It seems like common sense that an increase in tweets can drive an increase in live TV viewership, but until now there's been scant proof of such correlation. A study released by Nielsen has found just such a relationship. In fact, Nielsen went so far as to use the other c-word: causation.
Relying on live TV ratings and tweets for 221 primetime broadcast episodes that were studied using SocialGuide -- a venture between Nielsen and McKinsey & Co. that captures Twitter activity for all U.S. TV programming -- the study found correlations between tweet and tune-in surges.
Not surprisingly, a lift in ratings often yields more tweets. According to the findings, a rise in live TV ratings drove up the number of tweets about the programming among 48% of the episodes sampled. But more interestingly, on the flip side, an increase in the volume of tweets drove up live TV ratings in 29% of the episodes included in the study.
To gauge the correlation in both directions, Nielsen conducted two separate analyses. First, it performed a minute-by-minute time series analysis to see if increases in TV ratings generated more tweets within a window of five minutes. Then it looked in the other direction to see if more tweets produced higher tune-in within the same window.
"We saw a statistically significant causal influence indicating that a spike in TV ratings can increase the volume of tweets, and, conversely, a spike in tweets can increase tune-in," said Nielsen's chief research officer Paul Donato, in a statement.
Intuitively it makes sense that heightened Twitter activity causes people to change channels. If you see tweets about remarkable athletic prowess being demonstrated in a basketball game or hilarious insults being traded in a presidential debate and you're already sitting on the couch flipping channels, it follows that you're likelier to check out the source.
But there are abundant high-profile examples of broadcasts whose ratings didn't live up to the massive chatter they drove on Twitter. Oprah Winfrey's interview with Lance Armstrong and MTV's Video Music Awards last fall are among them. And the first airing of "Sharknado" had underwhelming ratings compared to previous SyFy titles, despite its massive Twitter explosion (though the second and third airings did substantially better).
The absence of detailed information about Nielsen's methodology also raises a few questions. The correlation between Twitter and live TV events is well understood, but what about other genres and programming with an older audience? Did Nielsen's sample include a wide cross-section of programming, or did it focus on broadcasts that over-index for high social engagement? Nielsen didn't reply to request for comment by press time.
For Twitter, proof of the two-way causation between tweeting and tuning in could be helpful as it pitches its new TV ad targeting product, through which advertisers can show Twitter ads to people who've already seen their TV ads.
Separate from this research, Twitter and Nielsen are readying their previously announced Nielsen Twitter TV ratings to be available for the fall TV season. It will measure the total audience for social TV activity on Twitter, including both people who tweet and people who see those tweets. Twitter is supplying the data, but the product will be sold by Nielsen, which hasn't yet announced what the unit of measurement will be.