The Portrayal of African-American in Business-To-Business Direct Mail: A Benchmark Study.

The purpose of this article is to extend latitudinally the stream of research regarding racial minorities in industrial advertising and to provide a benchmark for future longitudinal studies. It utilizes content analysis to examine business-to-business direct mail advertising depiction of African-Americans.

Findings indicate that about 28 percent of the direct mail pieces depicted blacks, nearly six percent of all people portrayed were African-American, there was virtually no difference in occupational portrayals of blacks and whites, and blacks were most frequently portrayed separately from whites. Implications of these findings for users of industrial direct mail are discussed.

Introduction

There has been a long history of academic research into the frequencies and portrayals of racial minorities in advertising (for example, Choudhury and Schmid 1974; Gox 1970; Humphrey and Schuman 1984; Kassarjian 1969; Shuey, King and Griffin 1953; Stearns, Unger and Luebkeman 1987; Taylor and Steam 1997; Taylor, Lee and Stern 1995; Wheatley 1971; Zinkhan, Qualls and Biswas 1990). This stream of research has also been extended to focus on the presence of the black minority in business publication advertising (Stevenson 1991; 1993). However, there has been no analysis of the frequency and portrayal of blacks in the business-to-business direct mail medium despite the rapid increase in the use of this medium and the increase in numbers of African-Americans (the terms "African-American" and "black" are used interchangeably throughout this article) occupying positions as industrial purchasers, managers, entrepreneurs and engineers. This void in the literature should be filled because the portrayal of blacks in bus iness-to-business direct mail has both strategic and societal implications.

From a strategic standpoint, business-to-business marketers need to effectively respond to the increasing African-American presence to ensure that their messages are reaching such decision makers, and research has shown that black models may be more effective than whites in communicating with the black audience (Choudhury and Schmidt 1974; Gibson 1969). Indeed, it has been shown that blacks display more positive attitudes to-ward the advertisement and the product itself when black models are portrayed (Pitts et al. 1989; Whittler 1989; 1991). Moreover, there appears to be little risk in using visible minorities in ads because several studies have indicated that there have been generally positive audience responses to ads containing blacks (Petroshius and Crocker 1989; Singer 1983; WhittIer 1989). Further, from a societal point of view, appropriate media portrayals can support the movement of minorities into managerial positions, particularly if stereotypical portrayals are replaced with more realistic repres entations (Faber, O'Guinn and Meyer 1987). It is the purpose of this paper to expand the literature on the black presence in print advertising by examining the frequency and portrayals of African-Americans in business-to-business direct mail. A secondary purpose is to provide a benchmark for future studies of minorities in direct mail advertising.

Background

Buying Center Influences

Blacks are increasingly occupying buying center positions as professional buyers, executives, engineers, managers, and administrators in industry. From 1983 to 1995 for example, there was an increase of more than 100 percent, a 1,327,000 worker increase, in the number of African-Americans employed in managerial or professional occupations. In the executive, administrative and managerial ranks specifically, there was an increase of more than 731,000 blacks, to about 7.2 percent of those employed in this capacity. By 1995, nearly 9,000 purchasing agents were black, up from 5.1 to 6.6 percent, and the number of black engineers had risen by more than 48,000, from 2.7 to 4.7 percent, in comparison to 1983. Further, there were more than 250,000 black sales supervisors and proprietors in this country in 1995, a more than 136 percent increase from 1983 (Statistical Abstract of the United States 1996). The importance of these changes in the purchasing decision process is borne out by recent Simmons (1996) data showin g that 7.7 percent of those involved in industrial purchase decision making are black.

In response to these numbers, it behooves industrial marketers to develop advertising messages that can reach and influence the black segment of the industrial market. And there is ample opportunity to do so because business publication advertising is increasing dramatically as business marketers seek to offset the rising costs of direct sales calls (Hartley and Patti 1988). This is especially evident in the case of direct mail where sales have grown dramatically in the business-to-business sector of the market. Since 1985, annual growth rates have averaged more than seventeen percent, and total 1993 business-to-business mail order sales volume was expected to exceed $120 billion (Sroge 1991, p. 27). This growth continued as increasing numbers of business-to-business marketers sought to enhance their promotional efforts with direct mail (Goerne 1992). The Direct Marketing Association reports that around 1.7 million firms in the United States use business-to-business marketing techniques and more than 33 perc ent of that number are prone to use direct marketing (Gitlitz 1995). This is expected to generate nearly $115 billion in business-to-business sales specifically from direct mail advertising (Endicott 1995). Moreover, as industrial field sales costs continue to rise, recently averaging $292 per call ("Sales Calls Expensive..." 1992, p. 11), the appeal of direct mail should continue to grow, especially since it usually takes several face-to-face visits to complete a sale.

Advertising Appeals

Despite the increasing use of business-to-business direct mail and the numerous studies that have emphasized the importance of using African-American models when advertising to black consumers (e.g., Ortizano 1989; Stearns, Unger and Luebkeman 1987; Whittler 1991; Whittier and Dimeo 1991; Wilson and Biswas 1995; Zinkhan, Qualls, and Biswas 1990), there are relatively few studies that specifically address the presence of blacks in industrial advertising (Stevenson 1991; 1993), and none that examine the black presence in business-to-business direct mail. Nevertheless, if research conducted on African-American consumers can be generalized to blacks in the buying center, black models may prove more effective in communicating with this increasingly important purchasing segment. This is because peer appeals are more persuasive than those made by other groups (e.g. Cantor, Alfonso and Zillman 1976; Newcomb 1943). Indeed, in summarizing five decades of sociological and psychological research, Mackie, Asuncion and Wo rth (1990) stated that "research programs have almost without exception demonstrated that other members of an individual's in-group have significant power to persuade that individual" (p. 812). Perhaps this explains why Whittler (1991) found that blacks were more likely to purchase from companies that used black models in promoting their products. He concluded that advertisers "cannot afford to offend this consumer segment by not including blacks in their commercial messages" (Whittler and Dimeo 1991, p. 45), because race is a salient communicator cue. This may be particularly relevant in the case of industrial direct mail advertising where straight rebuy and modified rebuy purchasing decisions are apt to be low rather than high involvement (Hutt and Speh 1998). In such cases, peripheral cues (race) have been found to be an important factor in attitude formation and change (Petty, Caioppo and Schumann 1983).

Furthermore, it has long been recognized that, similar to consumer buyers, people involved in industrial buying are motivated by organizational as well as personal concerns (Hutt and Speh 1998), and that "each party in the buying process has subtle roles and needs" (Bonoma 1982, p. 112). As Cravens (1987) stated:

Like consumer buyers, organizational buyers are influenced by a variety of interpersonal factors, including needs, attitudes, perceptions, and descriptive characteristics (demographics, life-style, and personality) (p. 217).

One recent study of purchase motivations in family-owned industrial businesses--which represent the majority of business establishments--found that personal and social objectives strongly influence purchase decisions (File and Prince 1996).

Perhaps because of such motivations, the importance of peer appeals, and the growing presence of African-Americans in the buying center, there has been an increase in the portrayal of African-American models in business print media (Stevenson 1991; 1993). To illustrate, a recent Federal Express ad used a black female model to portray an executive in a TV commercial. The rationale from FedEx's marketing department was "that women -- often minority women -- frequently make a company's decision about which package service to use" (Wynter, 1993, B6).

Previous Studies of Minority Presence in Print Advertising

In researching the black consumer market, Gibson (1969) and Choudhury and Schmidt (1974) found that the use of black models in print media might determine, to a great degree, who gets the black segment of the consumer market. Despite this, some recent research on general magazine advertising indicates that a small percentage of magazine advertisements were using black models and that percentages of black portrayals trail population percentages. To illustrate, Ortizano (1989) analyzed 2,502 magazine ads published in 1984 and found that only 5 percent of the general audience magazine ads depicted blacks. Zinkhan, Qualls and Biswas (1990) found that the percentage of ads using black models in 1986 was only 4.37 percent. Taylor, Lee and Stern (1995) looked at more than 1,600 ads from magazines published in 1992/1993 and found that 11.4 percent of the sample depicted black models. In the consumer direct mail medium, Wilson and Biswas (1995) found that the representation of blacks in consumer specialty catalogs wa s about 4 percent. Each of these studies concluded that the percentages of blacks in consumer studies trailed their representation in society. This is surprising because there has been continuing pressure from black interest groups to increase black presence and improve African-American role portrayal in ads (Joseph 1986; Reiss 1983). Further, the continuing racial tension in this country indicates that attention to racial issues in the media is still needed. Indeed, one survey of marketing and media executives (the majority of whom were white) found that more than half felt that through stereotyping and under representation, advertising media were contributing to the racial unrest in the United States (Ward 1992).

With regard to specific studies of black representation in industrial advertising, Stevenson (1991) noted that the percentage of ads depicting blacks in trade publications had increased to 10.6 percent in 1986 from 4.9 percent in 1966. A later study of horizontal business publications showed that the percentage of ads with blacks portrayed increased from 3.1 percent in 1957 to 13.8 percent in 1987 (Stevenson 1993).

Beyond the question of frequency of appearance of blacks in ads, qualitative portrayals have also come under scrutiny. Two qualitative issues that have been examined in previous studies are occupational portrayal and racial composition in ads depicting blacks. With regard to occupational portrayal, Cox (1970) found a change from unskilled occupational portrayal to skilled occupational portrayal from the 1949-50 period to the 1967-68 period. Kassarjian (1969) found that black occupational status in advertisements had improved from 1946 to 1965 and that those portrayed in professional, managerial, and clerical occupations had risen from less than one percent to nine percent over the 20-year period. Humphrey and Schumann (1984) found only 14 percent of blacks portrayed as laborers and Stearns, Unger and Luebkeman (1987) found that occupational skill level of blacks portrayed was no different than that portrayed by whites. In the case of industrial trade journal advertising, the study that analyzed African-Ameri can occupational portrayals in trade journals found that there was a decline of African-Americans in blue collar portrayals and an increase in black professional portrayals from 1966 to 1986 (Stevenson 1991). In the study of black portrayals in horizontal business magazines (Business Week, Forbes, Fortune) Stevenson (1993) found a significant decline in the portrayals of blacks in blue collar occupations during the period from 1967 to 1987, and a significant increase in black portrayals as managers and professionals from 1957 to 1987.

In looking into the issue of racial composition in magazine advertising, Kassarjian (1969) found a significant decline in racially mixed non-peer ads and an increase in mixed peer ads over the period from 1946 to 1965. Humphrey and Schumann (1984) found that the percent of ads depicting blacks in mixed race peer settings had increased from virtually none in 1950 to 65 percent in 1980. In trade publications the ads that depicted blacks in mixed racial peer settings changed little from 43 percent in 1966 to 47 percent in 1986 (Stevenson 1991). However, in horizontal business publications, the most frequent portrayal over the 30-year period from 1957 to 1987 was mixed peer ads (Stevenson 1993).

Research Focus

The purpose of this study is to expand the research regarding the quantitative and qualitative portrayal of blacks in industrial advertising into a new medium, business-to-business direct mail, and to determine if the representation of African-Americans in this medium is consistent with that found in other print media. The specific questions to be examined are:

- What is the frequency of black portrayals in comparison to white portrayals in business-to-business direct mail advertising?

- What are the occupational portrayals in ads depicting blacks in business-to-business direct mail advertising, and how do they compare to the portrayals of whites?

- What is the racial composition depicted in ads portraying blacks in business-to-business direct mail advertising?

Methodology

Data Collection

The direct mail pieces analyzed in this study were obtained via the cooperation of "list houses." List houses compile lists of consumer or business-to-business customers on a national basis. For example, American Business Information, Inc. compiles lists of U. S. businesses from yellow page directories; annual reports, 10Ks, and other SEC data; federal, state, and municipal government data; leading business magazines, newsletters, and major newspapers; and postal service information. The result is a database of over 11 million businesses.

An organization that is interested in marketing beyond its current customers using direct mail to sell a product or service contracts with a list compiler to lease a list of potential customers. To verify that the purchasing organization uses the list only for as many times as it has paid to use it and to verify that the direct mail piece that was originally submitted was in fact the one that was sent, list compilers use "ghost names." One of the authors volunteered to become a ghost name to verify use and content for the list compiler. A variation in spelling of the author's last name was used for the ghost name. No data was kept by the list compiler as to how many times a list was sold with the ghost name. In effect, the author's misspelled name was merely added to the database for all business-to-business direct mail pieces. In this manner, more than 600 business-to-business direct mail packages were received, over a one year period, from the 16 list houses that cooperated in this study. All duplicate pie ces were immediately discarded, leaving a usable sample size of 486 unique direct mail pieces. It is possible that discarding duplicate pieces biased the results. However, including an identical piece with a disproportionally high or low racial representation could also introduce bias. According to Kassarjian (1969), there is no compelling reason that duplicate pieces be either counted or discarded. The decision to discard was consistent with the approach taken in a recent study of the minority presence in consumer direct mail catalog advertising (Wilson and Biswas 1995).

Direct mail pieces often consist of several distinct parts: the envelope, the letter, the brochure or catalog, the order form, and the reminder, each of which has a specific function in achieving the sale. Because multiple "parts" make up a direct mail "package" and each part may depict people, all pieces were content analyzed according to the following procedure. First, each piece was content analyzed in order to determine if there were any people depicted. Since no people were depicted on any of the direct mail envelopes, order forms or reminders, these were not analyzed further. Drawings, cartoons, and pictorial pieces not portraying actual people were also disregarded as were those that did not show the model's face. The remaining letters, brochures and catalogs that portrayed people were then analyzed to determine the number of people portrayed, the number of blacks portrayed, the occupations depicted, and the interactions portrayed in packages depicting blacks. This process is similar to that used in p revious studies of other print media (Cox 1970; Kassarjian 1969; Sterns, Unger and Luebkeman 1987; Stevenson 1991; 1993; Taylor and Stern 1997; Taylor, Lee and Stern, 1995); it resulted in the content analysis of nearly 2100 separate photographs. The items advertised were primarily operating supplies, business services, and accessory equipment.

Coding Categories

Frequencies of black portrayals were evaluated on three dimensions: percentage of pieces containing blacks; percentage of separate pictures portraying blacks; and blacks as a percentage of all people depicted in the ads. Most previous studies of minority presence have calculated minority presence as the percentage of ads portraying minorities compared to the total number of ads portraying people (e.g., Kassarjian 1969; Sterns, Unger and Luebkeman 1987; Taylor, Lee and Stern 1995). This percentage was then compared to the minority percentage of the U. S. population. Others (Colfax and Sternberg 1972; Stevenson 1991; 1993; Wilson and Biswas 1995) have additionally calculated blacks as a percentage of all people portrayed in the ads. Both approaches were used in this study to give a more complete picture of African-American representation.

Roles were evaluated on two dimensions: occupational portrayal and racial composition. Occupational status was coded using categories developed by the Department of Labor (U. S. Department of Labor 1982). The categories were occupational: blue collar; managerial/professional; entertainer; clerical; and non-occupational: family; recreational; other. These categories were used in previous industrial print media studies (Stevenson 1991; 1993). Additionally, since one objective of the study was to compare black occupational portrayal to that of whites, each black occupational portrayal was paired with a white portrayal in the next ad which contained a white model. This methodology is consistent with that of previous authors (Cox 1970; Stearns, Unger and Luebkeman 1987).

The dimension of racial composition was scaled as in previous print media content analysis studies (Kassarjian 1969; Humphrey and Schumann 1984; Stevenson 1991; 1993). Categories for racial composition were: all models black; separate pictures (panel ad, no blacks and whites in same scene); mixed nonpeer (secretary and manager, for example); mixed peer (two clerical workers, for example).

Coding Procedures

As in any content analysis study, there was the potential for methodological problems in gathering data. In anticipation of such problems, several procedures were followed to enhance the reliability and validity of the data collection process (Kolbe and Burnette 1991). Two judges, both MBA students, were employed by the authors to collect and analyze the data. The judges were carefully trained regarding the coding scheme and used a coding sheet containing the operational definitions employed (see Appendix). A pretest of judges' accuracy and the coding scheme was made prior to the actual collection of data for the study, then one judge independently analyzed all packages. Finally, the second judge independently analyzed ten percent of all the packages included in the study and 100 percent of the ads in which blacks were depicted. Once it was determined that satisfactory levels of interjudge reliability were obtained, utilizing Kassarjian's (1977) agreement standard for reliability of agreement, the second jud ge analyzed the complete data set.

Reliability

Interjudge reliability for the entire data set was calculated using Cohen's Kappa (Cohen 1960) to correct for chance agreement. The results were .987 for model race (all ads), .887 for model race black ads only, .886 for occupational role, and .884 for racial composition. Due to limitations of Cohen's Kappa (Rust and Cooil 1994), reliability was also calculated using the Perreault and Leigh (1987) index. Estimates were as follows: model race (all ads) .991; model race black ads .969; occupational role .948; and racial composition .943.

Findings

Frequency of Appearance

Table 1 reports the findings regarding racial frequency of appearances. There was a statistically significant difference in the relative frequency of racial representation in business-to-business direct mail advertising ([X.sup.2] = 132.84; p [less than] .001). There were 486 direct mail packages analyzed in this study; 181 of the packages depicted people. Each of these packages depicted white people somewhere in the piece, that is, in no case did a direct mail package that depicted people not include whites. Additionally, 51 (28.2 percent) of the 181 packages depicted some blacks; 26 (14.4 percent) also included another minority (Asian, Hispanic, Native American). These findings are summarized in row one of Table 1.

Row two in Table 1 presents the data in a different manner. It shows racial portrayals as a number and percentage of each separate picture that depicted people in the 181 direct mail pieces. In other words, ten direct mail pieces might include 40 different pictures. Each of these pictures would offer the opportunity to portray people in numerous settings. If two of the ten pieces depicted an African-American, then 20 percent (2/10) would have been counted as containing blacks (row one of Table 1). However, if the two pieces that portrayed blacks represented only two of the 40 different pictures of people, the percentage of black portrayals (2/40) would be five percent. The percentage for this study is shown in row two (Table 1).

Analyzing each separate picture is consistent with scores of studies of the presence of minorities in magazine advertising (e.g. Bush, Hair and Solomon 1979; Humphrey and Schuman 1984; Ortizano 1989; Stevenson 1991, 1993; Zinkhan, Quails and Biswas 1990) where each separate picture was content analyzed for its racial portrayals. The results of this analysis showed that 136 (6.5 percent) of the 2,092 pictures analyzed portrayed blacks; other minorities were depicted in 56 (2.7 percent) of the pictures; and whites were depicted in 2,052 (98.1 percent) of the pictures (Table 1 row two).

Also shown in Table 1 (row three) is the number of blacks portrayed in comparison to the number of people of all races depicted in the ads, i.e., what percentage of the people portrayed are African-American. The preponderance of minority presence studies has compared the number of ads portraying blacks to the population of blacks in the United States. Yet the a priori criterion for "acceptable" frequency levels has been that the proportion of blacks in ads should equal the proportion of blacks in the population (e.g., Bush, Resnick and Stern 1980; Wheatley 1971), perhaps more accurately gauged by the analysis which generated row three in Table 1. This more detailed analysis is consistent with the approach taken in recent studies of the minority presence in business-to-business print advertising (Stevenson 1991; 1993) and in some studies in consumer media (e.g. Colfax and Sternberg 1972; Wilson and Biswas 1995).

It can be seen (row three Table 1) that the percentage of models who were black (5.8 percent) depicted in the direct mail packages analyzed in this study was much lower than both the percentage of separate pictures depicting blacks (6.5 percent) and the percentage of packages depicting blacks (28.2 percent). There were 2,561 people portrayed in the 181 packages analyzed; 149 (5.8 percent) were blacks, 63 (2.5 percent) were other minorities (Asians, Hispanics, Native Americans); the remainder (91.7 percent) were white.

Occupational Portrayals

Table 2 shows the occupational roles portrayed by African-Americans, and compares them to the portrayals of whites. As can be seen, there were virtually no differences between the races in occupational portrayals. The most frequent portrayal of both whites and blacks was in blue collar roles; 76 (51 percent) of the black models and 74 (49.7 percent) of the white models were portrayed thus. The second most frequent portrayal of blacks (26 percent) was in managerial/professional roles; 28.2 percent of the matched sample of whites were also portrayed in this manner. The only other occupational portrayal used was clerical; it was used in about eight percent of the portrayals of both blacks and whites. The balance of portrayals of both blacks and whites was in the non-occupational/recreational category where, for example, an ad might portray a person in an outdoor setting, or exercising.

Racial Composition

In Table 3 is shown the racial composition of the 136 separate pictures that portrayed African-Americans. It can be seen that the most frequent portrayal (63 percent) in such ads was blacks portrayed separately from whites with or without another black model in the ad. Less frequent (33 percent) was the portrayal of blacks and whites together as peers (all as professional or all as warehouse workers, for example). The least frequent of the portrayals (4 percent) in ads where blacks and whites were shown together was a non-peer portrayal (e.g. manager with production worker). However, in each of the mixed non-peer portrayals, the white model was always portrayed in the higher status role. For example, the white was portrayed as a manager, the African-American as the secretary.

Discussion

Historically, studies of the black minority presence in advertising have relied on comparisons to the black percentage of the overall population to judge adequacy of minority portrayal. Applying that percentage, currently 12.2 percent, to the findings of this study might be misleading, however. It is true that 28.2 percent of the direct mail packages portrayed African-Americans. Yet the portrayal of a single black in a direct mail piece that in total portrayed 50 people in 20 separate pictures would yield black representation percentages of two percent (1/50) and five percent (1/20) respectfully, perhaps more accurately reflecting the true percentage representation. On the other hand, if the actual percentages of blacks portrayed in the direct mail pieces in this study (6.5 percent of the separate pictures and 5.8 percent of all the people portrayed) were compared to the general population standard, a picture of inadequate representation would emerge. But would this be accurate, and would it reflect the perc entages in the target business-to-business population?

To answer that question, it is necessary to return to the occupational statistics available from the U. S. Bureau of Labor Statistics (U. S. Department of Commerce 1996). Those statistics indicate that African-Americans represented 7.2 percent of people employed as executives, administrators and managers in 1995, 6.6 percent of those employed as purchasing agents, and 4.7 percent employed as engineers. These numbers compare quite well to those of this study. That is, the study showed that 5.8 percent of all people portrayed were black, and that 6.5 percent of all pictures portrayed blacks. Thus it might be argued that the direct mail pieces in this study represented fairly the actual presence of blacks occupying professional roles in buying centers.

Interestingly, the percentages of black portrayals found in this study are somewhat different from those regarding ads depicting blacks in trade publication advertising (Stevenson 1991). In trade publication advertising only 4.4 percent of the people portrayed in 1986 were black, the corresponding percentage from the present study is 5.8 percent. However, a decade has passed since the previous study, and the percentage of blacks in the managerial and professional ranks was lower then. If the study were replicated today, perhaps the African-American percentage would be higher. On the other hand, in comparison to a study (Stevenson 1993) of the black presence in ads in Business Week, Forbes and Fortune, where 7.8 percent of people portrayed were black, there is a smaller black presence (5.8 percent) today in industrial direct mail. Further, the percentage of ads that portrayed blacks in direct mail pictures was found in the present study to be 6.5 percent; this also trails the 13.8 percent of ads portraying bl acks in horizontal business publications (Stevenson 1993). Nevertheless, in comparison to some recent studies on consumer print media, the incidence of blacks in industrial advertising suggests that business direct mail advertisers are doing a better job of representing blacks proportionately than are consumer advertisers. To illustrate, Zinkhan, Qualls and Biswas (1990) found that only 4.37 percent of consumer ads used black models, and Wilson and Biswas (1995) found that only four percent of models in specialty catalogs were African-American. The greater presence of African-Americans in industrial advertising may give support to the notion that whites are more willing to accept integration in work rather than in social or residential settings (Humphrey and Schuman 1984). Perhaps this is why a recent study (Taylor, Lee and Stern 1995) of a cross-section of magazine ads found that nearly 35 percent of the settings where African-Americans were portrayed were business settings; only about five percent were soci al.

In looking at occupational portrayals, the findings of this study differ markedly from those regarding the black presence in industrial trade publications and horizontal business publications (Stevenson 1991; 1993). In those publications, blue-collar portrayals were much less frequent; the dominant African-American portrayal was managerial/professional, perhaps reflecting the readership of the publications examined. In the direct mail pieces analyzed for this study, however, the "hands on" nature of portrayals used for such products as uniforms, packing materials and warehouse equipment could explain the difference. This seems plausible given the fact that the most frequent occupational portrayal in direct mail was blue collar, regardless of the race of the model depicted. Nevertheless, this is a somewhat ironic example of the finding of Sterns, Unger and Luebkeman (1987), regarding occupational portrayals, that the skill level of blacks portrayed was no different than that of whites.

With regard to racial composition, this study found that when blacks and whites appeared together in an ad, the most frequent portrayal (88 percent) was as mixed peers (45/51 observations). In only 12 percent of such pictures were non-peer portrayals used. These percentages are consistent with those found (84 percent) in both horizontal business publications (Stevenson 1993) and trade journals (70 percent) (Stevenson 1991). They also mirror findings in consumer studies where mixed peer portrayals most frequently depicted blacks and whites as peers (e.g., Humphrey and Schuman 1984). Nevertheless, the fact that all nonpeer portrayals placed blacks in lower status roles is not reflective of workplace demographics that show rapid advances of African-Americans into managerial/professional positions. Lower status nonpeer black portrayals are also not consistent with portrayals found in trade publications (Stevenson 1991), where the lower status role in mixed racial nonpeer ads was frequently occupied by a white mo del.

The findings of this study can be interpreted in several ways. There are some that may feel that business-to-business direct mail advertisers are attempting to overcome criticism of racial bias in the advertising media (Ward 1992) by supporting the acculturation into middle management of the rapidly growing numbers of blacks in such positions. Such acculturation can be enhanced by appropriate media portrayals (Faber, O'Guinn and Meyer 1987). Advertisers may also be seeking to influence the attitudes of whites who may have little prior workplace experiences with blacks as managerial co-workers (Bristor, Lee and Hunt 1995). On the other hand, perhaps direct marketers are simply responding tactically by attempting to appeal to an emerging demographic segment. At any rate, there is little current evidence of the early 1970's concern about a "white backlash" to the increasing frequency of blacks in ads (Cagley and Cardozo 1970; Block 1972). Even in the face of conflicting research findings that whites may react p ositively (Whittler 1989) or negatively (Whittler 1991; Whittler and Dimeo 1991) to the presence of black actors in ads, Whittler (1991) concluded that portraying black actors in mixed audience advertisements makes "sound economic sense" (p. 59). Perhaps this attitude is shared by some business-to-business direct mail advertisers. Nevertheless, there is evidence that not all direct mail advertisers agree, because more than 70 percent of the direct mail pieces analyzed in this study did not portray any blacks.

Implications, Limitations and Future Research

There are implications for direct mail advertisers in these findings. First, it is apparent that many business-to-business users of direct mail employ African-American racial portrayals. Nevertheless, the majority of the pieces analyzed portrayed no blacks whatsoever. Given the changing face of the buying center, enlightened self-interest would suggest that business-to-business users of direct mail advertising consider the inclusion of minorities. Second, attention should be devoted to occupational portrayals that reflect the work place. Unfortunately, this places advertisers in a "Catch-22" situation. If a high proportion of product users is blue collar, then it is realistic to feature models thus portrayed. Doing so, however, exposes advertisers to criticism for using stereotypical or demeaning portrayals. On the other hand, while higher status portrayals would deflect criticism, they would not reflect the business world. Nevertheless, as blacks increasingly occupy buying center positions as deciders, infl uences, and purchasers, this should be reflected. Perhaps advertisers should err in the direction of more professional portrayals, especially during this period of rapid movement of African-Americans into such positions. Finally, consistent portrayal of blacks as subordinates in mixed peer settings is unwise and inconsistent with work place trends. Advertisers in the direct mail medium should give consideration to modifying non-peer portrayals so that the lower status role is not always occupied by an African-American.

These suggestions, however, should be considered in light of the limitations of the present study. For example, some categories of industrial products such as raw materials, fabricating parts, and capital equipment were infrequently represented in the direct mail pieces received for this study. Thus, these findings are less generalizable to those categories of products than they are to operating supplies, accessory equipment and business services. Additionally, the study provides only a "snapshot" of direct mail practices because the data collection period was one year, 1996; thus it is unable to assess direct mail advertisers performance longitudinally as have many other media studies. Last, because study data were gathered using content analysis, the study can only evaluate what was done, not why the images selected were chosen, or how the target audience responded to the images. Studies of audiences and advertisers, rather than the ads themselves, would be necessary to provide this information.

There are also implications for future research in these findings and limitations. This study fills a void in the literature regarding black portrayals in print media advertising; it could serve as a benchmark for future longitudinal studies into business-to-business direct mail. In addition to exploring percentages of black portrayals overtime, other factors could be considered. For example, will occupational portrayals of African-Americans continue to parallel those of whites, and will there be changes in mixed peer portrayals? Industry specific studies could be carried out that are representative of the full range of business products, as could studies regarding particular types of direct mail, catalogs or brochures, for example. Attitudinal studies could be carried out with those responsible for the creation of business-to-business direct mail and experimental studies could be employed with direct mail recipients. And because the United States is becoming increasingly multiracial and multicultural, these issues could be extended to the consideration of other minorities, Hispanics for example, or Asians. Such studies also could be made more industry specific.

Conclusions

This study examined the frequency of appearance and racial portrayals of blacks in business-to-business direct mail. Results showed that the percentage of ads portraying blacks was quite close to the presence of blacks employed in the business world. Racial portrayals were similar between blacks and whites in terms of occupational portrayals, and mixed peer racial compositions were frequently used. Nevertheless, it was found that the qualitative portrayals employed in business-to-business direct mail differed from those found in other industrial media.

Thus, it appears that business-to-business direct mail advertisers are responsive to the increasing presence of African-Americans in the buying center. This assertion is supported by other studies regarding the increased presence of blacks in advertisements in the business press. Nevertheless, there are some opportunities for improvement in racial portrayals, and these should be considered. Last, additional studies of the direct mail industry would be useful in furthering our understanding of the role that racial and cultural portrayals play in the direct mail segment of business-to-business advertising. Such studies should be undertaken to aid advertisers in developing effective advertising messages and to provide necessary benchmarks for meaningful dialog regarding racial concerns.

Thomas H. Stevenson (Ph.D., Case Western Reserve University) is the Cullen Professor of Marketing in the Belk College of Business Administration at the University of North Carolina at Charlotte.

Linda E. Swayne (Ph.D., The University of North Texas) is the Chair of the Department of Marketing and a Professor of Marketing in the Belk College of Business Administration at the University of North Carolina at Charlotte.

References

Block, Carl E. (1972), "White Backlash to Negro Ads: Fact or Fantasy?" Journalism Quarterly, 49 (Summer), 253-62.

Bonoma, Thomas V. (1982), 'Major Sales: Who Really Does the Buying," Harvard Business Review, (May-June), 111-119.

Bristor, Julia M., Renee Gravois Lee and Michelle R. Hunt (1996), "African-American Images in Television Advertising: Progress or Prejudice?" Journal of Public Policy and Marketing, 14 (Spring), 48-62.

Bush, Ronald F., Joseph F. Hair, Jr. and Paul J. Solomon (1979), "Consumers Level of Prejudice and Response to Black Models in Advertisements," Journal of Marketing Research, 16 (August), 341-345.

Cagley, James W., and Richard N. Cardozo (1970), "White Responses to Integrated Advertising," Journal of Advertising Research, 10 (April), 35-39.

Cantor, Joanne R., Herminia Alfonso and Dolf Zillman (1976), "The Persuasive Effectiveness of the Peer Appeal and a Communicator's First-Hand Experience," Communication Research, 3 (3) (July), 293-3 10.

Choudhury, Pravat K. and Lawrence S. Schmid (1974), "Black Models in Advertising to Blacks," Journal of Advertising Research, 19-22.

Cohen, Jacob (1960), "A Coefficient of Agreement for Nominal Scales," Educational and Psychological Measurement, 20, 37-46.

Colfax, David, and Susan F. Steinberg (1972), "The Perpetuation of Racial Stereotypes," Public Opinion Quarterly, 36, 8-18.

Cox, Keith (1970), "Changes in Stereotyping of Negroes and Whites in Magazine Advertisements," Public Opinion Quarterly, (Winter), 603-606.

Cravens, David W. (1987), Strategic Marketing, Homewood, IL: Richard D. Irwin, Inc.

Endicott, R. Craig (1995), "Direct Ads Spur 12% of Consumer Sales," Advertising Age, 66 (October 9), 34.

Faber, Ronald J., Thomas C. O'Guinn and Timothy P. Meyer (1987), 'Televised Portrayals of Hispanics: A Comparison of Ethnic Perceptions," International Journal of Intercultural Relations, 11, 155-169.

File, Karen M. and Russ A. Prince (1996), "A Psychographic Segmentation of Industrial Family Business," Industrial Marketing Management, 25, 223-234.

Gibson, D. Parke (1969), $30 Billion Negro, New York: MacMillan Co. Gitlitz, Jonah (1995), "Direct Marketing Boosts Business-to-Business Selling," Business Marketing, 80, 8 (August), B2.

Goerne, Carrie (1992), "Business Marketers Find Increasing Success As They Turn to Direct Mail" Marketing News, 26 (July), 3.

Hartley, Steven W., and Charles H. Patti (1988), "Evaluating Business-to-Business Advertising: A Comparison of Objectives and Results," Journal of Advertising Research, (April/May), 21-27.

Humphrey, Ronald, and Howard Schuman (1984), "The Portrayal of Blacks in Magazine Advertisements: 1950-1982," Public Opinion Quarterly, 44, 551-563.

Hutt, Michael D. and Thomas W. Speh (1998), Business Marketing Management, New York: The Dryden Press.

Joseph, W. Franklyn (1986), "Despite Growth in Market, Strong Resistance Continues," Advertising Age, (August 25), 5-1, S-2.

Kassarjian, Harold H. (1977), "Content Analysis in Consumer Research," Journal of Consumer Research, 4, 8-18.

__________ (1969), "The Negro and American Advertising, 1946-1965," Journal of Marketing Research, 29-39.

Kolbe, Richard H., and Melissa S. Burnette (1991), "ContentAnalysis Research: An Examination of Applications with Directives for Improving Research Reliability and Objectivity," Journal of Consumer Research, 18 (September), 243-50.

Labor Force Statistics Derived from the Current Population Survey: A Databook (1982), U. S. Department of Labor, Bureau of Labor Statistics, Washington, D.C., 1.

Mackie, Diane M., Arlene G. Asuncion and Leila T. Worth (1990), "Processing of Persuasive In Group Messages," Journal of Personality and Social Psychology, 58, 5, 812-822.

Newcomb, Theodore (1943), Personality and Social Psychology, New York: Dryden.

Ortizano, Giacomo L. (1989), "Visibility of Blacks and Whites in Magazine Photographs," Journalism Quarterly, 66 (Autumn), 718-722.

Perreault, William D. and Laurence E. Leigh (1989), "Reliability of Nominal Data Based on Qualitative Judgments," Journal of Marketing Research, 26 (May), 135-148.

Petroshius, Susan M. and Kenneth E. Crocker (1989), "An Empirical Analysis of Spokesperson Characteristics on Advertisement and Product Evaluations," Journal of the Academy of Marketing Science, 17 (Summer), 2 17-225.

Petty, Richard E., John T. Cacioppo and David Schumann (1983), 'Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146.

Pitts, Robert E., D. Joel Whalen, Robert O'Keefe and Vernon Murray (1989), "Black and White Response to Culturally Targeted Television Commercials: A Values-Based Approach," Psychology and Marketing, 6, 4 (Winter), 311-323.

Reiss, Craig (1983), "Black Media Association Cites Offensive Ads," Advertising Age, (September 19), 78.

Rust, Roland and Bruce Cooil (1994), "Reliability Measures for Qualitative Data: Theory and Implications," Journal of Marketing Research, 21 (February), 1-14.

"Sales Calls Expensive but Still Effective," Industrial Distribution, 81, (13) (October 1992), 11-12.

Shuey, Audrey, Nancy King and Barbara Griffin (1953), "Stereotyping of Negroes and Whites: An Analysis of Magazine Pictures," Public Opinion Quarterly, (Summer), 281-287.

Simmons Market Research Bureau (1996), Simmons Study of Media and Markets, (Fall).

Singer, Benjamin D. (1983), "The Case for Using 'Real People' in Advertising," Business Quarterly, (Winter), 32-37.

Sroge, Maxwell (1991), The United States Mail Order Industry, Homewood, IL: Business One Irwin.

Stearns, James M., Lynette S. Unger and Steven G. Luebkeman (1987), 'The Portrayal of Blacks in Magazine and Television Advertising," AMA Educators' Proceedings, Susan R. Douglas et al., eds., Chicago: American Marketing Association, 198-203.

Stevenson, Thomas H. (1991), "How Are Blacks Portrayed in Business Ads?" Industrial Marketing Management, 20, (3) (August), 193-200.

__________ (1993), 'The Portrayal of Blacks in Business Publication Advertising," Journal of Marketing Management, 2, 2,26-33.

Taylor, Charles R. and Barbara B. Stern (1997), "Asian-Americans: Television Advertising and the 'Model Minority' Stereotype," Journal of Advertising, 26 (Summer), 47-61.

__________ , Ju Yung Lee and Barbara B. Stern (1995), "Portrayals of African, Hispanic, and Asian-Americans in Magazine Advertising," American Behavioral Scientist, 38 (February), 608-62 1.

U. S. Department of Commerce, Bureau of the Census (1996), Statistical Abstract of the United States.

Ward, Adrienne (1992), "What Role Do Ads Play in Racial Tension?" Advertising Age, 63 (August 10), 1, 35.

Wheatley, John J., (1971), "The Use of Black Models in Advertising," Journal of Marketing Research, 8, (August), 390-392.

Whittler, Tommy E. (1991), "The Effects of Actors' Race in Commercial Advertising: Review and Extension," Journal of Advertising, 20 (1), 54-60.

__________ (1989), "Viewers' Processing of Source and Message Cues in Advertising Stimuli," Psychology and Marketing, 6 (4), 287-309.

__________ and Joan Dimeo (1991), "Viewers' Reactions to Racial Cues in Advertising Stimuli," Journal of Advertising Research, 20 (1), 54-60.

Wilson, Elizabeth J., and Abhijit Biswas (1995), "The Use of Black Models in Specialty Catalogs," Journal of Direct Marketing, 9, 4, 47-56.

Wynter, Leon E. (1993), "Minorities Play the Hero in More TV Ads as Clients Discover Multicultural Sells," Wall Street Journal, (November 24), B1, B6.

Zinkhan, George M., William J. Qualls and Abhijit Biswas (1990), "The Use of Blacks in Magazine and Television Advertising: 1946 to 1986," Journalism Quarterly, 67 (Autumn), 547-554.

                 Racial Depictions In Business-to-Business
                        Direct Mail Advertisements
                 Racial Portrayal
                      Black          White       Other
Category            Number (%)     Number (%)  Number (%)
Direct Mail          51 (28.2%)    181 (100%)  26 (14.4%)
Piece (n = 181)
Separate Picture    136 (6.5%)    2052 (98.1%) 56 (2.7%)
(n = 2092)
People in Ads       149 (5.8%)    2349 (91.7%) 63 (2.5%)
(n = 2561)
[x.sup.2] = 132.84; p [less than] .001.
              Occupational Portrayals in Business-to-Business
                        Direct Mail Advertisements
                           Racial Portrayal
                                Black              White
                                Number       (%)   Number  (%)
Occupational Category
   Blue Collar                    76         (51)    74    (50)
   Clerical                       12         (8)     13    (9)
   Entertainer                    0          (0)     0     (0)
   Managerial/Professional        39         (26)    42    (28)
Non-Occupational Category
   Family                         0          (0)     0     (0)
   Recreational                   22         (15)    20    (13)
Total                            149        (100%)  149   (100%)
                Racial Composition In Business-to-Business
                        Direct Mail Advertisements
Ad Composition             Number  (%)
Separate Picture/All Black   85   (63%)
Mixed Peer                   45   (33%)
Mixed Nonpeer                6     (4%)
Total                       136   (100%)

Appendix

Operational Definitions of Occupational Status and Role Portrayals

Occupational Status

Occupational Portrayals

Blue collar: A character shown in a business/employment setting performing a non-skilled task such as laborer, waiter, warehouse person, maintenance employee.

Managerial/professional: A character shown in a business/employment setting performing a skilled task or supervising others, such as an engineer, sales professional, physician or corporate manager.

Entertainer: A character shown in employment as a professional athlete, singer, movie or television personality, or other artist.

Clerical: A character shown in a business/employment setting performing clerical duties such as word processing, typing, inventorying, switchboard.

Non-occupational Portrayals

Family: A character shown in a non-business setting depicted as a parent, child or any other relationship among relatives.

Recreational: A character shown in non-business social or recreational settings other than a family setting, a sole jogger, for example.

Other: A character shown in any other non-employment setting other than those listed above.

Racial Composition

All models black: Only African-American characters portrayed in ad.

Mixed non-peer ad: African-American and non-African-American characters together in non-equal status portrayals, secretary and manager, for example.

Mixed peer ad: African-American and non-African-American characters together in equal status portrayal, for example, two secretaries interacting.

Separate picture: African-American and non-African-Americans in separate panels or separate scenes, not interacting.