-- In Part IV (Data analysis: Basics) and Part V (Data analysis: Techniques)
   by the term `cohort study', `LONGITUDINAL cohort study' is always 
   intended. In such studies the same sample of units is followed over a 
   long period of time. 

-- There is a serious typo on page 349 (Chapter 14, Initial steps in data
   analysis):   
   In section 14.3.1, `circumambulations' should read: CROSSTABULATIONS

-- In chapter 17 (Regression analysis and beyond):
   on page 426 the definition of part correlation is not completely
   correct, it should read: 

   Part correlation. Correlation between the dependent variable and the
   residual of a predictor when it is regressed on the other predictors
   in the regression model.

-- Figure 17.12 on page 452 (Analysis techniques for interobserver agreement)
   has been adapted and now also includes the intraclass correlation
   coefficient.