+ Reply to Thread
Results 1 to 7 of 7

Thread: Can anyone help me choose appropriate form(s) of analysis?

  1. #1
    Points: 111, Level: 2
    Level completed: 22%, Points required for next Level: 39

    Posts
    7
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Can anyone help me choose appropriate form(s) of analysis?




    Hi,

    I was wondering if anyone could possibly help advise me on choosing an appropriate form of analysis

    I have 2 groups of participants.
    Group 1: Comparison group drawn from the community
    Group 2: Hospital patients drawn from four hospital sites

    Each participant received a booklet which contained:

    1) A page collecting demographic information

    2) An 18 item questionnaire to which participants responded on a Likert-type rating scale (scored 1-
    5) on which a higher score represents negative thought processes (i.e. higher score is bad)

    3) ) An 14 item questionnaire to which participants responded on a Likert-type rating scale, scored 1-to-5) on which a higher score represents better mental health (i.e. higher score is better)

    4) A 59-item questionnaire, to which participants responded True or False to each item, and which consists of nine sub-scales – for which the participant obtains a score on each – and provides estimates of prevalence of different personality types.

    My main focus has been on the latter questionnaire and the nine estimates of prevalence which it provides per participant (i.e. percentages).

    I would like to supplement this with a statistical investigation of (for example):
    a) Do respondent demographics have any effect upon responses?
    b) Is there a difference between the two groups?
    c) Is there a difference between the groups across the four hospital sites?
    d) Is there any patterns in responding across the questionnaire measures?

    And so to my question, would I be right in thinking that a regression analysis (e.g. multiple regression) would be the most appropriate form of analysis to use? (apologies if this is cringingly simplistic, but I really am a statistics novice!)

    Thanks in advance for any help received!
    Kenny

  2. #2
    Points: 111, Level: 2
    Level completed: 22%, Points required for next Level: 39

    Posts
    7
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Can anyone help me choose appropriate form(s) of analysis?

    Yikes! 98 views and no responses - I must be way of in my choice of test

    Just incase its a case of TLDR here is the short version:

    Two groups completed three querstionnaires, and I would like to examine for differences between groups - What test could I use?

    Thanks

  3. #3
    TS Contributor
    Points: 17,779, Level: 84
    Level completed: 86%, Points required for next Level: 71
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,542
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Can anyone help me choose appropriate form(s) of analysis?

    What analysis you can perform depends on the
    sample size you have, to didn't mention it.

    Assuming that your sample size is large enough,
    you can consider t-tests, Analysis of Variance with
    several factors, Analysis of Covariance, or multiple
    linear regression (ANOVA/ANCOVA and linear
    regression are very closely related, so it's often just
    a matter of convenience which one to prefer).

    With kind regards

    K.

  4. #4
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts

    Re: Can anyone help me choose appropriate form(s) of analysis?

    Quote Originally Posted by Kenny View Post
    4) A 59-item questionnaire, to which participants responded True or False to each item, and which consists of nine sub-scales – for which the participant obtains a score on each – and provides estimates of prevalence of different personality types.
    Can you explain what you mean here by prevalence of different personality types? Prevalence has a conventional meaning in epidemiology, and I think you may mean something else here, so it's probably worth clarifying this first. Knowing something about the dependent variable will help guide the choice of analysis.

    a) Do respondent demographics have any effect upon responses?
    b) Is there a difference between the two groups?
    c) Is there a difference between the groups across the four hospital sites?
    d) Is there any patterns in responding across the questionnaire measures?
    It might be worth trying to specify these questions in more detail. E.g., when you say "a difference", what kind of difference are you thinking of? A difference in mean scores? When you say "patterns in responding", what kind of phenomena are you interested in in specific?

  5. #5
    Points: 111, Level: 2
    Level completed: 22%, Points required for next Level: 39

    Posts
    7
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Can anyone help me choose appropriate form(s) of analysis?

    Hi,

    Thanks for replying,

    My group sizes are:
    Group 1 = 70
    Group 2 = 152
    2a(Hospital site one)=30 (All male)
    2b(Hospital site two)=40 (All male)
    2c (Hospital site three)=60 (All male)
    2d(Hospital site four)=22 (All female)
    (numbers proportional to the size of the population of the hospital sites)

    When I said ‘prevalence’ I meant that that questionnaire identifies respondents as falling into particularly personally categories (e.g. 87% of respondents possessed personality traits A, 68% personality traits B etc. [Respondents can fall into multiple groupings]).

    In terms of difference, I guess differences in mean scores would be as good as any. I suppose, that my basic question is just to see if there is a difference between groups? When I referred to ‘pattern’ of responding I just meant in terms of, for example, did group 2a respond differently to 2d, and were 2 different to group 1? etc.

    Sorry, I’m probably not describing this too well!
    Last edited by Kenny; 10-22-2014 at 08:39 AM.

  6. #6
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts

    Re: Can anyone help me choose appropriate form(s) of analysis?

    Quote Originally Posted by Kenny View Post
    questionnaire identifies respondents as falling into particularly personally categories (e.g. 87% of respondents possessed personality traits A, 68% personality traits B etc. [Respondents can fall into multiple groupings]).
    So the questionnaire simply gives a binary indication (yes/no) of whether or not a participant has the given trait? And participants can 'test positive' for one, none, or multiple traits?

  7. #7
    Points: 111, Level: 2
    Level completed: 22%, Points required for next Level: 39

    Posts
    7
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Can anyone help me choose appropriate form(s) of analysis?


    The respondent responds true or false to each item, which is then scored 1 or 0 and assigned to one of nine subcategories (i.e. types of personality). Each of these subcategories are then summed to produce a total score of how much that individual exhibits that form of personality. Each participant thus generates nine scores (although I have also entered and retained the raw scores) indicative of how much they exhibit nine forms of personality.

+ 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