The Solar Eclipse Generated Some Hilarious Search Spikes on Google Trends
A week ago today, North America was treated to a rare celestial spectacle when the moon passed between the sun and Earth, momentarily blocking the sun. Hordes of people flocked outside to witness the solar eclipse take place — the first of its kind in the contiguous US since 1979.
Of course, a lot has changed in technology over the past 38 years, leading to a new outlook on the event. With the convenience of virtually perpetual internet access and lightning-fast search engines, many of us have now become totally reliant on tools such as Google — in fact, it’s almost become a knee-jerk reaction to hit up Google whenever a question starts to bubble away on the tips of our tongues.
As such, it was entertaining to observe how people interacted with the beloved search engine around the time of the eclipse. Whilst there were plenty of search results for the keywords “solar eclipse”, there were a number of others that unfolded as a consequence of the celestial event happening overhead.
“My eyes hurt” generated bulk search results an hour after the eclipse took place
According to Google Trends, the term “my eyes hurt” soared in search queries shortly after the solar eclipse occurred on Monday, 21 August. This was accompanied by related phrases such as “eclipse headache”, “looking at sun”, “I looked at the sun”, “I looked at the sun for a few seconds”, “what happens if you stare at the sun” and “am I blind”.
Apparently, even though people were advised to avoid looking directly at the solar eclipse, many took the chance anyway. In the wake of their grave mistake, they then decided to do what any ordinary person in the digital age would do — consult Doctor Google. This more or less led to the following results, as compiled by a Reddit user:
Is Google Trends data really this clear-cut?
Certain sites have counter-argued this data, stating that it’s “misleading”. In the above chart, it appears that search queries for “my eyes hurt” matched the number of queries for “solar eclipse”, but this is not quite the case. For instance, the results for eye-related phrases may only represent 100 searches, while the results for “solar eclipse” might indicate 1,000 searches.
Quartz Media took this into account and decided to compare the two search results on the same scale, subsequently discovering that the former search term actually dwarfed the latter as illustrated below:
So, what does all this mean?
As a search metric, Google Trends does not indicate the number of searches conducted
While terms such as “my eyes hurt” certainly did rise in Google search results, it’s important to note that an element of context needs to be applied when using this tool. Sure, Google Trends is great for pinpointing and displaying interesting trends, but in this case, the charts that are being circulated make the spike seem much larger than it actually is.
In other words, it’s important to exercise caution when using Google Trends— after all, it doesn’t take much to manipulate data in one’s favour.
This does not mean Google Trends data is untrustworthy
Before you go leaping to conclusions and dismissing the value of Google Trends altogether, there is certainly a lot of integrity to be given to the tool. Since it introduced its real-time aspect two years ago, people around the world have managed to explore global reactions to major events, thereby acquiring a unique perspective into what residents are interested in across multiple regions.
According to Medium Corporation, “Trends data is an unbiased sample of our Google search data. It’s anonymised (no one is personally identified), categorised (determining the topic for a search query) and aggregated (grouped together). This allows us to measure interest in a particular topic across search, from around the globe, right down to city-level geography”.
Given the amount of Google searches conducted every second, it only makes sense to take a sample of the trillions of entries rather than the total number. To do so would make it way too difficult to process the enormous amount of data. Eliciting a sample of the data, on the other hand, gives scope for insights to be processed within just a matter of minutes of an event unfolding in the real world. Pretty cool, huh?
At the crux of Google Trends is the value of digging further into highly granular data, making it an ideal tool for combining with other data analyses. For instance, drawing a dataset from Google Trends can be highly complemented by demographic data generated by a census. Again, Medium Corporation explains this eloquently: “When we look at search interest over time for a topic, we’re looking at that interest as a proportion of all searches on all topics on Google at that time and location.”
Regardless of how much you trust Google Trends, there’s no denying that the latest solar eclipse is an amusing sign of the times
Whether you want to anally pick apart the data or not, the proof is essentially in the pudding: there was indeed a spike in search terms relating to eye pain following the eclipse, just not quite to the degree that is being suggested. It just goes to show a) how much Google has become a first instinct for sourcing information; and b) how little people pay attention to essential health and safety advice. #RIPeyesight.