Instagram’s Quest to ban Certain Words is Futile – Study

Instagram ban on certain words is rather futile

Instagram ban on certain words is rather futile

Instagram has been moderating the use of some words, hashtags or emojis for a while now and was once even rumoured to be working on an R rated version of the platform. This has not stopped people from using the banned items in cheeky ways and a study proved that.

Researchers from Georgia Institute of Technology decided to conduct a study about Pro Eating Disorder (Pro-ED) communities on Instagram. Apparently, Instagram was banning searches that were linked to these communities on its platform and was issuing advisories on others. The researchers were able to gather 2.5 million posts between 2011 and 2014 which was used for the study.

So what did they find out? Well first, they identified 17 “root tags” out of the top 200 tags from the data set which had the pro-ED content. Thanks to Instagram’s ban policy, the use of these tags dropped sharply. “For the banned tags thighgap, thinspo and thinspiration , the proportion of posts sharply drops when Instagram reported changing its community policies,” the researchers found out. “This is consistent across the other banned tags, the use of banned tags decreased 13-78% after April 2012 (mean 52%)”

Although the ban worked for the specific tags, users came up with other variants where the researchers were able to find 672 variants from the top 17 tags. Variants of certain tags like secretsociety were found to increase by 4000%, thinspo by 200% and thighgap by 500%. This means that even if Instagram’s ban worked where the use of certain tags dropped, activity on the platform on Pro-ED content was still driving.

Instagram’s strategy seems not to be working and the report gave an alternative solution to the problem:

Rather than suppressing such content, social media platforms need to consider alternative intervention techniques that provide this safety valves and promote recovery from pro-ED.

This solution can be applied to all social networks where positive content is encouraged to replace the negative ones because clearly, banning them does not solve the problem at all.