+ Reply to Thread
Results 1 to 2 of 2

Thread: analysis over all items OR by-subject OR by-items analysis?

  1. #1
    Points: 738, Level: 14
    Level completed: 38%, Points required for next Level: 62

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    analysis over all items OR by-subject OR by-items analysis?




    Hi,

    Let's say you have a data set of for instance 20 subjects who all answered 15 questions. Every question is one items that is either correct or incorrect.
    Is it better to do a
    1. by-subjects analysis (calculate % correct for all participants and then use those participant percentages to do the analysis)
    2. by-item analysis (calculate for each item how many % of the participants answered it correctly and then use those item percentages to do the analysis)
    3. OR is it o.k. to just take all the items on one big pile and do the analysis on the whole set of items?
    I am pretty sure the last option is not the best option, but are there situations thinkable where this IS accepted, like when you have very few items or very few participants (or very few of both)?

    Thanks in advance for your help!

  2. #2
    Points: 20,006, Level: 89
    Level completed: 32%, Points required for next Level: 344

    Posts
    568
    Thanks
    50
    Thanked 20 Times in 19 Posts

    Re: analysis over all items OR by-subject OR by-items analysis?


    Quote Originally Posted by fruppel View Post
    Hi,

    Let's say you have a data set of for instance 20 subjects who all answered 15 questions. Every question is one items that is either correct or incorrect.
    Is it better to do a
    1. by-subjects analysis (calculate % correct for all participants and then use those participant percentages to do the analysis)
    2. by-item analysis (calculate for each item how many % of the participants answered it correctly and then use those item percentages to do the analysis)
    3. OR is it o.k. to just take all the items on one big pile and do the analysis on the whole set of items?
    I am pretty sure the last option is not the best option, but are there situations thinkable where this IS accepted, like when you have very few items or very few participants (or very few of both)?

    Thanks in advance for your help!
    It depends on the aim of your analysis. Are you interested in measuring subject performance on the test or are you interested in learning about the properties of the items that comprise the test? If the former, is it plausible to do a total or composite score?

+ Reply to Thread

           




Tags for this 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