Many forms of misinformation have a clear source — authors or journalists misquoting a study, or a famous person giving an extreme quote. That allows us to check the underlying source, as in this example.
But some misinformation is in the form of an urban legend, which spreads through the grapevine with no clear source. After The New York Times bought Wordle, many people claimed — quite vehemently, given the addictiveness of Wordle — that the NYT had unfairly made it harder. This may have been driven by confirmation bias, since many players disliked the NYT’s political stance. They made increasingly strong claims based only on their own (or others’) subjective assessments.
Without a clear source for the difficulty of Wordle, you can’t check a person or a study. But, if you’re willing to get your hands dirty, you can investigate the underlying claim itself. Carnegie Mellon computer science professor David Andersen dug into the underlying source file behind Wordle to download the actual word list. In what’s below, it’s the ‘main.4d41d2be.js’ file.
He then ran a Python script comparing the old and new lists, and found this:
What does this mean in plain English? That the NYT removed exactly six words (fibre, lunch, agora, pupal, slave, and wench) and added no new ones, so Wordle became slightly easier, not harder.
Now the person on the street can’t be expected to go through the steps of a Carnegie Mellon computer science professor and conduct the checks ourselves. But what we can do is to see if there’s any evidence behind a claim before believing it, and certainly before sharing it — the third and final tip in my TED talk was to ‘pause before sharing’. Yet if we’re biased against the New York Times, we’re willing to believe claims that it made Wordle harder based on no more evidence that a few people saying ‘I think it’s become trickier’. As this Twitter thread explains, this was a great example of how confirmation bias causes conspiracy theories to spread.
