I 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.
I'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?
4 comments
[ 3.1 ms ] story [ 18.7 ms ] threadI 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.
http://coding.pressbin.com/88/How-my-new-site-Correlated-wor...
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?