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.

50 Anchoring—technique used to enable respondents to see how they previously responded to the same question. Branching—technique used to direct respondents to specific subsequent questions depending on how they responded to an initial question. Cascading style sheets—language to describe the presentation of a web-based document (i.e., colors, fonts, layout) enabling the separation of the document content from the presenta- tion, thereby improving accessibility and flexibility while reducing complexity and repetition in structural content. Commuter sheds—patterns (often in the form of geographic information system plots) of home originations for the journey to work to a particular location. Convenience sampling—sample where the respondents are selected based on the convenience of access and availabil- ity for the researcher. Coverage error—error that results when every unit in the population does not have a known, non-zero chance of being included in the sample.* Cross-sectional panel—panel comprised of a subset of the population at one point in time. E-mail blacklist—list of e-mail addresses that have been identified by a blacklist organization as sending unwanted or otherwise abusive e-mail. E-mail sender authentication—tool to help verify the Internet domain in an e-mail sender’s address, and thereby verify the sender’s identity. Fare media—non-cash transit fare payment products. Geocoding—process of assigning geographic coordinates such as longitude and latitude to street addresses and other points and features. With geographic coordinates, the fea- tures can then be mapped and entered into geographic information systems. Geographic information systems (GIS)—systems for creating and managing spatial data and associated attributes. Geo-location—science of determining geographic location. Heuristic—technique designed to solve a problem, providing a good solution to a simpler problem intersecting with the solution to a more complex problem. Intercept surveys—surveys conducted in person to randomly selected respondents at a site-specific location. Item nonresponse—error caused by respondents skipping questions or failing to complete a questionnaire. Longitudinal panel—panel observed at intervals over long periods of time. Multi-method administration—survey approach incorporating multiple methods, such as surveys administered through the web, telephone, on paper, or through in-person interviews. Multi-method recruitment—survey respondent recruitment strategy incorporating multiple recruitment sources such as e-mail, telephone, in-person, mail, etc. Panel—already collected sample source. Panel survey—group of individuals or businesses that are surveyed repeatedly. Point-in-polygon—determination whether a given point in the plane, such as a respondent’s address, lies inside, out- side, or on the boundary of a polygon, such as a trans- portation analysis zone. Psychographic questions—questions designed to charac- terize respondents based on attitudes, interest, behavior, and preferences to create detailed portraits of the respondents. Sample—all units of the population that are drawn for inclu- sion in the survey.* Sample bias—error that arises where data are skewed when the sample is not representative of the target population. Sample frame—list from which a sample is to be drawn to represent the survey population.* Sample population—all of the units (individuals, house- holds, and organizations) to which one desires to general- ize survey results.* Sampling error—result of collecting data from only a subset, rather than all, of the members of the sample frame.* Screener—part of a survey that establishes the criteria for respondents. Sender authentication—program similar to a bonded sender program whereby one applies for approval, and once e-mail practices have been audited and accepted, applicant is added to a whitelist, allowing e-mail messages to pass by spam filters. Split sample—sample split into two independent groups for purposes of comparison. Stated preference survey—survey designed to measure the relative importance of different attributes to understand consumer preferences. Survey nonresponse error—error that represents a failure to obtain information from elements of the population that were selected. Target population—population that the researcher wants to survey. Total population sampling—survey of all the respondents in the sample frame. Unit nonresponse—error caused by over- or under-representation of groups in the survey sample. * Definitions adapted from Dillman, Mail and Internet Surveys: The Tailored Design Method (1). GLOSSARY