This isn't data porn, nor is it for true connoisseurs.
A lot of this is seemingly random correlation with very little explanatory power.
Good info-vis and data summarization provides insight, or at least reveals relationships that are otherwise unseen or obscured.
These are polls without conclusions.
My psycholinguistics prof at uni was fond of saying that the more factors one considers the more likely you are to find spurious correlations. This sentiment makes me both lukewarm to and skeptical of any of the conclusions drawn on the hunch blog.
I have to agree with you. This showed up on my twitter feed from @Hunch a few hours ago (data porn title was theirs, not mine) and I posted it to HN. I didn't have a chance to dig further into the "reports" until a few minutes ago. Nothing to write home about! :-(
I feel like a stick in the mud doing so, but i feel obligated to ask... then why post it?
I don't think the point of HN is to throw things at the wall and see what sticks :\ there's been a lot of talk of the decline of HN recently, and i can't see how posting w/o reading helps. But perhaps that's just me.
A lot of this is seemingly random correlation with very little explanatory power.
Good info-vis and data summarization provides insight, or at least reveals relationships that are otherwise unseen or obscured.
These are polls without conclusions.
My psycholinguistics prof at uni was fond of saying that the more factors one considers the more likely you are to find spurious correlations. This sentiment makes me both lukewarm to and skeptical of any of the conclusions drawn on the hunch blog.