+ Reply to Thread
Results 1 to 4 of 4

Thread: Comparing Improvement Between Groups - Best Approach?

  1. #1
    Points: 4,048, Level: 40
    Level completed: 49%, Points required for next Level: 102

    Posts
    20
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Comparing Improvement Between Groups - Best Approach?




    Perhaps this is an easy issue, but I wanted to throw it out anyway.

    I'm working on a dataset where there are two groups, both having pre- and post-test scores. I want to see if one group has a significantly greater improvement gain from pre- to post-test than the other.

    The way I'm looking at it is as follows: to compute a new variable in SPSS which is the difference from pre- to post-test and then run an ANOVA on this computed variable to assess between group differences in improvement. Is this the best approach or should I be analyzing the pre- and post-test scores directly?

  2. #2
    Points: 2,595, Level: 30
    Level completed: 97%, Points required for next Level: 5

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts
    perhaps you can use the difference scores and use t-test?

  3. #3
    Points: 3,124, Level: 34
    Level completed: 50%, Points required for next Level: 76

    Posts
    13
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Smile Do a Power study on pre and post test scores.

    Quote Originally Posted by Barnzilla View Post
    Perhaps this is an easy issue, but I wanted to throw it out anyway.

    I'm working on a dataset where there are two groups, both having pre- and post-test scores. I want to see if one group has a significantly greater improvement gain from pre- to post-test than the other.

    The way I'm looking at it is as follows: to compute a new variable in SPSS which is the difference from pre- to post-test and then run an ANOVA on this computed variable to assess between group differences in improvement. Is this the best approach or should I be analyzing the pre- and post-test scores directly?

    I suggest use the pre and post test scores and then do a hypothesis test on the POWER of the study.

  4. #4
    Points: 3,182, Level: 34
    Level completed: 88%, Points required for next Level: 18

    Posts
    27
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Quote Originally Posted by SVM View Post
    I suggest use the pre and post test scores and then do a hypothesis test on the POWER of the study.
    How would you do that? The power analysis that is. I am also interested in this...Thank you.

+ Reply to Thread

           




Similar Threads

  1. Comparing within groups and across conditions
    By sifry37 in forum Statistical Research
    Replies: 0
    Last Post: 03-14-2011, 05:35 PM
  2. comparing age groups
    By ferdinand in forum Statistics
    Replies: 0
    Last Post: 10-25-2010, 04:32 AM
  3. Comparing and ranking many groups
    By kanan in forum Applied Statistics
    Replies: 0
    Last Post: 08-06-2010, 02:35 PM
  4. Replies: 12
    Last Post: 06-17-2009, 10:29 AM
  5. Comparing groups - z-scores
    By Alsvartr in forum Statistics
    Replies: 2
    Last Post: 08-16-2006, 11:59 AM

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