+ Reply to Thread
Results 1 to 3 of 3

Thread: Can subject be a factor?

  1. #1
    Points: 506, Level: 10
    Level completed: 12%, Points required for next Level: 44

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Can subject be a factor?




    Let's say we have an epidemiology research with 5 subjects (for simplicity): John, Mike, Andy, Frank and Philip. They contacted meningitis and their fever was measured at day and night. One might be interested in checking whether some subjects have fever only at day to guide treatment or further research into the disease. So first one would need to make a two-way ANOVA with subject and daynight as factors. Doing so (I have so far tried with SPSS and SigmaStat) gives 0 degrees of freedom for the error term and in order to proceed with the ANOVA one has to forsake testing the interaction of subject with daynight. However, the test for the interaction is the most interesting test in this case. It is of little interest to test whether meningitis causes fever or whether day and night are different sorts of things. The main effects are trivial.

    Surely there must be a statistical way to test this, with a Tukey post-hoc test on Subject to see if the subjects fall into two groups based on such interaction (such that some have lower fever at night) and to know who these subjects are to help guide their treatment.

    How could one carry out an ANOVA with subject as the first factor and check the interaction with the second factor (daynight in this example)? Any thoughts?

  2. #2
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Can subject be a factor?

    Not following what you want to do. Can you just write your question out again in a single sentence.

    What is your dependent variable??
    Stop cowardice, ban guns!

  3. #3
    Points: 506, Level: 10
    Level completed: 12%, Points required for next Level: 44

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Can subject be a factor?


    The dependent variable is fever (body temperature). Independent variable, is daynight (two levels), and ideally subject should be a second independent variable with as many levels as subjects. Some subjects will have more fever than others - that is not terribly interesting. The question of interest is the interaction of subject with daynight. For example, can one identify the patients that only have fever during the day for a different course of treatment than for the patients that have similar fever both day and night? Or perhaps some patients will only have fever at night.

+ 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