Fit Binary Logistic Model - Coding
main topic     see also 

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

Use to:

·    Specify how much change in the continuous predictors to display for the odds ratios

·    Specify the coding scheme for categorical predictors

·    Specify the reference level for each categorical predictor

·    Standardize the continuous predictors

Standardizing the continuous predictors can improve the interpretation of the model for specific conditions. You can standardize the continuous predictors using the following methods:

·    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.

·    Adjust the scale of the continuous predictors by dividing by the standard deviation: This method allows you to compare the size of the coefficients because they use a comparable scale. 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.

Dialog box items

Increments for odds ratios: Change the units for calculating the odds ratio from the units in the data for a continuous predictor. The coefficient for a variable when the model uses the logit link function represents the change in the log odds for an increase of one unit in the predictor. In some cases, a change of one unit in the data is too small to be meaningful. For example, if the predictor is mass in grams, a change of 1 gram can be too small to matter. Enter 1,000 to see the change in the odds ratio for a kilogram.

Continuous predictor: Shows all of the names of continuous predictors in your model. This column does not take any input.

Increment: Enter the amount of change for the continuous predictor.

Coding for Categorical Predictors: Use when you have categorical predictors. To perform the analysis, Minitab needs to recode the categorical data. Base your decision on whether you want to compare the levels to the overall mean or to the mean of a reference level.

(-1, 0, +1): Choose to estimate the difference between each level mean and the overall mean.

(1, 0): Choose to estimate the difference between each level mean and the reference level's mean.

Categorical predictor reference level: If you choose the (1, 0) coding scheme, the reference level  table becomes active in the dialog box. Minitab compares the means of the nonreference level(s) to the reference level. Changing the reference level does not affect the overall significance. But the change can make the results more meaningful. For information on why you might want to change the reference level, see Interpreting parameter estimates.

Categorical predictor: Shows all the names of categorical predictors in your model. This column does not take any input.

Reference level:  Choose a reference level for each categorical predictor.

Standardize continuous predictors: Use to control whether Minitab standardizes the continuous predictors. The standardized predictors are only used to fit the model and are not stored in the worksheet.

Do not standardize: Choose to use your original data for the continuous predictors.

Subtract the mean, then divide by the standard deviation: Choose to both center the predictors and to place them on a comparable scale.

Subtract the mean: Choose to center the predictors.

Divide by the standard deviation: Choose to use a comparable scale for all predictors.

Subtract a specified value, then divide by another: Choose to specify values rather than the mean and standard deviation estimates from the sample.

Continuous predictor: Shows all of the names of continuous predictors in your model. This column does not take any input.

Subtract: Type the value to subtract from each continuous predictor.

Divide by: Type the value that Minitab uses to divide the result of the subtraction.

Specify low and high levels to code as -1 and +1: Choose to transform the data linearly. All data values that fall between the Low and High values that you specify are coded to fall between -1 and +1. Designed experiments (DOE) use this scheme.

Continuous predictor: Shows all of the names of continuous predictors in your model. This column does not take any input.

Low: Type a value to code as -1. The default is the minimum value in the sample.

High: Type a value to code as +1. The default is the maximum value in the sample.