© John Wiley & Sons, Inc.

FIGURE 18-4: Typical output from a logistic regression model. The output on the left (a) shows statistical results output for the

model, and the output on the right (b) shows predicted probabilities for each individual that can be output as a data set.

Seeing summary information about the variables

At the top of Figure 18-4a, you can see information about the variables under Descriptives. It can

include means and standard deviations of predictors that are numerical variables, and a count of how

many individuals did or did not have the outcome event. In Figure 18-4a, you can see that 15 of the 30

individuals lived and 15 died.

Assessing the adequacy of the model

In Figure 18-4a, the middle section starting with Deviance and ending with AIC provides model fit

information. These are measures that indicate how well the fitted function represents the data, which is

called goodness-of-fit. Some have test statistics and an associated p value (see Chapter 3 for a

refresher on p values), while others produce metrics. Although there are different model fit statistics,

you will find that they usually agree on how well a model fits the data.

You may see the following model fit measures, depending on your software:

A p value associated with the decrease in deviance between the null model and the final

model: This information is shown in Figure 18-4a under Deviance. Under α = 0.05, if this p value

< 0.05, it indicates that adding the predictor variables to the null model statistically significantly

improves its ability to predict the outcome. In Figure 18-4a, p < 0.0001, which means that adding

radiation dose to the model makes it statistically significantly better at predicting an individual

animal’s chance of dying than the null model. However, it’s not very hard for a model with any

predictors to be better than the null model, so this is not a very sensitive model fit statistic.