3 ms·
Show HN: Correlated - Help us discover surprising correlations
- gkchesterton 16y agoIt's like "Freakonomics" for the masses.
- edithstein 16y agoI especially like the post on your "Coding Blog" that's linked at the bottom, which explains exactly how you make the correlations -- and even gives the SQL you use. We stats nerds eat that stuff up. I can't give you any points for the rigors of your methods, but I'm assuming you're not planning to present any of these statistics as reliable.
- edithstein 16y agoThe post I'm referring to is here: http://coding.pressbin.com/88/How-my-new-site-Correlated-works/ http://coding.pressbin.com/88/How-my-new-site-Correlated-wor...
- jawns 16y agoI'd love to get some feedback about how I have addressed the "first-time bounce rate." In an earlier draft of the site, users had to create an account before they could even cast a vote on the daily survey question. Now, they can cast their vote -- and then they're asked to create an account, and I give them a concise explanation of why it's important that they create an account (correlations can only be made by tracking their responses to various survey questions over time). Any suggestions on how I can address the "bounce" problem even more effectively?