A preference-opportunity-choice framework with applications to intergroup friendship.

Area of Science:

• Social Psychology
• Sociology
• Network Science

Background:

• Understanding factors influencing intergroup friendship choice is a key goal in social science research.
• Existing methods for separating personal preference from structural opportunity in friendship formation have limitations.
• A systematic comparison of these methods is needed to advance the field.

Purpose of the Study:

• To propose a general discrete choice framework for modeling friendship formation.
• To introduce a modified conditional logit model for estimating preference parameters independently of opportunity structure.
• To compare the proposed method with existing approaches for disentangling preference and opportunity in friendship networks.

Main Methods:

• Development of a general discrete choice framework where choice probability is proportional to the product of preference and opportunity.
• Modification of the conditional logit model to estimate preference parameters unbiased by opportunity structure.
• Empirical application using data from the National Longitudinal Study of Adolescent Health (Add Health).

Main Results:

• The proposed modified conditional logit model provides a robust method for estimating preference parameters.
• The study demonstrates how to conduct sensitivity analyses to assess the impact of opportunity structure specification.
• Hypotheses regarding homophily and status asymmetry in friendship choice were tested using the new framework.

Conclusions:

• The proposed discrete choice framework and modified conditional logit model offer a more accurate way to study intergroup friendship formation.
• This approach allows for a clearer understanding of the interplay between individual preferences and social structures in shaping social networks.
• The findings have implications for research on social influence, network dynamics, and the formation of diverse social connections.