Big numbers from the moderation process

Less Hate: 20% of the 600,000 published comments on all 4 websites needed moderation.

More Speech: 26,000 commenters interacted among themselves or with the moderators and journalists during the supervised moderation period in the project.

In brief, this report presents the big trends and effects of moderation from the whole process deployed on 4 websites (GSP.ro, Tolo.ro, Paginademedia.ro, Blogsport.ro) through comparisons between different periods of time (before and during the moderation period). These trends reflect mainly the impact the moderation had on the websites (in quantifiable terms like traffic, consumption, number of comments) and on the commenters and patterns of commenting.

A few things that are worth remembering from all our analysis:

• A downward trend can be seen from the daily evolution graphs on all websites regarding the percentage of moderated comments, which might suggest that the need for moderation went down over time.
• So much so as it cannot be a plausible explanation that moderators, for example, became more lenient in their manner of moderation towards the end of the project. On the contrary, since this was a learning process, they became more agile at spotting problems rather than less careful and they also became more involved.
• On GSP.ro when we compared the one year pre-moderation period with the one year moderation period we saw an effect of the moderation on the website: more commenters commented less. In other words, the application of the moderation system was necessary and its effects can be seen in the number of moderated comments as a percentage of total comments in the two periods: from 18% to 21% during the one year supervised moderation.
• The moderation system (SIDECO) put in place on these 4 websites cannot be identified as a factor that had a negative influence on the websites, as long as both increases and decreases can be explained by other factors.
• For example, on GSP.ro, despite the fact that there was a drop in the number of unique users (13%) between the two periods mentioned above, which cannot be attributed to the moderation as several independent factors were at play (the arrival of a new digital player in the sports field – DigiSport – or the drop in the number of Romanian athletes participating in international competitions), there was also a constant rise in the number of page views. Which rose by 58% between the beginning and end of the one year supervised moderation period.
• The websites that were non-sport-dependent (Tolo.ro and Paginademedia.ro) saw their audience rise importantly during the moderation period, but that was due mainly to editorial decisions.

For more on the big numbers from the moderation process, read the full report.

More on how the moderation worked in the project, here.

A useful set of rules for any online community based on our research in the project, here.