Marketing Research Ethics
Slide 1
Although many people believe that marketing ethics is an oxymoron, my goal for this lecture is to convince you otherwise and to provide you a basic framework and perspective for making more ethical decisions regarding marketing research.
Slide 2
This slide shows a basic model for ethics and marketing research that I find especially useful. I am a Kantian at heart, so I tend to believe in rights and duties. Here we can see that the rights and obligations involved in understanding marketing research concern subjects, researchers, and research clients.
Slide 3
Researchers have clear obligations to respondents; in particular, researchers should avoid deceiving respondents needlessly. That said, it may be necessary to temporarily disguise the true purpose of a research study to avoid biasing respondents. If respondents are somewhat aware of the hypothesis we are testing, then they may alter their otherwise normal behaviors, so it may be necessary to temporarily deceive respondents but unnecessary to deceive them needlessly.
Researchers should avoid invading respondents’ privacy. That doesn’t mean we don’t occasionally ask respondents for some personal background information. It may be necessary, as part of our study, to understand something about respondents’ profiles, so we may ask them about their education, their occupation, or their income. Nonetheless, marketing researchers should avoid needlessly prying into the private lives of respondents.
Generally, researchers should be concerned abouteach respondent’s well being. Researchers should guarantee that each respondent who decides to participate in a study is properly informed about that study, can make an informed decision about participation, and has granted a proper, informed, consent to participate. This is of particular concern if we’re conducting research on children. In that case, it is necessary to secure parents’ informed consent.
Slide 4
Unfortunately, there are many ways to needlessly deceive respondents. Here are just a few of them. A researcher can promise respondents anonymity but fail to do so. For example, researchers can use ultraviolet ink or ID numbers that uniquely identify respondents, yet claim in a cover letter that eachrespondent’s identity will be anonymous. Note that there is a difference between anonymity and confidentiality. I can promise that I won’t reveal that you answered a set of questions in a certain way or responded to an experiment in a certain way. If I promise not to reveal or to link you with your responses, then that’s a guarantee of confidentiality, which is the thing attorneys and physicians grant their clients or patients. They know who you are; they just won’t reveal anything that occurred as a result of your professional directions, with them or with anyone else. It is inappropriate to promise anonymity but not grant it.
Another deceptive practice is falsifying sponsor identification; that is, misrepresenting the sponsor of the research. Often, researchers do this to boost response rates. People are more willing to respond, for example, to surveys conducted by universities then by commercial organizations, so by asking your field workers to pretend to be students working ontheirprofessor’s research might encourage respondents whootherwise would have opted out of your study. Nonetheless, that’s inappropriate because you should never lie about true sponsor of the research.
These next two sets of practices are related and although they sound obscene, they are not. Selling under the guise of research and fundraising under the guise of research are referred to as sugging and frugging. Unfortunately, these practices, especially for telephone interviewing, have negatively affected researchers’ ability to collect survey data. The widespread usage of sugging and frugging have primed potential respondents to assume that any phone solicitationasking them to participate in a studywill ultimately entail a request either to buy something or to contribute to some charitable organization. To avoid those sorts of telephone interactions, people just automatically reject the caller; as a result, response rates for telephone interviews have dropped precipitously.
Slide 5
The telephone is not the only way in which marketers use sugging and frugging. In the next two slides you will see examples of sugging via email. In neither case is there truly a legitimate effort to collect marketing research data.
Slide 6 (No Audio)
Slide 7
Misrepresenting the research procedures is another way in which researchers deceive potential study participants. Suppose it will take respondents an hour to complete your questionnaire. Rather than telling themup front that it will take an hour of their time—and greatly reducing the likelihood they will agree to participate,you suggest that the questionnaire only takes a half hour to complete. Misrepresenting the procedure in this way to boost your response rate is unethical.
Misrepresenting the need for possible phone re-contacts also is problematic. When I agree to participate in a study, I assume that I am fully aware of what participation entails before I begin. If there are likely or mandatory follow ups that entail additional time and effort, then I should be alerted to that fact before I agree to begin the interview process.
Another way to misrepresent a research study is to misrepresent its purpose. As I mentioned earlier, it may be necessary temporarily to disguise the true purpose of a study in order to collect unbiased responses. Any academic study I would conduct would be screened by a human subjects committee. To ensure that no harm befalls study participants, such a committee would require that they be debriefed,especially if I fib about the purpose of my study. Regardless of the social desirability of my research goals, I should avoid misrepresenting the purpose of my study if possible. For example, if I tell people the purpose of my study is to figure out ways to reduce auto emissions, when in fact the real goal of my study is to boost mass transit revenues,then that’s a problem despite my socially desirable goal. After all, it’simpossible for respondents to provide informed consent to participate in a study if they are unaware of its true purpose.
Another way in which a study process could be misrepresented is to hide how the results will be used. If the results will be used for a commercial purpose, but I encourage respondents to believe they’re participating in an academic study, then that’s a problem.
Finally, failing to deliver promised compensation—for example, promising people that if they participate in my study and provide their name and address, then I will mail them a summary of the research results, but I fail to do so—is a problem, not only because I lied but because my failure will discourage my respondents from participating in subsequent studies. Obviously, they will doubt that the promise of compensation will ultimately result in receiving it.
Slide 8
Researchers must be sensitive to respondent privacy issues. There are many ethical issues that deal with observation without informed consent. If I use hidden microphones or hidden video cameras, then study participants will be unaware that their behaviors are being recorded. As a result, it’s impossible for them to give fully informed consent because they’re unaware of the full procedures and the implications of having agreed to participate in the study.Voice pitch analysis is another method of permanently recording behaviors; in this case, verbal behaviors that may be analyzed subsequently without consent. Garbology studies, in which aresearcher combs through other people’s trash to assess the things they have consumed, is another example of observation without informed consent.
I don’t mean to implythat researchersshould never conduct studies using hidden cameras and microphones, or never conduct voice pitch analyses or garbology studies; rather,that they must be careful when doing so and one way for academic researchers to ensure greater care is by having their studies reviewed by human subjectscommitteesthat will, in addition to the researcher, carefully consider privacy issues.
Another privacy issue relates to qualitative research techniques. You may recall from the lecture onqualitative techniques that a researcher may ask people to projecttheir opinions onto various ambiguous stimuli. In essence, the research is getting people to reveal their true feelingsthrough some guise. The researcher knows they are revealing their true feelings, but they may be unaware that they are doing so. This entails a privacy issue, so researchers should be careful when using qualitative research techniques.
Researchers certainly should avoid asking overly personal questions. Although demographic and socioeconomic questions are often a requirement of any study, overly personal questions that don’t truly pertain to the research needs are problematic.
Finally, and this is a far greater concern in Europe than in the U.S.—in fact, there are laws precluding this in Europe—is the issue of merging data from several sources. I might be unconcerned if my health care provider has certain information about me and my mortgage lender has different information about me and the government has still different information about me; none of the information in the separate databases maintained by those separate organizations concern me. However, if someone were to merge all that data into one aggregate and more powerful profile of me, then I’d be concerned. Hence, there is a privacy issue associated with merging data from multiple sources into a single aggregated profile.
Slide 9
Inadequate concern for respondents can take many forms, such as contacting respondents at an inconvenient time; for example, calling people at dinner time to conduct a forty-five minute telephone interview is inappropriate. Incomplete or insensitive interviewers are another way in which I, as a researcher, may show inadequate concern. I must be careful to properly train my interviewers in terms of their demeanor, the quality of their probing questions,and their general approach tointeracting with respondents. If I’ve trained my interviewers inadequately, then I’ve shown inadequate concern for respondents. Certainly, as I’ve mentioned previously, failure to debrief after any session or use of any guise shows inadequate concern.
I shouldn’t ask questions that might needlessly depress respondents subsequently. That’s not to say that I shouldn’t ask questions aboutpeople’s own funeral arrangements if I’m conducting marketing research for a funeral home. Clearly, asking questions about one’s demise is depressing, but that’s not needlessly depressing respondents.
Researchers should avoid going to the well too frequently; in this case, the well of research and polling. If researchersabuse the public, if they contact people repeatedly, to the point where they’re no longer gracious about lending us their time and energies, then that will just reduce response rates, reduce the quality of responses, and make the expense of marketing research prohibitive. Researchers shouldn’t query respondents too frequently and reduce their willingness to answer our questions.
Finally, non-disclosure of research procedures, as mentioned before—the length of the survey, follow-ups—reflects inadequate concern for respondents.
Slide 10
To this point I’ve focused on researchers obligations to respondents. Now I’ll briefly discuss researcher’s obligations to research clients. Basically, there are two types of obligations. One is to avoid abusing research designs or methods or results. For example, as a researcher I might develop a new procedure for collecting some sort of research data, and in my agreement with the client, I might agree that any methods I develop are proprietary. In other words, only the client has subsequent right to use those procedures. Well, if I subsequentlyconduct another study for another client using the same procedures, then I’ve clearly violated my obligations to the client. Even worse, if I’ve conducted a study in which collected information that would be useful, not only to the client but to competitors, and then approach competitors about a similar study that reveals the same information, that effort would represent a clear conflict of interest.
There are other ways in which the researchers can abuse clients: acting unprofessionally, not delivering what was promised, missing deadlines, and low-ball pricing. Such behaviors are clear abuses of clients and should be avoided by researchers.
Slide 11
This slide shows specific abuses of research design, methods, and results. One abuse is conducting unnecessary research—which is analogous to a physician performing unnecessary surgery or a dentist performing unnecessary dental work—merely to recover fees. If a researcher is to conduct research, the research ought to be necessary in the sense that it provides information to help marketing managers choose among viable alternative courses of action. The research should reduce, in a meaningful fashion, the uncertainty of the marketing manager. If the researcher knows in advance the study cannot succeed in this way, then that’s what I mean by unnecessary research and that’s a problem.
Researching the wrong or irrelevant problem is analogous to the aforementionedabuse. It’s still unnecessary research, but in this case it’s both unnecessary and wrong. Such research givesthe erroneous impression that it has provided relevant information. As a result, the client may delay a decision needlessly, and that delay may be costly. Researching the wrong or irrelevant problem is an abuse.
Continuing a study after spotting an error in the process is an abuse. For example, suppose I mail questionnaires to a thousand people—an expensive proposition—but then, once the questionnaires begin to return, I recognize that there is a major flaw in the questionnaire. However, I fail to mention this flaw to the client. Ijust continue to accept the questionnaires, enter respondents’ answers into my data analysis software, analyze those answers, prepare a report and then to present it, all without indicating to anyone that I spotted an error early in the process. That’s a clear problem. Researchers are not perfect and their procedures are not perfect, but researchers who spot a problem with their study are obligated to correct that problem as soon as it’s spotted.
Slide 12
There are several additional abuses related to the use of unwarranted shortcuts to secure a contract or to save expenses. For example, altering the sample design to obtain enough respondents, even if that sample design will not create a representative or projectable sample of the overall population,is an issue. Not meeting accuracy requirements, by reducing the sample size to thepoint that random sampling error becomes an unacceptable threat or there are insufficient numbers of respondents in each category to make meaningful cross-category comparisons, is an issue. Improper verification of subjects is an issue; it is nonsensical to query people who are unqualified to participate in a study, so researchers must ensure that all subjects are qualified. Finally, inadequate pretestingof questionnaires, which is likely to lower their reliability and validity, is an issue. The pretesting stage takes time, energy, and a bit of money,but it’s necessary to ensure that the subsequent study is done correctly.
Slide 13
Other researcher abuses of research design, methods, and results, are as follows.Researchers could misrepresent the limitations of their research efforts; for example, they could hide errors caused by non-response or sampling. Researchersare obliged to sensitize their clientsto the artificiality of an experimental design if a laboratory experiment is involved. In my lectures about experimentation, I talked about the riskiness of extrapolating from a laboratory experiment—with its low external validity—to people’s behaviors in real time.Similarly, overstating the validity or reliability of any study is an issue. The manager ought to know to what degree the results of any research study are reliable and valid, and thus know how much to weigh the results of that study in decision making.
Another abuse is the application of an inappropriate analytical technique. Unfortunately, learningaboutsophisticated marketing research methods may be time consuming, and the software, computers, and other materials required to use these methodsmay be expensive. As a result, researchers may encounterfinancial pressures to apply inappropriate techniques merely toamortize their costs, or be encouraged to think if they have this hammer, then every research problem is a nail. Another way that inappropriate techniques may be applied is by using more sophisticated, analytical methods than required to answer the research questions. In other words, impressing the client is an inappropriate reason to use asophisticated analytical technique. Similarly, using overly technical language in reports is a problem. Dazzling clients with B.S. is an abuse of clients. Researchers are obliged to communicate clearly and to provide clients with honest assessments of the information they collected and how it can be used to make better decisions.
Related to this issue is insufficient expertise to conduct research. As a researcher, perhapsI have alongtime client who suddenly has an interest in a type of research study about which I am not technically versed. Rather than indicatethat I don’t have sufficient expertise to conduct this type of study, and perhaps suggest other researchers with such expertise, I might try to muck along and attempt to conduct the study regardless. If I lack the expertise to conduct a study, then I shouldn’t attempt it, anymore than you would want to hire a surgeon to perform open heart surgery if that surgeon was a podiatrist.
Slide 14
Finally, here are six more ways in which researchers may abuse clients. I hope by going through all of these ways I will sensitize you to the things you need to watch for, such as overbilling or double-billing for projects. The problems with overcharging should be self-evident. Low-ball pricing, or winning a bid to conduct a research study knowing it cannot be completed at the bid price and then immediately raising the price after winning the bid, is problematic.
Failure to maintain client confidentiality is a problem. If I collect information for a client, then that client owns that information; hence,I have no right to provide that information to others.
Failure to avoid a conflict of interest can be an issue when researchers have multiple clients in the same industry. Things that researchers learn in conducting a study for one client should not be shared with for other clients. In other words, one client just paid for research results provided to another client, which clearly is inappropriate.
Reusing data collected for another client is another issue, especially if it’s proprietary. Clients charged for data collection should be safe in assuming that they’re paying for newly collected data.