Advanced

Is 5 votes stastically significant?

Story: New LXer feature: All users are editors; vote on unposted storiesTotal Replies: 1
Author Content
chris

Jul 26, 2004
4:39 PM EDT 		I like the idea of voting, and of using that as feedback for which articles are posted. However, I think the statisticians who thought up the "law of large numbers" were thinking of a few more digits in the sample size. We're just too small of a group to have numerical stability if 5 votes can change the results entirely. We're going to be subject to pretty bad jitter in the results, even without malicious intent. We need a lot more voters before voting results should be given much weight. At the current level, they probably should be no more than a small percent of the decision factor.

When 2 or 3 people can sway the vote entirely, and there is no restriction on who creates accounts, there is just too much potential for a small group of people to abuse the system.

(I'm not trying to be chicken little, but I feel better voicing a concern of a problem before it actually happens.)
dave

Jul 26, 2004
7:27 PM EDT 		Hi Chris,

I just threw out the number 5. :) I'd prefer it was more like 50.

Still yet, we don't have those kinds of daily logged in members willing to vote, so the entire process is still manual and directly through me. Nothing occurs without my approval, so all is still well in LXland. :)

dave

Post a reply to this thread:

A free account is needed to post. Write your comment here first: your text is kept in this browser while you sign up or log in, and is waiting for you when you come back.

Your message:
|||||
10
Markdown: **bold**, *italic*, [text](https://link), - lists, > quotes, $x^2$ for math. Formatting help

Sign up and post (it is free)  or  Log in and post