“Too Much Money Off Other People’s Backs”:
Status in Late Modern Societies*
Michèle Ollivier
Canadian Journal of Sociology 25, 4 (Fall 2000): 441-70.
Abstract: This paper revisits debates on occupational prestige in light of recent shifts in sociological conceptions of status, culture, and identity. Using multidimensional scaling and clustering of data collected from electricians, university professors, and students in adult education, it explores how people differently located in social structure perceive occupations. Results indicate that prestige scales are one among many cognitive schemata available in collective consciousness for representing social structure, drawing symbolic boundaries, and evaluating others. How this is done, however, varies with social location. Compared to electricians and students, professors see more congruence between occupational prestige and worth. Electricians and students use normative evaluations of worth as alternative criteria for evaluating occupations and drawing boundaries, in an effort to enhance their own social position and downgrade others.
Résumé: Cet article examine les débats sociologiques entourant le prestige occupationnel à la lumière des transformations récentes des concepts de statut, de culture et d’identité. Sur la base d’une analyse multidimensionnelle et typologique de données recueillies auprès d’électriciens, d’étudiant-es et de professeur-es d’université, il explore comment des acteurs différemment situés dans l’espace social perçoivent les occupations. Les résultats indiquent que les échelles de prestige constituent dans la conscience collective un schéma cognitif parmi d’autres permettant de représenter la structure sociale, d’y tracer des frontières symboliques et d’évaluer les occupations. L’utilisation de ces schémas cognitifs varie cependant selon la position sociale. Les professeur-es perçoivent une forte corrélation entre le prestige et certains critères normatifs d’évaluation des occupations. Les électriciens et les étudiant-es, par contre, utilisent ces derniers pour tracer différentes frontières symboliques qui rehaussent leur position sociale.
Contents:
Introduction | What is Status? | Occupational Prestige and Normative Evaluations of Worth | Research Design | Results | Conclusion | Notes | References | Appendix |
Introduction
The development of occupational prestige scales by North and Hatt, in the 1940s, gave rise to heated debates about the meaning of these scales and the inferences that could be drawn from them. Sociologists emphasizing the consensual nature of occupational ratings argued that they expressed collective beliefs about the worthiness of occupations (Shils, 1975; Treiman, 1977). Positions which ranked highest on the scales were those which were most highly valued by a collectivity, either because of their functional importance (Davis and Moore, 1945) or because of their proximity to order-creating values and institutions (Shils, 1975). This interpretation fit in very well with neo-Durkheimian and functionalist theories of stratification, according to which consensual evaluations of positions constitute a necessary foundation for social life. They provide a shared normative framework which ensures social integration and legitimates inequality.
Normative interpretations, however, were never very popular among sociologists who used prestige scales in empirical research. While most remained agnostic on the assumptions and implications underlying their use in research, some argued that occupational prestige reflects factual knowledge about the material rewards attached to occupations (Featherman and Hauser, 1976; Nam and Terrie, 1982). A commonly accepted interpretation, proposed by Goldthorpe and Hope (1974), is that occupational prestige scales measure the “desirability” of occupations in terms of socioeconomic rewards. Other things being equal, they argue, people generally agree that positions offering high rewards are more desirable than those affording low rewards. High levels of consensus among raters simply mean that people are generally aware of the economic and cultural advantages attached to occupations and value them as desirable accordingly. Goldthorpe and Hope (1974) acknowledge that their concept retains an evaluative component, but contend that this component has no normative connotation and no legitimating significance with regard to social inequality. After years of disputes the debate was finally settled. Prestige scales are now widely believed to reflect factual evaluations of the economic and cultural advantages afforded by occupations rather than normative evaluations of their worth (DiMaggio and Mohr, 1985; Featherman and Hauser, 1976; Goldthorpe and Hope, 1974; Grusky and Rompaey, 1992; Sørensen, 1992).
This position, however, leaves many questions unanswered, namely concerning the nature of prestige in contemporary societies as well as the relationship between socioeconomic status and normative evaluations of worth.[1] If occupational prestige scales are based on factual evaluations of material rewards, how do people conceive of the worthiness of occupations? Was Durkheim mistaken when he argued that a normative ordering of positions is an essential integrative feature of all stratified societies, or have society-wide, consensual evaluations become irrelevant in late modern times? Have they given way to a multiplicity of local status orders, as suggested by much contemporary sociological theory?
Making sense of occupational prestige in light of changing conceptions of status and identity is the main objective of this paper. The first section discusses how recent shifts in sociological theory have altered how status is viewed in contemporary societies. The neo-Durkheimian assumption that occupational prestige forms the backbone of a consensual and tightly-woven pattern of norms and values no longer appears tenable in light of the multiplicity of identities and hierarchies displayed in contemporary social life. Section 2 reviews existing research on occupations which shows, first, that occupational structure and occupational prestige are generally perceived as multidimensional, secondly, that the distinction between factual and normative components of occupational ratings is not always clear, and, thirdly, that beliefs about normative aspects of occupations differ according to people’s location in social structure. The remainder of the paper presents data collected during interviews with electricians, university professors, and students in adult education, in a research explicitly designed to examine how people differently located in social structure draw boundaries in occupational space and how they assess the worthiness of occupations. I argue that occupational prestige scales reflect what people perceive as a dominant system of assessment of social positions. Along with normative and other criteria of evaluation, they are used in various ways by groups differently located in social structure for drawing boundaries between themselves and others.
What is Status?
One of the main arguments justifying the rejection of normative interpretations of occupational prestige is that respondents to prestige surveys overwhelmingly mention material advantages of occupations when asked to explain their ratings (Coleman and Rainwater, 1978; Goldthorpe and Hope, 1974; Reiss, 1961). The rejection of normative interpretations, however, also reflects a major shift in sociological theory. In light of the complexity and anonymity of social life in advanced market societies, few sociologists now accept the Durkheimian assumption that a single, overarching status order permeates social life and exercises a systematic influence on social relations.
A first argument supporting this view is that contemporary societies are highly complex and segmented, so that people have multiple social affiliations which do not necessarily coincide (Blau and Schwartz, 1984; DiMaggio and Mohr, 1985; Melucci, 1996). Complexity and segmentation contribute to the emergence of a diversity of status cultures, which people learn to selectively display in appropriate social contexts (DiMaggio, 1987; Erickson, 1996). A second argument is that contemporary societies have witnessed a phenomenal increase in the breadth and anonymity of social relations, as more people come into more fleeting contact with others in a diversity of social settings (Calhoun, 1991). High levels of geographical and social mobility, made possible by the rapid expansion of means of transportation and new communication technologies, coupled with the extraordinary increase in the amount and diversity of information available through the mass media, have led to major changes in processes of identity formation and status competition. As opposed to relatively small communities, where status is linked to personal knowledge of fixed identities based on family name, reputation, kinship, and place of residence, enlarged and anonymous social spaces provide individuals with vastly increased possibilities of experimenting with strategies of self-presentation in everyday life. Combined together, segmentation and anonymity lead to the emergence of multiple and heterogeneous status cultures, anchored in constantly shifting identities and hierarchies (Hall, 1992).
Late modern societies are thus seen as comprising multiple, segmented, and sometimes incommensurable normative orders, in which status operates predominantly at micro or local levels: among different fractions of classes (Bourdieu, 1979; Coleman and Rainwater, 1978; Hope, 1982; Stehr, 1974), between adjacent or competing occupational groups (Blau and Duncan, 1967:63–70), and within small communities such as popular music artists (Ollivier, 1994), carnival strippers (Meiselas, 1976), upper class philanthropists (Ostrower, 1996), and adolescents (Milner, 1996). Rather than representing a society-wide moral classification, status is linked to the multilayered symbolic boundaries of race, gender, class, and lifestyle that people draw between themselves and others in processes of inclusion and exclusion (Lamont and Fournier, 1992). Occupations, in this sense, are only one of numerous dimensions of identity and status inequality.
Even if occupations are no longer considered the most salient dimension of inequality, questions remain on whether and how they are used for making status distinctions and drawing boundaries in everyday life. How do people draw boundaries around occupations? Do these boundaries have normative connotations? Are incumbents of occupations at the top of the scales considered more deserving than those at the bottom? Is their success an indication of superior moral qualities or abilities? Are they viewed as making more substantial contributions to the well-being of society than those at the bottom? In the following section, I review research on how people think about occupations in search of partial answers to some of these issues.
Occupational Prestige and Normative Evaluations of Worth
There are good reasons to believe that people perceive occupational structure as having more than one dimension. Research using multidimensional scaling (MDS), in particular, generally shows that people do not think of occupations in terms of a linear scale (see Burton, 1972; Coxon and Jones, 1978; Magana, Burton and Ferreira-Pinto, 1995; Krauss, Schild and Hodge, 1978). MDS studies invariably show that occupational prestige is one of the most salient dimensions of people’s representations of occupational structure, but most of them also find evidence of additional dimensions. These dimensions are interpreted in many different ways: as service orientation and skills/trade by Coxon and Jones (1978), as job autonomy and training by Burton (1972), and as an opposition between power-oriented or bureaucratic occupations on the one hand and the healing, academic, and autonomously-oriented professions on the other by Magana et al. (1995).
Furthermore, research also suggests that occupational prestige itself should be understood as comprising more than one dimension. Reviewing this issue, Grusky and Rompaey conclude that occupational prestige scales should be interpreted as the best one-dimensional representation of “an intrinsically two-dimensional occupational structure”, whose main dimensions are economic and sociocultural resources (1992:1719). What is meant exactly by sociocultural resources, however, is open to divergent interpretations. One interpretation is that occupational prestige compounds in a single scale different types of resources. Supporting evidence for this argument is garnered from the work of Bourdieu (1979), who conceptualizes occupational structure as a two-dimensional space with cultural and economic capital as its main dimensions, from the empirical work of DeGraff, Ganzeboom, and Kalmijn (1989), who operationalise and test the usefulness of these two dimensions for analysing lifestyles, and even from Duncan’s (1961) finding that occupational prestige represents a combination of income and education.
A competing interpretation, however, is that prestige scales have both a factual and a normative component. Using polar coordinate analysis, Hope (1982) shows that occupational prestige is an average of two dimensions: the standard of living afforded by occupations and an assessment of their value to society (Hope, 1982). Similarly, Grasmick (1976) uses multidimensional scaling to show that vectors measuring the social standing of occupations reflect a combination of material rewards and value to society, although the value to society dimension only operates in the middle range of the material rewards scale. Interestingly, occupations at opposite ends of Grasmick’s social value dimension (1976:101) are very similar to those which Bourdieu places at opposite ends of his cultural-economic capital dimension (1979:140–141). In both cases, the middle range of occupational space is characterized by an opposition between occupations such as teachers and social workers on the one hand, and more commercial occupations such as small store owners and managers on the other hand. The “value to society” gradation and the “cultural vs economic capital” dimensions could thus be considered as different interpretations of the same dimension of occupational space.
These two interpretations, however, have very different implications. While cultural resources may be considered as advantages attached to occupations, the same cannot be said of social usefulness. Social usefulness is neither a material nor a symbolic resource, but rather a measure of intangible contributions that people make to the general well-being of society. Which of these interpretations is correct cannot be resolved here, but Hope (1982) and Grasmick’s (1976) research suggests that the separation between normative and factual interpretations of prestige scales may not be as clear-cut as previously thought. As argued by Hope (1982), this does not mean that occupations which rank high on prestige scales command deference in everyday life, but it suggests that prestige scales do have some legitimating significance with regard to social inequality.
In this respect, studies of people’s perception of inequality show that in advanced industrial societies, the unequal distribution of rewards is largely perceived as legitimate (Cuneo, 1996; Lamont, forthcoming; Kluegel and Smith, 1986; Pammett, 1996). Belief in equality of opportunity is widespread, and to the extent that inequality of condition is perceived as reflecting individual qualities such as ambition, talent, and hard work, it is generally considered as fair. Several studies, however, also show that congruence between occupational prestige and normative conceptions of worth is much stronger at the top of the socioeconomic status hierarchy than at the bottom. People in high-status occupations are more likely to perceive their society as open and egalitarian (Cuneo, 1996; Kelley and Evans, 1993) and to explain existing inequality in terms of individual as opposed to structural variables (Kluegel and Smith, 1986). People at the bottom are more likely to dissociate socioeconomic status and worth, namely by emphasizing alternative sets of moral and cultural standards based on personal integrity, sincerity, and quality of interpersonal relationships (Lamont, forthcoming). Similarly, Guppy (1984) has shown that the level of consensus on occupational prestige is much higher among more privileged respondents than among less privileged ones. These findings fit in well with Parkin’s argument (1971) that different classes need to account for social inequality in different ways and thus develop different meaning systems: the dominant value system is a “moral framework which promotes the endorsement of existing inequality” (1971:81), the subordinate value system is oriented towards accommodation to dominant values, and the radical system promotes an oppositional stance towards social inequalities.
Occupational prestige scales, I argue, appear to tap into what people perceive as a dominant value system. In egalitarian and secular contemporary societies, this system largely reflects material advantages but it may also include evaluations of the social value of positions. How closely people endorse this value system, however, shows some variation which is at least in part a function of social location. Building on these insights, the remainder of this paper examines how people differently located in social structure perceive occupational space as well as the relationship between socioeconomic status and worth.[2] I expect that people will perceive occupations in terms of a multidimensional structure of which occupational prestige will be a major dimension. I also expect that beliefs about the worthiness of occupations will vary according to people’s location in social structure, with greater congruence between socioeconomic status and worth at the top than at the bottom. Whether people differently located in social structure draw the same or different boundaries in occupational space and how these boundaries are affected by normative evaluations of worth remain open issues.
Research Design
Multidimensional scaling (MDS) and hierarchical clustering of free-sorting data provide an inductive yet systematic approach for mapping out cognitive domains and measuring some of their dimensions. MDS transforms data on the perceived similarities between objects into spatial distances. It produces a visual representation, in two or more dimensions, of the relationships among objects. Objects which are most similar appear close together and those which are most dissimilar are far apart. MDS provides a visually attractive representation of similarities and distances among objects. It can also be used as a support for representing clusters into space.
Following standard procedures in MDS (Coxon and Jones, 1978, 1979; Burton, 1972; Magana, Burton and Ferreira-Pinto, 1995; Ollivier, 1997), I first asked three groups of respondents to sort 60 cards marked with occupation titles into any number of piles, in the way which felt most natural to them. In open-ended interviews, I asked them to describe their piles, to locate themselves in one of them, and to collapse them, if possible, into broader categories. Finally, I asked them to rate occupations on three different scales: their prestige or standing in society in general, their social usefulness, and the respect and admiration they personally felt for them. With regard to occupational prestige, respondents were instructed to rate occupations according to how they thought they were viewed in society in general. The other two scales are based on respondents’ personal views and are thought to more closely reflect their perception of the worthiness of occupations (for a similar use see Coxon and Jones, 1978).
Sixty occupations were selected from the Canadian 1980 Standard Occupational Classification (Statistics Canada, 1980), following a two-step procedure. In order to ensure that the occupations selected would be roughly representative of their distribution in the Canadian labour force, I first computed the proportion of the Canadian labour force in each of the Major Groups of the 1980 National Occupational Classification. I then selected the corresponding number of occupations within each of the Major Groups. The main criteria guiding the selection process at this point was that occupation titles be easily recognizable and different from one another. For example, there would have been no point in including both Chemical Engineers and Aerospace Engineers, since these occupations would likely be considered highly similar by most people, except perhaps by chemical and aerospace engineers themselves.
Because a main objective of this research is to explore whether people differently located in social structure perceive and evaluate occupations in different ways, I selected respondents among different occupational groups with a view of maximizing social distance between them. Keeping in mind Bourdieu’s (1979) two-dimensional representation of social space, in which the vertical dimension reflects total amount of capital and the horizontal dimension refers to an opposition between economic and cultural capital, I selected respondents in what I hypothesized would be different regions of social space. Interviews were conducted from January to June 1997 with electricians in a union training centre and university professors in a Canadian university. University professors were selected because of their overall high level of economic and cultural capital. Professors from the Departments of Mechanical and Aerospace Engineering, Business, English, and Religious studies were solicited by email and 19 interviews were conducted with those agreed to participate.[3] My intention was to contrast professors in the Humanities with those in the more technical fields of Business and Engineering. However, since only three respondents in the Humanities agreed to be interviewed compared to nine in Business and seven in Engineering, the results presented here are aggregated. Socio-demographic data collected during the interviews indicate that all respondents have completed graduate school and their mean income is $74,000. Only 4 out of 19 are women and their mean age is 47 years old.
Electricians were selected because they represent a traditional blue-collar occupation. They have lower levels of total capital compared to university professors, but higher levels of cultural capital in terms of formal and on-the-job training compared to many blue-collar and service occupations. Electricians were contacted through their local union, which organizes regular training sessions for its members. Group interviews were conducted by myself and a research assistant with a total of 19 electricians. Respondents worked in the same room and short individual interviews were conducted with each respondent as they finished sorting the cards. The electricians interviewed have lower levels of cultural and economic capital compared to university professors, but still relatively high levels of education. All but one have at least some post- secondary education and 65% hold a vocational college degree. Their mean household income is $47,000. All of them are male with a mean age of 36 years old.
Students in adult education were interviewed one year later. Because they were working towards a high school degree, I expected these students to have lower levels of economic and cultural capital compared to electricians, and to hold jobs requiring little education and offering low benefits. I was especially interested in comparing their views with those of electricians, who emphasized their high levels of skills in differentiating themselves from service and white-collar occupations requiring little or no training. With the help of a research assistant, I interviewed 18 students in a grade 12 mathematics class. Answers to socio-demographic questions revealed lower levels of education and household incomes compared to electricians, but also more diversity than I expected. Only one student had not already completed a high school degree at the time of the interview and 61% had at least some post-secondary education. Many were taking high school math to reorient themselves towards new careers in health care and computer science. They were equally divided between males and females and they were on average younger than the other two groups, with a mean age of 30 years old. Their mean household income was $39,000. They were also more diverse in terms of ethnicity and language compared to the other groups. Fifty percent were born outside Canada and one third (33%) spoke languages other than English or French at home. By contrast, almost all electricians (95%) were born in Canada and only two (11%) spoke languages other than English or French at home. A large percentage (42%) of university professors were born outside of Canada, but only 16% spoke a language other than English or French at home. Many of the students interviewed were recent immigrants, with an imperfect knowledge of English and high expectations of upward mobility. This introduced an unanticipated element in the analysis, since immigrant status has been shown to have an effect both on belief in the ideology of equality of opportunity and on occupational aspirations (Hou and Balakrishnan, 1996; Macleod, 1987).
For each respondent, a square co-occurrence matrix was computed, in which pairs sorted in the same pile received a score of 1 and pairs sorted in different piles received 0. Individual matrices were summed to produce an aggregate matrix for each group of respondents. The three aggregate matrices were then separately submitted to multidimensional scaling using the non-metric Minissa program.[4] The resulting configurations were interpreted using Property Fitting (PROFIT) and used as a support for displaying the results of hierarchical clustering.[5] PROFIT analysis is a regression technique which fits selected attributes of the objects into MDS configurations. PROFIT creates a set of coordinates showing the orientation of each attribute as an axis in the configuration. It provides a multiple R measure, which is taken as an indication of the salience of each variable in the configuration. As opposed to PROFIT, which identifies continuous dimensions in each configuration, cluster analysis allows for the identification of discrete groups. The two techniques provide complementary pictures of the same data set.
Results
Respondents’ Comments
One of the advantages of using the free sorting method to obtain similarity data is that it allows respondents to explain the criteria they used when sorting items into piles. Respondents used a variety of labels and criteria to describe their piles and almost all of them combined several criteria. There were many similarities between the three groups but also noteworthy differences. In all groups, the most often used criteria were levels of education and occupational/industrial sectors (e.g. primary resources, construction, travel, health care, etc.). Occupational sector was used to label at least one category by 15 professors, 12 students, and 10 electricians, while education and skills was used in at least one category by 11 electricians, 10 students, and 9 professors. A typical sorting by an electrician combining these two criteria is: a) actor; b) managerial; c) service-oriented; d) work alone; e) financial things; f) manual labour; g) higher learning; h) skilled trades with apprenticeship program*.[6] A few respondents stated that they could have classified occupations in different ways, using either occupational sector or education as the main criterion.
While education and occupational/industrial sectors were mentioned by a majority of respondents in the three groups, other criteria were specific to each group. Electricians and students, in particular, were much more likely than university professors to use criteria with normative connotations. Eight electricians and three students used criteria referring to respect or contempt, with categories labelled “crooks”, “overblown”, “overrated” or “overpaid”. Five electricians and three students mentioned criteria dealing with occupations’ usefulness to society or to themselves, and two electricians classified their cards according to their personal preferences, e.g. “jobs I would do as opposed to jobs I wouldn’t like to do”. These normative criteria were also often used in combination with one another as well as with education and occupational sectors. The following is a typical example: a) people we need; b) not too much skills; c) no impact on my life (almost useless, like dust on the shelf); d) help things run efficiently; e) people who provide help; f) too much money off other people’s backs; g) lots of schooling, do the best you can*. Another example is: a) white collar that I respect; b) blue-collar that I respect*; c) blue and white collar that I don’t respect and wouldn’t like to do.
Only one professor referred to the usefulness of occupations and none used labels such as “overblown” or “overpaid”. Four professors and five students used a high vs low metaphor to describe some of their categories, for example: “middle-range people, neither top nor bottom”; “high class”; “top of the rank* vs end of the pole”; “I don’t like terms like higher and lower but...”. None of the professors, however, even among those who used a high-low metaphor, used labels suggesting that some occupations were evaluated too highly.
Categories frequently used by professors but more rarely by electricians and students include the degree of authority or responsibility (10 professors), the existence of professional registration or some other form of certification (3 professors), and gender (3 professors). Furthermore, 12 professors out of 19 used the term “professionals” to label one of their piles compared to only 2 electricians and 5 students. This term was often associated with high income, high levels of skills, and professional registration. In addition, several professors made a distinction between “experts”, which they defined as highly skilled professionals who actually use their expertise in their work, and “managers” who are also highly educated but need more general skills to do their job.
Very few people explicitly mentioned income as a criteria. The nine respondents who mentioned income did so in relation to some type of evaluative criteria, for example “overpaid” or “skilled, but not enough to make big bucks”. The two respondents who used preferences as the dominant criterion mentioned income along with prestige, type of work, type of person, challenge, etc. when asked to explain further. Several respondents used criteria which were directly linked to their jobs. One professor in the Department of Religion made two piles: “people who work alone or inside themselves vs people who work with others or outside themselves”. Similarly, one respondent from the Department of Aerospace Engineering used the “severity of consequences for doing a job badly” as a main criterion when making his piles. He explained that an aerospace engineer who does a job badly endangers human life, while a janitor who performs badly only produces dirty premises.
These comments suggest the existence of common grounds between the three groups, since criteria such as education and occupational sector are widely used by all respondents. These comments also indicate that there is variation in how groups describe and evaluate occupations: electricians and, to a lesser extent, students are more likely than professors to consider that some occupations do not deserve the rewards and/or prestige that accrue to them. Similarly, professors place more emphasis on authority and responsibility compared to electricians and students. These comments will be used in the following sections in interpreting the MDS configurations and clusters.
Multidimensional Scaling and Profit Analysis
The aggregate co-occurrence matrices for the three groups were scaled in two and three dimensions. As often occurs in MDS studies, the interpretation of a third dimension proved difficult and for the sake of simplicity, only the two-dimensional solution is discussed here. Stress levels, which measure how well the configuration fits the data, are .139 for university professors, .148 for electricians, and .155 for students.[7] Results are displayed in Figures 1 to 3 in appendix. The three vectors in each configuration were generated by PROFIT analysis. They represent the three scales on which ratings were obtained from respondents. As expected, the most salient dimension in each configuration, as measured by multiple R, is occupational prestige. Multiple R on prestige is .937 for professors, .857 for electricians, and .814 for students. All groups place professional occupations at one extreme of the prestige scale (university professor, lawyer, electrical engineer, psychologist, physician, etc.) and low-skilled, low-paying occupations at the other extreme (janitor, hairdresser, service station attendant, courier, freight handler, etc.).[8]
The second dimension is more difficult to interpret, especially in the absence of precise measures of other attributes of occupations. Visual examination of the configuration in light of respondents’ comments suggests that the second dimension could be interpreted as an opposition between expert/technical/blue-collar occupations such as geologist, architect, electrical engineer, firefighter, plumber, welder, fisherman, and tailor on the one hand, and white-collar/administrative occupations such as hospital administrator, accountant, executive secretary, sales supervisor, travel agent, and bank teller on the other. The three configurations are highly similar in this respect. They are also highly similar to the configurations produced by Magana et al. (1995), who interpret their second dimension as an opposition between power-oriented or bureaucratic occupations on the one hand and the healing, academic, and autonomously-oriented professions on the other. This indicates that groups differently located in social structure do not have radically different views of occupational space.
The vectors measuring admiration and usefulness are relatively salient in all three configurations, but their orientation differs markedly in the three groups. The three vectors are close together and oriented in the same direction for professors in Figure 2. For electricians in Figure 1, by contrast, the prestige vector is oriented at an angle from both the usefulness and admiration vectors. The orientation of these vectors indicates that university professors generally think that occupations which have highest prestige in society are also those which are most socially useful and for which they have the most admiration. Electricians, by contrast are more likely to believe that occupations with high prestige in society are not necessarily those which are most useful and deserve highest admiration. This conclusion is supported by electricians’ comments that some occupations are overpaid and overrated. As for students, they occupy a middle position between the other two groups. They see more discrepancy between the three scales than university professors but less than electricians. Pearson correlations between the three scales, displayed in Table 1, support this interpretation. Occupational prestige is more highly correlated with the normative scales of admiration and usefulness among professors than among the other two groups. Pearson correlations between prestige and admiration are .89 for professors, .79 for students, and .72 for electricians. Correlations between prestige and usefulness are .79 for professors, .64 among students, and .50 among electricians.
Table 1 - Pearson Correlations between prestige, admiration, and usefulness -
electricians, professors, students
Prestige
Admiration
Usefulness
prof.
stud.
elec.
prof.
stud.
elec.
prof.
stud.
elec.
Prestige
professors
1.00
.953
.958
.895
.825
.683
.788
.640
.461
students
1.00
.954
.815
.792
.626
.708
.635
.429
electricians
1.00
.872
.815
.715
.775
.648
.503
Admiration
professors
1.00
.887
.810
.931
.762
.628
students
1.00
.846
.906
.884
.744
electricians
1.00
.853
.819
.909
Usefulness
professors
1.00
.875
.771
students
1.00
.854
electricians
All correlations are significant at the .001 level [9]
Three main conclusions can be drawn from these results. First, people differently located in social structure do not have radically different pictures of this structure, at least in the artificial context of the present research. When people are asked to sort into piles the same set of occupations, they produce pictures whose general layout is remarkably similar. Second, the congruence between occupational prestige and respondents’ personal evaluations of occupations is relatively high, with correlations ranging from .72 among electricians to .89 among professors. This is in line with research on perception of inequality, which shows that the unequal distribution of rewards in advanced industrial societies is widely perceived as legitimate and fair. Third, where one is located in social structure does influence normative evaluations of worth. Congruence between socioeconomic status and worth is highest among professors, who rank at the top of the three groups in terms of income and education. It is lowest among electricians, whose views closely match Parkin’s (1971) subordinate value system. Electricians largely accept as legitimate the dominant value system, as evidenced by the relatively high correlation between occupational prestige and personal admiration. They do however emphasize alternative criteria of evaluation, as shown by the lower correlation between prestige and social usefulness. Their comments that some occupations are “overblown” or “overpaid” also indicate that they are more likely than professors and students to think that the actual distribution of rewards, which underlies occupational prestige scales, is not entirely justified. While students have the lowest levels of education and income among the three groups, they occupy a middle position between high-status university professors and the working class subculture of electricians. They do not share the oppositional stance of electricians, but neither do they wholeheartedly embrace professors’ close association between socioeconomic status and worth. Given that most of these students were taking high-school math to improve their current employment situation, we can assume that they did not expect to remain in low-income and low-education positions.
These results, while interesting, are in line with previous research on perception of inequality, which shows that association between socioeconomic status and worth is greater among people located at the top of the occupational hierarchy than among those at the bottom. In the following section, cluster analysis shows that respondents in the three groups draw very different boundaries within social space, and that how boundaries are drawn is closely linked to normative evaluations of worth.
Hierarchical Clustering
The results of hierarchical clustering at the two-cluster and five-cluster levels are presented in Figures 4 to 6 in the appendix.[10] The two-cluster solutions are represented by solid lines and labelled 1 and 2, while the five-cluster solutions are drawn in dotted lines and labelled a to e. I first discuss the results for electricians and professors and then the results for students. The most interesting feature of the five-cluster solutions in Figures 4 and 5 is the striking similarity between electricians and professors at this level of analysis. Both groups used very similar categories, which on the basis of respondents’ comments I labelled in the following way: a) professions; b) managerial/white-collar occupations; c) skilled trades; d) services/lower-skilled occupations; e) not elsewhere classified. These categories are the same for the two groups although there are differences in their composition. For example, professors include house painter among the skilled trades while electricians do not. Electricians include registered nurse and dental assistant among skilled trades while professors place them among the professions.[11]
The two-cluster solutions, by contrast, show interesting differences between the two groups. While their five-cluster solutions are nearly identical, electricians and professors do not draw the same boundaries among these clusters. In Figure 4, electricians place the professions and skilled trades together in one cluster (cluster 1) while lumping together managerial/white collar and service/lower-skilled occupations in a different cluster (cluster 2). This pattern had already been apparent during the interviews. When electricians were asked to group their piles into fewer categories, many placed skilled trades and professions together by opposition to lower-skilled occupations which “any John Doe could do”. In Figure 5, by contrast, professors make one cluster (cluster 1) containing professional and managerial/white-collar occupations and another cluster (cluster 2) with skilled trades and service/lower-skilled occupations. Their emphasis on levels of responsibility and authority in labelling their piles provides a clue as to the rationale for this classification, since responsibility and authority are characteristics shared by professionals and managers in their first cluster as opposed to service and trade occupations in their second cluster.
In clustering occupations, electricians emphasize the high educational requirements and usefulness common to both skilled trades and professions, by opposition to service/lower-skilled occupations, which require minimal training, and to managerial/white-collar occupations, which they neither admire nor perceive as very useful. On the personal respect and admiration scale, in Table 2, electricians give higher scores to the professions (7.16) and skilled trades (7.03) than to managerial/white collar (5.58) and service/lower-skilled (5.33) occupations. They also give these two groups higher scores on the scale of social usefulness, with 7.93 for skilled trades and 7.48 for professionals as opposed to 5.78 for service/lower-skilled and 5.67 for managers/white-collar occupations. With regard to prestige, electricians give highest scores to the professions (7.81), followed by managerial/white collar occupations (6.31), skilled trades (5.74), and service/lower-skilled occupations (4.47). Electricians acknowledge that managerial/white collar occupations have relatively high prestige in society, even though they don’t find these occupations very useful and they do not admire them very highly. Conversely, they admire skilled trades and professions equally and find them equally useful, while being aware that skilled trades have lower prestige in society in general. In a clear case of occupational egoism, electricians use normative evaluations of worth as an alternative to socioeconomic status to enhance the position of their own occupational group.
Table 2 - Mean prestige, admiration, and usefulness by cluster - Electricians
(N=20)
Cluster
Cluster
Mean
Mean
Mean
Name
Size
Prest.
s.d.
Adm.
s.d.
Use.
s.d.
a. Professions
11
7.81
0 .95
7.16
1.02
7.48
1.24
b. Managers/White-collar
8
6.31
1.14
5.58
0.90
5.67
0.90
c. Skilled Trades
13
5.74
1.25
7.03
0.82
7.93
0.87
d. Services/Low-skilled
27
4.47
0.88
5.33
0.54
5.78
0.76
e. Actor
1
7.75
5.80
5.00
Total
60
5.66
1.62
6.08
1.11
6.53
1.33
F:23.3 p<.001
F:18.2 p<.001
F:18.0 p<.001
Professors also give highest average scores to professionals (6.98) on the scale of respect and admiration (see Table 3), followed by skilled trades (4.83), managerial/white-collar (4.71), which they place on an equal footing, and service/lower-skilled occupations (3.66), which come last. A similar pattern is repeated with regard to social usefulness, with 7.61 for professionals, 6.20 for skilled trades, 5.81 for managers/white collar, and 4.95 for service/low-skilled occupations. On the prestige scales, the average scores of professors are 7.06 for professionals, 5.30 for managers/white-collar, 4.01 for skilled trades, and 3.21 for service/low-skilled occupations. University professors also exhibit some measure of occupational egoism, by placing their own occupational group at the top of the three scales and giving lower rankings to all other groups. Like electricians, they rank managerial/white collar occupations higher on occupational prestige than on the scales of usefulness and admiration.[12] However, by clustering together professional and managerial occupations, which are closer in terms of prestige than in terms of admiration and usefulness, they do not express the same reluctance as electricians in accepting prestige as a principle for organizing occupational space.
Table 3 - Mean prestige, admiration, and usefulness by cluster - Professors
(N=19)
Cluster
Cluster
Mean
Mean
Mean
Name
Size
Prest.
s.d.
Adm.
s.d.
Use.
s.d.
a. Professions
14
7.06
1.20
6.98
0.85
7.61
1.00
b. Managers/White-collar
7
5.30
1.29
4.71
1.00
5.81
0.68
c. Skilled Trades
12
4.01
0.60
4.83
0.63
6.20
0.51
d. Services/Low-skilled
25
3.21
1.11
3.66
1.01
4.95
1.05
e. Actor/Computer Prog.
2
6.34
0.11
6.43
0.21
6.39
0.37
Total
60
4.62
1.88
4.88
1.59
5.97
1.37
F:32.2 p<.001
F:32.4 p>.001
F:19.7 p<.001
Results for students are not radically different from those obtained for professors and electricians, but neither do they show the same level of similarity. While the overall layout of their configuration is similar to that of the other two groups, students draw slightly different boundaries between clusters. Like electricians and professors, students make a distinction at the five-cluster level between professions, managerial/white-collar occupations, and blue-collar/service jobs. As opposed to both professors and electricians, however, they make a more detailed differentiation among professional occupations and a less detailed one among blue-collar and service occupations. They distinguish two different clusters among professional occupations: cluster 1b includes health-related, education, and social services occupations, while cluster 1a includes miscellaneous professions such as computer programmer, lawyer, geologist, and electrical engineer.[13] Cluster 2c includes managerial-white collar occupations and is highly similar to the cluster made by the other two groups, but cluster 2d agglomerates together service/lower-skilled and skilled trades/blue-collar occupations. It is only at the six-cluster level of analysis that these two groups are differentiated.
At the two-cluster level of analysis, students lump together the two professional groups in cluster 1 and place all other occupations, including managerial, white-collar/service, and skilled trades, in cluster 2. Professional occupations in cluster 1 are those which they most personally admire, find most socially useful, and which also score relatively high on occupational prestige. Interestingly, students give much higher scores on usefulness (8.13) and admiration (7.63) to the health/education/social services occupations compared to what they perceive as their general level of prestige in society (6.95) (see Table 4).
Table 4 - Mean prestige, admiration, and usefulness by cluster - Students
(N=18)
Cluster
Cluster
Mean
Mean
Mean
Name
Size
Prest.
s.d.
Adm.
s.d.
Use.
s.d.
a. Misc. Professionals
7
7.82
1.35
7.40
1.10
8.27
1.10
b. Health/Soc./Education
9
6.95
1.38
7.65
0.92
8.13
1.01
c. Managers/White-collar
11
6.02
1.23
5.32
0.71
6.14
0.86
d. Services/Blue-collar
32
4.58
1.35
5.42
1.07
6.75
1.32
e. Actor
1
7.11
4.83
5.50
Total
60
5.62
1.78
5.96
1.36
7.00
1.38
F:12.4 p<.001
F:13.9 p<.001
F:6.3 p<.001
Like electricians, students thus consider that one occupational category is unfairly evaluated by society, although this category is not the same for the two groups. For electricians, it is the trades that are unfairly evaluated, while for students it is health/education/social service occupations. This might also be a case of occupational egoism, since several students expressed the ambition of pursuing careers in health and computer sciences. It also reflects the well- documented tendency among respondents to draw finer boundaries among groups which are closest to them in social space than among groups which are farther apart (Coleman and Rainwater, 1978; Gilbert and Kahl, 1982; Warner et al., 1960). The picture drawn here in fact closely resembles that of the class structure in Yankee City (Warner et al. 1960), in which respondents identified the same class structure but used normative criteria to draw different boundaries within it.
Cluster analysis reveals interesting patterns that would have been obscured had occupations been ranked on linear scales. It reveals a fair amount of occupational egoism, since all groups tend to upgrade their own (actual or anticipated) occupational group and downgrade others. Normative evaluations of worth are especially important in this respect, since they provide alternative criteria for evaluating occupations and drawing boundaries in social space, especially among working class respondents. It is mainly on the basis of these alternative criteria that respondents enhance their own occupational group and downgrade others. The discrepancy between socioeconomic status and worth is especially noteworthy in the case of managerial/white-collar occupations, since the three groups of respondents consistently rank these occupations lower on admiration and usefulness than on occupational prestige.
Conclusion
Durkheim’s claim that consensual evaluations of social positions constitute a moral framework which ensures social integration has long been set aside as a convincing interpretation of occupational prestige scales. Visions of a single, overarching status order conferring meaning to social life has given way to pictures of multiple spaces of distinction anchored in incommensurable criteria of evaluation. In the sociology of culture, symbolic representations are no longer seen as part of a coherent and unified pattern of norms and values, transmitted through socialization and unreflexively enacted in everyday life. Rather, culture is increasingly seen as a “grab bag of odds and ends” (DiMaggio, 1997: 267), consisting of disparate images, narratives, and categorical schemes which may or may not be activated in different social contexts.
The view of status as exclusively anchored in fragmented identities and individualized inequality, however, does not account for basic characteristics of prestige scales and for some of the major findings of the present research. It cannot explain the extraordinarily high level of consensus across classes, gender, and ethnic/racial groups concerning the position of occupations on prestige scales (Treiman, 1977) as well as the relatively high level of congruence between occupational prestige and the personal respect and admiration that electricians, professors, and students feel towards occupations. Nor does it explain that respondents produce strikingly similar pictures of occupational space, not only in this study but in similar research in different countries (see Magana et al. 1995). These findings suggest that Durkheim was not entirely wrong when he claimed that social positions are graded in the moral consciousness of societies in a way which maintains social cohesion and legitimates inequality (Durkheim, 1967:275ff). Prestige scales may be seen as widely-shared representations of what people perceive as dominant evaluations of social positions. In market-driven and ostensibly egalitarian late modern societies, these evaluations primarily reflect the material resources and advantages afforded by occupations.
The line between factual and normative elements, however, is not always clear. As evidenced by the relatively high correlation between prestige and worth among the three groups of respondents, occupational ratings do appear to have some legitimating significance with regard to social inequality. Yet the existence of intersubjective evaluations of positions does not necessarily entail uniform consensus on their legitimacy. University professors, who are privileged in terms of income and education, more readily accept them as legitimate compared to students and electricians. They see more congruence between socioeconomic status and worth and the boundaries they draw between themselves and others closely mirror occupational prestige ratings. Electricians do not entirely subvert the dominant prestige order, since they express high admiration for the professions which also rank highest on occupational prestige. Rather, they use normative criteria to draw symbolic boundaries between themselves and others in such a way that they place their own occupational group at the top of the hierarchy along with the professions. Occupational prestige, then, should be seen as one among many cognitive schemata available in collective consciousness for representing social structure, drawing boundaries, and evaluating others. These schemata are not enacted unreflexively by oversocialized social actors (Wrong, 1961) and they may not be relevant for organizing all aspects of everyday life. Along with normative evaluations of worth and a host of other criteria, they are tools for building “fences or bridges” (Douglas and Isherwood 1979: 12) in processes of inclusion and exclusion.
Notes
* This research was supported by a postdoctoral fellowship from the Social Sciences and Humanities Research Council and by research grants from Carleton University and the University of Ottawa. I thank Aage Sørensen, Terry Nosanchuck, Michèle Lamont, François Gobeil, and two anonymous CJS reviewers for their comments on earlier drafts. [back to text]
1 In this paper, the terms status and prestige are used interchangeably. Socio-economic status and occupational prestige refer to gradations of positions with regard to socio-economic resources. The more general terms of prestige and status refer to intersubjective evaluations of superiority and inferiority which have acquired an existence beyond individual beliefs and which influence social relations in various ways, e.g. through patterns of deference, association, and exclusion. [back to text]
2 Because of the very small scale of this study, the only aspect of social location examined here is occupation. This doesn’t mean that occupation is considered a priori more relevant than other dimensions of social location such as gender, race/ethnicity, or region. [back to text]
3 None of these groups constitutes a random sample and the number of interviews is arguably very small. Because of these limitations, this research should be considered “an exploratory exercise in idea-generation” (Markovsky, 2000: 556) rather than a formal attempt at hypothesis testing. [back to text]
4 The matrices and MDS configurations were produced with Anthropac 3.0 (Borgatti, 1990). [back to text]
5 PROFIT was performed with Anthropac 3.0 and clusters were produced using the between-group average method in SPSS 6.1.3. [back to text]
6 Asterisks indicate where respondents placed themselves. [back to text]
7 These levels are well within those reported in similar research. See Burton (1972), Kruskal and Wish (1978), and Ollivier (1997). [back to text]
8 The orientation of the scales and layout of occupations differ in each configuration but rotating the three figures to align the arrows would produce highly similar results. Readers should pay attention to the location of occupations relative to one another rather than to their absolute position in a fixed space. [back to text]
9 Data in this and all other tables were not collected from random samples. Levels of significance are provided for heuristic purposes only. [back to text]
10 In selecting the number of clusters which most adequately represent respondents’ perception of occupations, my main concern was not to determine the “true” number of clusters present in the data but rather to find a compromise between the need for detail and the need for economy. Research on free-sorting has shown that respondents tend to sort objects at different levels of generality: some respondents make fine distinctions while others create only a few broad categories (Coxon and Jones, 1979). As a result, a two-cluster solution is not necessarily better than the six-cluster one, but simply less detailed. In the present research, respondents classified their cards into 2 to 19 piles and they generally had no difficulty grouping their piles into broader categories when asked to do so. [back to text]
11 One electrician used the term “skilled trades for women” to describe these occupations. [back to text]
12 A similar pattern was noted by Gusfield and Schwartz (1963). Banker and lawyer ranked higher on prestige than on usefulness while the reverse was true for plumber and schoolteacher. [back to text]
13 The inclusion of word processing operator in this cluster appears somewhat anomalous. This could reflect the tendency among students to group more occupations according to occupational/industrial classification (e.g. travel agent, flight attendant, and air pilot), and thus to associate word processing operator with computer programmer. [back to text]
References
Blau, Peter M and O. D. Duncan
1967 The American Occupational Structure. New York: John Wiley and Sons.
Blau, Peter and Joseph E. Schwartz
1984 Crosscutting Social Circles. Orlando: Academic Press.
Borgatti, Stephen P.
1990 Anthropac Provisional Documentation.
Bourdieu, Pierre
1979 La distinction : critique sociale du jugement. Paris: Les Éditions de Minuit.
Burton, Michael
1972 “Semantic Dimensions of Occupation Names.” In Roger N. Shepard, A. Kimball Romney and Sara Beth Nerlove (Eds.), Multidimensional Scaling: Theory and Applications in the Behavioral Sciences, vol. 2. New York and London: Seminar Press.
Calhoun, Craig
1991 “Indirect Relationships and Imagined Communities: Large-Scale Social Integration and the Transformation of Everyday Life.” In Pierre Bourdieu and J. S. Coleman (Eds.), Social Theory for a Changing Society. Boulder: Westview Press.
Coleman. Richard P and Lee Rainwater
1978 Social Standing in America. New York: Basic Books.
Coxon, Anthony and Charles L. Jones
1979 Measurement and Meanings: Techniques and Methods of Studying Occupational Cognition. London: MacMillan.
1978 The Images of Occupational Prestige. London: MacMillan.
Cuneo, Carl J.
1996 “International Images of Social Inequality: A Ten-Country Comparison.” In Alan Frizzel and Jon H. Pammett (Eds.), Social Inequality in Canada. Ottawa: Carleton University Press.
Davis, Kingsley and Wilbert E. Moore
1945 “Some Principles of Stratification.” American Sociological Review 10:242–249.
Degraaf, Paul M., Harry B. Ganzeboom and Matthijs Kalmijn
1989 “Cultural and Economic Dimensions of Occupational Status.” In W. Jansen, J. Dronkers, and K. Verrips (Eds.), Similar or Different? Amsterdam: SISWO.
DiMaggio, Paul and John Mohr
1985 “Cultural Capital, Educational Attainment, and Marital Selection.” American Journal of Sociology 90(6):1231–1261.
DiMaggio, Paul
1987 “Classification in Art.” American Sociological Review 52:440–455.
1997 “Culture and Cognition.” Annual Review of Sociology 23:263–287.
Douglas, Mary and B. Isherwood
1979 The World of Goods. New York: Basic Books.
Duncan, O. D.
1961 “A Socio-Economic Index for all Occupations.” In A. J. Reiss (Ed.), Occupations and Social Status. New York: Free Press.
Durkheim, Émile
1967 Le suicide. Paris: Presses universitaires de France.
Erickson, Bonnie
1996 “Culture, Class, and Connexions.” American Journal of Sociology 102(1):217–235.
Featherman, David L. and Robert M. Hauser
1976 “Prestige or Socioeconomic scales in the Study of Occupational Achievement?” Sociological Methods and Research 4(4):403–22.
Gilbert, Dennis and Joseph Kahl
1982 The American Class Structure: A New Synthesis. Homewood, Ill.: Dorsey Press.
Goldthorpe, John H. and Keith Hope
1974 The Social Grading of Occupations: A New Approach and Scale. Oxford: Clarendon Press.
Grasmick, Harold G,
1976 “The Occupational Prestige Structure: A Multidimensional Scaling Approach.” The Sociological Quarterly 17:98–108.
Grusky, David B. and Stephen E. Van Rompaey
1992 “The Vertical Scaling of Occupations: Some Cautionary Comments and Reflections.” American Journal of Sociology 97(6):1689–1711.
Guppy, L. Neil
1984 “Dissensus or Consensus: A Cross-National Comparison of Occupational Prestige Scales.” Canadian Journal of Sociology 9(1):69–83.
Gusfield, J.R. and M. Schwartz
1963 “The Meanings of Occupational Prestige: Reconsideration of the NORC scale.” American Sociological Rewiew 28: 265–271.
Hall, John R.
1992 “The Capital(s) of Culture: A Nonholistic Approach to Status Situations, Class, Gender, and Ethnicity.” In Michèle Lamont and Marcel Fournier (Eds.), Cultivating Differences: Symbolic Boundaries and the Making of Inequality. Chicago and London: The University of Chicago Press.
Hope, Keith
1982 “A Liberal Theory of Prestige.” American Journal of Sociology 87(5): 1011–1031.
Hou, Feng and T. R. Balakrishnan
1996 “The Integration of Visible Minorities in Contemporary Society.” Canadian Journal of Sociology 21(3):307–326.
Kelley, Jonathan and M. D. R. Evans
1993 “The Legitimation of Inequality: Occupational Earnings in Nine Nations.” American Journal of Sociology 99(1):75–125.
Kluegel, James R. and Eliot R. Smith
1986 Beliefs About Inequality. New York: Aldine de Gruyter,
Krauss, Vered, O. E. Schild and Robert W. Hodge
1978 “Occupational Prestige in the Collective Conscience.” Social Forces 56(3):900–917.
Krushkal, Joseph and Myron Wish
1978 Multidimensional Scaling. Beverly Hills: Sage Publications.
Lamont, Michèle
2000 The Dignity of Working Men: Morality and the Boundaries of Race, Class, and Immigration. Cambridge: Harvard University Press and Russell Sage Foundation.
Lamont, Michèle and Marcel Fournier
1992 Cultivating Differences: Symbolic Boundaries and the Making of Inequality. Chicago and London: University of Chicago Press.
Macleod, Jay
1987 Ain’t No Making It: Leveled Aspirations in a Low-Income Neighborhood. Boulder: Westview Press.
Magana, Raul, Michael Burton and Joao Ferreira-Pinto
1995 “Occupational Cognition in Three Nations: Relationships between Classification and Evaluation.” Journal of Quantitative Anthropology 5(2):149–168.
Markovsky, Barry
2000 “Review of Karin Knorr Cetina’s Epistemic Cultures: How the Sciences Make Knowledge.” Contemporary Sociology. 29(3): 556–557.
Meiselas, Susan
1976 Carnival Strippers. New York: Farrar, Strauss & Giroux.
Melucci, Alberto
1996 The Playing Self: Person and Meaning in the Planetary Society. Cambridge: Cambridge University Press.
Milner, Murray
1996 “Geeks, Weirdos, and Cool Kids: Status Among American Adolescents.” Paper presented at the Annual Meeting of the American Sociological Association, New York.
Nam, Charles B. and E. Walter Terrie
1982 “Measurement of Socioeconomic Status from United States Census Data.” In Mary G. Powers (Ed.), Measures of Socioeconomic Status: Current Issues. Bouldner: Westview Press.
Ollivier, Michèle
1994 Prestige and Social Closure in the Arts. Unpublished PhD Dissertation, Department of Sociology, Harvard University, Cambridge, MA.
1997 “Measuring Symbolic Boundaries Among Artists,” Poetics 24:299–328.
Ostrower, Francie
1996 “Status and Elite Philanthropy: Hierarchies and Processes.” Paper presented at the Annual Meeting of the American Sociological Association, New York.
Pammett, Jon H.
1996 “Getting Ahead around the World,” In Alan Frizzell and Jon H. Pammett (Eds.), Social Inequality in Canada, Ottawa: Carleton University Press.
Parkin, Frank
1971 Class Inequality and Political Order. London: MacGibbon & Kee.
Reiss, Albert J.
1961 Occupations and Social Status. New York: Free Press.
Shils, Edward
1975 Center and Periphery: Essays in Macrosociology. Chicago and London: University of Chicago Press.
Sørensen, Aage
1992 “Concluding Commentary: Class, Status and Power.” In David G. Grusky (Ed.), Social Stratification: Class, Gender and Race in Sociological Perspective. Boulder: Westview Press.
Statistics Canada
1980 Standard Occupational Classification. Catalogue no. 93–327, Ottawa.
Stehr, Nico
1974 “Consensus and Dissensus in Occupational Prestige.” British Journal of Sociology 25(4):411–426.
Treiman, Donald J.
1977 Occupational Prestige in Comparative Perspective. New York: Academic Press.
Warner, Lloyd W, Marcia Meeker and Kenneth Eels
1960 Social Class in America. New York: Harper and Row.
Wrong, D.
1961 “The Oversocialized Conception of Man in Modern Sociology.” American Sociological Review 26:183–193.
Appendix
(click on title to view figure full-size in a new window)
“Too Much Money Off Other People’s Backs”:
Status in Late Modern Societies*
Michèle Ollivier
Canadian Journal of Sociology 25, 4 (Fall 2000): 441-70.
Abstract: This paper revisits debates on occupational prestige in light of recent shifts in sociological conceptions of status, culture, and identity. Using multidimensional scaling and clustering of data collected from electricians, university professors, and students in adult education, it explores how people differently located in social structure perceive occupations. Results indicate that prestige scales are one among many cognitive schemata available in collective consciousness for representing social structure, drawing symbolic boundaries, and evaluating others. How this is done, however, varies with social location. Compared to electricians and students, professors see more congruence between occupational prestige and worth. Electricians and students use normative evaluations of worth as alternative criteria for evaluating occupations and drawing boundaries, in an effort to enhance their own social position and downgrade others.
Résumé: Cet article examine les débats sociologiques entourant le prestige occupationnel à la lumière des transformations récentes des concepts de statut, de culture et d’identité. Sur la base d’une analyse multidimensionnelle et typologique de données recueillies auprès d’électriciens, d’étudiant-es et de professeur-es d’université, il explore comment des acteurs différemment situés dans l’espace social perçoivent les occupations. Les résultats indiquent que les échelles de prestige constituent dans la conscience collective un schéma cognitif parmi d’autres permettant de représenter la structure sociale, d’y tracer des frontières symboliques et d’évaluer les occupations. L’utilisation de ces schémas cognitifs varie cependant selon la position sociale. Les professeur-es perçoivent une forte corrélation entre le prestige et certains critères normatifs d’évaluation des occupations. Les électriciens et les étudiant-es, par contre, utilisent ces derniers pour tracer différentes frontières symboliques qui rehaussent leur position sociale.
Contents:
Introduction | What is Status? | Occupational Prestige and Normative Evaluations of Worth | Research Design | Results | Conclusion | Notes | References | Appendix |
Introduction
The development of occupational prestige scales by North and Hatt, in the 1940s, gave rise to heated debates about the meaning of these scales and the inferences that could be drawn from them. Sociologists emphasizing the consensual nature of occupational ratings argued that they expressed collective beliefs about the worthiness of occupations (Shils, 1975; Treiman, 1977). Positions which ranked highest on the scales were those which were most highly valued by a collectivity, either because of their functional importance (Davis and Moore, 1945) or because of their proximity to order-creating values and institutions (Shils, 1975). This interpretation fit in very well with neo-Durkheimian and functionalist theories of stratification, according to which consensual evaluations of positions constitute a necessary foundation for social life. They provide a shared normative framework which ensures social integration and legitimates inequality.
Normative interpretations, however, were never very popular among sociologists who used prestige scales in empirical research. While most remained agnostic on the assumptions and implications underlying their use in research, some argued that occupational prestige reflects factual knowledge about the material rewards attached to occupations (Featherman and Hauser, 1976; Nam and Terrie, 1982). A commonly accepted interpretation, proposed by Goldthorpe and Hope (1974), is that occupational prestige scales measure the “desirability” of occupations in terms of socioeconomic rewards. Other things being equal, they argue, people generally agree that positions offering high rewards are more desirable than those affording low rewards. High levels of consensus among raters simply mean that people are generally aware of the economic and cultural advantages attached to occupations and value them as desirable accordingly. Goldthorpe and Hope (1974) acknowledge that their concept retains an evaluative component, but contend that this component has no normative connotation and no legitimating significance with regard to social inequality. After years of disputes the debate was finally settled. Prestige scales are now widely believed to reflect factual evaluations of the economic and cultural advantages afforded by occupations rather than normative evaluations of their worth (DiMaggio and Mohr, 1985; Featherman and Hauser, 1976; Goldthorpe and Hope, 1974; Grusky and Rompaey, 1992; Sørensen, 1992).
This position, however, leaves many questions unanswered, namely concerning the nature of prestige in contemporary societies as well as the relationship between socioeconomic status and normative evaluations of worth.[1] If occupational prestige scales are based on factual evaluations of material rewards, how do people conceive of the worthiness of occupations? Was Durkheim mistaken when he argued that a normative ordering of positions is an essential integrative feature of all stratified societies, or have society-wide, consensual evaluations become irrelevant in late modern times? Have they given way to a multiplicity of local status orders, as suggested by much contemporary sociological theory?
Making sense of occupational prestige in light of changing conceptions of status and identity is the main objective of this paper. The first section discusses how recent shifts in sociological theory have altered how status is viewed in contemporary societies. The neo-Durkheimian assumption that occupational prestige forms the backbone of a consensual and tightly-woven pattern of norms and values no longer appears tenable in light of the multiplicity of identities and hierarchies displayed in contemporary social life. Section 2 reviews existing research on occupations which shows, first, that occupational structure and occupational prestige are generally perceived as multidimensional, secondly, that the distinction between factual and normative components of occupational ratings is not always clear, and, thirdly, that beliefs about normative aspects of occupations differ according to people’s location in social structure. The remainder of the paper presents data collected during interviews with electricians, university professors, and students in adult education, in a research explicitly designed to examine how people differently located in social structure draw boundaries in occupational space and how they assess the worthiness of occupations. I argue that occupational prestige scales reflect what people perceive as a dominant system of assessment of social positions. Along with normative and other criteria of evaluation, they are used in various ways by groups differently located in social structure for drawing boundaries between themselves and others.
What is Status?
One of the main arguments justifying the rejection of normative interpretations of occupational prestige is that respondents to prestige surveys overwhelmingly mention material advantages of occupations when asked to explain their ratings (Coleman and Rainwater, 1978; Goldthorpe and Hope, 1974; Reiss, 1961). The rejection of normative interpretations, however, also reflects a major shift in sociological theory. In light of the complexity and anonymity of social life in advanced market societies, few sociologists now accept the Durkheimian assumption that a single, overarching status order permeates social life and exercises a systematic influence on social relations.
A first argument supporting this view is that contemporary societies are highly complex and segmented, so that people have multiple social affiliations which do not necessarily coincide (Blau and Schwartz, 1984; DiMaggio and Mohr, 1985; Melucci, 1996). Complexity and segmentation contribute to the emergence of a diversity of status cultures, which people learn to selectively display in appropriate social contexts (DiMaggio, 1987; Erickson, 1996). A second argument is that contemporary societies have witnessed a phenomenal increase in the breadth and anonymity of social relations, as more people come into more fleeting contact with others in a diversity of social settings (Calhoun, 1991). High levels of geographical and social mobility, made possible by the rapid expansion of means of transportation and new communication technologies, coupled with the extraordinary increase in the amount and diversity of information available through the mass media, have led to major changes in processes of identity formation and status competition. As opposed to relatively small communities, where status is linked to personal knowledge of fixed identities based on family name, reputation, kinship, and place of residence, enlarged and anonymous social spaces provide individuals with vastly increased possibilities of experimenting with strategies of self-presentation in everyday life. Combined together, segmentation and anonymity lead to the emergence of multiple and heterogeneous status cultures, anchored in constantly shifting identities and hierarchies (Hall, 1992).
Late modern societies are thus seen as comprising multiple, segmented, and sometimes incommensurable normative orders, in which status operates predominantly at micro or local levels: among different fractions of classes (Bourdieu, 1979; Coleman and Rainwater, 1978; Hope, 1982; Stehr, 1974), between adjacent or competing occupational groups (Blau and Duncan, 1967:63–70), and within small communities such as popular music artists (Ollivier, 1994), carnival strippers (Meiselas, 1976), upper class philanthropists (Ostrower, 1996), and adolescents (Milner, 1996). Rather than representing a society-wide moral classification, status is linked to the multilayered symbolic boundaries of race, gender, class, and lifestyle that people draw between themselves and others in processes of inclusion and exclusion (Lamont and Fournier, 1992). Occupations, in this sense, are only one of numerous dimensions of identity and status inequality.
Even if occupations are no longer considered the most salient dimension of inequality, questions remain on whether and how they are used for making status distinctions and drawing boundaries in everyday life. How do people draw boundaries around occupations? Do these boundaries have normative connotations? Are incumbents of occupations at the top of the scales considered more deserving than those at the bottom? Is their success an indication of superior moral qualities or abilities? Are they viewed as making more substantial contributions to the well-being of society than those at the bottom? In the following section, I review research on how people think about occupations in search of partial answers to some of these issues.
Occupational Prestige and Normative Evaluations of Worth
There are good reasons to believe that people perceive occupational structure as having more than one dimension. Research using multidimensional scaling (MDS), in particular, generally shows that people do not think of occupations in terms of a linear scale (see Burton, 1972; Coxon and Jones, 1978; Magana, Burton and Ferreira-Pinto, 1995; Krauss, Schild and Hodge, 1978). MDS studies invariably show that occupational prestige is one of the most salient dimensions of people’s representations of occupational structure, but most of them also find evidence of additional dimensions. These dimensions are interpreted in many different ways: as service orientation and skills/trade by Coxon and Jones (1978), as job autonomy and training by Burton (1972), and as an opposition between power-oriented or bureaucratic occupations on the one hand and the healing, academic, and autonomously-oriented professions on the other by Magana et al. (1995).
Furthermore, research also suggests that occupational prestige itself should be understood as comprising more than one dimension. Reviewing this issue, Grusky and Rompaey conclude that occupational prestige scales should be interpreted as the best one-dimensional representation of “an intrinsically two-dimensional occupational structure”, whose main dimensions are economic and sociocultural resources (1992:1719). What is meant exactly by sociocultural resources, however, is open to divergent interpretations. One interpretation is that occupational prestige compounds in a single scale different types of resources. Supporting evidence for this argument is garnered from the work of Bourdieu (1979), who conceptualizes occupational structure as a two-dimensional space with cultural and economic capital as its main dimensions, from the empirical work of DeGraff, Ganzeboom, and Kalmijn (1989), who operationalise and test the usefulness of these two dimensions for analysing lifestyles, and even from Duncan’s (1961) finding that occupational prestige represents a combination of income and education.
A competing interpretation, however, is that prestige scales have both a factual and a normative component. Using polar coordinate analysis, Hope (1982) shows that occupational prestige is an average of two dimensions: the standard of living afforded by occupations and an assessment of their value to society (Hope, 1982). Similarly, Grasmick (1976) uses multidimensional scaling to show that vectors measuring the social standing of occupations reflect a combination of material rewards and value to society, although the value to society dimension only operates in the middle range of the material rewards scale. Interestingly, occupations at opposite ends of Grasmick’s social value dimension (1976:101) are very similar to those which Bourdieu places at opposite ends of his cultural-economic capital dimension (1979:140–141). In both cases, the middle range of occupational space is characterized by an opposition between occupations such as teachers and social workers on the one hand, and more commercial occupations such as small store owners and managers on the other hand. The “value to society” gradation and the “cultural vs economic capital” dimensions could thus be considered as different interpretations of the same dimension of occupational space.
These two interpretations, however, have very different implications. While cultural resources may be considered as advantages attached to occupations, the same cannot be said of social usefulness. Social usefulness is neither a material nor a symbolic resource, but rather a measure of intangible contributions that people make to the general well-being of society. Which of these interpretations is correct cannot be resolved here, but Hope (1982) and Grasmick’s (1976) research suggests that the separation between normative and factual interpretations of prestige scales may not be as clear-cut as previously thought. As argued by Hope (1982), this does not mean that occupations which rank high on prestige scales command deference in everyday life, but it suggests that prestige scales do have some legitimating significance with regard to social inequality.
In this respect, studies of people’s perception of inequality show that in advanced industrial societies, the unequal distribution of rewards is largely perceived as legitimate (Cuneo, 1996; Lamont, forthcoming; Kluegel and Smith, 1986; Pammett, 1996). Belief in equality of opportunity is widespread, and to the extent that inequality of condition is perceived as reflecting individual qualities such as ambition, talent, and hard work, it is generally considered as fair. Several studies, however, also show that congruence between occupational prestige and normative conceptions of worth is much stronger at the top of the socioeconomic status hierarchy than at the bottom. People in high-status occupations are more likely to perceive their society as open and egalitarian (Cuneo, 1996; Kelley and Evans, 1993) and to explain existing inequality in terms of individual as opposed to structural variables (Kluegel and Smith, 1986). People at the bottom are more likely to dissociate socioeconomic status and worth, namely by emphasizing alternative sets of moral and cultural standards based on personal integrity, sincerity, and quality of interpersonal relationships (Lamont, forthcoming). Similarly, Guppy (1984) has shown that the level of consensus on occupational prestige is much higher among more privileged respondents than among less privileged ones. These findings fit in well with Parkin’s argument (1971) that different classes need to account for social inequality in different ways and thus develop different meaning systems: the dominant value system is a “moral framework which promotes the endorsement of existing inequality” (1971:81), the subordinate value system is oriented towards accommodation to dominant values, and the radical system promotes an oppositional stance towards social inequalities.
Occupational prestige scales, I argue, appear to tap into what people perceive as a dominant value system. In egalitarian and secular contemporary societies, this system largely reflects material advantages but it may also include evaluations of the social value of positions. How closely people endorse this value system, however, shows some variation which is at least in part a function of social location. Building on these insights, the remainder of this paper examines how people differently located in social structure perceive occupational space as well as the relationship between socioeconomic status and worth.[2] I expect that people will perceive occupations in terms of a multidimensional structure of which occupational prestige will be a major dimension. I also expect that beliefs about the worthiness of occupations will vary according to people’s location in social structure, with greater congruence between socioeconomic status and worth at the top than at the bottom. Whether people differently located in social structure draw the same or different boundaries in occupational space and how these boundaries are affected by normative evaluations of worth remain open issues.
Research Design
Multidimensional scaling (MDS) and hierarchical clustering of free-sorting data provide an inductive yet systematic approach for mapping out cognitive domains and measuring some of their dimensions. MDS transforms data on the perceived similarities between objects into spatial distances. It produces a visual representation, in two or more dimensions, of the relationships among objects. Objects which are most similar appear close together and those which are most dissimilar are far apart. MDS provides a visually attractive representation of similarities and distances among objects. It can also be used as a support for representing clusters into space.
Following standard procedures in MDS (Coxon and Jones, 1978, 1979; Burton, 1972; Magana, Burton and Ferreira-Pinto, 1995; Ollivier, 1997), I first asked three groups of respondents to sort 60 cards marked with occupation titles into any number of piles, in the way which felt most natural to them. In open-ended interviews, I asked them to describe their piles, to locate themselves in one of them, and to collapse them, if possible, into broader categories. Finally, I asked them to rate occupations on three different scales: their prestige or standing in society in general, their social usefulness, and the respect and admiration they personally felt for them. With regard to occupational prestige, respondents were instructed to rate occupations according to how they thought they were viewed in society in general. The other two scales are based on respondents’ personal views and are thought to more closely reflect their perception of the worthiness of occupations (for a similar use see Coxon and Jones, 1978).
Sixty occupations were selected from the Canadian 1980 Standard Occupational Classification (Statistics Canada, 1980), following a two-step procedure. In order to ensure that the occupations selected would be roughly representative of their distribution in the Canadian labour force, I first computed the proportion of the Canadian labour force in each of the Major Groups of the 1980 National Occupational Classification. I then selected the corresponding number of occupations within each of the Major Groups. The main criteria guiding the selection process at this point was that occupation titles be easily recognizable and different from one another. For example, there would have been no point in including both Chemical Engineers and Aerospace Engineers, since these occupations would likely be considered highly similar by most people, except perhaps by chemical and aerospace engineers themselves.
Because a main objective of this research is to explore whether people differently located in social structure perceive and evaluate occupations in different ways, I selected respondents among different occupational groups with a view of maximizing social distance between them. Keeping in mind Bourdieu’s (1979) two-dimensional representation of social space, in which the vertical dimension reflects total amount of capital and the horizontal dimension refers to an opposition between economic and cultural capital, I selected respondents in what I hypothesized would be different regions of social space. Interviews were conducted from January to June 1997 with electricians in a union training centre and university professors in a Canadian university. University professors were selected because of their overall high level of economic and cultural capital. Professors from the Departments of Mechanical and Aerospace Engineering, Business, English, and Religious studies were solicited by email and 19 interviews were conducted with those agreed to participate.[3] My intention was to contrast professors in the Humanities with those in the more technical fields of Business and Engineering. However, since only three respondents in the Humanities agreed to be interviewed compared to nine in Business and seven in Engineering, the results presented here are aggregated. Socio-demographic data collected during the interviews indicate that all respondents have completed graduate school and their mean income is $74,000. Only 4 out of 19 are women and their mean age is 47 years old.
Electricians were selected because they represent a traditional blue-collar occupation. They have lower levels of total capital compared to university professors, but higher levels of cultural capital in terms of formal and on-the-job training compared to many blue-collar and service occupations. Electricians were contacted through their local union, which organizes regular training sessions for its members. Group interviews were conducted by myself and a research assistant with a total of 19 electricians. Respondents worked in the same room and short individual interviews were conducted with each respondent as they finished sorting the cards. The electricians interviewed have lower levels of cultural and economic capital compared to university professors, but still relatively high levels of education. All but one have at least some post- secondary education and 65% hold a vocational college degree. Their mean household income is $47,000. All of them are male with a mean age of 36 years old.
Students in adult education were interviewed one year later. Because they were working towards a high school degree, I expected these students to have lower levels of economic and cultural capital compared to electricians, and to hold jobs requiring little education and offering low benefits. I was especially interested in comparing their views with those of electricians, who emphasized their high levels of skills in differentiating themselves from service and white-collar occupations requiring little or no training. With the help of a research assistant, I interviewed 18 students in a grade 12 mathematics class. Answers to socio-demographic questions revealed lower levels of education and household incomes compared to electricians, but also more diversity than I expected. Only one student had not already completed a high school degree at the time of the interview and 61% had at least some post-secondary education. Many were taking high school math to reorient themselves towards new careers in health care and computer science. They were equally divided between males and females and they were on average younger than the other two groups, with a mean age of 30 years old. Their mean household income was $39,000. They were also more diverse in terms of ethnicity and language compared to the other groups. Fifty percent were born outside Canada and one third (33%) spoke languages other than English or French at home. By contrast, almost all electricians (95%) were born in Canada and only two (11%) spoke languages other than English or French at home. A large percentage (42%) of university professors were born outside of Canada, but only 16% spoke a language other than English or French at home. Many of the students interviewed were recent immigrants, with an imperfect knowledge of English and high expectations of upward mobility. This introduced an unanticipated element in the analysis, since immigrant status has been shown to have an effect both on belief in the ideology of equality of opportunity and on occupational aspirations (Hou and Balakrishnan, 1996; Macleod, 1987).
For each respondent, a square co-occurrence matrix was computed, in which pairs sorted in the same pile received a score of 1 and pairs sorted in different piles received 0. Individual matrices were summed to produce an aggregate matrix for each group of respondents. The three aggregate matrices were then separately submitted to multidimensional scaling using the non-metric Minissa program.[4] The resulting configurations were interpreted using Property Fitting (PROFIT) and used as a support for displaying the results of hierarchical clustering.[5] PROFIT analysis is a regression technique which fits selected attributes of the objects into MDS configurations. PROFIT creates a set of coordinates showing the orientation of each attribute as an axis in the configuration. It provides a multiple R measure, which is taken as an indication of the salience of each variable in the configuration. As opposed to PROFIT, which identifies continuous dimensions in each configuration, cluster analysis allows for the identification of discrete groups. The two techniques provide complementary pictures of the same data set.
Results
Respondents’ Comments
One of the advantages of using the free sorting method to obtain similarity data is that it allows respondents to explain the criteria they used when sorting items into piles. Respondents used a variety of labels and criteria to describe their piles and almost all of them combined several criteria. There were many similarities between the three groups but also noteworthy differences. In all groups, the most often used criteria were levels of education and occupational/industrial sectors (e.g. primary resources, construction, travel, health care, etc.). Occupational sector was used to label at least one category by 15 professors, 12 students, and 10 electricians, while education and skills was used in at least one category by 11 electricians, 10 students, and 9 professors. A typical sorting by an electrician combining these two criteria is: a) actor; b) managerial; c) service-oriented; d) work alone; e) financial things; f) manual labour; g) higher learning; h) skilled trades with apprenticeship program*.[6] A few respondents stated that they could have classified occupations in different ways, using either occupational sector or education as the main criterion.
While education and occupational/industrial sectors were mentioned by a majority of respondents in the three groups, other criteria were specific to each group. Electricians and students, in particular, were much more likely than university professors to use criteria with normative connotations. Eight electricians and three students used criteria referring to respect or contempt, with categories labelled “crooks”, “overblown”, “overrated” or “overpaid”. Five electricians and three students mentioned criteria dealing with occupations’ usefulness to society or to themselves, and two electricians classified their cards according to their personal preferences, e.g. “jobs I would do as opposed to jobs I wouldn’t like to do”. These normative criteria were also often used in combination with one another as well as with education and occupational sectors. The following is a typical example: a) people we need; b) not too much skills; c) no impact on my life (almost useless, like dust on the shelf); d) help things run efficiently; e) people who provide help; f) too much money off other people’s backs; g) lots of schooling, do the best you can*. Another example is: a) white collar that I respect; b) blue-collar that I respect*; c) blue and white collar that I don’t respect and wouldn’t like to do.
Only one professor referred to the usefulness of occupations and none used labels such as “overblown” or “overpaid”. Four professors and five students used a high vs low metaphor to describe some of their categories, for example: “middle-range people, neither top nor bottom”; “high class”; “top of the rank* vs end of the pole”; “I don’t like terms like higher and lower but...”. None of the professors, however, even among those who used a high-low metaphor, used labels suggesting that some occupations were evaluated too highly.
Categories frequently used by professors but more rarely by electricians and students include the degree of authority or responsibility (10 professors), the existence of professional registration or some other form of certification (3 professors), and gender (3 professors). Furthermore, 12 professors out of 19 used the term “professionals” to label one of their piles compared to only 2 electricians and 5 students. This term was often associated with high income, high levels of skills, and professional registration. In addition, several professors made a distinction between “experts”, which they defined as highly skilled professionals who actually use their expertise in their work, and “managers” who are also highly educated but need more general skills to do their job.
Very few people explicitly mentioned income as a criteria. The nine respondents who mentioned income did so in relation to some type of evaluative criteria, for example “overpaid” or “skilled, but not enough to make big bucks”. The two respondents who used preferences as the dominant criterion mentioned income along with prestige, type of work, type of person, challenge, etc. when asked to explain further. Several respondents used criteria which were directly linked to their jobs. One professor in the Department of Religion made two piles: “people who work alone or inside themselves vs people who work with others or outside themselves”. Similarly, one respondent from the Department of Aerospace Engineering used the “severity of consequences for doing a job badly” as a main criterion when making his piles. He explained that an aerospace engineer who does a job badly endangers human life, while a janitor who performs badly only produces dirty premises.
These comments suggest the existence of common grounds between the three groups, since criteria such as education and occupational sector are widely used by all respondents. These comments also indicate that there is variation in how groups describe and evaluate occupations: electricians and, to a lesser extent, students are more likely than professors to consider that some occupations do not deserve the rewards and/or prestige that accrue to them. Similarly, professors place more emphasis on authority and responsibility compared to electricians and students. These comments will be used in the following sections in interpreting the MDS configurations and clusters.
Multidimensional Scaling and Profit Analysis
The aggregate co-occurrence matrices for the three groups were scaled in two and three dimensions. As often occurs in MDS studies, the interpretation of a third dimension proved difficult and for the sake of simplicity, only the two-dimensional solution is discussed here. Stress levels, which measure how well the configuration fits the data, are .139 for university professors, .148 for electricians, and .155 for students.[7] Results are displayed in Figures 1 to 3 in appendix. The three vectors in each configuration were generated by PROFIT analysis. They represent the three scales on which ratings were obtained from respondents. As expected, the most salient dimension in each configuration, as measured by multiple R, is occupational prestige. Multiple R on prestige is .937 for professors, .857 for electricians, and .814 for students. All groups place professional occupations at one extreme of the prestige scale (university professor, lawyer, electrical engineer, psychologist, physician, etc.) and low-skilled, low-paying occupations at the other extreme (janitor, hairdresser, service station attendant, courier, freight handler, etc.).[8]
The second dimension is more difficult to interpret, especially in the absence of precise measures of other attributes of occupations. Visual examination of the configuration in light of respondents’ comments suggests that the second dimension could be interpreted as an opposition between expert/technical/blue-collar occupations such as geologist, architect, electrical engineer, firefighter, plumber, welder, fisherman, and tailor on the one hand, and white-collar/administrative occupations such as hospital administrator, accountant, executive secretary, sales supervisor, travel agent, and bank teller on the other. The three configurations are highly similar in this respect. They are also highly similar to the configurations produced by Magana et al. (1995), who interpret their second dimension as an opposition between power-oriented or bureaucratic occupations on the one hand and the healing, academic, and autonomously-oriented professions on the other. This indicates that groups differently located in social structure do not have radically different views of occupational space.
The vectors measuring admiration and usefulness are relatively salient in all three configurations, but their orientation differs markedly in the three groups. The three vectors are close together and oriented in the same direction for professors in Figure 2. For electricians in Figure 1, by contrast, the prestige vector is oriented at an angle from both the usefulness and admiration vectors. The orientation of these vectors indicates that university professors generally think that occupations which have highest prestige in society are also those which are most socially useful and for which they have the most admiration. Electricians, by contrast are more likely to believe that occupations with high prestige in society are not necessarily those which are most useful and deserve highest admiration. This conclusion is supported by electricians’ comments that some occupations are overpaid and overrated. As for students, they occupy a middle position between the other two groups. They see more discrepancy between the three scales than university professors but less than electricians. Pearson correlations between the three scales, displayed in Table 1, support this interpretation. Occupational prestige is more highly correlated with the normative scales of admiration and usefulness among professors than among the other two groups. Pearson correlations between prestige and admiration are .89 for professors, .79 for students, and .72 for electricians. Correlations between prestige and usefulness are .79 for professors, .64 among students, and .50 among electricians.
Table 1 - Pearson Correlations between prestige, admiration, and usefulness -
electricians, professors, students
| Prestige | Admiration | Usefulness | ||||||||
| prof. | stud. | elec. | prof. | stud. | elec. | prof. | stud. | elec. | ||
| Prestige | ||||||||||
| professors | 1.00 | .953 | .958 | .895 | .825 | .683 | .788 | .640 | .461 | |
| students | 1.00 | .954 | .815 | .792 | .626 | .708 | .635 | .429 | ||
| electricians | 1.00 | .872 | .815 | .715 | .775 | .648 | .503 | |||
| Admiration | ||||||||||
| professors | 1.00 | .887 | .810 | .931 | .762 | .628 | ||||
| students | 1.00 | .846 | .906 | .884 | .744 | |||||
| electricians | 1.00 | .853 | .819 | .909 | ||||||
| Usefulness | ||||||||||
| professors | 1.00 | .875 | .771 | |||||||
| students | 1.00 | .854 | ||||||||
| electricians | ||||||||||
All correlations are significant at the .001 level [9]
Three main conclusions can be drawn from these results. First, people differently located in social structure do not have radically different pictures of this structure, at least in the artificial context of the present research. When people are asked to sort into piles the same set of occupations, they produce pictures whose general layout is remarkably similar. Second, the congruence between occupational prestige and respondents’ personal evaluations of occupations is relatively high, with correlations ranging from .72 among electricians to .89 among professors. This is in line with research on perception of inequality, which shows that the unequal distribution of rewards in advanced industrial societies is widely perceived as legitimate and fair. Third, where one is located in social structure does influence normative evaluations of worth. Congruence between socioeconomic status and worth is highest among professors, who rank at the top of the three groups in terms of income and education. It is lowest among electricians, whose views closely match Parkin’s (1971) subordinate value system. Electricians largely accept as legitimate the dominant value system, as evidenced by the relatively high correlation between occupational prestige and personal admiration. They do however emphasize alternative criteria of evaluation, as shown by the lower correlation between prestige and social usefulness. Their comments that some occupations are “overblown” or “overpaid” also indicate that they are more likely than professors and students to think that the actual distribution of rewards, which underlies occupational prestige scales, is not entirely justified. While students have the lowest levels of education and income among the three groups, they occupy a middle position between high-status university professors and the working class subculture of electricians. They do not share the oppositional stance of electricians, but neither do they wholeheartedly embrace professors’ close association between socioeconomic status and worth. Given that most of these students were taking high-school math to improve their current employment situation, we can assume that they did not expect to remain in low-income and low-education positions.
These results, while interesting, are in line with previous research on perception of inequality, which shows that association between socioeconomic status and worth is greater among people located at the top of the occupational hierarchy than among those at the bottom. In the following section, cluster analysis shows that respondents in the three groups draw very different boundaries within social space, and that how boundaries are drawn is closely linked to normative evaluations of worth.
Hierarchical Clustering
The results of hierarchical clustering at the two-cluster and five-cluster levels are presented in Figures 4 to 6 in the appendix.[10] The two-cluster solutions are represented by solid lines and labelled 1 and 2, while the five-cluster solutions are drawn in dotted lines and labelled a to e. I first discuss the results for electricians and professors and then the results for students. The most interesting feature of the five-cluster solutions in Figures 4 and 5 is the striking similarity between electricians and professors at this level of analysis. Both groups used very similar categories, which on the basis of respondents’ comments I labelled in the following way: a) professions; b) managerial/white-collar occupations; c) skilled trades; d) services/lower-skilled occupations; e) not elsewhere classified. These categories are the same for the two groups although there are differences in their composition. For example, professors include house painter among the skilled trades while electricians do not. Electricians include registered nurse and dental assistant among skilled trades while professors place them among the professions.[11]
The two-cluster solutions, by contrast, show interesting differences between the two groups. While their five-cluster solutions are nearly identical, electricians and professors do not draw the same boundaries among these clusters. In Figure 4, electricians place the professions and skilled trades together in one cluster (cluster 1) while lumping together managerial/white collar and service/lower-skilled occupations in a different cluster (cluster 2). This pattern had already been apparent during the interviews. When electricians were asked to group their piles into fewer categories, many placed skilled trades and professions together by opposition to lower-skilled occupations which “any John Doe could do”. In Figure 5, by contrast, professors make one cluster (cluster 1) containing professional and managerial/white-collar occupations and another cluster (cluster 2) with skilled trades and service/lower-skilled occupations. Their emphasis on levels of responsibility and authority in labelling their piles provides a clue as to the rationale for this classification, since responsibility and authority are characteristics shared by professionals and managers in their first cluster as opposed to service and trade occupations in their second cluster.
In clustering occupations, electricians emphasize the high educational requirements and usefulness common to both skilled trades and professions, by opposition to service/lower-skilled occupations, which require minimal training, and to managerial/white-collar occupations, which they neither admire nor perceive as very useful. On the personal respect and admiration scale, in Table 2, electricians give higher scores to the professions (7.16) and skilled trades (7.03) than to managerial/white collar (5.58) and service/lower-skilled (5.33) occupations. They also give these two groups higher scores on the scale of social usefulness, with 7.93 for skilled trades and 7.48 for professionals as opposed to 5.78 for service/lower-skilled and 5.67 for managers/white-collar occupations. With regard to prestige, electricians give highest scores to the professions (7.81), followed by managerial/white collar occupations (6.31), skilled trades (5.74), and service/lower-skilled occupations (4.47). Electricians acknowledge that managerial/white collar occupations have relatively high prestige in society, even though they don’t find these occupations very useful and they do not admire them very highly. Conversely, they admire skilled trades and professions equally and find them equally useful, while being aware that skilled trades have lower prestige in society in general. In a clear case of occupational egoism, electricians use normative evaluations of worth as an alternative to socioeconomic status to enhance the position of their own occupational group.
Table 2 - Mean prestige, admiration, and usefulness by cluster - Electricians
(N=20)
| Cluster | Cluster | Mean | Mean | Mean | |||
| Name | Size | Prest. | s.d. | Adm. | s.d. | Use. | s.d. |
| a. Professions | 11 | 7.81 | 0 .95 | 7.16 | 1.02 | 7.48 | 1.24 |
| b. Managers/White-collar | 8 | 6.31 | 1.14 | 5.58 | 0.90 | 5.67 | 0.90 |
| c. Skilled Trades | 13 | 5.74 | 1.25 | 7.03 | 0.82 | 7.93 | 0.87 |
| d. Services/Low-skilled | 27 | 4.47 | 0.88 | 5.33 | 0.54 | 5.78 | 0.76 |
| e. Actor | 1 | 7.75 | 5.80 | 5.00 | |||
| Total | 60 | 5.66 | 1.62 | 6.08 | 1.11 | 6.53 | 1.33 |
F:23.3 p<.001
F:18.2 p<.001
F:18.0 p<.001
Professors also give highest average scores to professionals (6.98) on the scale of respect and admiration (see Table 3), followed by skilled trades (4.83), managerial/white-collar (4.71), which they place on an equal footing, and service/lower-skilled occupations (3.66), which come last. A similar pattern is repeated with regard to social usefulness, with 7.61 for professionals, 6.20 for skilled trades, 5.81 for managers/white collar, and 4.95 for service/low-skilled occupations. On the prestige scales, the average scores of professors are 7.06 for professionals, 5.30 for managers/white-collar, 4.01 for skilled trades, and 3.21 for service/low-skilled occupations. University professors also exhibit some measure of occupational egoism, by placing their own occupational group at the top of the three scales and giving lower rankings to all other groups. Like electricians, they rank managerial/white collar occupations higher on occupational prestige than on the scales of usefulness and admiration.[12] However, by clustering together professional and managerial occupations, which are closer in terms of prestige than in terms of admiration and usefulness, they do not express the same reluctance as electricians in accepting prestige as a principle for organizing occupational space.
Table 3 - Mean prestige, admiration, and usefulness by cluster - Professors
(N=19)
| Cluster | Cluster | Mean | Mean | Mean | |||
| Name | Size | Prest. | s.d. | Adm. | s.d. | Use. | s.d. |
| a. Professions | 14 | 7.06 | 1.20 | 6.98 | 0.85 | 7.61 | 1.00 |
| b. Managers/White-collar | 7 | 5.30 | 1.29 | 4.71 | 1.00 | 5.81 | 0.68 |
| c. Skilled Trades | 12 | 4.01 | 0.60 | 4.83 | 0.63 | 6.20 | 0.51 |
| d. Services/Low-skilled | 25 | 3.21 | 1.11 | 3.66 | 1.01 | 4.95 | 1.05 |
| e. Actor/Computer Prog. | 2 | 6.34 | 0.11 | 6.43 | 0.21 | 6.39 | 0.37 |
| Total | 60 | 4.62 | 1.88 | 4.88 | 1.59 | 5.97 | 1.37 |
F:32.2 p<.001
F:32.4 p>.001
F:19.7 p<.001
Results for students are not radically different from those obtained for professors and electricians, but neither do they show the same level of similarity. While the overall layout of their configuration is similar to that of the other two groups, students draw slightly different boundaries between clusters. Like electricians and professors, students make a distinction at the five-cluster level between professions, managerial/white-collar occupations, and blue-collar/service jobs. As opposed to both professors and electricians, however, they make a more detailed differentiation among professional occupations and a less detailed one among blue-collar and service occupations. They distinguish two different clusters among professional occupations: cluster 1b includes health-related, education, and social services occupations, while cluster 1a includes miscellaneous professions such as computer programmer, lawyer, geologist, and electrical engineer.[13] Cluster 2c includes managerial-white collar occupations and is highly similar to the cluster made by the other two groups, but cluster 2d agglomerates together service/lower-skilled and skilled trades/blue-collar occupations. It is only at the six-cluster level of analysis that these two groups are differentiated.
At the two-cluster level of analysis, students lump together the two professional groups in cluster 1 and place all other occupations, including managerial, white-collar/service, and skilled trades, in cluster 2. Professional occupations in cluster 1 are those which they most personally admire, find most socially useful, and which also score relatively high on occupational prestige. Interestingly, students give much higher scores on usefulness (8.13) and admiration (7.63) to the health/education/social services occupations compared to what they perceive as their general level of prestige in society (6.95) (see Table 4).
Table 4 - Mean prestige, admiration, and usefulness by cluster - Students
(N=18)
| Cluster | Cluster | Mean | Mean | Mean | |||
| Name | Size | Prest. | s.d. | Adm. | s.d. | Use. | s.d. |
| a. Misc. Professionals | 7 | 7.82 | 1.35 | 7.40 | 1.10 | 8.27 | 1.10 |
| b. Health/Soc./Education | 9 | 6.95 | 1.38 | 7.65 | 0.92 | 8.13 | 1.01 |
| c. Managers/White-collar | 11 | 6.02 | 1.23 | 5.32 | 0.71 | 6.14 | 0.86 |
| d. Services/Blue-collar | 32 | 4.58 | 1.35 | 5.42 | 1.07 | 6.75 | 1.32 |
| e. Actor | 1 | 7.11 | 4.83 | 5.50 | |||
| Total | 60 | 5.62 | 1.78 | 5.96 | 1.36 | 7.00 | 1.38 |
F:12.4 p<.001
F:13.9 p<.001
F:6.3 p<.001
Like electricians, students thus consider that one occupational category is unfairly evaluated by society, although this category is not the same for the two groups. For electricians, it is the trades that are unfairly evaluated, while for students it is health/education/social service occupations. This might also be a case of occupational egoism, since several students expressed the ambition of pursuing careers in health and computer sciences. It also reflects the well- documented tendency among respondents to draw finer boundaries among groups which are closest to them in social space than among groups which are farther apart (Coleman and Rainwater, 1978; Gilbert and Kahl, 1982; Warner et al., 1960). The picture drawn here in fact closely resembles that of the class structure in Yankee City (Warner et al. 1960), in which respondents identified the same class structure but used normative criteria to draw different boundaries within it.
Cluster analysis reveals interesting patterns that would have been obscured had occupations been ranked on linear scales. It reveals a fair amount of occupational egoism, since all groups tend to upgrade their own (actual or anticipated) occupational group and downgrade others. Normative evaluations of worth are especially important in this respect, since they provide alternative criteria for evaluating occupations and drawing boundaries in social space, especially among working class respondents. It is mainly on the basis of these alternative criteria that respondents enhance their own occupational group and downgrade others. The discrepancy between socioeconomic status and worth is especially noteworthy in the case of managerial/white-collar occupations, since the three groups of respondents consistently rank these occupations lower on admiration and usefulness than on occupational prestige.
Conclusion
Durkheim’s claim that consensual evaluations of social positions constitute a moral framework which ensures social integration has long been set aside as a convincing interpretation of occupational prestige scales. Visions of a single, overarching status order conferring meaning to social life has given way to pictures of multiple spaces of distinction anchored in incommensurable criteria of evaluation. In the sociology of culture, symbolic representations are no longer seen as part of a coherent and unified pattern of norms and values, transmitted through socialization and unreflexively enacted in everyday life. Rather, culture is increasingly seen as a “grab bag of odds and ends” (DiMaggio, 1997: 267), consisting of disparate images, narratives, and categorical schemes which may or may not be activated in different social contexts.
The view of status as exclusively anchored in fragmented identities and individualized inequality, however, does not account for basic characteristics of prestige scales and for some of the major findings of the present research. It cannot explain the extraordinarily high level of consensus across classes, gender, and ethnic/racial groups concerning the position of occupations on prestige scales (Treiman, 1977) as well as the relatively high level of congruence between occupational prestige and the personal respect and admiration that electricians, professors, and students feel towards occupations. Nor does it explain that respondents produce strikingly similar pictures of occupational space, not only in this study but in similar research in different countries (see Magana et al. 1995). These findings suggest that Durkheim was not entirely wrong when he claimed that social positions are graded in the moral consciousness of societies in a way which maintains social cohesion and legitimates inequality (Durkheim, 1967:275ff). Prestige scales may be seen as widely-shared representations of what people perceive as dominant evaluations of social positions. In market-driven and ostensibly egalitarian late modern societies, these evaluations primarily reflect the material resources and advantages afforded by occupations.
The line between factual and normative elements, however, is not always clear. As evidenced by the relatively high correlation between prestige and worth among the three groups of respondents, occupational ratings do appear to have some legitimating significance with regard to social inequality. Yet the existence of intersubjective evaluations of positions does not necessarily entail uniform consensus on their legitimacy. University professors, who are privileged in terms of income and education, more readily accept them as legitimate compared to students and electricians. They see more congruence between socioeconomic status and worth and the boundaries they draw between themselves and others closely mirror occupational prestige ratings. Electricians do not entirely subvert the dominant prestige order, since they express high admiration for the professions which also rank highest on occupational prestige. Rather, they use normative criteria to draw symbolic boundaries between themselves and others in such a way that they place their own occupational group at the top of the hierarchy along with the professions. Occupational prestige, then, should be seen as one among many cognitive schemata available in collective consciousness for representing social structure, drawing boundaries, and evaluating others. These schemata are not enacted unreflexively by oversocialized social actors (Wrong, 1961) and they may not be relevant for organizing all aspects of everyday life. Along with normative evaluations of worth and a host of other criteria, they are tools for building “fences or bridges” (Douglas and Isherwood 1979: 12) in processes of inclusion and exclusion.
Notes
* This research was supported by a postdoctoral fellowship from the Social Sciences and Humanities Research Council and by research grants from Carleton University and the University of Ottawa. I thank Aage Sørensen, Terry Nosanchuck, Michèle Lamont, François Gobeil, and two anonymous CJS reviewers for their comments on earlier drafts. [back to text]
1 In this paper, the terms status and prestige are used interchangeably. Socio-economic status and occupational prestige refer to gradations of positions with regard to socio-economic resources. The more general terms of prestige and status refer to intersubjective evaluations of superiority and inferiority which have acquired an existence beyond individual beliefs and which influence social relations in various ways, e.g. through patterns of deference, association, and exclusion. [back to text]
2 Because of the very small scale of this study, the only aspect of social location examined here is occupation. This doesn’t mean that occupation is considered a priori more relevant than other dimensions of social location such as gender, race/ethnicity, or region. [back to text]
3 None of these groups constitutes a random sample and the number of interviews is arguably very small. Because of these limitations, this research should be considered “an exploratory exercise in idea-generation” (Markovsky, 2000: 556) rather than a formal attempt at hypothesis testing. [back to text]
4 The matrices and MDS configurations were produced with Anthropac 3.0 (Borgatti, 1990). [back to text]
5 PROFIT was performed with Anthropac 3.0 and clusters were produced using the between-group average method in SPSS 6.1.3. [back to text]
6 Asterisks indicate where respondents placed themselves. [back to text]
7 These levels are well within those reported in similar research. See Burton (1972), Kruskal and Wish (1978), and Ollivier (1997). [back to text]
8 The orientation of the scales and layout of occupations differ in each configuration but rotating the three figures to align the arrows would produce highly similar results. Readers should pay attention to the location of occupations relative to one another rather than to their absolute position in a fixed space. [back to text]
9 Data in this and all other tables were not collected from random samples. Levels of significance are provided for heuristic purposes only. [back to text]
10 In selecting the number of clusters which most adequately represent respondents’ perception of occupations, my main concern was not to determine the “true” number of clusters present in the data but rather to find a compromise between the need for detail and the need for economy. Research on free-sorting has shown that respondents tend to sort objects at different levels of generality: some respondents make fine distinctions while others create only a few broad categories (Coxon and Jones, 1979). As a result, a two-cluster solution is not necessarily better than the six-cluster one, but simply less detailed. In the present research, respondents classified their cards into 2 to 19 piles and they generally had no difficulty grouping their piles into broader categories when asked to do so. [back to text]
11 One electrician used the term “skilled trades for women” to describe these occupations. [back to text]
12 A similar pattern was noted by Gusfield and Schwartz (1963). Banker and lawyer ranked higher on prestige than on usefulness while the reverse was true for plumber and schoolteacher. [back to text]
13 The inclusion of word processing operator in this cluster appears somewhat anomalous. This could reflect the tendency among students to group more occupations according to occupational/industrial classification (e.g. travel agent, flight attendant, and air pilot), and thus to associate word processing operator with computer programmer. [back to text]
References
Blau, Peter M and O. D. Duncan
1967 The American Occupational Structure. New York: John Wiley and Sons.
Blau, Peter and Joseph E. Schwartz
1984 Crosscutting Social Circles. Orlando: Academic Press.
Borgatti, Stephen P.
1990 Anthropac Provisional Documentation.
Bourdieu, Pierre
1979 La distinction : critique sociale du jugement. Paris: Les Éditions de Minuit.
Burton, Michael
1972 “Semantic Dimensions of Occupation Names.” In Roger N. Shepard, A. Kimball Romney and Sara Beth Nerlove (Eds.), Multidimensional Scaling: Theory and Applications in the Behavioral Sciences, vol. 2. New York and London: Seminar Press.
Calhoun, Craig
1991 “Indirect Relationships and Imagined Communities: Large-Scale Social Integration and the Transformation of Everyday Life.” In Pierre Bourdieu and J. S. Coleman (Eds.), Social Theory for a Changing Society. Boulder: Westview Press.
Coleman. Richard P and Lee Rainwater
1978 Social Standing in America. New York: Basic Books.
Coxon, Anthony and Charles L. Jones
1979 Measurement and Meanings: Techniques and Methods of Studying Occupational Cognition. London: MacMillan.
1978 The Images of Occupational Prestige. London: MacMillan.
Cuneo, Carl J.
1996 “International Images of Social Inequality: A Ten-Country Comparison.” In Alan Frizzel and Jon H. Pammett (Eds.), Social Inequality in Canada. Ottawa: Carleton University Press.
Davis, Kingsley and Wilbert E. Moore
1945 “Some Principles of Stratification.” American Sociological Review 10:242–249.
Degraaf, Paul M., Harry B. Ganzeboom and Matthijs Kalmijn
1989 “Cultural and Economic Dimensions of Occupational Status.” In W. Jansen, J. Dronkers, and K. Verrips (Eds.), Similar or Different? Amsterdam: SISWO.
DiMaggio, Paul and John Mohr
1985 “Cultural Capital, Educational Attainment, and Marital Selection.” American Journal of Sociology 90(6):1231–1261.
DiMaggio, Paul
1987 “Classification in Art.” American Sociological Review 52:440–455.
1997 “Culture and Cognition.” Annual Review of Sociology 23:263–287.
Douglas, Mary and B. Isherwood
1979 The World of Goods. New York: Basic Books.
Duncan, O. D.
1961 “A Socio-Economic Index for all Occupations.” In A. J. Reiss (Ed.), Occupations and Social Status. New York: Free Press.
Durkheim, Émile
1967 Le suicide. Paris: Presses universitaires de France.
Erickson, Bonnie
1996 “Culture, Class, and Connexions.” American Journal of Sociology 102(1):217–235.
Featherman, David L. and Robert M. Hauser
1976 “Prestige or Socioeconomic scales in the Study of Occupational Achievement?” Sociological Methods and Research 4(4):403–22.
Gilbert, Dennis and Joseph Kahl
1982 The American Class Structure: A New Synthesis. Homewood, Ill.: Dorsey Press.
Goldthorpe, John H. and Keith Hope
1974 The Social Grading of Occupations: A New Approach and Scale. Oxford: Clarendon Press.
Grasmick, Harold G,
1976 “The Occupational Prestige Structure: A Multidimensional Scaling Approach.” The Sociological Quarterly 17:98–108.
Grusky, David B. and Stephen E. Van Rompaey
1992 “The Vertical Scaling of Occupations: Some Cautionary Comments and Reflections.” American Journal of Sociology 97(6):1689–1711.
Guppy, L. Neil
1984 “Dissensus or Consensus: A Cross-National Comparison of Occupational Prestige Scales.” Canadian Journal of Sociology 9(1):69–83.
Gusfield, J.R. and M. Schwartz
1963 “The Meanings of Occupational Prestige: Reconsideration of the NORC scale.” American Sociological Rewiew 28: 265–271.
Hall, John R.
1992 “The Capital(s) of Culture: A Nonholistic Approach to Status Situations, Class, Gender, and Ethnicity.” In Michèle Lamont and Marcel Fournier (Eds.), Cultivating Differences: Symbolic Boundaries and the Making of Inequality. Chicago and London: The University of Chicago Press.
Hope, Keith
1982 “A Liberal Theory of Prestige.” American Journal of Sociology 87(5): 1011–1031.
Hou, Feng and T. R. Balakrishnan
1996 “The Integration of Visible Minorities in Contemporary Society.” Canadian Journal of Sociology 21(3):307–326.
Kelley, Jonathan and M. D. R. Evans
1993 “The Legitimation of Inequality: Occupational Earnings in Nine Nations.” American Journal of Sociology 99(1):75–125.
Kluegel, James R. and Eliot R. Smith
1986 Beliefs About Inequality. New York: Aldine de Gruyter,
Krauss, Vered, O. E. Schild and Robert W. Hodge
1978 “Occupational Prestige in the Collective Conscience.” Social Forces 56(3):900–917.
Krushkal, Joseph and Myron Wish
1978 Multidimensional Scaling. Beverly Hills: Sage Publications.
Lamont, Michèle
2000 The Dignity of Working Men: Morality and the Boundaries of Race, Class, and Immigration. Cambridge: Harvard University Press and Russell Sage Foundation.
Lamont, Michèle and Marcel Fournier
1992 Cultivating Differences: Symbolic Boundaries and the Making of Inequality. Chicago and London: University of Chicago Press.
Macleod, Jay
1987 Ain’t No Making It: Leveled Aspirations in a Low-Income Neighborhood. Boulder: Westview Press.
Magana, Raul, Michael Burton and Joao Ferreira-Pinto
1995 “Occupational Cognition in Three Nations: Relationships between Classification and Evaluation.” Journal of Quantitative Anthropology 5(2):149–168.
Markovsky, Barry
2000 “Review of Karin Knorr Cetina’s Epistemic Cultures: How the Sciences Make Knowledge.” Contemporary Sociology. 29(3): 556–557.
Meiselas, Susan
1976 Carnival Strippers. New York: Farrar, Strauss & Giroux.
Melucci, Alberto
1996 The Playing Self: Person and Meaning in the Planetary Society. Cambridge: Cambridge University Press.
Milner, Murray
1996 “Geeks, Weirdos, and Cool Kids: Status Among American Adolescents.” Paper presented at the Annual Meeting of the American Sociological Association, New York.
Nam, Charles B. and E. Walter Terrie
1982 “Measurement of Socioeconomic Status from United States Census Data.” In Mary G. Powers (Ed.), Measures of Socioeconomic Status: Current Issues. Bouldner: Westview Press.
Ollivier, Michèle
1994 Prestige and Social Closure in the Arts. Unpublished PhD Dissertation, Department of Sociology, Harvard University, Cambridge, MA.
1997 “Measuring Symbolic Boundaries Among Artists,” Poetics 24:299–328.
Ostrower, Francie
1996 “Status and Elite Philanthropy: Hierarchies and Processes.” Paper presented at the Annual Meeting of the American Sociological Association, New York.
Pammett, Jon H.
1996 “Getting Ahead around the World,” In Alan Frizzell and Jon H. Pammett (Eds.), Social Inequality in Canada, Ottawa: Carleton University Press.
Parkin, Frank
1971 Class Inequality and Political Order. London: MacGibbon & Kee.
Reiss, Albert J.
1961 Occupations and Social Status. New York: Free Press.
Shils, Edward
1975 Center and Periphery: Essays in Macrosociology. Chicago and London: University of Chicago Press.
Sørensen, Aage
1992 “Concluding Commentary: Class, Status and Power.” In David G. Grusky (Ed.), Social Stratification: Class, Gender and Race in Sociological Perspective. Boulder: Westview Press.
Statistics Canada
1980 Standard Occupational Classification. Catalogue no. 93–327, Ottawa.
Stehr, Nico
1974 “Consensus and Dissensus in Occupational Prestige.” British Journal of Sociology 25(4):411–426.
Treiman, Donald J.
1977 Occupational Prestige in Comparative Perspective. New York: Academic Press.
Warner, Lloyd W, Marcia Meeker and Kenneth Eels
1960 Social Class in America. New York: Harper and Row.
Wrong, D.
1961 “The Oversocialized Conception of Man in Modern Sociology.” American Sociological Review 26:183–193.
Appendix
(click on title to view figure full-size in a new window)
