Original Article
Pregnancy posting: exploring characteristics of social media posts around pregnancy and user engagement
Abstract
Background: Pregnant women use the Internet and social networking sites to meet a variety of pregnancy-related needs as well as to help make decisions regarding their pregnancy and/or parenting. Knowing more about the current landscape of social networking sites as it relates to pregnancy can inform future work that wants to leverage social media for education or support.
Methods: Across a 7-day period, we conducted a content analysis of 4 Facebook pregnancy-related groups and 4 Instagram pregnancy-related channels. The posts were coded based on the nature, content, and purpose of each post as well as the number of likes, shares, and comments for each. After establishing inter-rater reliability, univariate statistics were used to describe the nature, characteristics, and popularity of the posts for each group or channel. We then used logistic regressions to predict which characteristics were associated with being popular or unpopular with users.
Results: There were 288 eligible posts within the 7-day window. Only a small proportion of the large (182,000 to 2,527,712) amount of followers responded with likes or comments (comments: mean =0.02%, SD =0.04%; likes: mean =0.36%, SD =0.89%). Facebook posts tended to give more general pregnancy-related information or opportunities for personal sharing while Instagram posts often had emotional support themes or posts meant to make pregnancy relatable. Popular posts tended to try to make pregnancy more relatable (OR =4.21, P<0.01) or offer emotional support (OR =4.62, P<0.01), while unpopular posts tended to provide general pregnancy-related information (OR= 2.17, P<0.05).
Conclusions: In general, there were clear characteristics that differentiated the two social networking site posts that garner user attention or not. This can inform how to provide health information and social support via social media to maximize its influence. This work suggests which features are desirable for followers and can help those interested in leveraging Instagram and Facebook to provide pregnancy-related health information.
Methods: Across a 7-day period, we conducted a content analysis of 4 Facebook pregnancy-related groups and 4 Instagram pregnancy-related channels. The posts were coded based on the nature, content, and purpose of each post as well as the number of likes, shares, and comments for each. After establishing inter-rater reliability, univariate statistics were used to describe the nature, characteristics, and popularity of the posts for each group or channel. We then used logistic regressions to predict which characteristics were associated with being popular or unpopular with users.
Results: There were 288 eligible posts within the 7-day window. Only a small proportion of the large (182,000 to 2,527,712) amount of followers responded with likes or comments (comments: mean =0.02%, SD =0.04%; likes: mean =0.36%, SD =0.89%). Facebook posts tended to give more general pregnancy-related information or opportunities for personal sharing while Instagram posts often had emotional support themes or posts meant to make pregnancy relatable. Popular posts tended to try to make pregnancy more relatable (OR =4.21, P<0.01) or offer emotional support (OR =4.62, P<0.01), while unpopular posts tended to provide general pregnancy-related information (OR= 2.17, P<0.05).
Conclusions: In general, there were clear characteristics that differentiated the two social networking site posts that garner user attention or not. This can inform how to provide health information and social support via social media to maximize its influence. This work suggests which features are desirable for followers and can help those interested in leveraging Instagram and Facebook to provide pregnancy-related health information.
