The other thing that is great about Google Trends data is that is essentially a window into "things people are actually searching for when they think nobody is looking." Yes, I realize the irony in that every Google click is tracked, but you'll get much more "honest" results looking at Google Trends data vs. plain interviews. For example, the country with the highest volume of "gay porn" searches per capita is Kenya, a country that is pretty vehemently anti-gay. The reasoning behind that is open to interpretation (I suspect many people are just genuinely curious given images of gay relationships in the media are so much less acceptable there), but for a society that basically deems gay sex extremely deviant, there sure are a lot of folks interested in what it looks like.
Re:Russians are trying to flee – data from Google Trends
by daenz on Saturday March 05, 2022 @12:25PM
(#30569566)
At the time of my post, I was unable to verify the Google Trends searches (I was using it incorrectly, and have verified it, see other comments). My main point about using the Trends data is that if there are so many more clearly substantive sources (videos, interviews, history), why lean mainly on Google Trends.
Re:Russians are trying to flee – data from Google Trends
by hn_throwaway_99 on Saturday March 05, 2022 @12:53PM
(#30569905)
Google Trends is actually a much better data source for this kind of thing, precisely because it's so difficult to game. The examples you give (videos, interviews, history) are trivial to use to give a biased view - just show only the videos that back your viewpoint, for example. It's easy to find a hundred videos on YouTube with "proof" the Earth is flat.
On one hand I do agree that there is a lot of propaganda on all sides, so one should be especially critical about data sources.
On the other hand, I think then that using Google Trends data for this is a pretty excellent source for this data, and a prime example of something that would be difficult to game. It's publicly available, so anyone can verify it. I guess a true conspiracist might think Google is faking the data, but that seems extremely unlikely and bizarre, given that there are whole swaths of terms you could search for to make similar points.
You may reach a different conclusion based on what a 4-5x increase in "Visa" searches means, but if anything this would be an example of data I would have a ton of confidence in.