More than Two Groups
and our alternative hypothesis is simply: Now, if we were to set about comparing the means5 with a t-test, a problem would arise. We can do only two at a…
BIOCHEMISTRY
Elements of Statistical Inference
So, the little squiggles aren’t all that mysterious; most stand for the same quantity in the sample and the population. Sample means begin with M, population means begin with Greek…
Describing the Data with Numbers
. When speaking to another statistician, we can say either the “mean” or the “X bar.”2 In recent years, we’ve seen a new abbreviation for the mean: M. This is…
Factorial ANOVA
Algebraically: where j is the subscript for the columns (brand), I is the number of rows (in this case, 2), and n is the sample size in each cell (in…
Simple Regression and Correlation
ANDCORRELATION CHAPTER THE THIRTEENTH Simple Regression and Correlation The previous section dealt with ANOVA methods, which are suitable when the independent variable is nominal categories and the dependent variable approximates…
Measures of Impact
CHAPTER THE TWENTY-SIXTH Measures of Impact This chapter introduces you to measures of “how much” of something—how many people have a disorder at any given time, how much greater a…
Multiple Regression
This is just longer than what we had before, not fundamentally different. A reasonable next step would be to graph the data. However, no one has yet come up with…
Measures of Association for Categorical Data
The Phi Coefficient, Contingency Coefficient, Yule’s Q and Cramer’s V The most obvious approach is to pretend that the data are actually interval and go ahead and calculate a Pearson…
Logistic and Poisson Regression
We can write a linear regression equation like this, but we shouldn’t. First, probabilities don’t go in a straight line forever; they are bounded by zero and one. Second, there’s…
Looking at the Data
FIGURE 2-1 Bar chart of the five least popular courses FIGURE 2-2 Figure 2-1 redrawn so that the categories are in order of preference and the tick marks are outside…
