The ethical framework of advertising and marketing research practitioners: a moral...

At the beginning of the last decade, Krugman and Ferrell (1981) noted within the pages of this journal that advertising agency personnel had questions about the ethical behavior of their peers. A following study by Hunt and Chonko (1987) found that advertising agency executives were particularly

worried about issues of equity and honesty. An ad agency's reputation with clients exists only to the extent that clients trust the integrity of the agency. Trust is associated with honesty (Zaltman and Moorman 1988); therefore, if clients are suspicious of the honesty of the ad agency, they are unlikely to trust the agency, thus harming the agency's ability to attract future business. Thus, doubts about the trustworthiness of advertising researchers, as expressed in the Hunt and Chonko (1987) study, merit further attention.

Marketing/advertising research is one area that continually raises ethical concerns (e.g., Bogart 1962; Tybout and Zaltman 1974; Frey and Kinnear 1979; Hunt, Chonko and Wilcox 1984; Zinkhan, Bisesi and Saxton 1989). While the work on ethics conducted thus far has been concerned with a specific moral behavior, i.e., honesty, the issue of business ethics can be viewed from lenses other than those of specific unethical behaviors (e.g., dishonesty, theft). One such lens is the consideration of the differing principles marketing/advertising researchers use in conducting and presenting their work. For example, marketing/advertising researchers may be making decisions using the principle of the maximization of personal welfare or they may be making decisions using the principle of the maximization of social welfare. Regardless of the perspective taken, the issue of ethics is important to all those involved in the profession.

The practice of marketing/advertising research can be portrayed as a chain of requestors and providers. Members of the chain include final clients, ad agencies, research suppliers, focus group moderators and data collection services. The concerns about ethics and research are really twofold. The first concern involves the degree to which all members of the research chain are perceived to be unethical by the public, and the second involves the degree to which members of the chain perceive other members of the chain as being less ethical. Researchers may find the reputation of the profession in danger if the public views their work with distrust. The same reputational damage may occur if members of the chain perceive each other to be untrustworthy. Indeed, perceptions of what is ethical may differ among members of the chain (Crawford 1970; Coney and Murphy 1976; Hunt, Chonko and Wilcox 1984; Ferrell and Skinner 1988; Akaah and Riordan 1989). For example, a research supplier might feel compelled to "cover up" nonresponse and sampling error hoping to please its client. Or, a data collection agency may significantly alter the sampling design in order to obtain responses. The client in these situations is apt to find such tactics unethical.

While one source of differing perceptions of what is ethical is the position of the actor vis a vis the act (Bok 1979), another possible source of different perceptions is differing philosophical approaches to the topic of ethics (Murphy and Laczniak 1981; French and Ebner 1986; Ferrell and Skinner 1988). Yet another source of different perceptions may be the differing cability of individuals to reason about moral problems. Hunt, Chonko and Wilcox (1984), in fact, have suggested that the next step in investigating the topic might be to adopt an approach such as that initiated by Kohlberg (1976) in the area of moral development. Kohlberg's (1969) work on moral development has been concerned with the stages individuals progress through in their development of the cognitive capability to reason about moral problems.

Moral Reasoning

Piaget (1932) hypothesized that the skills involved in moral (ethical) decision making are developed over time, dependent, in part, on interaction and collaboration with others. Kohlberg (1969, 1976) extended Piaget's work through a longitudinal study that tracked individuals' moral reasoning from childhood into adulthood. He identified six stages of moral development.

In the first two stages, labeled the preconventional level, the focus is on personal rather than social well being. In the second two stages, labeled the conventional level, social convention is the benchmark for ethical decisions. In the last two stages, labeled the postconventional level, the perspective for moral judgment is individualistic but with the goal of enhancing societal welfare. Each of these levels represents different cognitive reasoning abilities. The levels are increasingly abstract, suggesting increasingly sophisticated abilities to reason about moral issues. The stages as described by Kohlberg (1969) are sequential meaning that people must progress through one stage before they will reason at the next highest stage. Thus, Stage 1 is a lower order stage of cognitive reasoning ability than Stage 2 and Stage 2 is a lower order stage than Stage 3. It should be noted that not all decisions will be made using the highest stage of reasoning attained by the individual. In making any given judgement, an individual may well utilize reasoning from several of the stages. However, research suggests that individuals tend to rely on one of the stages as their primary reasoning mode (Rest 1986). Those operating at higher stages of moral development are more likely to act on their decisions than those reasoning at lower stages of moral development (Kohlberg, Levine and Hewer 1983). Consequently, if members of the research chain are operating at different levels of moral reasoning capabilities, they are likely to take different actions on the basis of their decisions. Similarly, if those involved in the marketing/advertising research process exhibit a tendency to make decisions using a lower stage of moral reasoning than that used by society at large, there is a decreased likelihood that they will engage in behaviors which are consistent with their ethical decisions. This may result in a public perception that researchers are prone to engage in unethical practices.

While much work has been done investigating the process of moral development, little has been done investigating the effect of the use of different stages of reasoning abilities within differing populations. The concerns expressed regarding the ethical practices of marketing/advertising researchers suggest that it would be worthwhile knowing if research practitioners routinely used more or less sophisticated methods of moral reasoning. Thus, exploratory research investigating this area is called for. Specifically, the questions posed in this research study are:

1. Do researchers make decisions using reasoning styles characteristic of a lower level of moral development than society? While there is some suggestion that the incidence of unethical decisions is not negligable (Akaah and Riordan 1989), there is no a priori reason to assume that marketing researchers in general, and advertising agency researchers in particular, are less morally developed than society as a whole. This question is of interest in light of the continuing criticism directed toward researchers.

2. Do different members of the research chain exhibit different levels of moral development? Previous studies have found differences in ethical opinions within the research industry. However, no studies have looked specifically at moral development.

Research Method

Members of the research community and leaders of several research organizations (e.g., Marketing Research Association) were consulted to determine appropriate groups to study in the research chain. Based on these discussions, five types of research employment groups were sampled: (1) advertising agency researchers, chosen from agencies with the largest billings for the previous year; (2) full service suppliers, randomly selected from the American Marketing Association (AMA) and Marketing Research Association (MRA) membership lists; (3) field services researchers, chosen from the MRA list; (4) focus group moderators, randomly selected from the rolls of the Qualitative Research Council of America (QRCA); and (5) clients, random~y selected using AMA and MRA lists.

The survey instrument included Rest's Defining Issues Test, as well as questions designed to determine, among other things, the subjects' job title, type of business, and age. The survey instrument and all supporting documents were pretested among members of the designated populations, producing only slight changes in the wording of the final cover letter.

The Defining Issues Test (DIT) has been used in over 500 studies in over 20 countries and has been extensively validated (Rest 1979a, 1986). The DIT uses scenarios that describe moral dilemmas. The subject is asked to indicate the importance of twelve factors in arriving at a solution to each dilemma. The subject is then asked to rank order the top four factors (a copy of the instrument is available from the authors). Output from the DIT includes stage scores, a D score, and various consistency check measures. Stage scores identify the three stages of moral reasoning used by a respondent when presented with the instruments' moral dilemmas. Thus, the stage scores identify not only the primary stage of moral reasoning used to make a decision, but also the second most used stage as well as the third most used stage. The D score is an index indicating the subjects' general level of moral reasoning. Since subjects use multiple stages of moral reasoning when making a decision it is difficult to assign an individual to an overall stage. Should respondents be categorized according to the highest stage of moral reasoning they exhibited or the one they used most frequently? The D score, a calibrated measure of the salience of each TABULAR DATA OMITTED stage of moral reasoning in an individual's judgment process, is a response to that question (See Rest 1979a, 1979b for a complete discussion of how D scores are computed). In terms of interpretation, higher D scores represent higher degrees of moral judgment, although they are not associated directly with Kohlberg's stages of moral development.

The DIT is an appropriate instrument for the current study because it uncovers the underlying process of how a researcher arrives at a moral decision. That underlying reasoning process would be the same whether one is considering an ethical problem in one's professional or personal life. Thus, there is reason to believe that the results of the survey instrument can be interpreted to profile the responding researcher's fundamental approach to decisions with moral implications.

A total of 100 individuals in each of the five research groups received a cover letter describing the purpose of the study and promising anonymity, the instrument, and a postage paid return envelope. Since the mailing went to specific people/companies, we can be sure that each group was mutually exclusive (i,e., no individual or company was included in more than one group). All subjects, except for moderators, were also asked to identify someone in their firm who dealt more directly with data collection or analysis than with managerial problems and have that person fill out a second copy of the DIT. After four weeks, follow-up letters were sent to 237 subjects in those groups with relatively low response rates. For example, since the focus group moderators had a relatively low response rate, we sent follow-up letters to those sample members. Note that since all replies were anonymous, it was impossible to identify actual non-respondents.

Of the 500 researchers selected for the sample, 28 could not be located with the mailing. A total of 118 completed questionnaires were received from the 472 researchers who were contacted--a response rate of 25%. This level of response rate is fairly common for ethics studies of this nature (e.g., Greyser and Reece 1971; Krugman and Ferrell 1981; Chonko and Hunt 1985; Ferrell and Skinner 1988). An additional 67 questionnaires were received from the pass-along second copies which were added to the other responses to form a total sample of 185.

Characteristics of the sample suggest a rather broad representation of the population, Age of subjects ranged from 21 to 67 years old, with number of years of experience ranging from 4 months to 40 years. The number of full time people employed in research activities in the subjects' firms ranged from one to 700 people. As Table 1 indicates, each group consisted of a wide spectrum in terms of demographic characteristics.

Results

Rest (1979b) developed national norms for the D scores (based on empirical evidence) and reported them in equal thirds. The lower third has D scores of 0-16, the middle third has D scores of 17-25, and the high third has D scores of 26 and above. The national median score is approximately 20. Recall that the higher the D score the higher the level of moral development. The breakdown of D scores for the sampled researchers results in 1% of the sample falling into the 0-16 category, 58% of the sample falling into the 17-25 category and 41% of the sample falling into the |is greater than~ 26 category, instead of falling into the predicted equal thirds. Thus, there is a disparity between the national norm and moral development scores of the sampled researchers (Chi square = 73.04, df = 2, p |is less than~ 0.001). This sample of researchers as a group scored higher than the national sample. Therefore, the answer to Research Question 1 is "no": marketing researchers do not reason about moral problems at a lower level of moral development than society. Although the sample of researchers does not permit accurate inferences about the ethics or moral standards of all researchers, it does provide an indication that many of those researchers willing to respond show a higher degree of moral reasoning than a national sample of the general population.

One possible explanation for this finding is that it is an artifact of the sampling method. Questions of ethical attitudes are a sensitive issue and many members of organizations may be reluctant to participate in such surveys. It may be that only those individual who consider themselves to be particularly ethical would be willing to participate. If this is the case, however, we would still expect to have received a fairly representative sample since most people consider themselves to be more ethical than others (Baumhart 1961; Posner and Schmidt 1987; Vitell and Festervand 1987; DeConninck 1990). Thus, if most of the recipients of the questionnaire consider themselves to be more ethical then their colleagues, then those who responded are probably fairly representative of the profession.

Another explanation for this finding is that, in general, those involved in research tend to be fairly highly educated. The average years of formal education of those in the sample was very high, indicating that, on average, researchers had at least a Masters level education. Research shows a high degree of correlation between years of formal education and level of moral reasoning (Rest 1986). In addition to being empirically verified, this makes conceptual sense. Moral reasoning is a special case of cognitive reasoning (Piaget 1932), and cognitive reasoning abilities would reasonably be expected to increase as a function of amount of education received. Consequently, it is not surprising that those with more education will also tend to use higher levels of moral reasoning. While the sample may be skewed toward the high end, not only of moral reasoning but of education, the occupation of research, nonetheless, demands higher levels of education. Certainly there would be tasks that could be performed by an employee with a high school diploma alone. However, many of the tasks associated with research, such as design of the sample, design of the data collection instruments, choosing methods of data analysis, and interpreting the results of the data analysis, require high levels of cognitive skills and, usually, high levels of training/education. Thus, we would expect to find that those involved in marketing/advertising research would have, in general, higher levels of education and, consequently, higher levels of moral reasoning skills.

While the subjects in this sample exhibited high levels of moral reasoning, that does not mean that they never rely on preconventional, self-centered methods to resolve moral dilemmas. As Table 2 indicates, approximately 12% of the subjects report that they occasionally use a stage 2 reasoning framework to resolve ethical issues. Four of the five groups that constitute the research chain are represented in this 12%.

Having examined differences between researchers and the national norm, focus can be turned to differences between the advertising agency researchers and the other groups that constitute this research chain. Table 3 displays average D scores for each group. As can be seen, there is very little difference between the groups, as is reflected in the ANOVA test results. Therefore, the answer to research question 2 is no; different members of the marketing/advertising research chain do not exhibit different levels of moral reasoning skills.

The research of Hunt, Chonko, and Wilcox (1984), as well as that of Ferrell and Skinner (1988), found that the importance weightings of a series of possible causal factors which might explain ethical practices TABULAR DATA OMITTED TABULAR DATA OMITTED differed within levels of the research chain. Their research, however, did not directly test for differences in ethical behavior between levels of the chain. While such differences in behavior may exist, the D scores indicate that the reasoning underlying those behaviors may not be as dissimilar as one might imagine.

Discussion

The results of this study suggest that researchers do not make decisions using lower levels of moral reasoning skills than those used by people in general. Also, different members of the research chain do not vary significantly in their moral reasoning skills. This does not mean, however, that those involved in the research process routinely make decisions of which the rest of society would approve. This study examined the ability of researchers to reason about moral problems, but we do not suggest that the reasoning process will necessarily lead to ethical decisions. There has been some research which suggests that individuals can be influenced in their assessment of right and wrong by superiors (Hegerty and Sims 1978, 1979). Those reasoning at Stage Four, for example, may tend to obey the rules of the organization of which they are a member and wish to promote the welfare of that organization as a way of promoting social order. Those who firmly believe in the social utility of a free-market system, while reasoning at stage five, may conclude that the welfare of society is dependent upon the welfare of the firm and that, consequently, those actions which benefit the firm necessarily benefit society. Thus, the fact that researchers are reasoning at normal and higher than normal levels does not allow one to conclude that their decisions will necessarily be considered ethical by the general public or even by other members of the research chain.

In the aggregate, the data show that researchers are utilizing a level of moral reasoning that emphasizes social utility rather than personal reward. As indicated in a study by Rotzoll and Christians (1980), advertising agency employees commented that their first concern when faced with an ethical problem was the potential effect of their behavior on the client and on the mores of the agency. This is an example of the same social or conventional morality that was demonstrated by the sampled researchers. According to Kohlberg (1969), making decisions in the best interest of society is a higher order, more abstract, idea than making decisions for personal benefit. Because they are reasoning at more abstract levels, marketing/advertising researchers may accept the notion that firm reputation is a long-term consequence of honest behavior which improves firm reputation via increased trust and cooperation (Bok 1979; Zaltman and Moorman 1988). This suggests, contrary to Bogart's (1962) misgivings, that a code of ethical standards for research might help in this endeavor by stating the norms of the profession and making clear that characteristics such as honesty are not over-ridden by the need to please. A code of ethical standards could, then, serve a similar function as the medical code of ethics, circumventing the tendency to see the employer as being the rule-making authority to which the decision maker must accede.

It is our hope that this study has shed new light on a very important topic in the research industry. Perhaps these exploratory results will foster additional research and cause researchers to seek common ground to avoid and resolve ethical problems.

References

Akaah, Ishmael P. and Edward A. Riordan (1989), "Judgments of Marketing Professionals About Ethical Issues in Marketing Research: A Replication and Extension," Journal of Marketing Research, 26 (February), 112-120.

Bogart, Leo (1962), "The Researcher's Dilemma," Journal of Marketing, 26 (January), 6-11.

Bok, Sissela (1979), Lying: Moral Choice in Public and Private Life, New York: Vintage Books.

Baumhart, Raymond (1961), "How Ethical Are Businessmen?" Harvard Business Review, 39, (July-August), 6-19, 156-166.

Chonko, Lawrence B. and Shelby D. Hunt (1985), "Ethics and Marketing Management: An Empirical Examination," Journal of Business Research, 13 (August), 339-359.

Colby, A., Lawrence Kohlberg, J. Biggs and M. Liberman (1983), "A Longitudinal Study of Moral Judgement," SCRD Monograph #48 (1-2, Serial # 200).

Coney, Kenneth A. and John H. Murphy (1976), "Attitudes of Marketers Toward Ethical and Professional Marketing Research Practices," in Proceedings: Southern Marketing Association, Henry W. Nash and Donald P. Robin, eds., 172-174.

Crawford, Merle C. (1970), "Attitudes of Marketing Executives Toward Ethics in Marketing Research," Journal of Marketing, 34 (April), 46-52.

DeConinck, James B. and David J. Good (1989), "Perceptual Differences of Sales Practitioners and Students Concerning Ethical Behavior," Journal of Business Ethics, 8 (September), 667-676.

Ferrell, O.C. and Steven J. Skinner (1988), "Ethical Behavior and Bureaucratic Structure in Marketing Research Organizations," Journal of Marketing Research, 25 (February), 103-109.

French, Warren A. and Myra Ebner (1986), "A Practical Look at Research Ethics," Journal of Data Collection, 26 (Fall), 49-53.

Frey, Cynthia J. and Thomas C. Kinnear (1979), "Legal Constraints and Marketing Research: Review and Call to Action," Journal of Marketing Research, 16 (August), 295-302.

Greyser, Steven A. and Bonnie Reece (1971), "Business Men Look Hard at Advertising," Harvard Business Review, 49 (May-June), 18-36.

Hegerty, W. Harvey and Henry P. Sims, Jr. (1978), "Some Determinants of Unethical Decision Behavior: An Experiment," Journal of Applied Psychology, 63 (August), 451-457.

----- and ----- (1979), "Organizational Philosophy, Policies, and Objectives Related to Unethical Decision Behavior: A Laboratory Experiment," Journal of Applied Psychology, 64 (June), 331-338.

Hunt, Shelby D. and Lawrence B. Chonko (1987), "Ethical Problems of Advertising Agency Executives," Journal of Advertising, 16 (4), 16-24.

-----, Lawrence B. Chonko, and James B. Wilcox (1984), "Ethical Problems of Marketing Researchers," Journal of Marketing Research, 21 (August), 309-324.

Kohlberg, Lawrence (1969), "Stages and Sequence: the Cognitive-Developmental Approach to Socialization," in Handbook of Socialization Theory and Research, David A. Goslin, ed., Chicago: Rand McNally, 347-480.

-----(1976), "Moral Stages and Moralization: The Cognitive-Developmental Approach," in Moral Development and Behavior, Thomas Lickona, ed., New York: Holt, Reinhart and Winston.

-----, Charles Levine and Alexandra Hewer (1983), Moral Stages: A Current Formulation and a Response to Critics, Basel, Switzerland: Karger.

Krugman, Dean M. and O. C. Ferrell (1981), "The Organizational Ethics of Advertising: Corporate and Agency Views," Journal of Advertising, 10 (1), 21-30.

Murphy, Patrick E. and Gene R. Laczniak (1981), "Marketing Ethics: A Review with Implications for Managers, Educators and Researchers," in Review of Marketing, 1981, Ben M. Enis and Kenneth J. Roering, eds., Chicago: American Marketing Association, 251-266.

Piaget, Jean (1932), The Moral Judgement of the Child, Marjorie Gabain, trans., New York: The Free Press.

Posner, Barry Z. and Warren H. Schmidt (1987), "Ethics in American Companies: A Managerial Perspective," Journal of Business Ethics, 6 (July), 383-391.

Rest, James R. (1979a), Development in Judging Moral Issues, Minneapolis: University of Minnesota Press.

-----(1979b), Revised Manual for the Defining Issues Test, Minneapolis: Minnesota Moral Research Projects.

-----(1986), Moral Development: Advances in Research and Theory, New York: Praeger.

Rotzoll, Kim B. and Clifford G. Christians (1980), "Advertising Agency Practitioners' Perceptions of Ethical Decisions," Journalism Quarterly, 58 (Autumn), 425-443.

Tybout, Alice M. and Gerald Zaltman (1974), "Ethics in Marketing Research," Journal of Marketing Research, 11 (November), 357-368.

Vitell, Scott J. and Troy A. Festervand (1987), "Business Ethics: Conflicts, Practices and Beliefs of Industrial Executives," Journal of Business Ethics, 6 (February), 111-122.

Zaltman, Gerald and Christine Moorman (1988), "The Importance of Personal Trust in the Use of Research," Journal of Advertising Research, 28 (May), 16-23.

Zinkhan, George M., Michael Bisesi, and Mary Jane Saxton (1989), "MBA's Changing Attitudes Toward Marketing Dilemmas: 1981-1987," Journal of Business Ethics, 8 (December), 963-974.

Stephen B. Castleberry is a Professor of Sales and Marketing in the School of Business and Economics at the University of Minnesota at Duluth. Professor Castleberry received his Ph.D. from the University of Alabama.

Warren French is a Professor of Marketing in the Terry College of Business at the University of Georgia, Athens, GA. Professor French received his Ph.D. from the Pennsylvania State University.

Barbara A. Carlin is an Assistant Professor of Management in the Terry College of Business at the University of Georgia, Athens, GA. Professor Carlin received her Ph.D. from the University of Texas at Austin.

The authors would like to thank the Marketing Research Association for funding this project and the reviewers for their helpful and insightful comments on previous versions of this manuscript.