Area based stratified random sampling using geospatial technology in a community-based survey.

Area of Science:

• Public Health
• Epidemiology
• Health Disparities

Background:

• Obesity research in Hispanic populations often overlooks community and environmental factors.
• Community-based studies require representative sampling to avoid selection bias.
• Novel sampling methods are needed to ensure inclusivity of disparate populations in health research.

Purpose of the Study:

• To describe a novel Geographic Information System (GIS)-based sampling strategy for a rural community-based health study.
• To ensure an ethnically balanced sample of Hispanic and non-Hispanic participants.
• To examine social determinants of health and their association with obesity in a rural population.

Main Methods:

• An area stratified random sampling procedure was designed using GIS software.
• The sampling area was divided into strata based on Hispanic household proportion.
• A three-stage sampling process involved selecting Census blocks and then housing units within blocks.

Main Results:

• The study successfully recruited an ethnically balanced sample of 106 Hispanic and 111 non-Hispanic participants.
• The completed surveys per stratum closely matched the proposed distribution, indicating successful implementation.
• The GIS-based sampling minimized selection bias in this rural, community-based study.

Conclusions:

• A standardized, area-based randomized sampling approach, integrating GIS, enabled the recruitment of a balanced sample in rural community research.
• This methodology is effective for reaching disparate populations and should be considered for future community-based studies.
• The findings highlight the importance of robust sampling techniques in addressing health disparities.