Twitter NOT a Reliable Indicator of Public Opinion

Seemingly dozens of startups exist which try to make sense of Twitter sentiment, and Twitter itself has been trying hard to portray its sentiment data as an accurate reflection of public opinion. However, Pew Research Center — among the gold standards of public research polling — says it might be all for naught.Pew compared traditional public poll results with Twitter sentiment data around eight of the most significant political events over the last year, often finding significantly divergent results.According to Pew, in some instances — Barack Obama’s reelection, the first presidential debate and a federal court ruling on California’s same-sex marriage ban — the reaction on Twitter was “more pro-Democratic or liberal than the balance of public opinion.” However, other events — Obama’s second inaugural speech, John Kerry’s nomination as Secretary of State and Obama’s 2012 State of the Union address — elicited a more conservative response on Twitter than they did in opinion polls.

Pew also pointed to the general negativity of political tweets, which perhaps suggests people are more likely to tweet about something about which they disapprove rather than vice-versa. That would affect Twitter sentiment data vis-a-vis public polling data, as Twitter data is comprised of opinions from people who weren’t directly prompted to share an opinion whereas public opinion polls rely on respondents’ answers to a series of questions.

Why else is there a difference between Twitter sentiment and public opinion polling? Three reasons, according to Pew: Demographics, sampling and grouping.

  • On demographics, only 13% of adults use Twitter, and only 3% said they regularly or sometimes tweet about the news, according to previous Pew studies — hardly enough to form a representative sample of voters. Additionally, Pew found that Twitter tends to skew young and left. “Twitter users are considerably younger than the general public and more likely to be Democrats or lean toward the Democratic Party,” reads the study.
  • There’s also a sampling issue: Twitter conversations can include non-voters (such as those under 18 or international citizens), while public opinion poll samples about domestic politics are limited to citizens aged 18 and older.
  • Finally, there’s a problem of grouping: Not everybody who tweets about politics tweets aboutevery political event.

“Those who tweeted about the California same-sex marriage ruling were likely not the same group as those who tweeted about Obama’s inaugural or Romney’s selection of Paul Ryan,” reads the Pew study, which pointed to the substantial difference in the number of people who tweeted about Obama’s relected compared to Kerry’s nomination as evidence.

“In the two days following Obama’s re-election on Nov. 6, there were nearly 14 million Tweets from people expressing their reaction,” said Pew. “And more than five million expressed their reactions to the first presidential debate. But other events, particularly the federal court ruling on same sex marriage in California last February and Obama’s nomination of John Kerry in December, drew a much lower volume of tweets.”

There is, of course, a potential conflict of interest here: Pew might not want Twitter infringing on its opinion-polling turf.

Mashable has contacted Twitter for its response to the Pew study, and we will update this post with any comment.

Do you think Twitter can be a reliable source of public opinion?

[Source: Mashable]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: