Comparing Cell Phone and Web for a Student Survey

• Activity Number: 242
• Type: Contributed
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Young-Je Woo*+ and Sun-Woong Kim and Mick Couper
• Companies: Dongguk University and Dongguk University and University of Michigan
• Keywords: cell phone survey ; web survey ; coverage ; response rate

Abstract

In many countries the proportion of the population with cell phones is higher than that with Internet access. This is particularly true of the college population. Given the recent attention focused on surveys of cell phone users, and the preference for using the Internet to survey college students, it is useful to compare these two modes of data collection. This paper reports on a mode experiment conducted in the 2010 Time Use Survey of students at Dongguk University in South Korea. The frame of registered students contains both cell phone numbers and e-mail address. A sample of students was selected from the list and randomly assigned to a CATI or Web survey mode. Students were notified of the survey in both modes using both text messaging and e-mail messages. Findings show that the cell phone survey has a distinct advantage over the Web survey concerning response rates, coverage of domains, and item nonresponse. Substantive differences between the two modes were found for about half the survey questions. This suggests that cell phone surveys may be useful to surveys in populations with universal or near-universal coverage, and where cell use may be more popular than Internet use.