Twitter May Not Be A Good Gauge For Public Opinion

March 6, 2013

This article is included in these additional categories:

Analytics, Automated & MarTech | Data-driven | Digital | Government & Politics | Social Media

Pew-Twitter-Gauge-of-Public-Opinion-March2013The Pew Research Center has released a study comparing reactions to  8 major political news events on Twitter to national public opinion polls, finding that they vary quite significantly. The Twitter conversation does not tend to be exclusively more liberal or conservative than general public opinion, but it does seem to be quite negative at points. The researchers attribute the differences to Twitter’s limited reach (13% of US internet users) and its users not being an accurate representation of the general public, among other factors. For example, according to a recent Pew study, Twitter users represent a younger age demographic (see link above). Also, the subset of Twitter users who share their views on events changes alongside the topic, as does the volume of tweets.

Looking at selected case studies from the report, it appears that Twitter reaction was more liberal than public opinion during Obama’s re-election and the California same-sex marriage ruling, while the opposite was true during Obama’s 2nd inaugural speech and the State of the Union address. During the final weeks of the presidential campaigns, the Twitter conversation about both candidates were markedly more negative than positive, as noted here.

In all, the study concludes that while the reaction to political events on Twitter “provides an interesting look into how communities of interest respond to different circumstances, it does not reliably correlate with the overall reaction of adults nationwide.” It’s an interesting point to consider, given that many brands are now measuring sentiment on social media sites, and particularly as social media users voice their opinions about their customer experiences on those sites. Clearly, the Pew findings cannot be extrapolated to apply to brand sentiment (political views being more divisive than perspectives on brand experiences, for one), but the findings do raise the point that those who are vocal on Twitter aren’t necessarily always representing the views of the wider public.

About the Data: A detailed explanation of the Pew study’s methodology can be found here.

Explore More Articles.

Pin It on Pinterest

Share This