Consuming 
                Interests 
                >> Focus 
                groups, brand image, and other staples of modern advertising all 
                sprang from the work of a group of Chicago social scientists. 
                These pioneering market researchers used tools from psychology, 
                anthropology, and sociology to study a once-neglected topic: why 
                people buy stuff.
              
               BEER, 
                ACCORDING TO EXPERTS, "is 
                not a prestige item." There's nothing distinctive or exclusive 
                about it. Cheap and widely available, it does not require stylish, 
                single-task glassware, like Champagne flutes or brandy snifters, 
                but can be consumed from plastic mugs, recycled jelly jars, even 
                straight from the can or bottle.
BEER, 
                ACCORDING TO EXPERTS, "is 
                not a prestige item." There's nothing distinctive or exclusive 
                about it. Cheap and widely available, it does not require stylish, 
                single-task glassware, like Champagne flutes or brandy snifters, 
                but can be consumed from plastic mugs, recycled jelly jars, even 
                straight from the can or bottle.
              Regular 
                beer drinkers, social scientists agree, "do not fall all 
                over themselves" to imitate high society. They rarely dress 
                for the occasion. They may not even wear shirts. They drink beer 
                to quench their thirst, to relax, to grease the wheels of social 
                interaction and unleash spontaneity.
              Beer 
                "marks the absence of power relations, authority, or social 
                striving," says the foremost study of the subject. Beer advertisements 
                that cater to status seekers-depicting sophisticates in elegant 
                settings and formal attire, sipping this chilled, golden beverage 
                while basking in the good life-evoke only hostility among beer 
                drinkers. "Oh, the Ritzy Bitches like beer" is a typical 
                response, or "I'll bet one stinking glass of beer costs half 
                a buck in a place like that."
              As 
                the price reveals, this is not a recent study. But what students 
                of consumption now call the "use code" of beer hasn't 
                changed much since a cluster of curious Chicago social scientists, 
                assisted by a handful of graduate students and 350 hard-drinking 
                Chicagoans, probed the beer-buying world in the early 1950s.
              The 
                researchers belonged to a small firm called Social Research, Incorporated-SRI 
                for short-housed in the Hyde Park Bank building on 53rd Street. 
                They veered off from the University in 1946 to form the company 
                almost by accident, when a series of coincidences and opportunities 
                made a venture into the commercial world seem like a very good 
                idea indeed.
              Over 
                the next two decades SRI helped to revolutionize the field of 
                market research, transforming its assumptions, methods, goals, 
                and consequences in ways that quickly redirected the world of 
                advertising and slowly, very slowly, made consumption-"the 
                most studied single phenomenon in American life," according 
                to Andrew Abbott, AM'75, PhD'82, chair of sociology at Chicago-a 
                focus of purely academic inquiry.
               The 
                SRI team did it by applying social-science techniques to what 
                was then a neglected topic: why people buy stuff. The researchers 
                were among the first to adapt tools developed by anthropologists 
                for the study of primitive societies and use them to investigate 
                modern industrial communities. They borrowed personality tests 
                invented by psychologists to probe the meaning of purchasing decisions. 
                And they applied evolving notions of social status, mobility, 
                and class aspirations to understand what buying, owning, and displaying 
                consumer goods meant. In its short life SRI left a lasting impact 
                on theworld of market research, influences still felt in everything 
                from the absolute devotion to "brand image," to reliance 
                on focus groups, to academic study of the meanings of consumption.
The 
                SRI team did it by applying social-science techniques to what 
                was then a neglected topic: why people buy stuff. The researchers 
                were among the first to adapt tools developed by anthropologists 
                for the study of primitive societies and use them to investigate 
                modern industrial communities. They borrowed personality tests 
                invented by psychologists to probe the meaning of purchasing decisions. 
                And they applied evolving notions of social status, mobility, 
                and class aspirations to understand what buying, owning, and displaying 
                consumer goods meant. In its short life SRI left a lasting impact 
                on theworld of market research, influences still felt in everything 
                from the absolute devotion to "brand image," to reliance 
                on focus groups, to academic study of the meanings of consumption.
              Market 
                research, up until the end of World War II, consisted almost entirely 
                of crunching census data to determine how much of a given product 
                how many customers might buy. Shaped by the Depression, the field 
                used simple demographics to gauge often-limited demand for specific 
                products and to measure response to advertisements. As production 
                capacity and buying power increased, researchers began to look 
                for ways to stimulate new rather than measure existing demand. 
                "Our problem," summarized a senator from Wisconsin, 
                "is not too much cheese produced, but rather too little cheese 
                consumed." One component of the post-war effort to peddle 
                more cheddar was to shift research away from how much consumers 
                could be expected to eat toward what made them hungry-and how 
                to make them hungrier-a field that was dubbed motivation research.
              "This 
                is a completely different kind of research," noted the introduction 
                to SRI's Study of Consumer Attitudes on Beer and Beer Advertising. 
                "Instead of counting noses, this is diagnostic research. 
                The analyst is trained to look for basic attitudes, just as a 
                doctor looks for symptoms." As social scientists, SRI looked 
                for the social meanings of beer. "The attitudes of different 
                class levels toward beer," was its driving focus, along with 
                "the impact of typical beer advertising on different classes."
              The 
                new approach-finding the unconscious, intuitive, emotional factors 
                that drive consumption at each class level and grafting the most 
                alluring of those factors onto the product-caught on. By the mid-1950s 
                most major advertising agencies had hired motivation-research 
                consultants or had recruited teams of "whiskers" from 
                academia. "More and more advertising and marketing strategists 
                are adapting their sales campaigns to the psychologists' findings," 
                noted the Wall Street Journal in 1954. "Top-drawer 
                advertising agencies," echoed the trade magazine Printer's 
                Ink, "favor the increased use of social sciences."
              The 
                commercial success of motivation research soon provoked a backlash. 
                In 1957, journalist Vance Packard sounded a very public alarm 
                in The Hidden Persuaders, a book that plumbed the moral 
                depths of depth analysis and mass marketing while selling more 
                than 500,000 copies, enough to make it No. 1 on nonfiction best-seller 
                lists for six weeks. 
              "The 
                use of mass psychoanalysis to guide campaigns of persuasion has 
                become the basis of a multi-million dollar industry," Packard 
                told his readers. "Large scale efforts are being made, often 
                with impressive success, to channel our unthinking habits, our 
                purchasing decisions, and our thought processes by the use of 
                insights gleaned from psychiatry and the social sciences." 
                These "awesome" psychological tools and "ingenious" 
                anthropological techniques have been designed by a "breed 
                of persuaders known in the trade as the depth boys," he warned. 
                "Many of us are being influenced and manipulated, far more 
                than we realize."
              Motivation 
                research was selling us not just goods but also political candidates, 
                social attitudes, career and personal goals, or states of mind. 
                "We move into the chilling world of George Orwell and his 
                Big Brother as we explore some of the extreme attempts at probing 
                and manipulating now going on," Packard wrote, a world where 
                the consumer, the citizen, and the voter "more and more is 
                treated like Pavlov's conditioned dog."
              A 
                caricature of the depth boys even made it into the movies. Putney 
                Swope, Robert Downey's 1969 dark comedy about the advertising 
                business, opens with a presentation by Dr. Alvin Weasely, billed 
                as "one of the most respected motivational researchers in 
                the world." He has been called in to help a Madison Avenue 
                advertising agency boost sales for an unpopular brew by explaining 
                the unconscious motives of beer drinkers. "Beer," Weasely 
                pronounces, "is for men who doubt their masculinity. That's 
                why it's so popular at sporting events and poker games. On a superficial 
                level," he continues, "a glass of beer is a cool, soothing 
                beverage. But in reality, a glass of beer is pee-pee dickie."
              Though 
                filled with the unexpected, SRI's own in-depth consumer studies 
                of everything from soap to soap operas, department stores to lumber 
                yards, and alcohol, tobacco, and fast cars, were never that shocking 
                or implausible. Its researchers were grounded by a commitment 
                to serious social research, access to the latest tools of academic 
                social science, and the bias of the founders toward social class, 
                not psychological factors, as the basis for understanding consumer 
                decisions.
              The 
                late-1940s and 1950s were an odd time to fixate on social class. 
                "The prevailing post-war view," said Kim A. Weeden, 
                assistant professor of sociology at Chicago, speaking at a campus 
                conference last fall about SRI and the history of consumer research, 
                "was that class divisions in America were gradually disappearing." 
                Many social scientists believed that, with increasing affluence, 
                the different classes were converging in their culture, lifestyle, 
                values, and standard of living, but, said Weeden, "there 
                were of course exceptions."
              Perhaps 
                the most noticeable of the exceptions was a social scientist named 
                Lloyd Warner-SRI's founder and by all accounts its generative 
                intellect. Warner, who joined the U of C as a sociology professor 
                in 1935, had taught SRI's other two founders, Burleigh Gardner 
                and William Henry, PhD'44. An anthropologist by training, he'd 
                spent three years as a graduate student doing field work in Australia, 
                scrutinizing the social structure of an aboriginal tribe. But 
                he grew less interested in "primitives" and increasingly 
                convinced that the tools of social anthropology might better be 
                applied to modern American society-an idea that would not become 
                popular until the 1970s.
              "His 
                platform," recalls SRI colleague Lee Rainwater, AM'51, PhD'54, 
                "was that all human life partakes of the same basic species 
                behavior." If so, then the tools used to understand sacred 
                tribal rituals or daily routines should work just as well to understand 
                the Fourth of July or breakfast cereal. When Warner returned to 
                the U.S. in 1929 to take a position at Harvard, he decided not 
                to finish his dissertation on kinship among aborigines thousands 
                of miles away but to focus instead on the social systems of a 
                nearby small town.
              He 
                quickly became involved in his "Yankee City" project. 
                With 18 fieldworkers, mostly volunteers, Warner spent four years 
                (1930-34) studying Newburyport, Massachusetts, using versions 
                of the methods he had mastered while working with a primitive 
                tribe: observation, close analysis of social networks, and unstructured 
                interviews that allowed the subjects to wander wherever their 
                interests led.
              The 
                result was five books, the Yankee City series. The first volume, 
                The Social Life of a Modern Community (1941), emphasized the 
                role of social class as the basic structuring principle of urban 
                society, as powerful as kinship among Australian tribes. Unlike 
                the Marxist view, Warner's definition of class was based less 
                on how people got money and more on how they spent it. The means 
                of production had given way to the meanings of consumption.
              "In 
                Yankee City," notes Michael Karesh, AM'95, whose well-researched 
                sociology master's thesis on SRI has triggered a revival of interest 
                in the group, "different social classes were distinguished 
                not simply by income but by different goods, behaviors, values, 
                and points of view. To achieve upward mobility an individual not 
                only had to earn more money or buy better goods, but perceive, 
                value, and use goods in a different way. The shiny new Cadillac 
                of the upper-lower class was no substitute for the history-laden 
                house and furniture of the upper-upper class. To Warner, these 
                symbols marked the various social positions in a way that both 
                sustained the social structure and, for some individuals, allowed 
                mobility within it."
              In 
                his small-town studies and later in a study of Chicago financed 
                by the Chicago Tribune, Warner split the population into 
                six classes-from upper-upper to lower-lower-a rough formula followed 
                by most sociologists. Although his work created only a "respectful 
                stir in academic circles," wrote Packard, it churned merchandising 
                circles to a frenzy and "came to be regarded as a milestone 
                in the sociological approach to the consumer."
              
              BY 
                THE TIME 
                the 
                first Yankee City volume appeared, Warner had been lured away 
                from Harvard by Chicago's greater enthusiasm for interdisciplinary 
                work. He was followed by Burleigh Gardner, a country boy from 
                Texas who had come to Harvard to study anthropology and wound 
                up working on the Yankee City studies. Described by Packard as 
                a "mop-haired, slow-speaking, amiable man," Gardner 
                was ill at ease with scholarly pretensions and preferred life 
                on the fringes of academe. But in 1942 he was enticed into teaching 
                in Chicago's newly created Committee on Human Relations in Industry.
              After 
                reeling Gardner in, Warner stepped halfway out, to consult for 
                a new company, the Office for the Study of Social Communication. 
                Despite its scholarly name, the OSSC was started by a greeting-card 
                magnate to learn more about his customers. Warner and another 
                former graduate student, William Henry, adapted a series of the 
                "projective" tests then in use as a way to ferret out 
                the card consumers' unconscious motives, dividing buyers into 
                12 distinct personality profiles.
              When 
                those studies were finished, the OSSC dissolved, but in 1946-this 
                time with backing from Sears, Roebuck-Warner and Gardner formed 
                their own consulting group, SRI, to help companies investigate 
                employee and customer attitudes. They brought in Henry, who had 
                joined the Chicago faculty in 1944, to run the psychological testing. 
                Gardner, who had quickly tired of academic politics and meetings, 
                resigned from the University to become SRI's executive director.
              It 
                was the ideal arrangement, argues Karesh. Informally connected 
                but formally separate from the University, SRI could "acquire 
                the latest conceptual and methodological tools in the social sciences 
                and apply them to commercial ends."
              The 
                company quickly made a name for itself in the emerging field of 
                consumer motivation research, pulling together Gardner's interest 
                in commercial applications of social science, Henry's expertise 
                in psychoanalytic testing, and Warner's faith in the crucial importance 
                of social class.
              As 
                "senior consultants," Warner and Henry spent most of 
                their time on campus, and Gardner devoted himself to courting 
                new clients and writing articles for advertising periodicals. 
                The quotidian work was done by the firm's junior members, grad 
                students from human development and sociology who needed the money 
                at a time when there was little funding for graduate study in 
                the social sciences.
              One 
                of the first students on board, Harriet Moore, became director 
                of research, supervising staff training and study design and providing 
                "much of the intellectual force in [SRI's] day-to-day operation," 
                according to Sidney Levy, PhB'46, AM'48, PhD'56, another insolvent 
                grad student who arrived at SRI in 1948. By the early 1950s, the 
                core staff was in place. Moore, Levy, and Lee Rainwater, who came 
                in 1950, formed a close trio. Key members Ira Glick, AM'51, PhD'57; 
                Richard Coleman, PhD'59; and others soon followed. The professional 
                staff never grew very large, however, topping out at 17 in 1957.
              The 
                period, said Levy, was "the most exciting and intensely absorbing 
                in my life. We lived SRI from breakfast until bedtime, brooding 
                over methods and data gathering and seeking penetrating insights."
              "Much 
                of the excitement," notes Karesh, "followed from the 
                feeling on the part of those involved that they were part of a 
                pioneering team composed of brilliant minds exploring new intellectual 
                terrain." Because Gardner had a tendency to accept assignments 
                without knowing whether SRI could perform them, its members had 
                to be especially creative. "New concepts and methods were 
                generated internally," says Karesh, "or borrowed from 
                the University and then combined and applied in novel ways."
              "We 
                did the first qualitative study for the Coca-Cola company," 
                Levy recalled, "on why people drink soft drinks; the first 
                qualitative study for AT&T on the meaning of the telephone. 
                For the Wrigley Company we studied what baseball meant to Cubs 
                fans. A study for FTD, the flower delivery system, analyzed the 
                poignancy of flowers as symbolic of the life cycle, representing 
                and celebrating its beauty and fragility and the inevitability 
                of death."
              The 
                basic approach, said Levy, began with the so-called depth interview, 
                a free-style open-ended conversation. Sometimes SRI researchers 
                also interviewed two of three people at a time, or even larger 
                groups, a technique now known as focus groups. "Within this 
                more or less non-directive approach we embedded various projective 
                devices," he recalled, including variants on such clinical 
                techniques as the TAT, the Rorschach, sentence completion, word 
                association, draw a person, "and even the curious Szondi 
                test," now discredited, in which an individual was shown 
                eight photographic portraits and asked to choose the person he 
                would most, and least, like to sit next to on a long trip. The 
                twist was that each picture portrayed a person with a major psychiatric 
                disorder. A subject's selection was thought to reveal something 
                about the chooser's psychological needs.
              The 
                hallmark of SRI was its compulsion to assess the social status 
                of every subject. "We took pictures of people's houses and 
                living rooms," said Levy. "We sent interviewers to spend 
                whole days observing. We classified all our respondents so we 
                could examine the effects of social class on consumer behavior. 
                And we taught our clients about social stratification and the 
                structure of American society."
              
              
              SRI 
                WAS NOT THE NATION'S 
                first motivation research firm. There was a rival New York academic 
                group, the Bureau of Applied Social Research. Led by Paul Lazarsfeld, 
                a psychiatrist and mathematician from Vienna, BASR was closely 
                tied to Columbia University. Lazarsfeld, arguably the field's 
                first scholar, began writing about the psychology of market research 
                as early as 1934. But he got there "too early" for commercial 
                success, suspects Karesh. The research tools and the market weren't 
                ready. BASR survived into the 1960s but never developed SRI's 
                commitment to clients or freedom from university bureaucracy. 
                It didn't help that its best-known project was the Edsel.
              One 
                of Lazarsfeld's students had better timing. Ernest Dichter came 
                to the U.S. from Vienna in 1938. In 1946, after a series of jobs, 
                he started his own consumer research firm, the Institute for Motivation 
                Research. By the early 1950s the firm's success could be seen 
                in its glitzy headquarters, a 30-room, hilltop mansion just up 
                the Hudson from Manhattan. By 1956 it had conducted more than 
                500 studies, and by 1964 more than 2,500. Dichter, "a jaunty, 
                exuberant, balding man," who called himself Mr. Mass Motivation, 
                was the prime player in Packard's book, bragging that he maintained 
                a standby "psycho-panel" of several hundred families, 
                "whose members have been carefully charted as to their emotional 
                make up."
              The 
                Chicago version of Dichter was Louis Cheskin, director of the 
                Color Research Institute of America. Cheskin claimed to have begun 
                motivational research in 1935, when he backed into the field from 
                the area of package design, where he emphasized the role of color. 
                (We have him to thank for tinted toilet tissue.) His most famous 
                contribution was the sex change he performed for Marlboro cigarettes. 
                The brand was originally designed to appeal to women, with red 
                paper to mask lipstick smears. But far more men smoked. So Cheskin 
                designed a manly package and helped devise an ad campaign based 
                on "man-sized flavor," featuring rugged men, mostly 
                cowboys on horses. The men all had tattoos, to lend them a "virile 
                and interesting-past look." The ads still run. And Cheskin 
                Research, now headquartered in Redwood Shores, California, still 
                exists.
              While 
                Dichter and Cheskin were all business, most staff at SRI, despite 
                the income that came with commercial success, remained scholars 
                at heart. "As social scientists," recalled Levy, "we 
                were not content just to write proprietary research reports for 
                our clients. We thought about the larger implications of our specific 
                research projects," publishing regularly and presenting at 
                scholarly conferences.
              In 
                combining commercial and academic pursuits, SRI managed to unite 
                two formerly unconnected words into a phrase that ultimately defined 
                consumer research, and has shaped marketing ever since. Levy officially 
                coined the term "brand image" for an article he and 
                Gardner wrote in 1955 at the request of the Harvard Business 
                Review. They defined it as the "sets of ideas, feelings 
                and attitudes that consumers have about brands." But the 
                term also took in consumer impressions about who might be expected 
                to buy the product and what buying it told the world about the 
                owner.
              "Each 
                product or brand exists in people's minds as a symbolic entity," 
                recalled Levy, "an integrated result of all their experiences 
                with it in the marketplace." This meant that advertisements 
                could no longer just be about the merits of a product or about 
                price but had to enhance the product's aura. They were "an 
                investment in the long-run reputation of the brand." Like 
                a good Chicago student, Levy traced this notion back to three 
                sources: Plato's concept of an idealized form, William James's 
                musings on the social self as a consequence of recognition by 
                others, and his own study of the pert and perky personality of 
                Betty Crocker, a kitchen-bound homemaker who existed only as a 
                picture on General Mills packaging.
              The 
                first SRI report to mention brand image was "Automobiles-What 
                They Mean to Americans," a 1954 study commissioned by the 
                Chicago Tribune and instantly scooped up by contemporary 
                marketing journals. This was SRI's optimal subject. "The 
                American prizes his car above every possession," proclaimed 
                Pierre Martineau, director of research at the Tribune and 
                a fervent advocate of motivation research. "No other consumer 
                good at the time," argues Karesh, "was more heavily 
                laden with symbolic qualities."
              The 
                report was the first study to delve into what a product revealed 
                about those who bought it. "We fit people into slots by the 
                kind of cars they drive," notes the report. "The automobile 
                has come to be one of the most important ways we have of revealing 
                characteristics and feelings and motives. The car tells what we 
                want to be-or think we are."
              The 
                researchers found that a product's personality could be just as 
                important as performance or price. In fact, very few consumers 
                in the study, mostly lower-class men, had any real interest in 
                the technical aspect of new cars. Rather, the buying process came 
                down to "an interaction between the personality of the car 
                and the personality of the individual."
              The 
                researchers presented personality profiles for 18 domestic cars, 
                assessing the social status associated with each and combining 
                that with personality factors tied to make, color, and accessories, 
                rating cars for prestige and ability to attract attention-and 
                how they did so. Was a car overtly ostentatious, fairly flashy, 
                slightly stylish, or consciously inconspicuous?
              Take 
                the Cadillac, "America's dream car." Cadillac conferred 
                the highest status, offered the most luxury, was seen as the best 
                built of any U.S. automobile. On the down side, many people resented 
                it as too snobbish or snooty. It was the car for new money, for 
                those who needed the status boost, especially "people of 
                deprived origins." The truly rich, SRI reported, the upper-upper 
                class, wouldn't go near it; they drove beat-up old station wagons, 
                displaying their indifference through deliberate downgrading.
              Buicks 
                were socially a notch below Cadillac but for those on the way 
                up. They were seen as reliable, sturdy cars for substantial people. 
                Fords were for the young, singles, hot rodders-fast, flexible 
                and rugged, chic, modern, and a bit showy. Plymouths were sensible, 
                inexpensive, small, with neutral styling. "Plymouth receives 
                very little criticism," noted the report, "but neither 
                does anybody get wildly excited about it." Shortly after 
                the study came out, the Chrysler Corporation, Plymouth's maker, 
                overhauled the image of all its cars.
              Thanks 
                to the press from studies like this and a flurry of customers 
                inspired by The Hidden Persuaders, SRI was extraordinarily 
                busy throughout the late 1950s. In 1960 the firm was able to return 
                Packard's favor; it made one of his most ominous predictions come 
                true.
              "To 
                get some publicity," recounted Lee Rainwater, "and to 
                satisfy our own curiosity, and to overcome the prejudice against 
                a Catholic in the White House," an SRI team analyzed the 
                1960 presidential candidates Nixon and Kennedy. They found that 
                Kennedy had one serious weakness, the perception that he was too 
                immature, too subordinate to his powerful family.
              The 
                key to changing this perception, they decided, was to build up 
                JFK's stature as the head of his own independent family through 
                the popular image of his spouse. The Democrats had to show that 
                Jacqueline Kennedy was a capable, intelligent, accomplished wife 
                and mother. Warner drew up a memo that sketched out ways to do 
                this and took his proposal to friends in the Kennedy campaign. 
                They went over it in detail with Mrs. Kennedy, then arranged for 
                her to do a series of televised interviews. Kennedy, of course, 
                won by a narrow margin. How much difference SRI made will never 
                be known; there was no funding for a follow-up evaluation.
              
              THROUGH 
                THE YEARS, 
                SRI's work largely remained outside the academic pale. Graduate 
                students could not get credit for their SRI research, no matter 
                how clever, or use SRI data for scholarly analysis. And interest 
                in the cultural properties of consumer goods was still seen as 
                "intellectual slumming" by serious scholars, said anthropologist 
                and consumption scholar Grant McCracken, AM'76, PhD'81, author 
                of Culture and Consumption (as well as a work with a less 
                intimidating title, Big Hair). Studying what people buy 
                was seen as applied rather than pure research, tainted by contact 
                with the business world, an elitist bias that has "kept social 
                scientists away from consumer research for decades."
              
              For 
                a while, SRI brought both sides together, but not for long. Several 
                staffers, those who never finished their dissertations, were devoted 
                to commercial work. A few left to start their own firms. Others, 
                like Warner, retained a primary commitment to scholarship. Lee 
                Rainwater and Richard Coleman managed to produce scholarship while 
                at SRI but were increasingly frustrated by the confidential nature 
                of most of their applied work.
              Rainwater 
                left in 1963 to teach sociology at Washington University in St. 
                Louis, joining Harvard in 1969. Coleman quit SRI in 1969 for the 
                Joint Center on Urban Studies at MIT and Harvard, moving on to 
                teach marketing at Kansas State University. Originally captivated 
                by commercial work, Levy migrated to the academic camp and joined 
                Northwestern's marketing department in 1961, becoming its chair 
                in 1980.
              As 
                social-sciences funding increased in the 1960s the firm's supply 
                of student talent dwindled, and in the 1970s newer, computer-driven 
                quantitative techniques triumphed over SRI's highly interpreted 
                qualitative studies. By the mid-1970s the firm existed in name 
                only and motivational research had lost the limelight. It had 
                answered the questions that it could, then gave way to other forms 
                of consumer research.
              Corporate 
                interest in consumer motives never wilted, however, and interest 
                in consumption research slowly crept into favor in academe. The 
                Association for Consumption Research, the first organization of 
                academic researchers in the field of consumer behavior, was founded 
                in 1971. Anthropologists began to take Warner's lead by studying 
                their own societies, and the social sciences as a whole began 
                to acknowledge consumption as a legitimate topic-although they 
                maintained their distance from applied or marketing research, 
                a stance that often forced them to reinvent established techniques. 
                
              Societal 
                and academic interest in social class also began to rise in the 
                1970s, but it focused more on poverty and extreme wealth, people 
                at the fringes rather than the middle majority, and thus had less 
                impact on marketing. Richard Coleman humbly noted at the SRI conference 
                that his 1983 article, "The Continuing Significance of Social 
                Class to Marketing," was ironically the last article ever 
                published by the Journal of Consumer Research to feature 
                social class as a major variable.
              Indeed, 
                Warner's six-category system of class now seems somewhat simplistic. 
                "Where," wondered Coleman, "would most Americans 
                place the high-income nerds of Silicon Valley?" Today researchers 
                stress the emergence of overlapping or competing hierarchies. 
                Just as there are now several hundred television channels catering 
                to very specific interests, instead of the former big three networks, 
                there are multiple vertical as well as horizontal social clusters, 
                each with its own internal rankings. "Class takes a back 
                seat to racial identities, gender, ethnic heritage, or religious 
                identities," said Weeder. "Labor is out, feminism is 
                in."
              Even 
                the car market is "no longer that useful in studying social 
                class," said Coleman. "Now, what people want has very 
                little to do with class. It has to do with whether one wants to 
                feel sybaritic or sexy or whatnot."
              "As 
                a practitioner," said Leo J. Shapiro, AB'42, PhD'56, founder 
                of a market research firm that competed with SRI, "and I 
                am about as pure a practitioner as you will find," he told 
                the conference, "I can tell you that Lloyd Warner's concepts 
                and techniques were adopted commercially and used by practitioners 
                and academicians, but they are no longer relevant." That 
                is not because they don't explain things about society or help 
                us anticipate the future. "They are no longer relevant because 
                they have been displaced by more economical techniques."
              Social 
                research by practitioners, he explained, is not about discovering 
                profound insights. It is about finding some tiny new thing that 
                gives the client a competitive edge, "a little bit of truth 
                that will let him get his money back." Why use something 
                as stunningly complex and costly as class if you can plug in a 
                much simpler variable and get similar results?
              A 
                much simpler-and thus more efficient-variable is knowing who the 
                people are who buy what you sell. This information is increasingly 
                available, thanks to the spread of catalog shopping, electronic 
                transactions, loyalty incentive cards in stores, and the Internet. 
                "The ultimate unit for analysis is not the person," 
                said Shapiro, "but the act."
              And 
                each act is being recorded. "There are data sets that include 
                every single electronic act of consumption for the last several 
                decades," noted Abbott. "Let me assure you," said 
                Jonathan Frenzen, AB'78, MBA'82, PhD'88, clinical professor in 
                the Graduate School of Business, "Big Brother is watching, 
                and he is watching increasingly effectively." Once you purchase 
                clothes via catalog, he emphasized, the vendor knows your address, 
                your VISA number, exactly what you purchased, and your measurements.
              Shopping 
                on the Web reveals even more, he added. Marketers can track your 
                click stream, "where you go and how long you pause on each 
                page." Amazon.com, for example, uses a technique called collaborative 
                filtering to chart your purchases and then suggests other books 
                that are somehow related, items already purchased by people who 
                look, at least to Amazon, a lot like you.
              It 
                works. Two weeks before a book on healthy eating by the University's 
                Michael Roizen was published, Amazon e-mailed everyone who had 
                bought the doctor's previous book on healthy aging, or any similar 
                book. Without a single ad or book review, even before the book 
                was released, it was a top-five seller for Amazon. "This 
                kind of behavior," predicted Frenzen, "is soon going 
                to follow you to the browser in your car."
              In 
                an odd way, such focused data gathering may lessen the hovering 
                menace of Big Brother's constant attention. While SRI's researchers 
                wanted to know everything about a few representative customers, 
                current commercial consumer research concentrates on very specific, 
                isolated acts of millions of essentially classless and faceless 
                customers.
              This 
                might be one more reason that consumption has remained so long 
                outside the academic universe. The consumer research industry 
                collects staggering volumes of information, figures that may be 
                extremely important if you're paid to sell a product but, details 
                that, said Frenzen, "may be viewed as being simply too trivial 
                for academics to worry about."
              If 
                social science has been slow to embrace commerce since the emergence 
                of firms like SRI, commerce has fallen head over heels for social 
                science. "Ten years ago I told Andy [Abbott] there was more 
                social science done outside academia than within universities," 
                said Eric Almquist, head of the customer management team for Mercer 
                Management Consulting. Today market research is a $14 billion 
                industry. By 2003 the total will reach $21 billion.
              "Social 
                science is central to business," argued Almquist, "and 
                growing even more central. Winning and losing in market capitalization 
                is increasingly determined by how well companies anticipate changes 
                in the marketplace, a social science issue."
              Almquist 
                recently completed a study that looked at sudden drops in corporate 
                share price. Over the last five years, 10 percent of the Fortune 
                1000 companies lost more than one-fourth of their shareholder 
                value within a month. "These are what we call value collapse," 
                Almquist said. "Customer research turns out to be the single 
                biggest thing companies 
                could have done better to avoid this catastrophe." Rubbermaid, 
                for example, lost touch with one customer: Wal-mart. For Readers' 
                Digest the major problem is age. Their consumers are aging 
                "faster than time," explained Almquist. "They have 
                retired permanently."
              His 
                own firm uses modern economic techniques such as discrete-choice 
                analysis to measure effects created by brand image as defined 
                by Levy and Gardner in the mid-1950s. "We've been trying 
                to push their insights further," Almquist said, "and 
                to link brand image to demand curves." Researchers now can 
                decipher exactly how much a brand's image is worth, how that image 
                shifts demand toward or away from the brand. If the brand's appeal 
                goes up, the company can raise the price, or sell more goods.
              The 
                best example is motorcycle maker Harley-Davidson, which, according 
                to Almquist, has the world's strongest brand image. While a Harley 
                costs $18,000, the Japanese equivalent runs about $6,000. Yet 
                Harley, at three times the price, "does not have the quality," 
                said Almquist. "Harleys have more problems. They require 
                more service. You have to wait six months to get one." But 
                if you're serious about the biker lifestyle, freedom, the open 
                road, and bugs in your teeth, you do not buy the Japanese knock-off 
                because it is not a Harley-Davidson. You don't see many people, 
                he added, with Yamaha or Suzuki tattoos.
              "The 
                men and women of SRI were exploring uncharted terrain when they 
                began their work," said Andreas Glaeser, assistant professor 
                of sociology, at the November conference. Decades later, "academic 
                cultural analysts have finally begun to take up some of their 
                concerns": the use of commodities as cultural symbols, the 
                global impact of brand names and logos (which have become more 
                readily identifiable than most religious symbols), and the absolute 
                requirement for multi-national companies to study culture and 
                cultural differences. 
              All 
                this should have been apparent 50 years ago, Glaeser concluded, 
                in SRI's study of beer. "In the parlance of contemporary 
                consumption research, SRI was investigating the practices of beer 
                consumption while investigating the semiotics of drinking," 
                Glaeser said. "SRI identified the major dimensions of the 
                use code of beer...teasing out, for example, the social resonance 
                of clinking beer glasses...and made concrete suggestions to their 
                clients on how to construct brand codes in accordance with use 
                codes, all this without using a language of identities, practices, 
                and semiotics, long before the academic world picked up an interest 
                in a related set of topics."
              Not 
                only that, he might have added, but SRI also said it in a way 
                clients and consumers easily understood. "Beer," SRI 
                pronounced, in the parlance of people who drink the stuff, "is 
                not a prestige item."
              
              
              
              
              