Click for next page ( 26


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 25
25 two-thirds of those conducting web surveys implementing Measuring Coverage Error Using Primary Data multi-method surveys to improve the response of their sam- pling population. The ideal way to measure coverage error is with primary data (i.e., information regarding the actual survey administered) from the sampling population that is being targeted. The typ- SURVEY ERROR CONSIDERATIONS ical sampling frame for most transit agency researchers is IN WEB-BASED TRANSIT SURVEYS their current and potential ridership; primarily people within There are a variety of survey error issues for researchers and the geographic area in which they operate. Information about agencies to consider when using web-based transit surveys. web penetration rates for people in the sampling frame can All survey methods have survey errors; therefore, each survey help determine whether there is reason to consider using method's errors must be understood in the broad context of all web-based research and how much of a concern coverage available survey methods, so that the transit researcher can error should be. A major finding from this synthesis research understand which survey method is best in a given situation. is that, for any research the transit agency conducts, web- It is also important to know when it is appropriate to use mul- based or otherwise, respondents be asked the following: tiple survey methods in a study to improve response and to whether they have web access; if that access is at home, at mitigate and minimize survey error for the entire study. Sur- work, or both; and the speed of their web connection (this can vey error can include coverage error, nonresponse error (both help to understand potential nonresponse as a result of slow connections). If they do have access, it is also critical for the unit nonresponse and item nonresponse), measurement error, transit agency to collect their e-mail addresses so that they and sampling error. This chapter focuses on coverage error can be put into a customer database that can easily be tapped and nonresponse error, which are both seen as critical issues to survey customers again in the future. for web-based transit surveys based on the results of the syn- thesis survey. Measurement error is discussed briefly as part Twenty-seven of the 36 respondents who completed the of the multi-method survey approach. However, the primary survey conducted for the synthesis were transit agencies. Of concern of transit researchers is coverage error in web-based those 27 agency respondents, 6 provided data on their transit surveys and therefore first addresses this topic. customers' web penetration and the remaining 21 (77% of the agencies surveyed) answered "did not know" to the Inter- net penetration question, which asked, "Do you know what COVERAGE ERROR percent of your customers have Internet access?" (Another 11 nonagency transit researchers who completed the survey Results of the synthesis survey make it clear that potential bias were not applicable to this analysis.) The average customers' from coverage error in web-based transit surveys is a primary web penetration reported by transit agencies in the survey concern of transit researchers and agencies. When asked was 71% (ranging from 50% to 90%), which is nearly iden- "What do you think are the disadvantages of web-based sur- tical to current national statistics that report web penetration veys?," all of the transit researchers currently conducting web at approximately 72% (7). It is interesting to note that some surveys cited coverage error and/or sampling bias owing to transit markets reported very high web penetration, up to coverage concerns. When asked "What do you feel are the rea- 90%, and the Tri-County Transportation District of Oregon sons your organization does not conduct web-based surveys?," (TriMet) noted that its research found web penetration levels two-thirds of researchers not currently conducting web sur- were higher for transit riders than for non-riders. Therefore, veys also cited coverage error/sampling bias as reasons. transit researchers must not assume that web-based surveys of their sample populations will necessarily result in high Coverage error occurs when a potential respondent within coverage error. a population cannot be accessed by the survey method being used. Good sampling practice aims to ensure that all mem- That the remaining 21 agency respondents answered "did bers of the population of interest have a chance of being sam- not know" to the Internet penetration question likely repre- pled for the study. For example, absence of Internet access sents the more important statistic of this analysis, because it for a potential respondent to a web-based survey would be shows that many agencies have not yet conducted research to considered coverage error, as would lack of telephone access determine their customers' web penetration numbers, includ- for a potential respondent to a telephone survey (1,6). ing some large urban agencies. In light of concerns about coverage error in web-based surveys, transit researchers must be able to measure potential Measuring Coverage Error Using Secondary Data coverage error in their target populations and understand how much importance to place on coverage error when If primary, or internal, data are incomplete or unavailable, choosing a survey method for their study. Coverage error potential coverage error in web-based surveys can be deter- varies depending on the respondents being targeted and the mined using secondary research on web penetration. This survey method. Coverage error can be measured using pri- research, often national in scope, can be particularly helpful in mary data and/or secondary data. determining web penetration of non-rider populations, a group

OCR for page 25
26 that most transit researchers may find more difficult to conve- TABLE 6 niently sample as opposed to sampling their own riders. A rea- U.S. INTERNET USAGE BY DEMOGRAPHICS sonable understanding of web penetration rates can be found from U.S. Census data (by state) and other secondary data, as well as from anecdotal research in the transit agency's geo- Demographics of Internet Users graphic area (e.g., research by businesses that have workers with web access, etc.). Businesses targeted as potential sources Use the Internet (%) Total Adults 72 of new riders will often be able to inform the transit researcher Women 69 about employees' web access at work. Men 75 Age There are a variety of sources for web penetration data, 1829 84 the most comprehensive being the U.S. Census' Computer 3049 83 Use and Ownership from the Current Population Survey. 5064 71 Although comprehensive, the Census data tends to be older 65+ 30 than other sources, making it less useful than more current Race/Ethnicity White, Non-Hispanic 73 data sources. Internet usage is growing at such a significant Black, Non-Hispanic 60 rate that even 3-year-old data may be considered out of date. English-speaking Hispanic 79 Furthermore, Census data only tracks computer/Internet Community Type usage at either home or work and does not provide one number Urban 75 that includes total access penetration regardless of location, Suburban 73 thereby underestimating the population's overall access rate. Rural 65 One very credible data source for the United States is the Pew Household Income Less than $30,000/yr 54 Internet & American Life Project, which tracks total cover- $30,000$49,999 78 age wherever this usage (access) occurs (7). Their September $50,000$74,999 87 2005 Tracking Survey contains statistics on Internet usage $75,000+ 94 (see Table 6). Educational Attainment Less than High School 38 As can be seen from this table, there are income, geo- High School 62 graphic, race, age, and gender factors in Internet penetration Some College 82 data; however, overall access penetration is relatively high at College+ 92 Dial-Up High-Speed 72% and growing quickly (Figure 17). Home Internet Users 39% 59% Actual effective access penetration may be slightly Source: Pew Internet & American Life Project September 2005 Tracking Survey (7). higher, as potential respondents may have Internet access at FIGURE 17 Computer and Internet access data from the U.S. Census (8).