Leave a comment

Comments 5

mathmanatm March 29 2008, 19:19:55 UTC
That "Me too" wedge is too large.
Unless a lot of people are lying.

Reply

mathmanatm March 29 2008, 19:20:51 UTC
Wait. WHAT IS THE POPULATION? This graph is meaningless without that information!!
What are n for all of those groups? Are we being lied to with "statistics"?

Reply


mathmanatm March 29 2008, 19:25:21 UTC
Your title is anecdotal evidence only!
Peer review journals would tell you to revise and resubmit.

Reply

lone_wanderer23 March 29 2008, 19:35:56 UTC
I don't know that this would be going to a peer journal, per se.

This is one of those graphs you see on the evening news.

As for population size, I'll run with "sufficiently large". (n >= 30)

Reply

mathmanatm March 29 2008, 20:10:52 UTC
Meh. With an n around thirty, you would still be looking at something with a rather large confidence interval, seeing as you'd be looking at the normal standard error only divided by about 5. (sigma over the square root of n, whoo!)

Reply


Leave a comment

Up