Will you pay $500 for the iPhone?

TalkBack 1 of 25:

Next »

• Thread View
• Flat View

Not a statistician?

>I'm not statistician but the fact that Compete Inc only quizzed >379 people seems like a flaw to me. A sample of 379 seems >incredibly small to me, too small to draw any real conclusions >from.

To obtain a 95% confidence level with a confidence interval of 5 on a population of, say, 300,000,000 (which is far more than the number of cellphone customers in the United States) requires a sample size of 384.

A better way to think about it is "If I choose N people randomly out of a population, what is the probability that they don't represent that population." And that probability falls dramatically with each additional new member of the sample set. If you ask 5 people a yes/no question and 4/5 say no, it's possible that they are a cluster of weirdos, but if you ask 100 and 75 say no it is much less likely that continued sampling will significantly (+/- 5 points) alter that percentage.

Posted by: abzdnet   Posted on: 02/24/07 You are currently: a Guest | Members login | Terms of Use

Alert moderator to an offensive message

Subscribe to this discussion via Email or RSS

SponsoredWhite Papers, Webcasts, and Downloads

• Server Consolidation and Containment With Virtual Infrastructure VMware To meet the constant demand to deploy, maintain and grow a broad array of ... Download Now
• Why Isn't Server Virtualization Saving Us More? A Few Small Changes May Dramatically Increase Your Efficiency VMware Companies have rapidly adopted server virtualization over the past few ... Download Now
• VMware Infrastructure: A Guide to Bottom-Line Benefits VMware Frustrated by the costs of maintain ever larger data centers?or building ... Download Now