• by Mark Walsh  @markfwal, March 20, 2013

Emerging from second-screen viewing, the marriage of TV and Twitter is becoming the digital era version of movies and popcorn. Some 10 million social media comments are made about TV content each day, with 95% made on Twitter, according to social TV analytics firm Bluefin Labs, which Twitter acquired last month.

The growing relationship between Twitter and TV was underscored in December when Nielsen struck a partnership with Twtter to create a new measurement tracking viewers’ engagement with programs they watch. The Nielsen Twitter TV Rating will provide a standard metric for conversations about TV shows on Twitter.

The new social TV rating will not roll out until the start of the fall 2013 season. But Nielsen separately today released findings pointing to a correlation between Twitter activity and TV ratings. The study, conducted with social TV analytics startup SocialGuide, found that Twitter was one of three factors -- including prior-year ratings and advertising spend -- “to positively impact TV ratings in a statistically significant way.”

“While prior year rating accounts for the lion’s share of the variability in TV ratings, Twitter’s presence as a top three influencer tells us that Tweeting about live TV is likely a significant indicator of program engagement,” said Andrew Somosi, CEO of SocialGuide, in which Nielsen has bought a stake. Other factors examined included promotional spend for a show, and differences in age and gender among audiences.

The study looked at the fall 2012 premiere and mid-season program ratings of over 140 broadcast and cable programs. For premiere episodes, it found an 8.5% increase in Twitter volume (number of tweets) was linked with a 1% increase in TV show ratings for 18- to-34-year-olds. For the 35-49 demo, a 14% increase in Twitter activity was associated with a 1% bump in ratings.

Furthermore, the study found that Twitter’s correlation to TV ratings for mid-season episodes strengthened for both age groups. A rise in Twitter volume of 4.2% and 8.4% was associated with a 1% increase in ratings for 18- to-34-year-olds and 35- to-49-year-olds, respectively. By mid-season, Twitter was responsible for more of the variance in ratings for 18- to-34-year-olds than ad spend.

Nielsen was quick to note that the research doesn’t prove a causal relationship between a spike in Twitter buzz and higher TV ratings. But the findings of an alignment between tweeting and TV ratings raises the question of whether Twitter could eventually be used to predict TV ratings, according to a SocialGuide spokesperson.

Not all signs point to Twitter as a hit maker. A survey released earlier this month by video search company Digitalsmiths found that social media buzz didn’t play a big role in TV viewership. Two-thirds (67%) of people asked whether they watched a TV show or movie based on buzz on Facebook, Twitter and other social networks said “no,” while 19% said “yes.” And three-quarters said they don’t post on social networks about what they’re watching.

The Nielsen study does focus on younger audience segments, which are more likely to take to Twitter while watching TV as well as being the ones most sought after by advertisers. Nielsen said it plans to conduct further research at the end of the current TV season to compare this study’s findings on premiere and mid-season episodes to finale episodes of the same programs.