Editor's note: Sharon Weissbach is data modeling manager, National City, Louisville, Ky.

Discussions about the Internet have moved well beyond the question of whether it will be a factor in the future of education, commerce, entertainment and communication. Now the only reasonable disagreement is about how soon it will reach mass popularity.

Marketing research is just one of the fields bound to be transformed as on-line technology catches up to the visions of the visionaries. As problems of speed and ease of access are overcome, more and more people will flock to the Internet. Marketing researchers need to be there with them if they want to keep the profession relevant. Articles in the professional magazines and journals have already begun to consider the Internet as part of the research mix. So it is important to begin creative thinking now about how marketing research can thrive in this new medium and how it can use fresh approaches to improve research quality.

Some of the barriers to conducting research on the Internet are temporary hindrances whereas others are inherent to the environment. Current users of the Internet are a non-representative sample of the population as a whole, but as more people use the Internet this becomes less and less of an issue. It will be a while before those on the Internet are representative for most marketing research studies, but there are many inquiries, such as research about the Internet itself or other computer topics, for which the Internet crowd is already a perfectly reasonable population from which to sample. Even Internet skeptics must admit that, in the long run, limiting surveys to Internet users may be no more problematic than relying on telephone interviews is today.

The CommerceNet/Nielsen Internet Demographics Study (http://www.nielsenmedia.com) is a World Wide Web survey done in conjunction with a phone survey to assess the bias inherent in the Web data. As the discrepancy between the two survey formats decreases the Web becomes a mass market vehicle. This may be quite a few years away but in the age of rapid change and widespread adoption of new technologies, many are betting on it happening sooner rather than later.

Worried about privacy

Security on the Internet is a major concern and users today are quite understandably worried about privacy issues. Given the commercial incentives for insuring that information such as credit card numbers can be transmitted safely, however, encryption methodology will be at the forefront of Internet developments. As people become comfortable buying goods, filling out application forms and transmitting private communications over the Internet, they won't be troubled by submitting survey responses. There is also the benefit, already noticed in computer-administered interviews, that people become very comfortable interacting with a computer and have a sense of privacy since they are working alone.

The lack of human contact can, of course, also be a drawback. There may not be enough motivation to continue through a long list of questions without interaction with another person on the other end of the line or the incentive of being paid and actually going to a research facility. An encouraging human voice can be most effective in convincing a person to agree to participate in a survey.

One solution is to combine techniques to attain good rates of cooperation. A standard approach to increase participation in telephone interviewing is to send a postcard first telling the subject to expect a call. The mails are also used for follow-up cards reminding people to send in their questionnaires. For Internet surveys, both regular and E-mail can be used similarly.
Using snail mail may seem primitive but for marketing researchers trying to maximize representativeness and participation in a research effort, it's useful to approach subjects from different angles. For example, instead of just putting a survey out on the Internet and ending up with a totally self-selecting sample, a researcher could mail letters to a targeted group which includes an ID and an Internet address so that only those solicited will be able to get onto the survey site. In this scenario survey responses can be tied back to information already known about the individuals recruited by matching data with their original identification codes.

Another way to motivate survey participation is to let respondents into a restricted area of a Web site once they complete the survey. Other enticements could include entry in a contest, the chance to download some personal software, participation in a game or a customized report based on information given.

Appealing graphics could increase involvement by making things interesting, but given the current state of most Internet connections, it's unwise to risk scaring respondents away by having them wait for the time-consuming loading of extraneous features. As new telephone, cable and satellite connections increase communication speed, this will become less and less of an issue and wonderful creative opportunities will be available for marketing researchers.

Much potential

The most obvious benefit of using the Internet for research is the ability to reach large numbers of people. It is hard to imagine another medium which can provide so much potential while remaining economically feasible. The Internet is an international arena where many barriers to communication have been erased. The Graphics, Visualization and Usability Center (GVU) and the Georgia Institute of Technology, which report conducting "the oldest and largest public service Web-based surveys" (http://www.cc.gatech.edu/gvu/user_surveys) are currently experimenting with surveys in French, German, Spanish and Japanese. Participation is encouraged by appealing to a sense of working for the public good when it is stated that "your time and effort is greatly appreciated and will help monitor and guide the growth of future Web developments."
Of course, as the software develops, Internet surveys will have all the benefits associated with other forms of computer surveying. One appealing aspect of interacting with computers is that they let people respond at their own speed. Techniques such as surveys on diskette offer freedom from strict time constraints. It would be hard to overestimate the advantages of flexibility and convenience in today's environment where it is harder and harder to reach busy, active potential respondents.

As with other computer-assisted surveys, the sophisticated Internet survey can tailor follow-up questions to responses and review missing answers and errors so that they can be corrected. It would even be possible for a respondent to establish an ID when starting a survey so that there is no need finish in one sitting. That approach, if successful, would be a plus when longer surveys are needed (and it would certainly be an area for potential research to determine whether the same rules for survey length apply when people can respond in bits and pieces).

These are all things that can be achieved today to some extent through either surveys on diskette, computer administration of surveys on-site or computer-assisted telephone interviewing, but it is important to remember the potential of the Internet for bringing together the best of a variety of approaches.

Internet as platform

One way to take advantage of the Internet environment for marketing research is to use it as a platform for questionnaires that would be inefficient to conduct using most current approaches. It isn't worthwhile to conduct a phone survey to ask two or three questions. But on the Internet, a survey component could unobtrusively be included within a general site which is used for marketing or business transactions. For example, if a person accesses a banking home page and then goes to the "Credit Card" link, he or she could be asked a few questions about the features of a credit card they find most important before moving along to the information component. You have to be careful not to antagonize people by getting in their way or taking up too much time, but wit and creativity can be the basis for attracting and keeping consumer interest. Of course, the best way to insure that a person will want to share opinions is to put the shared data to work right away in a practical application. If a person knows that answering questions will result in more personalized information from a site, the immediate payback is a great motivator.

Marketing researchers as a group are extremely trustworthy and have consistently had very high ethical standards in guarding confidentiality and only using data in ways approved by informed consent, but use of the Internet does demand the evaluation of some new capabilities in light of professional standards and privacy issues. Microsoft received negative publicity when it was rumored that the company was trying to discover pirated software by looking at the hard drive of each person who registered Windows 95. While Microsoft probably had no such agenda but was simply gathering consumer information for marketing purposes, the propriety of such unauthorized data capture is highly questionable. Organizations like the Electronic Frontier Foundation are working to guarantee civil liberties on the Internet and it is important that reputable researchers become knowledgeable enough about new technologies to define their principles in an enlightened and sensitive manner. Marketing researchers on the Internet will also have to be aware that there are now even more opportunities for selling under the guise of research and it will be a challenge to differentiate the reputable projects from those using deceptive practices.

Resistance will diminish

Many people are still nervous about the Internet and may be hesitant to divulge information electronically which they would freely give over the phone or to a researcher in the mall. This resistance will diminish over time. Once people become comfortable in using credit cards for Internet purchases and transmitting other forms of digital cash, they will also share their ideas without fear of misappropriation. All issues regarding confidentiality and informed consent must be meticulously considered so that each researcher is aware of ways in which this medium may affect research ethics.
The possibilities for abuse of trust on the Internet are not that much greater than through the mails or phone lines but in order to attract respondents, researchers will need to be more explicit than ever in their guarantees of safety and in supporting the validity of their projects. These efforts will be worthwhile in tapping into a communication network with so many exciting possibilities.