Announcement

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

  • Optimally-pooled t-test for partially pooled data

    I am using Stata 14.2. I have data consisting of 350 observations, 250 of which are from adolescents and 100 are from parents or caregivers. Among these, there are 90 parent/caregiver - adolescent pairs (i.e. within the same household). To compare differences in responses linked to individual or household characteristics (e.g. sex, household type, location) among these 350 individuals, I believe I should use an optimally-pooled t-test for partially paired data (based on Guo and Yuan 2017, full reference below). I would appreciate knowing if anyone knows of coding to do this in STATA.
    Thank you in advance!

    Guo, B. and Yuan, Y., 2017. A comparative review of methods for comparing means using partially paired data. Statistical methods in medical research, 26(3), pp.1323-1340.

  • #2
    HTML Code:
    https://www.tqmp.org/RegularArticles/vol18-1/p055/p055.pdf
    talks about the various methods.

    p. 60 has a fairly straightforward test if it suits your data.

    Comment


    • #3
      Thank you - very helpful.

      Comment

      Working...
      X