Validation of interaction-based egocentric elicitation using ecological momentary assessment with young adults currently or formerly homeless

IntroductionSocial network changes are common as individuals transition from homeless to Supportive Housing. Egocentric approaches to elicit network members have been utilized with young adults experiencing homelessness to better understand such socio-environmental contexts; however, such approaches are subject to recall bias. Momentary measurements of one’s social network via Ecological Momentary Assessment (EMA) can support in understanding the accuracy of egocentric elicitation methods.ObjectiveTo examine who is elicited using an interaction-based approach in egocentric social network analysis among currently homeless and formerly homeless young adults residing in Supportive Housing and explore which alters are reported in interactions and how often are they reported when egocentric network analysis is embedded in EMA.MethodsThe present study utilizes interaction-based elicitation embedded within Ecological Momentary Assessment (EMA) in two sub-samples: currently homeless young adults (n = 72) and formerly homeless young adults residing in Supportive Housing (n = 118).ResultsMost egos interacted with three of their top five alters, while only 8.9% interacted with all five. Conversely, only 2% of egos reported that they did not interact with any of their top five. Several differences in alter characteristics were identified by housing status. Alters indicated as being more supportive are elicited first and alters elicited first are interacted with the most.ConclusionsEMA findings suggest that a cross-sectional social network survey of top five alters will results in an ordered list, with the most important/frequent alter being named first and trend downward. Results have implications for future study design involving egocentric network analyses. Present findings indicate qualities of alters present and thus, offer plausible “best practices” when utilizing network elicitation methods, particularly interaction-based methods.