ERRATA for Shipley, B. 2016.  Cause and correlation in biology: A user’s guide to path analysis, structural equations and causal inference with R.  Cambridge University Press.

 

Changes are shown in red.

​

Page 24, Box 2.1 should read:

Given a causal graph G, if X and Y are two different vertices in G, and W is a set of vertices in G that does not contain either X or Y, then X and Y are not d-separated given W in G if and only if there exists at least one undirected path U between X and Y such that (a) every collider is either in W or else has a descendant in W and (b) no other vertex on U is in W.

 

Page 24.  There is a step missing at the bottom of the page.  This missing step is:

(4) If every undirected path between X and Y is blocked, then X is d-separated from Y given W.

 

Page 25, Figure 2.6 is wrong as presented.  The correct Figure 2.6 is:

​

 

 

 

Page 26, third line, second column in Table 2.1, should read:

Since S1 is a causal descendant of U, conditioning on S1 activates U along path XàUßV

 

Page 168.  The lines describing how to calculate the variance and 95% confidence intervals for the Monte Carlo estimates of the true probability for the maximum likelihood chi-squared distribution should read:

 

… around p by referring to a normal distribution whose confidence interval (Manly 1997) around p by referring to a normal distribution whose mean is (x-Np) and whose variance is Np(1-p).  Thus, the 95 per cent confidence interval is

 

 

 

 

Page 198:  The syntax to name groups should be: group: NAME.  So, for instance, the code at the bottom of this page should be:

My.mod<-“

Group: Male

y~a*x

z~b1*y

group: Female

y~a*x + c*z

“

Similar corrections should be made in the code given on pages 213-214.