Next Upcoming Google+ Hangout: Tuesday, August 27 @ 7PM (CST) - To Participate CLICK HERE

Search For Topics/Content
Help Me, Help You...
This form does not yet contain any fields.

    Misc. Stats Topics Discussion > confidence intervals vs. confidence levels and sample size

    I am having trouble understanding how to determine sample size for a quantitative study in which I will be using ANOVA for multiple group means. Some online calculators only ask for confidence interval, confidence level, and population size. Others also ask for within group standard deviation, and difference between groups . . . My population size is 246, my significance level is 0.o5, the power is 0.8. Is significance level the same as confidence level? Is confidence interval the margin of error? How do I know the margin of error before I determine the sample size which must occur before I even begin recruiting subjects for the study? How do I know within group standard deviation? How do I klnow difference between groups? Help!

    June 7, 2012 | Unregistered CommenterDana

    Hi Dana!

    First of all, thank you for your question (s)! You ask a lot of good questions, but I want to make sure I understand what your main question is:

    I'm a bit confused, because you first ask how to determine sample size, but then you indicate that you have 246. Are you trying to determine the power you have with that number of people? Please clarify and I'll be happy to help.

    On a broader level, I'd recommend checking out G*power, which is free software distributed by the University of Michigan, that handles various aspects of power analysis with several different types of models. I believe it would have no problem working with your design, and it is fairly easy to use/learn. Also, it is available for both Mac and PC, which is nice.

    Here is the link to that software:

    http://www.psycho.uni-duesseldorf.de/abteilungen/aap/gpower3/download-and-register

    June 7, 2012 | Registered CommenterJeremy Taylor