Errata for Interpreting and Visualizing Regression Models Using Stata

The errata for the Interpreting and Visualizing Regression Models Using Stata book are provided below. Click here for an explanation of how to read an erratum. Click here to learn how to determine the printing number of a book.

(1), (2) Chapter 2, p. 39, regress example
In a recent Stata update, a tolerance was changed that impacts whether a variable is dropped because of collinearity. An updated version of Stata drops the cubic term from the model that is illustrated on page 39. To avoid this, use yrborn40 instead of yrborn, yielding

. regress educ c.yrborn40##c.yrborn40##c.yrborn40

To make this model consistent with those shown on page 38, also change yrborn to yrborn40 for the models shown on page 38. The surrounding text remains correct, including the references to the R-squared values.
(1) Chapter 3, p. 54, last paragraph
We first need to run the marginsplot command to compute the adjusted means as a function of age (shown below). We first need to run the margins command to compute the adjusted means as a function of age (shown below).
(1) Chapter 4, p. 89, first line of second paragraph
The variable ed2 contains 0 for 12 or fewer ... The variable ed2m contains 0 for 12 or fewer ...
(1) Chapter 7, p. 193, second paragraph
We can specify p.dosegrp on the ... We can specify q.dosegrp on the ...
(1) Chapter 7, p. 198, third and fourth displayed equations
\[ H_0\!:\; (1/2)*\mu_{1} + (1/2)*\mu_{4} + -(1/2)*\mu_{2} + -(1/2)*\mu_{3}(1/2)= 0\\ H_0\!:\; (1/2)*\mu_{1} + -(1/2)*\mu_{2} + -(1/2)*\mu_{3} + (1/2)*\mu_{4}(1/2)= 0 \] \[ H_0\!:\; (1/2)*\mu_{1} + (1/2)*\mu_{4} + -(1/2)*\mu_{2} + -(1/2)*\mu_{3}= 0 \\ H_0\!:\; (1/2)*\mu_{1} + -(1/2)*\mu_{2} + -(1/2)*\mu_{3} + (1/2)*\mu_{4}= 0 \]
(1) Chapter 8, p. 216, last sentence of first paragraph
Among those who are nondepressed, ... Among those who are depressed, ...
(1) Chapter 10, p. 299, paragraph below figure
Likewise, the difference in the adjusted means comparing group 3 versus 1 are also all significant (see right panel of figure 10.12). Likewise, the difference in the adjusted means comparing group 3 versus 2 is also all significant (see right panel of figure 10.12).
(1) Chapter 18, p. 481, figure caption
Figure 18.8. Probability of being very unhappy by class and education Figure 18.8. Predicted number of children by education