Advancing efficient egocentric network data collection in behavioral health research: lessons from an adaptive online social network survey

Social network analysis (SNA) is valuable in behavioral health research, yet rigorous egocentric data collection remains limited due to participant burden, scalability challenges, and variability in network sizes. Face-to-face interviews provide detailed data but are resource-intensive, whereas web-based surveys offer scalability but require careful design for usability and data quality. This study presents an applied case study of developing and implementing an online questionnaire to efficiently collect egocentric network data, and, as a secondary methodological contribution, an approach for consolidating multiple family members’ independently collected reports into family-wide networks. Conducted as part of the Supporting Household Health through Family-led Promotion (SHAYP) study, the Qualtrics-based survey incorporated adaptive logic, structured response formats, and skip patterns. Participants identified key social connections and provided relational data on relationship type, health influence, and interaction frequency. A structured design facilitated merging egocentric networks into broader family-wide structures. We detail strategies to enhance usability, minimize burden, and maximize data quality, providing practical insights for researchers integrating SNA into web-based behavioral health studies. A total of 116 participants from 36 families completed the survey. Individually, participants identified 2–10 important people (mean = 4.1, SD = 2.2); after consolidating overlapping reports within each family, family-wide networks comprised 4–20 unique individuals (mean = 9.5, SD = 4.3) and 6–202 directed dyadic relationships (mean = 31.7, SD = 26.0). Most (94.8%) self-administered the survey, with a median completion time of 20.1 min. Dynamic question tailoring, structured node identification, and targeted redundancy reduction were designed to support data quality while minimizing burden. This study demonstrates a structured, adaptive online survey for egocentric network data collection, and offers an illustrative approach for consolidating multi-respondent family network data. Future advancements in survey tools, such as interactive network visualization and AI-assisted analysis, may further enhance data collection practicality and rigor.