A workbook/notes for an online course in beginning statistics can be found at
The main text for the course is
Statistics by Freedman, Pisani, and Purves.
We use the Data Program for analyzing
data, and the Box Model Program
for simulations.
Here are a couple of illustrations from the Box Model program.
The central limit theorem (histogram of sums)
To see the histogram of a sum of independent draws of tickets from a box (i.e., with replacement), enter
into the "Box" area some numbers for the tickets. For example, if you type
'0,0,0,0,0,0,1,2,3,4,5,6,10,10,10,' you create a box with six 0's, three 10's, and one
each
of 1, 2, 3, 4, 5, and 6. Press 'OK' to see the histogram of the tickets (one draw), as
well as the
histograms of the sum of two, three, four, and five draws.
Press "More draws" to see the histograms for higher sums. Are they getting more normal-curve
looking?
Confidence intervals
The example here uses the box with '1,2,3,4,5,6,' as if rolling a fair six-sided die. You can choose the number
of rolls ("# Draws per sample," whose default is 10), then press one of the "More sample" buttons to see
that number of confidence intervals. These are z-intervals. About 95% should have the
box average, 3.5, in them.
- Home: Stat.Istics.Net — Bringing statistics into the 20th century
- Data Program: Analyze data — Histograms, scatter plots, multiple regression, chi-square tests of independence, logistic regression.
- Box models: Randomly draws tickets from a box, to see the law of averages and the central limit theorem.
- Guessing Correlations: Match correlations with scatter plots.
- Monty Hall: Win a new car!
