ABSTRACT

This paper examines issues pertaining to the use of the World Wide Web for market research purposes. With an estimated present worldwide user population of over 50 million (Ellsworth and Ellsworth, 1997) and projections of 707 million users by 2001, the Internet offers an enormous opportunity for market research. There are many advantages that the Internet has over conventional market research methods; It is faster, has higher response rates, allows for quick follow-ups, is cheaper and less prone to human error. Instead of the long time lags involved in telephone or mail survey, Internet surveys can potentially gather thousands of responses within hours from a global sample frame. This results in reduced costs in distribution, postage, paper, salaries, telephone use and data entry but with a larger sample.

Higher Response Rates - Respondents are contacted through their email and questionnaires are completed at the respondent’s convenience. Keyboard and mouse replace the tedium of filling out printed questionnaires by hand (CustomerSat, 1997). Unlike mail surveys where even if the design of the questionnaire and the sampling are good, a low response rate is a concern, as much of it is relegated to the trash can (Costes, 1997). Electronic surveys are also found to be more enjoyable and more interesting than paper-based surveys, which also contributes to high response rates.

Higher Levels of Response - Confidentiality assurance of Internet surveying, especially on sensitive issues, may well increase the level of response in comparison to telephone and face-to-face interviews. Kiesler and Sproull (1986) found that subjects in a computer interview answered more questions than those in a paper based survey. The interactivity of the Internet, as Costes (1997) notes, may be a good way to deal with the problem of anonymity, as clarifying the study's purpose can increase response rate. Surveyonline, a Internet survey research company found that “people who answer questions via the Internet seem to enjoy the process, especially when compared with traditional pencil and paper formats.” The reasons behind this is unconfirmed, but is said to be a combination of novelty and the ability to reply at their own convenience. Follow-up to keep respondents happy is also faster and cheaper (CustomerSat, 1997), which could contribute to an overall positive feeling about online surveying.

Reduced costs-Using the Internet, costs of survey deployment, response tabulation, and capturing verbatim open-ended responses are a fraction of the cost of conventional interviews with open-ended responses (CustomerSat,1997). In addition, it is inexpensive to use the collected information: responses can be used directly without the need to print or retype (Costes, 1997).

Eliminates human error-Since there is no need to retype or tabulated the collected data. The data are transferred directly from the survey to the database in electronic form, human error in data entry is eliminated.

Geographical spread-The Internet expands the company's market to include global markets (Hodges, 1996; Mosley-Matchett1997; Unger, 1996), allowing those from around the world to participate in surveys. This is especially useful for companies operating globally. However, this could count as a negative for companies that operate on a local or national scale, in which additional steps need to be taken to screen out unwanted respondents.

Faster response time-In a study conducted by Bachmann, Elfrink and Vazzana (1996), the results clearly show the superiority in response time of e-mail surveys; it showed that email surveys returned 6.5 days faster than the paper-mail on average. Moreover, it showed a very rapid rise in the cumulative percent of e-mail responses received, with more than half of the total responses in the e-mail group arrived within two days, and more than 80% within one week.

No interviewer introduced biased- Golden, Beauclair and Sussman (1992) conclude “electronic mail ... is a cool medium, void of the non-verbal cues and nuances conveyed in face-to-face communications." This would suggest that email surveys are more reliable due to the absence of interviewer introduced biases that are found in telephone and face to face interviews.

The Downside

However, like all mediums of primary research, on-line surveying does have its shortcomings (Alreck and Settle, 1995; Hair et. al., 1995). For example, large portions of the population do not have access to the Internet at the moment. Those that do have access are highly educated, fairly young (average of 35), with an above average income and are the adopters and innovators in society (Ellsworth and Ellsworth, 1997). Hence issues such as to the level of bias, the validity of data collected and the ability to generalise data collected in Internet survey research are of concern to the Internet researcher.

Self-selection- one of the most serious disadvantages to surveying on the Internet is evident is that of self-selection. Using Internet surveying it is difficult to really do a qualification of the sample, and in most cases you are likely to enrage more people than you would qualify. Research sampling of any online population taken only from those willing to respond or give their personal contact information over the Internet is likely to result in wrong conclusions. As Kully (1997) puts it, “It is the equivalent to sampling only the top of the pyramid and trying to draw conclusions of the entire pyramid. Respondent bias can be minimised by conducting surveys using research companies or Universities. The offer of incentives to respondents, commonly used by commercial researchers, can increase response rate and can be used by all Internet researchers.

A Unique Segment-The Internet population is only a fraction of the total population; even in the USA where the Internet is most widely used, only 6% of the population are estimated to have access (Eaton, 1997). Moreover, half the Web users fall into a single consumer segment of the U.S. population--called Actualises in the VALS 2 system. Members of the segment are highly educated and work in academic or technical fields (SRI International, 1995). Currently, most of the Web's other half are students or recent graduates working in technical, managerial, or professional fields. This population is overwhelmingly a Generation X crowd, with some 70% of other-half respondents reporting ages under 30 (SRI International, 1995). As such, the Internet user population is clearly not representative of the population at large. Companies will have to take this into account when contemplating the use of the Internet as a data-gathering tool.

Proper Netiquette

There are several things to keep in mind when conducting market research over the Internet. First and foremost it is important to be courteous, polite and not intrusive. The respondent should be given a way to vacate the mailing list. It is also necessary to check for lists within the target list so as to avoid double mailings to the same email account. As with other forms of survey research, the identity of the researcher and the purpose should be provided to respondents.

Paper surveys are typically short, in order to improve response rates. The same can be said of Internet surveys, which should be kept short (one or two screens) in order to generate high levels of response. Additionally, it should be remembered by the researcher that many people still pay for each email message they receive and some pay according to the length of the message: short surveys are both wise and kind. And, speaking of consideration for the respondent, surveys should be personalised. People have names and like to have them used. Never rent lists or use other company’s lists. This is both bad netiquette and likely to reduce response rates. And high response rate is the key to surveying while the Internet is the key to high volumes of respondents.

References

Alreck, P. L. and Settle, R. B. (1995) The Survey Research Handbook, 2nd edn, Chicago: Irwin.

Bachmann, D., Elfrink, J., Vazzana, G. (1996). Tracking the Progress of E-Mail vs. Snail-Mail. Marketing Research, Vol 8, Iss 2, Summer, pp 31-35.

Costes, Y. (1997) “Effectiveness of e-mail vs. paper-mail as a survey media” Universite Paris Dauphine.

CustomerSat.com company brochure (1997). http://www.CustomerSat.com

Eaton, B. (1997) “The Problems And Pitfalls Of 'Net Survey Methodologies” Creative Research Systems Inc.

Ellsworth, J.H., and Ellsworth, M.V. (1997) Marketing on the Internet, New York: Wiley

Golden, P.A., Beauclair, R. & Sussman, L. (1992). Factors affecting electronic mail use. Computers in Human Behavior, 8, 297-311.

Hair, J. F., Anderson, R. E., Tatham, R. L. and Black, W. C. (1995) Multivariate Data Analysis, 4th edn, Englewood Cliffs: Prentice Hall.

Hodges, J. (1996, Feb.12). It's becoming a small World Wide Web after all: Publishers, advertisers regionalize content services. Advertising Age [On-line].

Available: http://www.adage.com

Kully, E.A. (1997) “Conducting Research via the Internet” Crestwood Associates.

Mosley-Matchett, J. D. (1997, Jan. 20). Remember: It's the World Wide Web.

Marketing News, 31(2), 16.

SRI International (1995) http://www.sri.com/

Sproull, L. & Kiesler, S. (1986). Reducing social context cues: Electronic mail in organizational communication. Management Science, 32(1), 1492-1512.

SurveyOnline.com (1997) http://www.surveyonline.com/

Unger, R. (1996, February). Brand with the New Technology: Part I. Paper presented at the Retail Advertising Conference, Chicago, IL.