+ Reply to Thread
Results 1 to 3 of 3

Thread: Power and Sample Size Calculation

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Power and Sample Size Calculation




    Hi, First post here--i'm having trouble figuring out how to calculate the sample size for a study I'm planning. I don't have much statistics background, so bear with me--

    My guess is that I should be doing a "One Way ANOVA Power analysis?"

    Here is my hypothetical example:
    I'm looking at backpack strap thickness, and the relationship with the backpack breaking.
    -All other variables controlled for--only looking at thickness and breakage.
    -There are 3 different thicknesses of backpack straps (1", 2", 3" thick).
    -Backpack breaking is an all-or-none event.

    How many backpacks do I need to test in each group?

    Based one similar studies, I have the breakage rates of 1"=20%; 2"=25%; 3"=14%. Am I doing the right test? I'm not sure how I plug in the breakage rates.

  2. #2
    TS Contributor
    Points: 17,779, Level: 84
    Level completed: 86%, Points required for next Level: 71
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,542
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Power and Sample Size Calculation

    This is not for an ANOVA, since interval scaled variables are nowhere involved.
    You could perform a Mann-Whitney U-test using break yes/no as grouping variable
    and thickness as ordinal scaled dependent variable. Or, you could treat thickness as
    categorical variable and perform a Chi² test.

    With kind regards

    K.

  3. #3
    Points: 1,340, Level: 20
    Level completed: 40%, Points required for next Level: 60

    Posts
    19
    Thanks
    9
    Thanked 1 Time in 1 Post

    Re: Power and Sample Size Calculation


    The one way ANOVA only allows for continuous outcomes. Here, you have a binary one. If you want associative go chi square, if you want predictive, go with logistic regression.

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats