2127-0597_SUPPORTING_STATEMENT_2012_Part_B[1]

SUPPORTING STATEMENT

FOR

STATE OBSERVATIONAL SURVEYS OF SEAT BELT USE

B.

Collections of Information Employing Statistical Methods

1. Describe potential respondent universe and any sampling selection method to be used.

Respondent universe are the drivers and front seat outboard passengers in passenger motor vehicles being operated on roadways of a given State during daylight hours. States may use simple or stratified random sampling to select the roadway segments that will be employed as observational sites. The roadway segments that are eligible for selection as observational sites must be drawn from counties or county-equivalent primary sampling units that collectively comprise at least 85 percent of the passenger motor vehicle occupant fatalities that occurred during a recent period of one to five years.

2. Describe procedures for collecting information, including statistical methodology for stratification and sample selection, estimation procedures, degree of accuracy needed and less than annual periodic data cycles.

The information is collected via direct visual observation of trained personnel stationed at the observation sites. Observers view passenger motor vehicles passing the site and note and record the shoulder seat belt use status (Belted or Not Belted or Unknown) of the driver and any front seat outboard passenger. States may elect to use simple or stratified random sampling to select the observation sites at which the information will be collected. Stratification variables may include but are not limited to the classification (rural versus urban) of the counties or county-equivalent primary sampling units, roadway segment length, roadway functional classification, estimated annual vehicle miles of travel on the roadway segments, average daily traffic counts on the roadway segments and many others. The data collection cycle is once yearly. Required precision is not more than 2.5 percentage points standard error in the estimation of the seat belt use rate of drivers and front seat outboard passengers. The non-response rate for shoulder seat belt use status (i.e., Unknown) is not to exceed ten percent of observations.

3. Describe methods to maximize response rate.

Observers receive hands on training with annual updates. Unannounced visits are made to at least five percent of the observation sites by quality control monitors to assure compliance with the specified observation procedures.

4. Describe tests of procedures or methods.

Verification that the observational procedures produce trustworthy assessment of the shoulder seat belt use status of drivers and front seat outboard passengers is conducted by experienced observational surveyors as they administer training to the observers. Throughout the data collection period (typically a two to eight week span) data analysts review the raw data to verify that the non-response rate is not exceeding ten percent of the observations.

5. Provide name and telephone number of individuals who were consulted on statistical aspects of the IC and who will actually collect and/or analyze the information.

Statisticians employed by the National Highway Traffic Safety Administration’s National Center for Statistics and Analysis developed the design criteria for the Observational Survey of Seat Belt Use. The Associate Administrator of the National Center for Statistics and Analysis is Ms Terry Shelton, who can be reached at (202) 366-4290, or via [email protected]. The responsibility for designing a survey in compliance with the criteria and for collecting and analyzing the data rests with the individual States, territories, District of Columbia and Puerto Rico.