Advances in Consumer Research Volume 14, 1987 Pages 492-496
SOCIAL CLASS AND CONSUMER BEHAVIOR: THE RELEVANCE OF CLASS AND STATUS
James E. Fisher, Saint Louis University
[The author gratefully acknowledges the financial support provided by Saint Louis University's Beaumont Faculty Development] Fund.
The use of social stratification in consumer research has been criticized for naive conceptualization. This paper examines the theoretical basis for asserting a close connection between social class structure and consumer behavior. Max Weber's seminal contribution to stratification theory provides the basis for this examination. Special consideration is given to the dimensions of class and status, which figure prominently in Weber's work. The relevance of this approach to consumer research is summarized in the form of several basic propositions.
In the late 1950's and early 1960's, several contributions to the marketing literature appeared that pointed to the significance of social class for understanding consumer behavior. These early writings drew heavily on Warner's conception of social class--one which focused largely on position and prestige of families within relatively small, self-contained communities. The connection between social class and consumer behavior that was outlined by such individuals as Martineau (1958), Coleman (1960), and Levy (1966) was a somewhat broad one stressing variations in values, lifestyles, and general consumption goals. Many of the findings presented were based on syntheses of proprietary studies and thus actual quantitative evidence was rarely included.
Interest in this broad subject area continued through the 1960's and early 1970's, although with a somewhat different orientation. For it was at this same time that segmentation research was growing in importance, and social class studies in marketing were largely diverted into this research stream. To oversimplify somewhat, one might say that in the search for the so-called superior correlate with buying behavior, social class was perceived as a likely candidate. More specifically, some argued that social class would prove superior to income as a basis for segmentation. Thus the social class vs. income issue emerged, and within a period of a few years about a dozen articles appeared in the marketing literature that joined this debate (e.g., Wasson 1969; Myers, Stanton, and Haug 1971; Myers and Mount 1973; Hisrich and Peters 1974). The basic social class vs. income issue remained clouded during the seventies, and interest in it eventually waned.
Although the direct application of social class theory to consumer behavior may seem slight in recent years, the usefulness of the concept continues to be demonstrated. There are several research currents--consumption symbolism (Levy 1981; Belk, Mayer, and Bahn 1982; Solomon 1983), the impact of women's roles on consumption (Schaninger and Allen 1981; Reilly 1982), and the cultural context of consumption (Reilly and Rathie 1985; Hirschman 1985; McCracken 1986)--that use or expand upon stratification concepts. The apparent ease with which social class is subsumed by these and other research efforts suggests the concept is a potentially valuable one, but one also subject to various interpretations.
Efforts to apply stratification theory more directly and precisely are becoming increasingly common. In this modest revival of interest, one finds the social class vs. income issue has been revisited, only this time with considerable more sophistication and methodological rigor (Schaninger 1981). Coleman, a pioneer in this area, has more recently (1983) published an article in the Journal of Consumer Research pointing to the continuing significance social class has for marketing and setting forth some methodological guidelines for research in this area. Also, Dominquez and Page (1981a, 1981b) have suggested how consumer behavior research in this area might best proceed, as have Shimp and Yokum (1981). Pulling these and other contributions together, the following criticisms warnings. and recommendations seem to emerge:
1. Social class is a construct with "surplus meaning." No consensus exists as to the number, cohesiveness, separability, and distinctiveness of social strata. And yet, marketing researchers have reflected little on the reliability and validity issues surrounding the social class construct. The availability of a variety of stratification scales and the widespread use of certain of these have sometimes short-circuited full consideration of what precisely is being measured.
2. Much of the empirical work that has been done has been characterized by naive conceptualization (cf. Dominquez and Page 1981a on this point). At its most simplistic, research has unfolded in the following manner: a standard social class index is used to determine a family's social class membership, and then a direct link is sought between this variable and some other consumption specific variable such as product, store, or brand choice behavior. This approach is subject to criticism on several points. For example: (a) intervening variables are frequently ignored (e.g., life-cycle stage), (b) inappropriate statistical techniques are used to measure the effect of stratification (often bivariate techniques which ignore the aforementioned intervening variables), (c) choice of dependent measures is frequently ill-suited for testing social class influence.
3. Reliance by marketing scholars on an essentially Warnerian framework--a view of social classes as discrete membership groups that evidence a high degree of cultural homogeneity--has produced research that has principally sought to demonstrate the existence of significant differences in consumption behavior across classes. Such an approach, while certainly legitimate and necessary, has for the most part left unexamined the diversity and dynamism of intra-class behavior. This flaw is especially serious given the rise in two-income families, female-headed households, and other types of living arrangements that depart from the traditional marital-couple mold. Indeed, many of the instruments used to measure social class assume a family in which the male head is the sole wage-earner and is most likely at the peak of his earning capacity. This approach seems especially ill-suited for handling issues surrounding (a) social mobility and status attainment, (b) life-cycle changes and their impact on social standing, and (c) status inconsistency and status crystallization.
CLASS AND STATUS DIMENSIONS
This paper makes no claim to resolve these longstanding problems. Instead it argues that a particular view of social class structure--one originally proposed by Max Weber and subsequently developed and researched at length by sociologists--can be especially helpful in understanding certain aspects of consumer behavior. Furthermore, it is a view that begins to address the aforementioned problem areas.
Weber's conceptualization of social class is appropriately delimited and yet avoids the inflexibility of the somewhat outdated Warnerian vies. In addition, this approach offers a compelling explanation of why one expects to find a close connection between social class position and consumption-related activities.
Weber (1946) recognized that social stratification [This paper will use the terms 'social class" and "social stratification" interchangeably, although strictly speaking a society may be stratified without having distinct social classes. The concepts of class and status, however, are distinguished from each other. This reflects Weber's use of the terms, but is frequently at odds with ordinary usage, which uses terminology like class level and social status interchangeably. The relationship between the concepts of class and status and the broader phenomenon of stratification will be developed further in this paper.] was a multidimensional phenomenon. He asserted that society was ordered along several hierarchies, with class and status being the principal stratification dimensions. Class, for Weber, was largely an economic category. It has been variously associated with occupation, wealth, or, more broadly, life chances. Status, on the other hand, has to do with social distinctions and thus has been commonly linked with lifestyles.
Consumer researchers have not usually found the distinction between class and status a useful one to make. Wilkie (1986) offers a possible explanation for this:
Marketers and consumer behavior professionals have tended to concentrate almost entirely on the life-style dimension, as this reflects most directly the purchasing patterns and consumption habits of different social classes.... [O]ther fields [have given attention] to the life-chances dimension of social stratification (p. 659).
To press this point a bit further, one might suggest that the connection between social class and lifestyle has been made too closely. For example, the assertion that lifestyle is "the essence of social class" (Myers and Guttman 1974) is an overstatement and may encourage the neglect of other social class aspects.
Dominquez and Page (1981a, 1981 ,) maintain that the distinction between class and status is highly relevant for consumer researchers working with social stratification issues. Indeed, they propose that at the outset of any study one must determine whether the consumption-related behavior is more likely to be related to class or status. Furthermore, they offer the following guidelines for making such a determination
Class centers on the individual and his/her occupation, while status revolves around the family and its position in the community based on home type, location and value, interactions, and memberships as well as education, family, background, and occupation Therefore class is best suited to those consumer decisions that are predominantly individually, rather than jointly, made or delegated to the family Class is also best suited to those values, lifestyles, and communication patterns that are centered on work, leisure (because of the impact of occupational role on availability, use of and spending for leisure time), investment, saving, and attitudes toward and perceptions of financial outlook (1981b p 156)
Dominquez and Page challenge consumer researchers to think more precisely about what social class is and why one expects it- to be related to consumer behavior In addition, they assert that class as well as status aspects of social class structure are relevant to consumption However, in framing the class/status question in somewhat stark terms as an either/or matter, they have underestimated the interdependent nature of these stratification dimensions
The connection between class and status is especially relevant when considering human behavior in the consumption role Indeed, according to Weber the dimensions of class and status are connected via the market mechanism
With some over-simplification, one might thus say that classes" are stratified according to their relations to the production and acquisition of goods; whereas 'status groups ' are stratified according to the principles of their consumption of goods as represented by special "styles of life" (Weber 1946, p 193)
The relationship between class and status is complex Weber went beyond the Marxist position that class is the primary determinant of status The social closure that is accomplished by various status groups can in turn have a profound influence on class and attendant economic realities Thus lifestyle does not simply flow from social standing but rather plays a central role in the establishment, maintenance and modification of structured social inequality
Lifestyles and the consumption patterns which support them can take on the character and function of exclusion mechanisms (Boskoff 1972) and thus serve to prevent mobility and to institutionalize privileges While it is certainly obvious that the absence of adequate economic resources can effectively prevent one from practicing a certain lifestyle, it is perhaps less obvious that the various values, skills, and aesthetic standards embodied in a particular lifestyle may also resist easy acquisition or imitation and therefore they can just as effectively limit one's economic and social opportunities Even for status groups not at the top, social closure accomplished by lifestyle can insulate and protect such groups from outside influence and thereby afford them a measure of autonomy
The preceding discussion has argued that the relationship between social class and consumption is a fundamental one We have held that style of consumption is not only an expression of a particular social class orientation, but also a mechanism that figures prominently in the actual shaping of the social structure
The essential elements of this perspective are summarized and expanded in the form of several basic propositions.
Proposition 1: Social class is a useful construct for explaining consumption behavior because it offers insight into both the various resources that limit consumer choice and the preferences that direct the allocation of those resources.
Perhaps the significance of this proposition can be illustrated by way of a comparison. Economic theory of consumer behavior starts with a consumer's budget constraint and his preferences for some assortment of goods and then sets about determining how the consumer will allocate his scarce financial resources so as to maximize utility. The theoretical apparatus inevitably leads to this point of equilibrium, but only if the consumer's tastes can be treated as given. Social class analysis steps into this vacuum, giving evidence that important differences in lifestyles and consumption patterns stem from corresponding differences in social standing. But not only does this perspective offer insight into consumer preferences and tastes, but it also deepens our understanding of the particular constraints under which the consumer acts in the marketplace. There are at least four resource dimensions that are encomPassed in social class structure:
Financial. The concentration of wealth at the very top of the social spectrum in the United States is enormous. Gilbert and Kahl (1982) estimate that the so-called capitalist class, which accounts for 1 percent of the U.S. population, controls about half the country's wealth. Considering the entire class structure, we find that the class-income correlation can be estimated at about 0.4 (Coleman 1983). This figure, while quite significant, underlines the fact that income class and social class are not interchangeable concepts.
Social. People tend to associate with social class equals in their friendships, marriages, organizational memberships, and residential choices (Warner 1949; Hollingshead 1950; Laumann 1966; Simkus 1978; Gilbert and Kahl 1982). Furthermore, as social class rank increases so too does level of social participation (Kahl 1957; Hodges 1964; Curtis and Jackson 1977). Thus higher social class rank suggests a wider social network than does lower, which is characterized by kin-oriented ties that are further restricted in a geographical sense (Coleman 1983). Social skill and occupational position seem inextricably linked; entrance and advancement into many middle and upper-middle class occupations depends significantly upon one's ability to demonstrated certain social attitudes and traits (Lynd and Lynd 1929; Whyte 1952; Kanter 1977). Similarly the occupational position of parents strongly colors the socialization of their children, thereby inculcating social values which may either facilitate or inhibit occupational achievement (Kohn 1969).
Cultural. Familiarity with cultural matters and the possession of cultural credentials appears to be strongly associated with higher social class standing (Bourdieu 1973). Cultural capital is sometimes acquired--typically through the acquisition of appropriate academic credentials or through participation in cultural affairs--in an effort to enhance or improve one's social standing (DiMaggio 1982).
Time. Perceptions of time appear to be strongly conditioned by social class membership. Three findings in particular emerge: (1) lover class members reveal a more restricted orientation to future events than those from the middle class (Davis and Dollard 1940; LeShan 1952; Bernstein 1960, 1968; Horton 1967; Liebow 1967); (2) lower class members with upwardly mobile aspirations and who have had an opportunity to gain anticipatory socialization into middle class values and standards are less likely to have constricted time horizons than their social class counterparts without such aspirations (Ellis and Lane 1966); (3) regardless of social class level, persons with constricted time perspectives are at a competitive disadvantage in meeting the institutionalized role demands of the middle-class world (Davis and Dollard 1940; Teahan 1958; Bernstein 1960, 1968; Cohen, Fraenkel, and Brewer 1968; Williams 1970). The value and availability of time also varies across social classes. Because their time is worth more in economic terms, individuals with higher social class standing frequently find themselves with little leisure time (Linder 1970). At the same time, higher class standing implies greater freedom and flexibility in managing time constraints (Douglas and Isherwood 1978).
Without at all denying the centrality of income and wealth as a determinant of consumer choice, most social class theorists would maintain that economic resources are typically mediated and augmented by other resource dimensions. In addition to making distinctions among various types of resources-financial, social, cultural and time-related-one should also make a distinction between the resources one brings to the market and the particular manner in which those resources are deployed. The latter distinction between consumption possibilities and actual consumption patterns mirrors the distinction Weber makes between class and status dimensions.
Proposition 2: Class structure is latent; status, manifest. Class position contributes significantly to status configurations. Status differentiation is in turn linked to variations in consumption activities via the lifestyle mechanism.
The consumption potentialities suggested in class indicators are realized in status groups; life-chances are transformed into lifestyles. This proposition can be further developed by another comparison to economic theory, in this instance, to the "new" economic theory of consumer behavior (Becker 1965; Gronau 1977). This perspective views consumption as a process that is not unlike the production process. Just as a firm inputs capital and labor to produce goods, so too does a household combine goods and time to produce commodities that in turn yield utility. And just as it is possible to speak of a technology associated with the production of goods, so too can one consider a household consumption technology. (Some households may exhibit higher levels of efficiency than others; some may have goods intensive technologies and others time or leisure intensive ones.) And, finally, just as capital and labor draws its value from its productive potential, so too is the social significance of the consumer's resources (class indicators) ultimately realized when they are fashioned into a particular lifestyle (status indicator).
Lifestyle, as used in the present context, is identified as an equilibrating mechanism by which potentialities for behavior are translated into actual behavior deemed appropriate to a particular social position. On the one hand, certain styles of life cannot be realized (or are realized with great strain and difficulty) in the absence of the requisite resources. Status is not independent of class. On the other hand, there is so-e uncertainty surrounding the precise manner in which class position is translated into status structures. Boskoff, who has given special consideration to the role of lifestyle in social -stratification, views lifestyle as "a trial-and-error cumulation of shared adaptations to a set of social opportunities and restrictions" (1972, pp. 159-160). Position, therefore,
does not necessarily "cause" style of life, but rather provides cues in specific recurring situations that tend to elicit an unplanned but -"logico-meaningful" set of attitudes, values, and actions. The functional or adjustive value to the actors of a style of life may be inferred in part from the tenacity of such values and their relative i 'unity from external attempts at change (1969, p. 263).
Another issue, suggested by the above quote, concerns the consistency and stability of lifestyles. If styles of life are conceived as adaptive mechanisms, then as conditions change or perceptions of conditions change so too will lifestyles. Lifestyle reflects an ongoing series of judgments about resources and rewards, limitations and opportunities that confront individuals and families. The process is not a haphazard one. The social class structure itself points to the existence of a set of lifestyles that serve as "fixed valuation points" (Boskoff 1972) or "ideal types" (Kahl 1957). This is no less true in a society characterized by social mobility. The argument can in fact be advanced that status achievement is facilitated by the presence of these value systems; they ascribe meaning to one's choice of goals and the means to those goals.
Proposition 3: Goods may be said to take on the properties of status symbols if the purchase and use of them is indicative of membership in a particular status group. Furthermore, status symbols are efficacious only insofar as there are restrictive mechanisms associated with their appropriation that serve to limit their "fraudulent" use.
Goffman's analysis (1951) of Status symbols reveals several different mechanisms that can limit the inappropriate use of such objects. Therefore, when considering which articles of consumption may function as status symbols, one must recognize that restrictions limiting their appropriation may arise from several different sources. Limited financial resources can of course restrict the use of many goods, but there are other resource dimensions that just as effectively limit consumption opportunities. Limitations or restrictions having to do with time, social skills, ability, and family history can also figure prominently in several areas of consumption
Advertisements that make appeals to status assume a variety of stances with respect to these restrictive mechanisms. For example, the following slogan makes clear the financial demands implied by ownership: "As long as there are people who can afford perfection, BMW will continue to build it." It may be possible to relax one restrictive mechanism, while introducing another--an approach taken by a paperback book club that offers its goods to "smart people who aren't rich." Yet another possibility is provided by the appeal which is unabashed in its circumvention of the restrictive mechanism. This is found in an advertisement for Levolor blinds that offers guidance on "how to decorate rich, rich, rich when your not, not, not."
Status symbols essentially embody the conjunction of Weber's class and status dimensions. An individual can employ consumer goods as status symbols and in so doing attempt to claim a particular social standing or to attribute social significance to his behavior. But the claim remains distinct from the basis for that claim, just as status aspirations remain distinct from the class realities
When class and status dimensions are not consistent either at the individual or the societal level, then instability and insecurity results, fostering an increased use of status symbols (Gerth and Mills 1953) Belk (1986) has recently speculated that the conspicuous consumption of yuppies may stem in part from their relatively slow occupational advancement and inadequate gains in real income Their consumption is compensatory in the sense that consumption gratification (a status concern) is being substituted for job satisfaction (a class consideration)
The precise focus of compensatory consumption also reflects the influence of class and status Simply put, some spheres of consumption are more restrictive than others, and some consumers may seek to emphasize those areas of consumption which are relatively speaking, less restrictive Of course, the term "restrictive is used in a relative sense; determinations as to which areas are restricted or not depends upon an assessment of one'-s control over a variety of resources (i e, money, time, etc ) vis-a-vis the type and level of resource needed to secure a particular consumption goal
For example, a family with limited financial resources may recognize that certain high-priced luxuries are beyond its means; however, that same family may find that it does have the resources required to engage in various forms of social or cultural consumption which provide important, alternative sources of satisfaction and status Many similar scenarios are possible Time and effort can frequently be substituted for money And money can be spent in a manner intended to compensate for limited access to highly restricted social circles.
When, by necessity or choice, consumption is focused in one area and not another, there seems to be a tendency to elevate the importance of the former, while denigrating that of the latter Tumin (1967) expands on this tendency
If (an individual's social rating] depends on his manifestation of a number of criteria, and if he fulfills more of some criteria and less of others, he will tend to stress the importance of those he fulfills while he minimizes the importance of those unfulfilled If his home is modest but his children attend good private schools, he will call attention to this fact and at the same time he will stress the importance of maintaining only a modest residence (p. 101).
Such behavior is similar in its aim to certain dissonance reductions efforts; it compensates for the inadequacies of one's own choice by minimizing the attractiveness of those options which were either foregone or denied Thus, in matters of consumption, the consumer may find it helpful to make a virtue out of necessity
Proposition 4 Just as consumption style can be used as a technique of exclusion, serving to maintain the privilege and status of those who belong, so too can consumption "finesse" or expertise provide a means of penetrating class and status barriers In other words, consumption can be enlisted as a strategy to advance social class standing.
Restrictive mechanisms are not impenetrable barriers For those possessing social aspirations and desiring upward mobility, the appropriate lifestyle and corresponding consumption style must be identified and learned One can hypothesize that genuine social class mobility consists of this complementary movement through both class and status dimensions objective success as measured by career advancement, educational credentials, or personal wealth must find distinctive expression in the appropriate "higher" style of life. Consumption styles aspiring to this new standard, but found wanting, risk the kind of scorn that has traditionally been heaped on the excesses of the nouveau riche. And yet, because validated acceptance into a higher stratum depends primarily on the successful manipulation of the signs and symbols of the corresponding lifestyle, social acceptance may sometimes be Rained without real objective ascent.
Consumption, therefore, can be viewed as a kind of skill. Scitovsky (1976) makes this point within an historical context: "Until the end of the eighteenth century, education was a privilege of the leisure class and consisted, appropriately enough, of training in consumption skills" (p. 229). In our present age, one can speculate whether an apparent proliferation of consumption guides, as well as the continued emergence of influential taste-makers (e.g., fashion experts, architects, interior designers) that develop and maintain the status machinery, does not reflect a prevalent status anxiety which stems in part from individuals cross-pressured by class and status inconsistencies.
In virtually all societies, there are visible and significant differences in the availability and possession of valued items. Social class theory asserts that this unequal access to scarce resources and desired rewards is by no means a random process. There are mechanisms or social processes that contribute to both the consistency and permanence of structured social inequality . One such mechanism--the most crucial one--is provided by the marketplace, defining as it does both consumption possibilities (i.e., class variables) and actual consumption patterns (i.e., status variables). This paper has attempted to develop this insight further, focusing discussion on the Weber's seminal contribution to stratification theory and suggesting its relevance for consumer research.
Belk, Russell W. (1986), "Yuppies as Arbiters of the Emerging Consumption Style," in Advances in Consumer Research, Vol. 13, ed. Richard Lutz, Provo, UT: Association for Consumer Research, 514-519.
Belk, Russell W., Robert Mayer and Kenneth Bahn (1982), "The Eye of the Beholder: Individual Differences in Perception of Consumption Symbolism," in Advances in Consumer Research, Vol. 9, ed. Andrew Mitchell, Ann Arbor, MI: Association for Consumer Research, 523-530.
Bernstein, B. (1960), "Language and Social Class," British Journal of Sociology, 11 (September), 271.
Bernstein, B. (1968), "Elaborated and Restricted Codes: An Outline," Sociological Inquiry, 36 (Spring), 254-261.
Boskoff, Alvin (1969), Theory in American Sociology, New York: Crowell.
Boskoff, Alvin (1972), The Mosaic of Sociological Theory, New York: Crowell.
Bourdieu, Pierre (1973), "Cultural Reproduction and Social Reproduction," in Knowledge, Education, and Cultural Change, Richard Brown, ed. London: Tavistock.
Cohen, R., G. Fraenkel, and J. Brewer (1968), "The Language of the Hard-Core Poor: Implications for Culture Conflict," Sociological Quarterly, 9 (Winter), 19-28.
Coleman, Richard P. (1960), "The Significance of Social Stratification in Selling," in Marketing: A Maturing Discipline, Proceedings of the American Marketing Association 43rd National Conference, ed. Martin L. Bell, Chicago: American Marketing Association, 171-184.
Coleman, Richard P. (1983), "The Continuing Significance of Social Class to Marketing," Journal of Consumer Research, 10 (December), 265-280.
Curtis, Richard F., and Elton F. Jackson (1977), Inequality in American Communities, New York: Academic Press .
Davis, Allison, and John Dollard (1940), Children of Bondage, Washington, D.C.: American Council on Education.
DiMaggio, Paul (1982), "Cultural Capital and School Success: The Impact of Status Culture Participation on the Grades of U.S. High School Students," American Sociological Review, 47 (April), 189-201.
Dominquez, Louis V. and Albert L. Page (1981a), "Stratification in Consumer Behavior Research: A ReExamination," Journal of the Academy of Marketing Science, 9 (Summer), 250-271.
Dominquez, Louis V. (1981b), "Use and Misuse of Social Stratification in Consumer Behavior Research, r Journal of Business Research, 9, 151-173.
Douglas, Mary and Baron Isherwood (1979), The World of Goods, New York: Basic Books.
Ellis, R.A., and W.C. Lane (1966), "Social Mobility and Career Orientation," Sociology and Social Research, 50 (April), 280-296.
Gerth, Hans and C. Wright Mills (1953), Character and Social Structure: The Psychology of Social Institutions, New York: Harcourt Brace Jovanovich.
Gilbert, Dennis and Joseph A. Kahl (1982), The American Class Structure: A New Synthesis, Homewood, Illinois: Dorsey Press.
Goffman, Erving (1951), "Symbols of Class Status, British Journal of Sociology, 2, 294-304.
Hirisch, Robert D. and Michael P. Peters (1974), "Selecting the Superior Segmentation Correlate," Journal of Marketing, 38 (July), 60-63.
Hirschman, Elizabeth C. (1985), "Primitive Aspects of Consumption in Modern American Society," Journal of Consumer Research, 12 (September), 142-154.
Hodges, Harold N. (1964), Social Stratification: in American, Cambridge, Mass.: Schenkman.
Hollingshead, August B. (1950), "Cultural Factors in Selection of Marriage Mates," American Sociological Review, 15, 619-627.
Horton, J. (1967), "Time and Cool People," Transaction, 4 (April), 5-12.
Kahl, Joseph A. (1957), The American Class Structure, New York: Rinehart.
Kanter, Rosabeth Moss (1977), Men and Women of the Corporation, New York: Basic Books.
Kohn, Melvin L. (1969), Class and Conformity A Study in Values, Homewood, IL: Dorsey Press.
Laumann, Edward 0. (1966), Prestige and Association in an Urban Community, Indianapolis: Bobbs-Merrill.
(For remaining references, please contact author)