Scribd Logo
Upload

Welcome to Scribd!

  • 99 views

    Ordinal Regression

    Uploaded by

    Vira Mulyawati

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Ordinal Regression

    Uploaded by

    Vira Mulyawati
    99 views4 pages

    Original Title

    ORDINAL REGRESSION

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    99 views4 pages

    Ordinal Regression

    Uploaded by

    Vira Mulyawati

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 4

    ORDINAL REGRESSION

    Author: Basile Marquier Date Amended: 15/04/2019

    The following resources are associated:


    SPSS dataset Graduate.sav’

    Ordinal regression in SPSS


    Dependent (outcome) variable: ordinal
    Independent (explanatory) variables: Continuous (scale) and/or Categorical
    Common Applications: Regression is used to (a) look for significant relationships
    between two variables or (b) predict a value of one variable for a given value of the other.
    Data: The data set ‘Graduate.sav’ looks at factors that influence the decision of whether to
    apply to graduate school. College juniors are asked if they are unlikely, somewhat likely, or
    very likely to apply to graduate school. Hence, our outcome variable has three categories.

    College juniors are asked if they are unlikely, somewhat likely, or very likely to apply to
    graduate school. Hence, our outcome variable has three categories. Data on parental
    educational status, whether the undergraduate institution is public or private, and current
    GPA is also collected.

    Assumptions for Ordinal regression


    Assumptions How to check
    Proportional Odds Test of parallel lines

    Steps in SPSS
    Analyze  Regression  Ordinal

    Move ‘Decision to apply’ to the Dependent box. The categorical independent variables
    ‘Education of parents’ and ‘Private or Public institution’ should be moved to the Factor(s)
    box. The scale independent variable GPA score should be moved to Covariate(s) box.

    We will keep the default options:

    © Basile Marquier
    University of Sheffield
    Ordinal regression in SPSS

    We will tick “Test of parallel lines” as it will check the assumption of proportional odds.
    Ordinal regression in SPSS

    Output

    Model Fitting Information


    -2 Log
    Model Likelihood Chi-Square df Sig.
    Intercept Only 557.272
    Final 533.091 24.180 3 .000
    Link function: Logit.

    The p-value of less than 0.001 shows that the model is a very good finding on how well does the
    model fits the data.

    Goodness-of-Fit
    Chi-Square df Sig.
    Pearson 400.843 435 .878
    Deviance 400.749 435 .879
    Link function: Logit.

    For this table, we would like to fail to reject the null hypothesis in order for our model to be a good
    fit. There are two tests (Pearson and Deviance). Both reject the null hypothesis.

    Test of Parallel Linesa


    -2 Log
    Model Likelihood Chi-Square df Sig.
    Null Hypothesis 533.091
    General 529.077 4.014 3 .260
    The null hypothesis states that the location parameters (slope coefficients) are
    the same across response categories.
    a. Link function: Logit.

    This tests the assumption of proportional odds and we want it to be greater than 0.05. This is the
    case here (p-value = 0.26). The main assumption of the ordinal regression is checked.
    Ordinal regression in SPSS

    Parameter Estimates
    95% Confidence Interval
    Estimate Std. Error Wald df Sig. Lower Bound Upper Bound
    Threshold [apply = 1] 1.214 .939 1.671 1 .196 -.627 3.055
    [apply = 2] 3.310 .952 12.076 1 .001 1.443 5.177
    Location gpa .616 .263 5.499 1 .019 .101 1.130
    [pared=.00] -1.048 .268 15.231 1 .000 -1.574 -.522
    [pared=1.00] 0a . . 0 . . .
    [public=.00] .059 .289 .041 1 .839 -.507 .624
    [public=1.00] 0a . . 0 . . .
    Link function: Logit.
    a. This parameter is set to zero because it is redundant.

    The most interesting aspect is the locations. We have a statistically significant result for GPA score
    (p-value = 0.019). We also have a statistically significant result for variable “pared” (p-value <
    0.001). On the other hand, the fact that the undergraduate institution is public or private does not
    have any statistically significant impact on the dependent variable “apply”.

    The value of the GPA coefficient is positive (0.616), which suggests that as GPA increases the
    likelihood of applying to a graduate school will increase. The value of [pared=0] is negative (-
    1.048), which suggests that if you are in the category “pared=0”, i.e. no parents have a degree, you
    are less likely to apply to a graduate school.

    You might also like