Hello,
I had a questionnaire conducted in which all participants had to answer four different questions in order to measure the level of satisfaction, motivation, collaboration, and help in seven different scenarios. The aim was to measure the dependent variables in the first scenario and to see how they evolved through the different scenarios. The dependent variables were measured on a 7 Likert scale ( from strongly disagree to strongly agree ). All participants answered my questionnaire, so it's a within-subject design.
I've created dummies for scenarios:
scenario 1, 2,3,4,5 6 and 7. And also for the gender variable ( female, male ) and for situation ( employed, unemployed, student, retired, other,self-empoyed).
First, I performed a Wilcoxon test. Then, I performed an ordinal regression. But I don't understand how I'm supposed to interpret the values in the "Estimate" column from te table "Parameter Estimate".
For example, the Wilcoxon tests showed that scenarios 2 and 3 had a significantly negative impact on satisfaction, whereas scenario 6 had a significantly positive impact. But here I see that for scenarios 2 and 3, the “Estimate” is positive, not negative? I don't understand.
When I run ordinal regression on spss, and select els dummies for scenario (except scenario 1 because I'm using it as a basis for comparison), I get this :
The table shows :
Scenario 2 = 0 Estimate :1.037 p:
The "Parameter Estimates" output in IBM SPSS for ordinal regression provides estimates of the effects of each predictor variable on the odds of moving from one category to the next in the ordinal response variable. Here's a breakdown of how to interpret the output:
Variable: The name of the predictor variable.
B: The estimated regression coefficient, which represents the change in the log-odds of moving from one category to the next for a one-unit change in the predictor variable.
Std. Error: The standard error of the regression coefficient, which indicates the variability of the estimate.
Wald Chi-Square: A statistical test of the significance of the regression coefficient. A significant value (p < 0.05) indicates that the predictor variable has a significant effect on the ordinal response variable.
df: Degrees of freedom for the Wald Chi-Square test.
p: The p-value associated with the Wald Chi-Square test.
Exp(B): The exponentiated regression coefficient, which represents the change in the odds of moving from one category to the next for a one-unit change in the predictor variable.
95% CI for Exp(B): The 95% confidence interval for the exponentiated regression coefficient.
For example, if the Exp(B) value is 1.5, it means that for every one-unit increase in the predictor variable, the odds of moving from one category to the next increase by a factor of 1.5.
By examining the Parameter Estimates output, you can determine which predictor variables have a significant effect on the ordinal response variable and the direction of those effects.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Control Theory
- Control Systems