Web-Based Survey Techniques (2006)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5The project survey was designed to determine the ways that web-based surveys are being used in the transit field and to understand how transit researchers are conducting their studies. The respondents who were included in the survey sample came from a variety of organizations: public transit agencies, consultancies, universities, MPOs, and other gov- ernment entities focused on transportation and transit- related issues (Table 1). The “convenience sampling” method was used for this synthesis owing to the relatively small number of researchers in the field and the limited scope of the syn- thesis project. The sampling list was therefore developed using easily available and relevant sources that were not necessarily exhaustive. Specifically, the sample list came from the TCRP synthesis topic panel, the APTA Market- ing and Communications Committee, the TRB Survey Methods Committee, the TRB Marketing and Fare Policy Committee, and other selected researchers in the transit field who were recommended by panel members and others affiliated with this project. The recruitment was conducted by means of e-mail, with each invitation con- taining a custom link with a unique embedded password, which limited a respondent to answering the survey only once. Survey recruitment took place in three stages: the first invitations went to the TCRP Synthesis Topic Panel, the TRB Survey Methods Committee, and the TRB Marketing and Fare Policy Committee. During the second stage, e-mail invitations were sent to selected researchers in the trans- portation field. In the last stage, invitations were sent to the APTA Marketing and Communications Committee. The fieldwork for this synthesis survey took place during Febru- ary and March 2006. Overall, the response rate for completed surveys was 21% (36 of 175 invited). As well as being a convenience sample, it should also be noted that the sample may have nonresponse bias as a result of the survey invitation being titled “TCRP Synthesis Topic SH-07—Web-Based Survey Techniques.” This nonresponse bias might have occurred because those who do not conduct web-based surveys may have elected not to respond, believ- ing that they were not relevant to the study. Therefore, it is possible that the survey actually overstates web-based survey research in transit. The synthesis survey was divided into four sections: • Screener – Type of organization in which respondent works. • General survey inventory – Frequency and types of surveys respondent is involved in conducting Origin–destination, Customer satisfaction, Mode choice, Planning, and Other. – Characteristics of those surveys; that is, panel, cross- sectional, etc. – Modes studied. – Recruitment, sampling, and administration methods. – Quality of results data set. – Purpose of survey. – To whom results were presented. • Web-based survey specifics – Advantages and disadvantages to conducting web- based surveys. – Likelihood of implementing web-based surveys (if not currently in use). – Objectives of web-based surveys. – Design, software used, and hosting of web-based surveys. – Features of field for web-based surveys. – Support required for conducting web-based surveys. – Recruitment methods. – Resulting data set characteristics. – Response rates. – Costs. – Successful practices. – Web use by organization. • Contact information. A copy of the survey questionnaire is included in Appen- dix A. The survey was converted into a web questionnaire using a proprietary software program developed by the research team. As with other computer-based survey methods, web- based surveys can be programmed to validate responses as they are entered. CHAPTER TWO SYNTHESIS SURVEY METHODOLOGY

Another favorable feature of web-based surveys is the abil- ity to follow skip and branching patterns that depend on pre- vious responses. In this regard, web-based surveys are much easier for the user than paper-based surveys, where skip pat- terns can become confusing. In the synthesis survey respon- dents were asked the frequency with which they conduct five different types of transportation-related surveys, as outlined earlier. To relieve respondent fatigue, respondents were asked detailed questions about a maximum of three types of sur- veys. If respondents had conducted more than three survey 6 types, then the types of surveys they were asked about were randomly balanced to ensure enough of each survey type was collected. If respondents provided information on only one or two types of surveys, they were asked only about those types. A successful practice for any web-based survey is to pro- vide support for respondents with a “help desk.” For the syn- thesis survey, a survey-specific e-mail account was created and monitored during the course of the survey’s fieldwork schedule. A toll-free telephone number was placed on every page of the survey so that a respondent could immediately call the research staff if confused by a question or in case of technical difficulties. Beyond direct contact through phone and e-mail, the live survey data were monitored for com- ments that warranted attention from survey administrators. Finally, the database was also checked regularly to see if respondents were encountering any technical errors that were trapped by the software system and written to the database. To complete the survey effort, the data were reviewed and any respondents that required or requested follow-up on their responses were contacted for further information. Organization Count Percent Public transit agency 25 69 Consultancy 4 11 University 2 6 Metropolitan planning organization 3 8 Federal government 1 3 City government transportation department 1 3 Total 36 100 TABLE 1 RESPONDENT ORGANIZATIONS TO SYNTHESIS SURVEY