+ Reply to Thread
Results 1 to 7 of 7

Thread: What is the difference between a factor and a covariate for multinomial logistic reg?

  1. #1

    What is the difference between a factor and a covariate for multinomial logistic reg?




    Hi,

    I am trying to run a multinomial logisitic regression and I am at the dialog box in SPSS 19 where it asks me to input my dependent, my factors, and my covariates. I have been looking everywhere for a good definition to indicate which of my variables are my factors and which are my covariates, but I am confused. here are examples of what I found:

    Multinomial Logistic Regression: This is used to determine factors that affect the presence or absence of a characteristic when the dependent variable has three or more levels. In the dialog box, you select one dependent variable and your independent variables, which may be factors or covariates. The dialog box has the following submenus:
    and this which is somewhat clearer

    In SPSS, non-metric independent variables are included as “factors.” SPSS will dummy-code non-metric IVs.

    In SPSS, metric independent variables are included as “covariates.” If an independent variable is ordinal, we will attach the usual caution.
    Further research lead to this:
    Metric data is any reading which is atleast at an interval scale. As opposed to Non Metric data which can be nominal or ordinal.

    Ex: weight, height, distance, revenue, cost etc., all of them are interval scales or above. Hence they are metric data.

    On the other hands, satisfaction ratings, Yes/No responses, Male/Female readings etc., are Non Metric Data.

    Read more: http://wiki.answers.com/Q/What_is_de...#ixzz1f8aQXOlA
    Here is what my data breakdown looks like:
    [AgeRange=0]
    [AgeRange=1]
    [AgeRange=2]
    [AgeRange=3]
    [AgeRange=4]
    [AgeRange=5]
    [Income=1]
    [Income=2]
    [Income=3]
    [Income=4]
    [Income=5]
    [Income=6]
    [OwnEducation=1]
    [OwnEducation=2]
    [OwnEducation=3]
    [OwnEducation=4]
    [Gender=1]
    [Gender=2]
    [MontherEducation=1]
    [MontherEducation=2]

    So just let me verify (because by explaining it here I think I understand it better): Are all of my 5 variables ordinal? The age, income, and educations are all lists (i.e: 18-24, 25-34, etc). If so they are all Factors and I have no covariates, right?

    Thanks. I'm also posting this to help future searchers with this dilemma.

    Edit: If this is in the wrong sub-forum please feel free to move it

  2. #2
    Point Mass at Zero
    Points: 7,788, Level: 59
    Level completed: 19%, Points required for next Level: 162
    ledzep's Avatar
    Location
    Berks,UK
    Posts
    650
    Thanks
    174
    Thanked 133 Times in 131 Posts

    Re: What is the difference between a factor and a covariate for multinomial logistic

    Ordinal Variables are when there is a natural ordering to the data [ hence ordinal]. E.g. Income as Low, Mid, High is a ordinal variable.

    If your variables have such orderings then they will be Ordinal variables.

    I guess, Gender won't have such order. It is just a factor or categorical variable.

    If your age is continuous (rather than age brackets) then it would be a covariate. If your age is given as age-brackets, then it wont be covariate.
    Oh Thou Perelman! Poincare's was for you and Riemann's is for me.

  3. #3
    R must die
    Points: 24,318, Level: 94
    Level completed: 97%, Points required for next Level: 32
    Awards:
    Activity Award
    noetsi's Avatar
    Posts
    4,448
    Thanks
    267
    Thanked 706 Times in 678 Posts

    Re: What is the difference between a factor and a covariate for multinomial logistic

    My guess, given other experiences with SPSS, is that factors are interval variables and covariates are categorical ones. SPSS often makes that distinction in its analysis.
    "Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise." John Tukey

  4. #4
    R must die
    Points: 24,318, Level: 94
    Level completed: 97%, Points required for next Level: 32
    Awards:
    Activity Award
    noetsi's Avatar
    Posts
    4,448
    Thanks
    267
    Thanked 706 Times in 678 Posts

    Re: What is the difference between a factor and a covariate for multinomial logistic

    Assuming metric means interval I got it exactly reversed...

    In SPSS, non-metric independent variables are included as “factors.” SPSS will dummy-code non-metric IVs.

    In SPSS, metric independent variables are included as “covariates.” If an independent variable is ordinal, we will attach the usual caution.
    "Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise." John Tukey

  5. #5
    Points: 647, Level: 12
    Level completed: 94%, Points required for next Level: 3

    Posts
    7
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: What is the difference between a factor and a covariate for multinomial logistic

    okay, so I seem to have 4 ordinal and one categorical (gender)... but all 5 are to be inputted as factors. correct?

  6. #6
    R must die
    Points: 24,318, Level: 94
    Level completed: 97%, Points required for next Level: 32
    Awards:
    Activity Award
    noetsi's Avatar
    Posts
    4,448
    Thanks
    267
    Thanked 706 Times in 678 Posts

    Re: What is the difference between a factor and a covariate for multinomial logistic

    If you consider ordinal variables to be categorical in nature. Many consider them to be interval (covariates apparently in SPSS). If you have ordinal variables with a lot of distinct levels you will end up with a lot of dummy variables.

    How many levels does each ordinal variable have?
    "Far better an approximate answer to the right question, which is often vague, than an exact answer to the wrong question, which can always be made precise." John Tukey

  7. #7
    Points: 647, Level: 12
    Level completed: 94%, Points required for next Level: 3

    Posts
    7
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: What is the difference between a factor and a covariate for multinomial logistic


    Hi noetsi,

    They are as follows:
    Are you male or female?

    Male
    Female

    In what age range do you fall?

    18-24
    25-34
    35-44
    45-54
    55-64
    65 or Above

    What range best describes your yearly income?

    $0 - $25,000
    $25,001 - $50,000
    $50,001 - $75,000
    $75,001 - $100,000
    $100,001 - $150,000
    $150,000 +

    What level of education have you completed?

    High School or Equivalent
    College
    Undergraduate
    Graduate

    Does your mother have an undergraduate degree or higher?

    Yes
    No

+ 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