The Ordinal Logistic Regression is used to perform logistic regression for ordinal response variable with one or more predictors. Ordinal variables are variables with 3 or more levels and these levels have a natural ordering.
1. Choose ProcessMA > Statistics > Regression > Ordinal Logistic Regression.
2. In Response, select the column containing the response values.
3. In Frequency (Optional), select the column containing frequency data.
4. In Model, select the columns containing variables to be included in the model.
5. In Factors (Optional), select the columns containing variables in the model that are factors.
6. In Link functions, select the link functions to fit the response model.
7. In Order of response values, enter the order of the response variable.
8. Click OK.
Note To select a column of data into a textbox, double-click on any of the column names shown in the list on the left of the dialog box while in the textbox.
Response: Numeric.
Frequency: Integer; Must be greater than zero; Must have the same number of data points as the variable.
Model: Text or Numeric; Must have the same number of data points as the variable.
Factors: Text or Numeric; Must have the same number of data points as the variable.
Order of response values: Text or Numeric; Must match the levels in the response variable; Must be separated by commas (,).
You want to understand the effect of amount of pesticides and humidity on the life span of mosquitoes. You have classified life span as 1 (< 2 day), 2 (2 to 4 days) and 3 (4 to 6 days).
1. Open worksheet ProcessMA > Tools > Data Files > Regression.xls.
2. Choose ProcessMA > Statistics > Regression > Ordinal Logistic Regression.
3. In Response, select M – Lifespan.
4. In Model, select N – Pesticide, O – Humidity.
5. In Factors (Optional), choose O – Humidity.
6. Click OK.
Interpretation
From the logistic regression table, you can obtain the coefficients for the regression model. Constant(1) and Constant(2) are estimated intercept for lifepan 1 and 2 respectively. The p-values for Humidity is 0.795 which indicates that there is insufficient evidence to conclude that humidity have an effect on mosquito population.