Both the papers ask questions along similar lines, however, the question raised by the Kelly is more subtle and in a sense a follow up to the one posed by Bakshy et al.
- Garrett, R. Kelly. “Echo chambers online?: Politically motivated selective exposure among Internet news users.” Journal of Computer-Mediated Communication2 (2009): 265-285.
- Bakshy, Eytan, Solomon Messing, and Lada A. Adamic. “Exposure to ideologically diverse news and opinion on Facebook.” Science6239 (2015): 1130-1132.
The first paper asks quite an interesting question that deals with the consequences of technology i.e. “Will selective new exposure or the ability to limit one’s exposure to articles or sources of a certain kind lead to a dystopian future where democracy will collapse?” The answer to which was, unsurprisingly, no.
The results of the paper suggest that prospective opinion-reinforcing information greatly influences article selection and that immediately raises a follow up question, “how does fake news framed as opinion-reinforcing content come into play in shaping perception of truth among people?”. If people are more likely to consume articles that support their ideological point of view, then such articles can serve as good tools for derailing the truth. Since there is potential of the truth being manipulated, “does this lead people to very extreme, one-sided and tunneled vision i.e. can people be indirectly be programmed to adopt a certain point of view?if so, how devastating can the effects be?”
The second question has more to do with the design of the study, “does voluntarity (if that’s a word) exaggerate behaviors?” i.e. does passive vs active instrumentation of people lead to different results? will a person’s choice of articles be different if they knew that they were being monitored?
The third question has to do with the choice of the population. People chosen from well known left or right leaning sights are generally quite enthusiastic and/or as compared to the general public. Hence the results that were reported here might be an artifact of the choice of the population. So i guess the question i’m asking is “whether people who read news from partisan sources are democrats for the sake of being democrats or republicans for the sake of being republicans?”. Hence the generalization of the results to the general public is uncertain.
The second paper asks a different question of whether people get isolated from a opinion-challenging information even if they choose to. Given, the amount of data available to the authors i’m surprised the paper is as fleshed out as it could be, but the paper does a good job of quantitatively proving that the fear is unfounded. An interesting thing that i noted was that the neutral group tended to have a smaller proportion of friends who were neutral as compared to liberal of conservative friend, the difference is almost ~20%.
So is it that neutral people have these proportions because they are neutral in leaning, or is that them being neutral is an affect of having equal proportions of conservative or liberal friends?
Another aspect that i think could’ve been easily studied in the paper but was overlooked, is that given hard news that was opinion challenging, how likely are people to share it? This statistic would give deeper insight into how open are each of the three groups to change and/or how likely they are to peruse the truth of the matter and whether they value truth over ego.
Perhaps, this study could also look into different but specific topics e.g. climate change, and look at similar statistics to determine if “micro echo chambers” (topic based echo chambers) are formed. This sounds certainly seems like an interesting direction to me!
