I really enjoyed reading the recent Pew Research piece on 2016 Election news coverage. The article illustrates the increase in election-related stories around next year’s presidential race. I found the article count approach – across 15 metro newspapers – a good starting point, but one which leaves a few critical questions unanswered.
How High is High, How Low is Low?
The first question is quite simple: what does an 82 article count for Chris Christie actually mean? Are 82 stories high? Are they high for a candidate? Is 82 an increase over the prior month? is it seasonal when compared to the same period last year? Is it part of an upward/downward trend over a longer time period?
Essentially, story counts/mentions need to be placed in a greater, macro context. This would enable you to determine how high is high, and how low is low.
[mediaQuant: to truly put a single count in the right context you need to count everything, or at least the most salient topics, trends, subjects, people, institutions and whatever else is on the news agenda that day. At mediaQuant we apply a consistent methodology to thousands of brands, influencers, trends, topics, politicians, athletes, entertainers, world leaders, etc. We measure all these story topics cross the same set of media segments. And we make sure we’re looking at the same time frame. At this point we can convert a raw mention or story counts into a relative index score, or media rating. If the top story count across those 15 newspapers was 83, then Chris Christie’s 82 is quite high. But if the top story count is 1,239, then 82 is very low. The only way to determine those ranges is to measure thousands of storylines, group them consistently across time. This is exactly what we do at mediaQuant.]
Media Segments and Audience Reach
Taking the media pulse of 15 newspapers skews the results to a metro-centric demographic that reads traditional newspapers. Domestic newspapers are just one of many media segments that readers (voters) are exposed to. Were the right 15 selected? What happens when you look at regional, rural, trade, consumer, business, international, broadcast, etc. channels?
Again, the media landscape is complex, multifaceted, and quite broad. It also has a social dimension that is growing fast and frequently skews to the opposite demographic of traditional news outlets.
[mediaQuant: One of the beautiful things about news is the sheer diversity of coverage that’s out there. From liberal leaning national newspapers like the NY Times to conservative broadcast networks like Fox, and everything in between. Assessing media prominence across this diversity of media outlets and allowing for segment-level comparisons provides the clearest lens for discerning story direction, magnitude, and coverage. Even more important is factoring in the audience reach at all these media outlets. A story in the NY Times reaches 21 million individuals. An election story in my local Oregonian reaches 300,000 readers. mediaQuant surveys nearly 20,000 media outlets across 24 different media segments (i.e. traditional, social and search networks) and factors in audience reach in its measurement methodology.]
Social and Search Data
Beyond traditional newspaper coverage, social media sources provide a kind of amplifier and a viral channel for election news discussion and distribution. The story counts from 15 newspapers provide one view into an election, but what occurs after a story breaks takes place predominantly online (i.e. via blogs, forums, social and search networks).
Also, news is becoming more conversational, and this couldn’t be more the case then with election coverage. These discussions and re-broadcasts (retweets) are more common now than during prior elections. Counting newspaper articles omits these segments and the amplification they have on election storylines.
[mediaQuant: Our topline media rating for each candidate (along with the thousands of brands, trends, topics, organizations, athletes and celebrities in our database) is a weighted aggregate of 24 separate media segment counts from both traditional, social and search media channels. We include search metrics since it offers an amazing “pull” metric that complements the “push” metrics from the traditional “publishing” news outlets. We also include a full online news outlet segment that scoops up off-the-radar news sites that are typically ignored in most media measurement programs.]
I see mediaQuant’s methodology as a kind of the Hubble Telescope of media measurement, while classic story counts are more akin to a terrestrial observatory. We bring additional clarity, field of vision, and detail that is simply not available through more traditional media measurement programs.
I invite everyone to take the mediaQuant dashboards for a ride. And check out our methodology and blog posts to see exactly how we approach measurement across 40+ market segments. Data is refreshed monthly and new brands, topics, trends, influencers and candidates are added continuously.