FEATURED:

Twitter selects 2 studies to measure 'conversational health'

Twitter selects 2 studies to measure 'conversational health'
© Getty

Twitter has selected two academic proposals to help measure the “conversational health” of its platform.

The company announced on Monday that it had selected the two measurement frameworks out of a pool of more than 230 proposals that had been submitted since Twitter began soliciting ideas to improve its platform in March.

One of the projects will study how echo chambers are formed around certain discussions and the problems that arise from them. The study was put forth by academics from Leiden University, Syracuse University, Delft University of Technology and Bocconi University.

ADVERTISEMENT

The other study, proposed by professors at the University of Oxford and University of Amsterdam, will focus on the impact that intergroup dialogue on Twitter has on decreasing prejudice.

Twitter has been dogged by criticism that it’s not doing enough to crack down on abusive behavior on its platform. It launched this initiative in order to better study the effects of the dialogue hosted by the company.

“Ensuring we have thoughtful, comprehensive metrics to measure the health of public conversation on Twitter is crucial to guiding our work and making progress, and both of our partners will help us continue to think critically and inclusively so we can get this right,” Twitter said in a blog post on Monday.

The company has recently been hit with accusations that it’s been “shadow banning” leading conservatives, which involves limiting their reach by making their accounts harder for users to come across. Last week, President TrumpDonald John TrumpDemocrats slide in battle for Senate Trump believes Kushner relationship with Saudi crown prince a liability: report Christine Blasey Ford to be honored by Palo Alto City Council MORE threatened an investigation into alleged incidents of shadow banning reported by Vice.

Twitter says that it does not engage in shadow banning and that the incidents were all part of a glitch that made it so some user names would not be discoverable in search suggestions.