Comment on Lemmy has a problem

user224@lemmy.sdf.org ⁨4⁩ ⁨days⁩ ago

• Alphane_Moon@lemmy.world ⁨4⁩ ⁨days⁩ ago

  I believe similarweb tracks web usage of thousands of validated users (i.e. they know their demographic profile) and they extrapolate on a per site basis form there.

  source

  • MonkderVierte@lemmy.zip ⁨4⁩ ⁨days⁩ ago

    So guesswork?

    source

    • rikudou@lemmings.world ⁨4⁩ ⁨days⁩ ago

      We call it statistics around these parts. But yeah, it’s guesswork.

      source

      • GreyEyedGhost@lemmy.ca ⁨4⁩ ⁨days⁩ ago

        Statistics is to guesswork what data is to anecdotes.

        source
  • rumschlumpel@feddit.org ⁨4⁩ ⁨days⁩ ago

    How could that possibly lead to a bias? /s

    source

    • Alphane_Moon@lemmy.world ⁨4⁩ ⁨days⁩ ago

      If you have 10,000+ users that are a representative sample of all internet users (or pretty close), your estimates should be pretty good for popular sites.

      Your sample might still be small for lemmy.ml.

      source

      • rumschlumpel@feddit.org ⁨4⁩ ⁨days⁩ ago

        I think the demographic for lemmy (and lemmy.ml in particular) is so niche and so biased towards privacy that even with a bigger sample, it’s next to impossible to get good data.

        Though I don’t doubt that the lemmy userbase heavily skews male.

        source

        • -> View More Comments