SPSS tests the proportional odds assumption, which is commonly referred to as the test of parallel lines because the null hypothesis states that the slope coefficients in the model are the same across response categories (and lines of the same slope are parallel). Since the ordered logit model estimates one equation over all levels of the response variable (as compared to the multinomial logit model), the test for proportional odds tests whether our one-equation model is valid. If we were to reject the null hypothesis based on the significance of the Chi-Square statistic, we would conclude that ordered logit coefficients are not equal across the levels of the outcome, and we would fit a less restrictive model (i.e., multinomial logit model). If we fail to reject the null hypothesis, we conclude that the assumption holds
PLUM
chist BY numcred othnstal housng WITH age duration
/LOCATION = numcred age duration
/CRITERIA = CIN(95) DELTA(0) LCONVERGE(0) MXITER(100) MXSTEP(5)
PCONVERGE(0)
/LINK = LOGIT
/PRINT = FIT TPARALLEL.
• chist is the dependent variable, numcred, othnstal, and housing are factors, and age and
duration are covariates.
• The location model is based on numcred, age, and duration. Note, however, that goodnessof-
fit statistics will be based on all of the factors and covariates on the variable list.
• CRITERIA specifies that the confidence level to use is 95, no delta value should be added
to cells with observed zero frequency, and neither the log-likelihood nor parameter estimates
convergence criteria should be used. This means that the procedure will stop when
either 100 iterations or 5 step-halving operations have been performed.
• LINK specifies that LOGIT function should be used.
• PRINT specifies the Pearson chi-square and the likelihood-ratio
chi-square statistics which are computed based on the classification
specified on the variable list.The Print subcomand also specified chi-squared score test of the parallellines assumption are tested.
发帖
雷达卡
京公网安备 11010802022788号
