The impetus behind the Digital News Test Kitchen’s “Civilizing User Comments” project and news website user-comments research was the Greeley Tribune shutting down its comment section in 2011. Following this event, we were fortunate to obtain all of the comments from the Greeley Tribune from early February to late April 2011, the period leading up to the shutdown.
We are currently working on a textual analysis of the most-commented stories from the last week of comments during that time period on the Tribune’s website. More details on this project will be shared shortly, but, in the meantime, feast your eyes on some preliminary research results by Shannon Sindorf, my co-researcher on this project.
The following results correspond to the three months of comment data in our possession:
- Total unique user IDs: 987
- Total comments: 13,798
- 6,238 comments (45.21% of the total) were made by 20 people
- 412 out of 987 users commented only once
These numbers are fascinating on their own and seem to affirm what’s commonly known as the “1% rule” or “90-9-1 principle,” which says that in online fora, individuals are more likely to lurk than participate. (I.e., 1% are the “super-users” who participate regularly and often; 9% participate regularly but not often; and 90% of users don’t or rarely participate.)
Additionally, one research question to extract from this with potentially significant implications is: How substantive and how civil were the comments made by the top 20 commenters compared to those made by those who only commented once? If it was trolls who were bringing down the level and civility of the conversation and the core community was otherwise civil, wouldn’t shutting down the entire commenting apparatus be too extreme of a response?
Watch for more fascinating data coming from us soon.
Yes. I have experienced it. People who comment most bring down the civility of the conversation. I am already in the 90% users and looking at comments that have nothing to do with the topic hinders me more from participating.
I would love to see such correlation between the users and comments, if there is any.