In the Little Apps, datasets are under the “Stat Using Tech” item in the “source package” selector.
The Sugar
data frame is available in the Little Apps.
Each Little App can take a simple random sample from any of the datasets. Choose a large dataset–use “All” for the sample size–and say that for the purpose of demonstration, we’ll consider this to be the population. Then show the process of selecting a simple random sample.
When a categorical explanatory variable is selected, you can also do a stratified sample. Ask what different between a stratified sample and a simple random sample. (You can turn stratification on and off to see the difference. This will be most obvious when the explanatory variable has very different numbers of points at the different levels.)
NOTE: There’s a StatPREP activity on sampling bias, but it is written in terms of the confidence interval, which has not yet been introduced in this book.
The Class_survey
, Eyeglasses
, defects
, Tuition
, Insurance
, data frames are available in the Little Apps.
NOTE TO KATE: I don’t find the Density
data set in the format described in the book.
NOTE TO KATE: Figure 2.5 caption should be “Jitter plot”. And narrow the width and height of the jittering.
NOTE TO KATE: The area
data frame is in the package, but I can’t see how it is about the number of deaths due to carbon monoxide poisoning as described in table 2.1.4.
NOTE TO KATE: The histograms shown in section 2.2 maybe would be better rendered using datasets and variables from the data available in the Little Apps.
NOTE TO KATE: The example in section 2.3.2 contains an awful lot of NAs.
ONLY GOT AS FAR AS Table 2.3.5 in entering documentation for data.
SEE THE new F Little App.