Austin van Loon is a PhD candidate in the Stanford Sociology department. Austin’s areas of interest include social networks, computational methods, social psychology, and the sociology of culture. He studies organizational culture; political polarization; and the mechanisms of social network formation, maintenance, and change (especially in organizations).
Organizational networks consist of both formal and informal ties. While the latter may result in productivity improvements for the firm, e.g. if information flows more efficiently, they may also circumvent the formal and legal rules that offer protection from discrimination for certain groups, such as women. To study whether and when informal ties within a workplace can circumvent these formal rules, and whether circumvention of formal rules amplifies gendered disadvantage, we propose to use internal email correspondence from three large US-based firms to identify expressive (e.g. friendship, sharing jokes) and instrumental (e.g. work-related, solving organizational tasks) communication between employees. We utilize machine-learning to classify each email as a case of expressive and/or instrumental interaction. Using these classified emails and network analysis, we first seek to ascertain whether expressive communication begets instrumental communication and vice versa. Secondly, we test whether this process happens unequally for men and women. These findings on how women experience relationships in the workplace differently from men would provide important insight to devising better diversity policies.