+ Reply to Thread
Results 1 to 4 of 4

Thread: Study design issue: using national survey data as control group

  1. #1
    Points: 7, Level: 1
    Level completed: 13%, Points required for next Level: 43

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Study design issue: using national survey data as control group




    I'm working on a descriptive study of children who have specific health issues. Unfortunately, there is no good internal comparison group, so I'd like to compare vision rates with existing national survey data. My question is: a) Is this even possible, given there will be some issues linking local level to national/state level and b) aside from chi square/t-tests to test differences in rates, are there any special tests I should be using?
    Last edited by adamsaa23; 11-03-2014 at 05:32 PM.

  2. #2
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Study design issue: using national survey data as control group

    It depends on what the issues in linking are. Obviously it is possible to do this but to address this issue you would need to know what the issues are how the literature has addressed this, that is what has past research done in this regard.

    To me this seems like the time to use a one sample t test. Depending on your type of data a test of proportion might also apply. You say you are using descriptive statistics, t test and chi square really are not descriptive in nature.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  3. #3
    Points: 7, Level: 1
    Level completed: 13%, Points required for next Level: 43

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Study design issue: using national survey data as control group

    Assuming I am able to link the national dataset using a geographic identifier (and one that I would then create for my local study), wouldn't this create some sort of sampling bias? Maybe I'm over thinking this.

  4. #4
    Points: 3,730, Level: 38
    Level completed: 54%, Points required for next Level: 70

    Posts
    155
    Thanks
    7
    Thanked 30 Times in 29 Posts

    Re: Study design issue: using national survey data as control group


    I think what noetsi is suggesting is to calculate the relevant measure using the national survey data, whether it be a mean or proportion, and use a one sample test to see if your own data are different.

+ 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