Interestingly, we found that closed triads based on shared organizational assignments were associated with a lower likelihood of tie formation and persistence over time, suggesting a diminishing return when shared foci and closure intersect.

Our findings challenge existing knowledge that attributes only matter in dyads. Instead, we show that attributes work at the group level!

In this study, we proposed a new network concept, “categorical closure,” to capture an increased tendency for closure in homogeneous triads. Such a phenomenon, we believe, is common in real-world networks but has surprisingly understudied.

RT @ntu_sociology: Our department is now accepting applications for a full-time faculty position starting Aug 2025. All ranks and specializ…

In addition to confirming the network spillover effect, this project also illustrates the utilities of social sensing data in exploring and designing public health intervention. The team is super strong and totally dedicated to being the best!

We found that changes in food purchase behaviors extended from employees actually enrolled in the program to their socially-tied coworkers who were not themselves intervention participants.

The database for this project involved records of millions of cafeteria purchases from thousands of employees for more than three years and included participants and non-participants in a workplace wellness program.

In this paper, we examined if a workplace eating intervention may benefit non-enrolled workers. We wanted to figure out how social connections affect health-related behavior, or the so-called network spillover effect.

RT @NAChristakis: Social contagion is a powerful force. People copy the thoughts, feelings, & actions of those to whom they are connected.…

In summary, to combat misinformation, it is crucial to examine both fake news publishers and online users. Their interactions on social media may shape the dynamics of misinformation diffusion, implying the importance of a relational perspective in understanding this phenomenon.

(3) More surprisingly, embedding with fake news publishers had an inverted U-shaped association with diffusion.

(2) Counterintuitively, social proximity to mainstream media was positively associated with more fake news tweets, retweets, and likes.

Using COVID-19 vaccine misinformation as a case study, the longitudinal network analyses suggest that: (1) as expected, ties to accounts with more followers were associated with more fake news tweets, retweets, and likes.

The paper examines the associations between network positions in digital misinformation market and misinformation diffusion to answer an empirical puzzle: why are only a few fake news publishers able to propagate their ideas on social media while most fail?

@jw_lockhart Congrats!!