Forums
New posts
Articles
Product Reviews
Policies
FAQ
Log in
Register
What's new
Search
Search
Search titles only
By:
New posts
Menu
Log in
Register
Install the app
Install
Forums
Apple Computing Products:
macOS - Desktop Hardware
What Percent of People Prefer Mac Over PC?
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Reply to thread
Message
<blockquote data-quote="MacInWin" data-source="post: 1821763" data-attributes="member: 396914"><p>Ian, while I agree with you that the Study Group looks like it has bias, the sample size of 2200 is certainly adequate, if the sample is truly random, for a statistical analysis. And if the results are significant enough, you can get to the 95/99% level to say the results did not occur by chance alone. Adding more to the sample size will reduce the difficulty of reaching the 95/99% level of confidence, but that alone does not disqualify the total result. Political polling uses tiny fractions of the voting population to make statements about the probability for candidates using the same statistical approach. Ditto for dividing into the groups you suggested. That division just makes it more difficult for the 95/99% level to be reached because the sample size goes down because you've divided the whole into smaller units.</p><p></p><p>For more, Wikipedia has a short article on Statistical Significance: <a href="https://en.wikipedia.org/wiki/Statistical_significance" target="_blank">Statistical significance - Wikipedia</a></p><p>And SPSS has a tutorial on it at: <a href="https://www.spss-tutorials.com/statistical-significance/" target="_blank">Statistical Significance - Beginners Tutorial and Examples</a></p></blockquote><p></p>
[QUOTE="MacInWin, post: 1821763, member: 396914"] Ian, while I agree with you that the Study Group looks like it has bias, the sample size of 2200 is certainly adequate, if the sample is truly random, for a statistical analysis. And if the results are significant enough, you can get to the 95/99% level to say the results did not occur by chance alone. Adding more to the sample size will reduce the difficulty of reaching the 95/99% level of confidence, but that alone does not disqualify the total result. Political polling uses tiny fractions of the voting population to make statements about the probability for candidates using the same statistical approach. Ditto for dividing into the groups you suggested. That division just makes it more difficult for the 95/99% level to be reached because the sample size goes down because you've divided the whole into smaller units. For more, Wikipedia has a short article on Statistical Significance: [url=https://en.wikipedia.org/wiki/Statistical_significance]Statistical significance - Wikipedia[/url] And SPSS has a tutorial on it at: [url=https://www.spss-tutorials.com/statistical-significance/]Statistical Significance - Beginners Tutorial and Examples[/url] [/QUOTE]
Verification
Post reply
Forums
Apple Computing Products:
macOS - Desktop Hardware
What Percent of People Prefer Mac Over PC?
Top