This Second Edition of The Tao of Statistics: A Path to Understanding (With No Math) provides a reader-friendly approach to statistics in plain English. Unlike other statistics books, this text explains what statistics mean and how they are used, rather than how to calculate them. The book walks readers through basic concepts as well as some of the most complex statistical models in use. The Second Edition adds coverage of big data to better address its impact on p-values and other key concepts; material on small data to show readers how to handle data with fewer data points than optimal; and other new topics like missing data and effect sizes. The book’s two characters (a high school principal and a director of public health) return in the revised edition, with their examples expanded and updated with reference to contemporary concerns in the fields of education and health.

# Both Sides Loaded—Canonical Covariance Analysis

### Both Sides Loaded—Canonical Covariance Analysis

• Multiple measures
• Multiple everything
• Related how?

Canonical covariance analysis relates multiple dependent variables to multiple independent variables all at once, rather like MANCOVA. Instead of looking for mean differences in scores, however, canonical covariance analysis looks at the correlations between all the dependent and all the independent variables. It produces the best weighting of each variable to produce the highest correlation between the two “sides” (dependent and independent variables). Some statisticians will spend their entire careers without needing (or wanting?) to run a canonical covariate analysis, except maybe when they were in graduate school. These models generally are big and filled with many assumptions. In fact, they are usually so loaded with variables and assumptions that there is little agreement on ...

locked icon