Sample and Respondent Provided County Comparisons Among Cellular Respondents Using Rate Center Assignments

Area of Science:

• Public Health
• Survey Methodology
• Geographic Information Systems

Background:

• The increasing reliance on cell phones for personal communication necessitates evaluating their impact on telephone survey sampling.
• Traditional landline surveys face challenges with cell phone-predominant populations, affecting geographic sampling accuracy.
• Cell phone numbers do not inherently provide accurate geographic location data, complicating survey research.

Purpose of the Study:

• To examine the county-level accuracy of cell phone samples within the 2016 Behavioral Risk Factor Surveillance System (BRFSS) dataset.
• To assess the reliability of using rate centers for determining respondent locations in cell phone surveys.
• To identify factors influencing the geographic accuracy of cell phone-based survey samples.

Main Methods:

• Analysis of the 2016 Behavioral Risk Factor Surveillance System (BRFSS) sample data.
• Evaluation of county-level geographic accuracy for respondents sampled via cell phones.
• Comparison of accuracy rates across different states, regions, metropolitan statuses, and demographic groups.

Main Results:

• Cell phone samples achieved 58% accuracy at the state and county level for the 2016 BRFSS.
• Accuracy at the state level alone was higher, reaching 93%.
• Significant variations in accuracy were observed based on state, region, metropolitan status, demographics, and survey items.

Conclusions:

• Cell phone samples present a notable challenge for achieving precise county-level geographic accuracy in large-scale surveys like the BRFSS.
• Researchers must consider the variability in cell phone sample accuracy when designing studies with geographic parameters.
• Further research is needed to improve methods for ascertaining accurate respondent locations in cell phone-based surveys.