National Survey of Drinking and Driving Attitudes and Behaviors: 2008: Volume 3: Methodology Report

National Survey of Drinking and Driving Attitudes and Behaviors: 2008: Volume 3: Methodology Report

  • Published Date:


  • Language:
Filetype[PDF-1.73 MB]

  • Alternative Title:
    Volume 3, methodology report. National survey of drinking and driving attitudes and behaviors : 2008
  • Publication/ Report Number:
  • Resource Type:
  • Geographical Coverage:
  • Edition:
    Methodology report; Final report; April 2009-April 2010.
  • Contracting Officer:
  • Abstract:
    This report presents the details of the methodology used for the 2008 National Survey of Drinking and Driving Attitudes and Behaviors conducted by Gallup, Inc. for the National Highway Traffic Safety Administration (NHTSA). This survey represents the eighth version in a series of periodic surveys that began in 1991. The objective of this survey was to assess the status and trends regarding the public’s attitudes, knowledge, and self-reported behavior related to drinking and driving. The Volume III: Methodology Report is prepared to supplement (i) Volume I: Summary Report which presents the key results of the survey including: basic frequencies on drinking and driving, perceptions of drinking and driving as a problem, actions taken to prevent drinking and driving, attitudes and experience with enforcement of drinking and driving laws, and the perceived effectiveness of different intervention strategies and (ii) Volume II: Findings Report providing an in-depth analysis of the topics presented in Volume I. This Volume III, Methodology Report, describes the methods used for sampling, data collection, data weighting, data analysis, and also includes copies of the questionnaires. The target population for the 2008 National Survey of Drinking and Driving Attitudes and Behavior was the general driving age public (aged 16 and older) in the 50 States and the District of Columbia. The mode of data collection, as in the previous rounds, was telephone. However, because of the rapidly expanding cell phone only population, as well as to ensure adequate coverage of young adults in the target population, the sample for the 2008 study also included interviews with respondents who use only cell phones and do not have a landline phone in their household. All interviews were conducted in both English and Spanish. The target population was geographically stratified into the four census regions (Northeast, Midwest, South, and West), and sampling was done independently within each stratum (region). The sample allocation across the four regions was proportional to the size of the target population in each region. This is a departure from the NHTSA 2004 and 1999/2000 surveys, where about 100 interviews were completed for each of the 50 States and D.C. Interviewing took place over a three-month period, from September 10, 2008 to December 22, 2008 and each interview averaged about 17.5 minutes in length. For the main study, a total of 50,448 landline and 32,049 cell phone numbers were dialed across all four regions. A total of 6,999 interviews were completed, including 1,607 (about 23%) interviews from the cell phone only sample. A minimum of seven plus seven call design was used to make a human contact and then to complete an interview, and the overall response rate was 24.1%. The procedure for response rate calculation was based on the standard guidelines established by the American Association for Public Opinion Research (AAPOR). The final telephone sample of persons aged 16 and older was weighted to U.S. population counts to account for the sample design, differential non-response, and under-coverage of some groups in the sample frame. For post-stratification weighting, target data were obtained from the Current Population Survey (March 2008). A non-response bias study was also planned in accordance with Office of Management and Budget (OMB) requirements in order to examine the potential for nonresponse bias. A total of 200 interviews were completed with a random sample of non-respondents. The questionnaire for the non-response bias study included a subset of questions from the main study. The analysis plan for the non-response bias study was to compare the respondents and the non-respondents on key variables (survey data on selected survey questions). In addition, respondents of the 2008 national survey were split into two groups: i) early or “easy to reach” and ii) late or “difficult to reach” respondents. The total number of calls required to complete an interview in the main study was used to define these groups. Overall, findings indicated that there was little difference between respondents and non-respondents. This suggests that the possibility of any serious non-response bias is minimal. Finally, an experiment was incorporated in this study to assess whether a cash incentive increased response rates among cell phone-only respondents. Out of the 1,209 who received the incentive offer, 970 (80.23%) completed the survey. Among the 793 who did not receive the incentive, 637 (80.32%) completed the survey. These nearly identical percentages suggest that the offer of a $10 cash incentive did not impact the propensity of completing the survey among cell phone-only respondents.
  • Format:
  • Main Document Checksum:
  • File Type:
No Related Documents.

You May Also Like:

Version 3.15