[CLEAN] Supporting Statement Part B_Online Survey on Safety Belt Usage of Commercial Motor Vehicle Drivers 09.11.24

Department of Transportation

Federal Motor Carrier Safety Administration

Quantitative Data on Safety Belt Usage of Commercial Motor Vehicle Drivers

OMB Control No.

INTRODUCTION

This is to request the Office of Management and Budget’s (OMB) Clearance for the Collection of Quantitative Data. (OMB Control No.).

Part B. Collections of Information Employing Statistical Methods

1. DESCRIBE POTENTIAL RESPONDENT UNIVERSE AND ANY SAMPLING SELECTION METHOD TO BE USED.
   
   

According to the FMCSA Carrier and CMV Driver Counts, as of December 31, 2021, there are approximately 8.7 million commercial motor vehicle (CMV) drivers operating in the United States. All CMV drivers who are at least 18 years old, reside in the United States, have access to the Internet, and are members of the Centiment panel are eligible to be included in the sampling universe for the study. DCG will use simple random sampling of the existing panel from Centiment and any available list samples from FMCSA partners as the sample frame to conduct an email-administered survey. The existing panel and list samples are proprietary information, and the total size and coverage of these sample sources are unknown. In previous studies that Centiment conducted for shipping and logistic companies, the panel was able to achieve between 3,000-8,000 completed responses. Given the modality and use of an existing panel, self-reported data from potential respondents will determine eligibility for active CMV drivers residing in the United States to complete the full survey.

To test the survey in the online survey platform, DCG will first conduct an informal pretest of the survey to ensure routing and skip-patterns function as intended. The pretest will not include any eligible sample of CMV drivers and will be conducted internally within DCG Communications for testing purposes only. Once the pretest has concluded and the survey is functioning as intended, DCG will conduct a formalized pilot of the survey to 50 CMV drivers through a panel provider, Centiment.

The formalized pilot will help the agency estimate the expected response rate when soliciting drivers and will help inform the total number of survey solicitations needed to reach the total sample size of 1,000 CMV drivers for the full survey. The survey is estimated to be in data collection for four weeks or until 1,000 CMV drivers are reached.

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.

Data collected in this survey is self-reported by respondents included in the existing panel and any list samples acquired for sampling. The survey is not intended to be representative of the overall universe of the U.S. CMV driver population. The intended sample size is 1,000 CMV drivers. Assuming 8.7 million CMV drivers are currently operating in the United States, a sample size of 1,000 respondents will yield a margin of error of ±3.10% at a 95% confidence level¹ to be generated from this study. Given the unknown demographic composition of the panel and of CMV drivers, estimates produced in the survey will be the mean value of each variable for each individual survey question in the survey instrument. Estimates specific to particular drivers’ categories or demographic groups will be based on fewer respondents and will necessarily have larger margins of error. These averages will give the agency a better understanding of the behavior and perceptions of safety belt use among CMV drivers in the sample frame.

Existing industry employment data that is self-reported by panel members exists within the Centiment’s panel and will be used for initial survey invitations to help target CMV drivers through simple random sampling of this subset of the panel. Given the panel's proprietary nature, the exact number of CMV drivers in panels is unknown, and the Centiment panel cannot provide any precise proportions of the panel's demographic composition. Centiment does note that their panel is reflective of the general U.S. population by age, gender and the nine census regions of the United States. Any available list samples of known CMV drivers that can be provided by FMCSA partners will be introduced through random sampling if the Centiment panel experiences low response rates of CMV drivers. Self-reported data within screening questions related to consent, age, state of residence, current employment, and CMV operators will be used to reach the target population for the full survey.

For general population studies, Centiment has noted that response rates can range from 35% to 50%, but targeted studies to subsets of the population can yield a lower response rate. In previous studies of industries that include CMV drivers, such as employees of shipping and logistics companies within the United States, Centiment’s panel yielded 3,000-8,000 completed responses with a response rate between 20% and 30%. Although the number of CMV drivers within the panel is unknown, the results from previous indicate that reaching 1,000 existing CMV drivers is likely feasible. FMCSA is confident that 1,000 CMV drivers within the Centiment panel will be willing to participate in the survey.

3. DESCRIBE METHODS TO MAXIMIZE RESPONSE RATE AND TO DEAL WITH THE ISSUES OF NON-RESPONSE.
   
   

Initial survey invitation messages via email will be sent to the panel and potentially to available list sample in small batches throughout data collection to avoid oversampling. Potential respondents who do not engage with the survey link in the invitation message will be sent reminder messages every 72 hours until a total of 5 messages are sent before the sample is considered unresponsive.

4. DESCRIBE TESTS OF PROCEDURES OR METHODS TO BE UNDERTAKEN.

Pre-testing of the approved survey instrument to ensure routing and skip patterns are working as intended will be implemented. Additionally, a pilot with 50 CMV drivers will be conducted. The pretest and pilot will help ensure that the survey instrument is working as intended and will provide insights on any adjustments needed to the questionnaire design and its functionality on the online platform.

5. PROVIDE NAME AND TELEPHONE NUMBER OF INDIVIDUALS WHO WERE CONSULTED ON STATISTICAL ASPECTS OF THE INFORMATION COLLECTION AND WHO WILL ACTUALLY COLLECT AND/OR ANALYZE THE INFORMATION.

Brian Ronk – Contracting Officer’s Representative and Government Project Manager – (202) 366-1072

Adam Cyr – Program Manager (Contractor) – (315) 405-6676

Tavian Mackinnon – Research Lead (Contractor) – (207) 650-8692

1 For margin of error, the standard formula: z * √p * (1 - p) / √(N - 1) * n / (N - n) with normal distribution

1