Platform Structures, Homing Preferences, and Homophilous Propensities in Online Social Networks

Abstract

This study investigates how the attributes of online social platforms (e.g., open vs. closed, symmetric vs. asymmetric social networks) and user preferences regarding platform diversity (e.g., single-homing vs. multi-homing) moderate the influence of homophily on user behaviors across varied social segments. On the basis of panel data, we delve into the interplay among homophily, structural diversities across various online social networking services (SNSs), and users’ homing preferences in such social platforms. The data feature of the SNS consumption behaviors of 10,172 individual users are obtained over a period of 134 days. An agent-based simulation model is developed to further validate and generalize the empirical findings. The simulation and empirical results consistently indicate that propensity toward homophily differs significantly across platform types and users’ channel adoption behaviors. Online homophily is more pronounced in closed, private social networks than in open, public social networks. Users of asymmetric and symmetric SNSs exhibit weak and strong homophily, respectively. Moreover, whereas users who adopt a single SNS channel tend toward homophily, those who subscribe to multiple SNSs confirm the idea that “opposites attract.” These findings suggest that users of online SNS channels manifest complex human interactions typified by the combination of homophily, heterophily, and asymmetric social preference, although homophily is the most prominent disposition. Our findings suggest that homophily should be considered a dynamically changing human characteristic rather than a static attribute, and stakeholders should exploit the complex nature of users’ homophilous behaviors to address social problems in online SNSs and to enhance the effectiveness of social advertising.

Key words and phrases:

• age homophily
• agent-based simulation
• gender homophily
• homophilous relationships
• multi-homing
• negative binomial regression
• panel data
• platforms
• SNS
• social networking services
• social segregation

Supplemental File

Supplemental data for this article can be accessed on the publisher’s website at 10.1080/07421222.2017.1373008

Notes

1. These private SNSs are intended to deepen relationships with a handful of close friends who share mobile phone numbers. Path, for example, limited the number of friends to 50 for such purposes.

2. According to the Korean Information Society Development Institute, Facebook, Twitter, and KAKAO Story together represent 91.9 percent of total SNS usage in Korea. KAKAO Story is very similar to Path (https://path.com/)—a popular private SNS site in the United States.

3. For robustness checks, we employed the total log-in time as an alternative dependent variable and reran the same analysis. Although the magnitudes of the coefficients vary, the results on the basis of the total log-in time are qualitatively consistent with those of the main findings with the frequency of log-ins as a dependent variable. The specific results on the total log-in time are available upon request.

4. We cannot simultaneously examine age and gender in one integrated model that incorporates all four covariates because of multicollinearity problems. Specifically, for example, the sum of $NSAg{e}_{i,t-1}$ and $NDAg{e}_{i,t-1}$ is exactly the same as the sum of $NSGende{r}_{i,t-1}$ and $NDGende{r}_{i,t-1}$at every t.

Additional information

Notes on contributors

Hyeokkoo Eric Kwon

Hyeokkoo Eric Kwon ([email protected]) is a Ph.D. candidate in information systems at Korea Advanced Institute of Science and Technology (KAIST). His research interests include economics of information systems, mobile addiction, social media platform competition, and patent trolling. He has published his work in Information Systems Research. He has also presented his work at the International Conference on Information Systems, Workshop on Information Systems and Economics, and Conference on Information Systems and Technology.

Wonseok Oh

Wonseok Oh ([email protected]; corresponding author) is the KCB Chair Professor at the College of Business, Korea Advanced Institute of Science and Technology. He received his Ph.D. in information systems from the Stern School of Business at New York University. His research interests include economics of information technology, mobile consumption and addiction, social networks, and e-book pricing. His work has been published in Information Systems Research, International Journal of Electronic Commerce, Journal of the AIS, Journal of Management Information Systems, Management Science, MIS Quarterly, and Production and Operations Management.

Taekyung Kim

Taekyung Kim ([email protected]) is an assistant professor of business administration at the University of Suwon, Korea. He received his Ph.D. in management information systems from Seoul National University. His publications appear in Information Systems Research, International Journal of Information Management, and Asian Pacific Journal of Information Systems.