Nielsen, the television ratings authority, released a study on August 6 claiming a statistically significant causal relationship between Twitter activity and a change in television ratings, using data provided by Nielsen subsidiary SocialGuide. The key missing element from the Nielsen/SocialGuide study is the magnitude of the change due to Twitter activity . The study claims statistical significance (some Tweets sometimes do change TV ratings), but never discusses by how much the ratings were changed. A statistically significant change doesn’t need to be large to be significant. Since they didn’t include magnitude, I’d bet the change is small rather than large.
Magnitude has been included by Nielsen/SocialGuide in a previous study on correlation between Twitter activity and Nielsen ratings, so its omission in this study may also indicate that the causation magnitude is small.
More than anyone, I want to see a connection between Twitter activity and TV ratings (it’s part of my job, after all). This study could provide the ammo needed by digital leaders in traditional media outlets to advocate for even greater integration of television content with social media. We just need to know the magnitude of the change in ratings that social media can achieve.