recruitment differ across jobs, as do patterns of socialization and the nature of work performed.

 Evidence of association between occupational status and adult mortality

Two major prospective investigations which demonstrate the relationship between occupational status and health are the Whitehall studies of British civil servants and the Wisconsin Longitudinal Survey of men and women who graduated from Wisconsin high schools in 1957. Whitehall I, the original 1967 project, documented for men a steep inverse relationship between occupational grade (using the British Registrar General’s Scale) and poor health outcomes, including mortality from many diseases. The 1985-1988 Whitehall II project examined a new cohort of British civil servants (Marmot, Davey Smith, Stansfeld, Patel, North, Head, White, Brunner, and Feeny 1991); and follow-up studies of this sample continue (Marmot 1999). In the 25 year follow-up of Whitehall I, men in the lowest civil service grade had a three-fold higher risk of death from all causes of mortality compared to men in the highest grade. Whitehall II has shown that there has been no diminution of the gradient found in Whitehall I in prevalence and incidence of many cardiovascular and other health outcomes. The Wisconsin Longitudinal Study was conducted among a sample of Wisconsin high school graduates who were aged 53 or 54 in 1992-1993, and used the Duncan Socioeconomic Index (SEI) as a measure of occupational status. An analysis comparing the Wisconsin and Whitehall studies showed that for self-perceived health, depression, psychological well-being and smoking there was a clear inverse relationship with occupational status (Marmot, Ryff, Bumpass, Shipley & Marks 1997).

Non-simplicity of the relationship between occupational status and health

Despite its utility, there are some difficulties associated with using occupational status as a marker of social position. These include the possibility of reverse causation, changes in occupational status over the life course, the difficulty of assigning occupational status to persons outside the formal labor force, sex and race/ethnic differences in the effect of occupational status on health, and difficulties comparing occupational status indicators across contexts.

 While cross-sectional data have been used to show an association between occupational status and health, the effect of occupational status over the entire life course has been relatively understudied (but see Kaplan et al. 1971; Mare 1990; Moore & Hayward 1990; Pavalko, Elder & Clipp 1993; Waitzman & Smith 1994). Merely inspecting the correlation between occupational status and health at one point in time may be misleading due to the problem of reverse causation; it is possible that instead of an individual’s occupation having an impact on her health, the choice of a particular occupation may actually depend on or be constrained by existing health conditions. Another problem associated with cross-sectional analyses is reliance on measurement of occupational status at a single point in time, when it can be a moving target as individuals change jobs over the course of a career. Job changes that occur relatively early in the career may lead individuals into higher-status jobs, while those in later years may be related to a decreasing capacity to perform high strain jobs with age. The optimal time to measure occupational status in a cross-sectional study is not clear, and the frequently used “final occupation,” or the occupation at the time of death, may not be a good indicator of conditions experienced over the course of the working life (Mare 1990; Moore & Hayward 1990). Some studies that have used measures of occupational status at several time points have shown that persistently low occupational status or downward status mobility can impact health (Williams 1990). A U.S. study found that compared with whites who remained in professional and technical jobs during the entire follow-up period, African American and white males who remained in lower occupational classes or made transitions into lower occupational classes had significantly higher rates of incident hypertension (Waitzman & Smith 1994). These studies point to the need for a life course approach to occupational status that incorporates risk assessment of a particular job within the context of a career (Pavalko, Elder & Clipp 1993).

 Sometimes it is difficult to assign an individual’s occupational status, particularly for those outside the recognized paid labor force, such as unemployed, nonretired adults, homemakers, those in the informal or illegal economy, and individuals, such as children and retired adults, who are not expected to be in the active labor force (Krieger, Williams & Moss 1997). In addition, standard occupational status indicators are probably more accurate for white men than for women or non-whites, because they were originally developed on the basis of a largely white and male labor force. For example, existing measures are not likely to be sensitive to the dynamics of women in the workplace, including their choices about full- and part-time employment, employment interruptions in response to family constraints, and the interrelationship between a woman’s and her spouse’s occupational statuses (Gregorio, Walsh and Paturzo 1997). Women also tend to be concentrated in a smaller number of occupations than men, and to be disproportionately represented in the low-paying positions with fewer opportunities for advancement (Mutchler and Poston 1983; Pugh and Moser 1990). Non-white workers are more likely than whites in the same occupation to be exposed to carcinogens or other damaging conditions at work, and are paid less for the same work even after work experience and educational attainment are taken into account (Krieger, Williams and Moss 1997).

 Finally, comparisons of occupational status measures across time and different populations can be problematic. In the 1980s, Don Treiman and others published a number of articles showing how similar prestige and status rankings of occupations were across race, sex and countries, however rankings of social position based on some occupational status indicators, such as prestige or income, have been shown by some other analysts to be unstable over time (Berkman and Macintyre 1997). These indicators were originally based on the occupational structure that held in the 1950s and 1960s, and job rankings and requirements have changed considerably since then, while the occupational status indicators have been only infrequently updated. International comparisons based on prestige rankings are difficult because of differences in cultural preferences that make the scales incomparable (Kunst and Mackenbach 1994; Berkman and Macintyre 1997). In addition, studies that have concentrated on more limited occupational groups within specific industries or employment settings (such as the British army or the civil service in London) have shown much larger differences in mortality between sub-groups than studies that cover the entire occupational spectrum (see Lynch and Oelman 1981; Davey Smith, Shipley and Rose 1990). This could suggest that conventional measures of occupational status are imprecise, and fail to capture the considerable variation within occupational categories in education, income, and health risks (Davey Smith and Egger 1992).  Some analysts have speculated that the reason the occupational gradient was stronger in the civil service in London is because grade is such a precise measure of occupational status: there is no issue of unreliability, and everyone knows everyone else’s grade. In contrast, not everyone knows everyone else’s occupation in society in general and there is probably less absolute consensus about the status ranking of census occupational categories as well as a fair bit of heterogeneity within occupational categories (e.g., assistant professors and full professors have the same census code and will all be assigned the same status).  

These and other reasons might explain why occupational status is used less frequently than education and income as a measure of socioeconomic status in research in the United States.  An important practical reason for this less frequent usage is the necessity of first coding occupations into Census Occupation Codes. This is a major undertaking, the benefit of which researchers outside of Sociology may question.  An alternative occupational status system used in health-related research is the Hollingshead Index. The Hollingshead Index combines an ordinal ranking of seven occupational categories, ranging from higher executives of large concerns, proprietors, and major professionals (1) to unskilled employees (7) with a seven category ranking of educational categories ranging from professional degree, such as MA, MD, PhD (1) to less than seven years of schooling (7), The occupational score is weighted by seven, the educational score weighted by four, and the two are added to create an overall score. Individuals with scores totaling more than 64 are assigned to the lower occupational class, while those scoring 17 or less are assigned to the upper occupational class, and those falling between these scores are categorized as upper-middle, middle, or lower-middle on the occupational spectrum.

 British and European research frequently uses social class position to examine occupational status differentials (Berkman and Macintyre 1997), but this captures a slightly different concept that is discussed below.

 Measurement Approaches

There are three main traditions in examining how occupational status is connected to health, each most representative of a particular discipline (Ettner & Grzywacz 2001). Occupational health research focuses on physical aspects of the job environment, such as exposure to toxic substances, and their relation to poor health outcomes (Slote 1987; and see Indicators of Work Environment: Physical Work in the Social Environment notebook for more detail). Research in the areas of occupational health psychology and social epidemiology examine the way that psychological and psychosocial features of the work environment, including decision latitude and job demands, as well as workplace social support, influence health outcomes (Karasek & Theorell 1990; and see Workplace Social Environment in the Social Environment notebook for more detail). Occupational status, the focus of this chapter, has been central to sociological research, and measures the effects of both objective (e.g., educational requirements) and subjective (e.g., social prestige) measures of status or hierarchical rank (see Ross & Mirowsky 1995). Measures of occupational status are based on several concepts: public opinion of the level of esteem associated with a job (prestige), the social relationships that create positions of social class (Wright’s measure of social class, Erikson-Goldthorpe social class index), or a combination of the educational requirements and monetary rewards associated with a particular position (British Registrar General’s Scale, Duncan SEI).

 Prestige

Occupational prestige is a measure that captures either a relationship of deference or derogation between role incumbents, or the general desirability or goodness of an occupation (Siegel 1971). Prestige is based on the rankings of occupations by survey respondents on the basis of goodness, worth, status, and power, and is a robust measure, showing little variation regardless of how people are asked to rate occupations (Kraus, Schild & Hodge 1978), whether occupations are rated by men or women (Bose & Rossi 1983), the race of raters (Siegel 1970), the date on which raters ranked occupations (Hodge, Siegel & Rossi 1964; Hauser 1982; Nakao & Treas 1994), or raters’ own social class standing (Treiman 1977; Haller & Bills 1979).

 Social Class

Measures of social class standing have been used as indicators of occupational status, including Wright’s classification and the Erikson-Goldthorpe social class index. Measures of social class differ from other measures of occupational status because they aim to capture the ongoing economic interactions between people, rather than identify the personal characteristics that determine an individual’s position within a stable hierarchy. Since the goal is to identify power relationships, identifying an individual’s social class position also requires the collection of information about supervisory or managerial activity and the size of the work establishment. In Wright’s typology, social class position is based on ownership of capital assets, control of organizational assets, and possession of skill or credential assets (Wright 1985; 1996). Wright’s early typology employed three primary classes within the capitalist system: the capitalist class, the working class, and between these in social class standing, the petty bourgeoisie (Wright 1979). Three kinds of power in capitalist systems relate to decisions about what is to be produced: control over the mean of production, control over how things are to be produced, and control over labor power. The capitalist class tends to have more control over the means of production, while managers and technicians have more control over how things are produced, and the working class has the least control or power of any group. Wright’s later work has focused on the distribution of different types of assets, but still strongly distinguishes the owners of the means of production from non-owners, and managers and supervisors from others.  The Erikson-Goldthorpe schema, like Wright’s measure, uses information on occupation, self-employment, number of employees, and supervisory status to classify individuals into an 11 category graded hierarchy (Erikson & Goldthorpe 1992). The conceptualization of class as a social position created by social relationships has only begun to be used in the public health literature (Soderfeldt, Danermark, & Larsson 1987; Krieger 1991).

An important unresolved question is whether the health/mortality differences associated with the Erikson-Goldthorpe class categories are greater than those associated with U.S. SES measures.  And if so, are these differences due to national differences (e.g., class is more important in England than in U.S.) or to measurement differences (such that if we could collapse U.S. census occupation codes into Erikson-Goldthorpe categories, we would find larger effects)? 

 Education and rewards

The majority of measures of occupational status, however, are based on some combination of the educational requirements and monetary rewards associated with the position. These include the British Registrar General’s Scale, the Duncan Socioeconomic Index, and United States Census occupation categories. The British Registrar General’s Scale, developed in 1913 by the Registrar General T.H.C. Stevenson and based on a graded hierarchy of occupations ranked by skill, has long been used in British public health surveillance and research. In this schema, occupations are a measure of “standing in the community” or “culture,” and there are five major grades: Social Class I (professional), Social Class II (intermediate), Social Class IIINM (skilled nonmanual), Social Class IIIM (skilled manual), Social Class IV (partly skilled), and Social Class V (unskilled) (Szreter 1984). The British Registrar General’s Scale has proven to be particularly useful in predicting differential risk of morbidity and mortality among employed men (Townsend, Davidson & Whitehead 1990; Marmot, Bobak & Smith 1995).

 The Duncan Socioeconomic Index (SEI), which dominates the research literature in the United States, is an amalgam of occupational prestige and census occupation score rankings. SEI scores were originally constructed by Duncan using data from the 1950 census and the 1947 National Opinion Research Center prestige study, and have been updated several times since (Duncan 1961). For example, in one commonly-used incarnation of the SEI, each U.S. census-defined occupation has an SEI score that is a weighted combination of “occupational education,” which measures the proportion of that occupation’s incumbents who had one or more years of college education in the 1970 census, and “occupational income,” indicating the percentage of incumbents who earned $10,000 or more in 1969. The SEI is the most frequently used indicator of socioeconomic and/or occupational status in sociological research because it best describes socioeconomic differences between occupations and has the highest criterion validity (Featherman & Hauser 1976; Hauser & Warren 1997; Warren, Sheridan & Hauser 1998).

 Occupational categorization and the United States Census

In the United States, the Census Bureau uses two detailed classification schemes for categorizing jobs according to occupation and industry of employment. Both classification schemes consist of several hundred categories and are modified every ten years to take into account changes in the labor market that occur between decennial censuses.   Occupational prestige and the SEI scoring systems are linked to the Census occupation codes; that is, each occupation code is assigned a prestige score and an SEI score and, therefore, all individuals in the same occupation are assigned the same scores.   The occupation and industry codes can be collapsed into broad categories and are sometimes linked to other data sources to obtain non-self-report measures of the average physical, environmental, and psychosocial working conditions associated with particular jobs or groups of jobs (see Schwartz, Pieper & Karasek 1988 and Roos & Treiman 1980; see also, Cain 1980 and Cain and Treiman 1981). Occupation coding can be done with the Standard Occupational Classification (SOC) System. http://bls.gov/soc.home.htm provides both a search function by occupation name and a link to a page that shows the structure of the system. These codes can be up to six digits.

 References

Berkman, L.F. and S. Macintyre. (1997). The measurement of social class in health studies: old measures and new formulations. In Social Inequalities and Cancer, eds. M. Kogevinas, N. Pearce, M. Susser, and P. Boffetta, pp. 51-64. Lyon: IARC Scientific Publications No. 138.

 Bose, C.E. & Rossi, P.H. (1983). Prestige standings of occupations as affected by gender. American Sociological Review, 48:316-330.

 Cain, Pamela. 1980. "An Assessment of The Dictionary of Occupational Titles as a Source of Occupational Information." Pp. 148-197 in Work, Jobs, and Occupations : A Critical Review of the Dictionary of Occupational Titles, edited by A. R. Miller, D. T. Treiman, P. S. Cain, and P. A. Roos. Committee on Occupational Classification and Analysis, Assembly of Behavioral and Social Sciences, National Research Council Washington, D.C., National Academy Press.

 Cain, Pamela and Donald Treiman. 1981. "The Dictionary of Occupational Titles as a Source of Occupational Data." American Sociological Review 46: (3) 253-278.

 Davey Smith, G., and M. Egger. (1992). Socioeconomic differences in mortality in Britain and the United States. American Journal of Public Health 82:1079-1081.

 Davey Smith, G., M.J. Shipley, and G. Rose. (1990). The magnitude and causes of socio-economic differentials in mortality; further evidence from the Whitehall study. Journal of Epidemiology and Community Health 44:265-270.

 Duncan, O.D. (1961). A socioeconomic index for all occupations. In Reiss, A., Jr., editor, pp. 109-138. Occupations and Social Status. New York: Free Press.

 Erikson R. & Goldthorpe, J.H. (1992). The Constant Flux: A Study of Class Mobility in Industrial Societies. Oxford: Clarendon Press.

 Ettner, S.L. & Grzywacz. (2001). Worker’s perceptions of how jobs affect health: A social ecological perspective. Journal of Occupational Health Psychology 6(2):101-113.

 Featherman, D.L. & Hauser, R.M. (1976). Prestige or socioeconomic scales in the study of occupational achievement? Sociological Methods and Research, 4:403-422.

 Gregorio, D.I., S.J. Walsh, and D. Paturzo. (1997). The Effects of Occupation-Based Social Position on Mortality in a Large American Cohort. American Journal of Public Health 87:1472-1475.

 Haller, A.O. & Bills, D. (1979). Occupational prestige in comparative perspective. Contemporary Sociology, 8:721-734.

 Hauser, R.M. (1982). Occupational status in the 19^th and 20^th centuries. Historical Methods, 15:111-126.

 Hauser, R.M. & Warren, J.R. (1997). Socioeconomic indexes for occupations: a review, update, and critique. Sociological Methodology, 27:177-298.

 Hodge, R.W., Siegel, P.M. & Rossi, P.H. (1964). Occupational prestige in the United States, 1925-1963. American Journal of Sociology, 70:286-302.

 Hollingshead, A.B. & Redlich, F.C. (1958). Social class and mental illness. New York: John Wiley & Sons.

 House, J.S., Stretcher, V., Metzner, H.L. & Robbins, C. (1986). Occupational stress and health in the Tecumseh Community Health Study. Journal of Health and Social Behavior, 27:62-77.

 House, J.S., Wells, J.A., Landerman, L.R., McMichael, A.J. & Kaplan, B.H. (1980). Occupational stress and health among factory workers. Journal of Health and Social Behavior, 20:139-160.

 Kaplan, B.H., Cassell, J.C., Tyroler, H.A., Cornoni, J.C., Kleinbaum, D.G. & Hames, C.G. (1971). Occupational mobility and coronary heart disease. Archives of Internal Medicine, 128:938-942.

 Karasek, R.A. & Theorell, T.(1990). Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York: Basic Books.

 Karasek, R.A., Baker, D., Marxer, F. Ahlbom, A. & Theorell, T. (1981). Job decision latitude, job demands, and cardiovascular disease: a prospective study of Swedish men. American Journal of Public Health, 71:694-705.

 Kraus, V., Schild, E. & Hodge, R.W. (1978). Occupational prestige in the collective conscience. Social Forces, 56:900-918.

 Krieger, N. (1991). Women and social class: a methodological study comparing individual, household, and census measures as predictors of black/white differences in reproductive history. Journal of Epidemiology and Community Health 45:35-42.

 Krieger, N., Williams, D.R. & Moss, N.E. (1997). Measuring social class in U.S. public health research: concepts, methodologies, and guidelines. Annual Review of Public Health, 18:341-378.

 Kunst, A., and J. Mackenbach. (1994). International variations in the size of mortality differences associated with occupational status. International Journal of Epidemiology 19:1001-1010.

 Lynch, P., and B.J. Oelman. (1981). Mortality from coronary heart disease in the British army compared with the civil population. British Medical Journal 283:405-407.

 Mare, R.D. (1990). Socio-economic careers and differential mortality among older men in the United States. Pp. 362-387 in Measurement and Analysis of Mortality, edited by Vallin, J., D’Souza, S. & Palloni, A. Oxford: Clarendon Press.

 Marmot, M. (1999). Multi-level approaches to understanding social determinants. In Berkman, L. and Kawachi, I. (eds.) Social Epidemiology. Oxford: Oxford University Press.

 Marmot, M., Bobak, M. & Smith, D.G. (1995). Explanations for social inequalities in health. In Society and Health, ed, B. Amick III, S. Levine, A.R. Tarlov, and D. Walsh, pp. 172-210. New York: Oxford University Press.

 Marmot, M.G., Davey Smith, G., Stansfeld, S., Patel, C., North, R., Head, J., White, I., Brunner, E. & Feeny, A. (1991). Health inequalities among British civil servants: The Whitehall II study. Lancet, June 8:1387-1393.

 Marmot, M.G., Ryff, C.D., Bumpass, L.L., Shipley, M. & Marks, N.F. (1997). Social inequalities in health: next questions and converging evidence. Social Science and Medicine 44(6):901-910.

 Miller, Ann, Donald Treiman, Pamela Cain, and Patricia Roos. 1980.  Work, Jobs, and Occupations : A Critical Review of the Dictionary of Occupational Titles. Committee on Occupational Classification and Analysis, Assembly of Behavioral and Social Sciences, National Research Council Washington, D.C. : National Academy Press.

 Moore, D.E. & Hayward, M.D. (1990). Occupational careers and the mortality of elderly men. Demography, 27:31-53.

 Mutchler, J.E., and D.L. Poston. 1983. Do females necessarily have the same occupation status scores as males? Social Science Research 12:353-362.

 Nakao, K. & Treas, J. (1994). Updating occupational prestige and socioeconomic scores: how the new measures measure up.” In: Marsden, P., editor, pp. 1-72. Sociological Methodology 1994. Washington, D.C.: American Sociological Association.

 Pavalko, E.K., Elder, G.H. & Clipp, E.C. (1993). Worklives and longevity: insights from a life course perspective. Journal of Health and Social Behavior, 34(4):363-380.

 Pugh, H., and K. Moser. 1990. Measuring women’s mortality differences. In Women’s Health Counts, ed. H. Roberts, pp. 93-112. London: Routledge.

 Roos, Patricia and Donald Treiman. 1980. "Appendix F: DOT Scales for the 1970 Census Classification." Pp. 336-389 in Work, Jobs, and Occupations : A Critical Review of the Dictionary of Occupational Titles, edited by A. R. Miller, D. J. Treiman, P. S. Cain, and P. A. Roos, Committee on Occupational Classification and Analysis, Assembly of Behavioral and Social Sciences, National Research Council, Washington, D.C.:  National Academy Press.

 Ross, C.E. & Mirowsky, J. (1995). Does employment affect health? Journal of Health and Social Behavior, 36:230-243.

 Schwartz, J.E., Pieper, C.F. & Karasek, R.A. (1988). A procedure for linking psychosocial job characteristics data to health surveys. American Journal of Public Health, 78(8):904-909.

 Siegel, P.M. (1970). Occupational prestige in the Negro subculture. Sociological Inquiry, 40:156-171.

 ____. (1971). Prestige in the American Occupational Structure [dissertation]. Chicago, IL: University of Chicago

 Slote, L. (1987). Handbook of occupational safety and health. New York: Wiley.

 Soderfeldt, B., Danermark, B. & Larsson, S. (1987). Social class and sickness absences: a comparative study of four ways to measure social class. Scand. J. Soc. Med. 15:211-217.

 Sorenson,G., Pirie, P., Folsom, A. Luepker, R., Jacobs, D. & Gillum, R. (1985). Sex differences in the relationship between work and health: The Minnesota Heart Survey. Journal of Health and Social Behavior, 26:379-394.

 Szreter, S.R.S. (1984). The genesis of the Registrar General’s social classification of occupations. British Journal of Sociology 35:522-546.

 Townsend, P., Davidson, N. & Whitehead, M. 1990. Inequalities in Health: The Black Report and the Health Divide. London: Penguin Books.

 Treiman, D.J. (1977). Occupational Prestige in Comparative Perspective. New York: Academic Press.

 Waitzman, N.J. & Smith, K.R. (1994). The effects of occupational class transitions on hypertension: racial disparities among working class men. American Journal of Public Health, 84:945-950.

 Warren, J.R., Sheridan, J.T. & Hauser, R.M. (1998). Choosing a measure of occupational standing: how useful are composite measures in analyses of gender inequality in occupational attainment? Sociological Methods and Research, 27:3-76.

 Williams, D.R.(1990). Socioeconomic differentials in health: a review and redirection.” Social Psychology Quarterly 53:81-99.

 Williams, D.R. & Collins, C. (1995). U.S. socioeconomic and racial differences in health: patterns and explanations. Annual Review of Sociology, 21:349-386.

 Wright, E.O. (1996). Class Counts: Comparative Studies in Class Analysis. New York: Cambridge University Press.

 _____. (1985). Class Counts. London: Verso.

 _____. (1979). Class Structure and Income Determination. New York: Academic Press.