Coverage Error Embedded in Self-Selected Internet-Based Samples: A Case Study of Northern Indiana

Abstract

The Internet is rapidly evolving as a new tool for conducting tourism research. Given known advantages, steps to examine the possible drawbacks need to be taken to verify Internet survey methods. This article focuses, in particular, on coverage error of self-selected e-samples. Based on the response to a question regarding willingness to provide contact information on the Internet, the respondents of two conventional tourism surveys were assigned to likely, somewhat likely, and not likely groups. Analyses were conducted to identify differences between these groups in terms of demographics, travel styles, travel-planning behaviors, and trip characteristics. The results indicate that Internet-based sampling procedures create a substantial potential for bias. Therefore, use of the Internet for tourism research requires substantial caution when generalizing the findings to a population that includes nonusers and self-selected nonparticipants.

Towards understanding members’ general participation in and active contribution to an online travel community

“…An important limitation of the study can be seen in the fact that the respondents were self-selected; these problems are difficult to avoid in studies of virtual communities since the members cannot be ''forced'' to fill out a survey questionnaire and the characteristics of the virtual community have not been documented (Hwang & Fesenmaier, 2003). Thus, this study was not able to address the question of whether respondents and non-respondents in the community differ in important characteristics.…”

Assessing Advertising in a Hierarchical Decision Model

“…The web-based travel survey was distributed to all inquirers based upon the date of contact (within 3 months of the request for travel information) and the destination from which information was requested. This is because it is generally argued that the advantages of online surveys (e.g., low cost, fast response, and wide accessibility of the Internet) enable tourism advertising researchers to send questions to the population of people who requested travel information, and therefore largely eliminate the use of complex structured sampling procedures (Fricker & Schonlau, 2002;Hwang & Fesenmaier, 2004). This approach enables us to obtain a sizeable sample which assures robustness of the parameter estimates (i.e., underlying behavioral response), which in turn, enables us to evaluate the relative impact of the hypothesized variables on advertising response.…”

A systematic comparison of first-time and repeat visitors via a two-phase online survey

“…First and foremost, this study suffers coverage error (Hwang & Fesenmaier, 2004), as most online surveys do. Hwang and Fesenmaier (2004) suggest that this bias comes from several sources.…”