Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • pwr with unequal sample sizes

    Hi,

    I want to calculate the MDE for a survey likert outcome measured on a 4-point scale, and I have the following groups w/ the respective sizes: 1) control (2000), 2) t1 (800), 3) t2 (1700), t3 (1700). I want to examine the coefficients on each of these t groups relative to the control via linear regression and also conduct t-test for each pair of treatment groups to examine whether there is any significance. In particular, I have a hypothesis that there is no difference between t1 and t2, but between t3 and t2 and t1. Without information on group means/SDs, is the following command the best way to proceed?

    Code:
    power oneway, power(0.8) n1(2000) n2(800) n3(1700) n4(1700)

  • #2
    The estimated effect size for 0.8 for this is 0.0419.

    Comment

    Working...
    X