Specify the coding scheme for Fit Binary Logistic Model

Stat > Regression > Binary Logistic Regression > Fit Binary Logistic Model > Coding

Increments for odds ratios

For models that use the logit link function, Minitab calculates the odds ratio. For some predictors, the default odds ratio for a change of 1 unit in the predictor is not useful. For example, if the odds ratio for a change of 1 gram is too small, enter 1,000 to see the odds ratio for a change of 1 kilogram instead.
Continuous predictor
Shows all of the names of the continuous predictors in your model. This column does not take any input.
Increment
Enter the amount of change in the continuous predictor that Minitab uses to calculate the odds ratio.

Coding for categorical variables

Coding for categorical predictors
To perform the analysis, Minitab needs to recode the categorical predictors using one of two methods. Consider changing the method based on whether you want to compare the levels of the predictor to a baseline or to a reference level. The coding scheme does not change the test of the overall effect of the predictor. For more information, go to Coding schemes for categorical predictors.
  • (-1, 0, +1): Choose to estimate the difference between each level and a baseline.
  • (1, 0): Choose to estimate the difference between each level and a reference level. If you choose the (1, 0) coding scheme, the reference level table becomes active in the dialog box.
Reference level table
Categorical predictor
This column of the table shows all the names of the categorical predictors in your model. This column does not take any input.
Reference level

Minitab compares the non-reference levels to the reference level. Changing the reference level does not affect the overall significance, but it can make the coefficients and odds ratios more meaningful to interpret.

For example, a categorical predictor about whether customers have children has the levels "Yes" and "No." The response event is that a customer buys a cereal on a shopping trip.

The reference event is in the denominator of the odds ratio. When you change the reference level, the odds ratio inverts. When the reference level is "No," the odds ratio follows:
The odds ratio of 5 indicates that the customer is 5 times more likely to buy the cereal when the factor is "Yes" than when the factor is "No."
When the reference level is "Yes", the odds ratio follows:
The odds ratio of 0.2 indicates that the customer is 0.2 times as likely to buy the cereal when the factor is "No" as when the factor is "Yes."

When you change the reference level, the sign of the coefficient changes. When the reference level is "Yes," the coefficient is -1.6. The negative coefficient indicates that the customer is more likely to purchase the cereal at the reference level of the factor. When the reference level is "No," the sign of the coefficient changes to become 1.6. The positive coefficient indicates that the customer is less likely to purchase the cereal at the reference level of the factor.

Standardize continuous predictors

You can choose to standardize the continuous predictors in your model. The standardized predictors are only used to fit the model and are not stored in the worksheet.

Standardizing the continuous predictors can improve the interpretation of the model for specific conditions.
  1. Center the continuous predictors by subtracting the mean: This method helps reduce multicollinearity, which improves the precision of the coefficient estimates. This method is helpful when your model contains highly correlated predictors, higher-order terms, and interaction terms. Each coefficient represents the expected change in the response given a one unit change in the predictor, using the original measurement scale.
  2. Standardize the scale of the continuous predictors by dividing by the standard deviation: This method makes the ranges of the predictors more homogenous so that you can compare the size of the coefficients. This approach is helpful when you want to know which predictors have a larger effect, while controlling for differences in scale. However, each coefficient represents the expected change in the response given a change of one standard deviation in the predictor.
Use one of the following methods to standardize your continuous predictors:
  • Do not standardize: Use your original data for the continuous predictors.
  • Specify low and high levels to code as -1 and +1: Use to both center the predictors and to place them on a comparable scale. Minitab uses this method in design of experiments (DOE). All data values that fall between the low and high values that you specify are transformed to fall between −1 and +1. In the table, enter low and high values or use the default minimum and maximum values in the sample.
    Continuous predictor
    Shows the names of all the continuous predictors in your model. This column does not take any input.
    Low
    Enter a value to code as −1. The default is the minimum value in the sample.
    High
    Enter a value to code as +1. The default is the maximum value in the sample.
  • Subtract the mean, then divide by the standard deviation: Use to both center the predictors and to place them on a comparable scale.
  • Subtract the mean: Use to center the predictors.
  • Divide by the standard deviation: Use a comparable scale for all predictors.
  • Subtract a specified value, then divide by another: Specify other values rather than using the mean and standard deviation estimates from the sample.
    Continuous predictor
    Shows the names of all the continuous predictors in your model. This column does not take any input.
    Subtract
    Enter the value to subtract from each continuous predictor.
    Divide By
    Enter the value that Minitab uses to divide the result of the subtraction.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy