Amid this week’s firestorm over Republicans’ attempt to weaken the Office of Congressional Ethics (and subsequently, backpedal on that attempt), you may have seen this chart floating around the Internet. It depicts data from Google Trends about Americans’ search interest in learning who their congressional representatives are, with a pronounced spike around 9 a.m. Tuesday.
The Hill reported this on Tuesday, with the The New York Times later referencing the trend. The conclusion some are drawing is that people looked up their representatives, then called them to complain about the ethics issue, and that the groundswell of support got Congress members’ attention. And, indeed, some Congress members reported that these phone calls did influence them.
But there’s a buyer-beware issue with this data, and with all Google Trends data — the problem of scale. While there’s a “spike” here, there’s no way to tell how big it is without comparing it to other popular — or unpopular — search terms. Depending on what you’re comparing it to, that spike is little more than a speed bump.
Here, for example, is the “who is my representative” search compared to searches for “Obamacare” in the U.S. over the past week.
The spike nearly disappears here — as of 9 a.m. Tuesday, “who is my representative” registered at 7 on this chart, compared to 60 for Obamacare.
But then, maybe that’s not fair. Obamacare is a really big topic in the news right now. Let’s try a Google search term that’s still politics-related but with less mass appeal, like infrastructure.
The Congress-member-search spike is a bit more apparent here, but it’s still dwarfed even by searches for the relatively nonsexy topic of infrastructure.
The really striking comparisons come when you compare this “spike” to searches for nonpolitics-related topics. How about a newly released, poorly reviewed movie?
What about a certain pop superstar who recently gave a catastrophic performance on national TV?
OK, so the spike for Mariah Carey happened a couple of days earlier, but you get the picture. It’s true that Google Trends data can say something about what people care about, and that can be super-useful in a democracy where politicians are theoretically supposed to represent those cares. For example, researchers reported in October that Trump searches outstripped Hillary Clinton searches. They subsequently concluded that “this could be evidence that Mr. Trump is doing better than the polls project.”
But when that data is presented without a reference point, it can overstate the importance of a particular search trend. In fact, to get a term that has comparable volume to the Congress member search, you have to get creative and do a lot of random Googling. As it turns out, searches for “who is my representative” are roughly equal to searches for a 1990s trend that people have mostly forgotten.
So searches for “who is my representative” spiked, surpassing Beanie Baby searches for the first time in … well, a long time.
None of this is to say that this spike didn’t matter; lots more people than usual did Google their Congress members’ offices, and anecdotal evidence suggests that there was an increase in people calling their Congress members.
But it is to say that Google data on only one search term can misrepresent the size of a phenomenon. Yes, plenty of people were Googling their Congress members this week. But plenty more were watching that Mariah video.