Estimating Contextual Effects from Ego Network Data

Abstract

Network concepts are often used to characterize the features of a social context. For example, past work has asked if individuals in more socially cohesive neighborhoods have better mental health outcomes. Despite the ubiquity of use, it is relatively rare for contextual studies to use the methods of network analysis. This is the case, in part, because network data are difficult to collect, requiring information on all ties between all actors. In this article the authors ask whether it is possible to avoid such heavy data collection while still retaining the best features of a contextual-network study. The basic idea is to apply network sampling to the problem of contextual models, in which one uses sampled ego network data to infer the network features of each context and then uses the inferred network features as second-level predictors in a hierarchical linear model. The authors test the validity of this idea in the case of network cohesion. Using two complete data sets as a test, the authors find that ego network data are sufficient to capture the relationship between cohesion and important outcomes, such as attachment and deviance. The hope, going forward, is that researchers will find it easier to incorporate holistic network measures into traditional regression models.

Network interventions for managing the COVID-19 pandemic and sustaining economy

“…Particularly, our model needs to be tailored and calibrated with relevant parameters and data to reliably assess the potential impacts of these strategies in specific settings and contexts. Once digital contact tracing data of large sections of populations becomes available, network sampling and other techniques should allow for more precise predictions of the effect size of each NPI (51)(52)(53).…”

Expanding Social Network Conceptualization, Measurement, and Theory: Lessons from Transnational Refugee Populations

“…Such analyses are more commonly found in studies using data from Add Health. However, given the demonstrated capacity to estimate global network properties with egocentric data (Smith and Gauthier 2020), future research that uses name generators to gather ego network data may be able to estimate a wider array of network variables that can be used to further understand and aid refugee resettlement. Implementing the array of methods, analyses, and graphical representations from Add Health and associated studies may provide valuable insight into refugees and social networks for future studies.…”

Exploring Naturalistic Diffusion of an Evidence-Based Mental Health Intervention across Peer Networks of Youth in Sierra Leone

“…In this study, we deployed a trained enumerator who led the personal network data collection in the field. We adapted and abridged PERSNET, a standard personal network survey instrument routinely used to collect personal network data [30][31][32][33][34][35]. The PERSNET was administered to the YRI and control index participants who had consented and enrolled in the study.…”

“…We adjusted for four compositional measures of personal network characteristics routinely used along with demographic characteristics to explain the diffusion of innovation in numerous studies [30][31][32][33][34][35][36]. We also adjusted for the following demographic variables: education (i.e., never attended, primary, junior secondary, and senior secondary), gender, and relationship status (i.e., partner, no partner).…”