Examples and Their Data Sets
David C. Howell
Over the years I have assembled additional examples that I use in
my classes or assign to students. I'm in the process of collecting those here. Each set
will eventually have at least a brief description of the study behind the data, a
reference, a description of the data file, and the data themselves. I encourage you to
download these files and play with them. Just remember where they came from. Additional
examples are found in the text and in the "Lecture and Lab ideas" link.
Generating Data with a Specified Pattern of Correlations
- I often see queries on how to generate a set of data with a specified intercorrelation
matrix. The best answer to this question was provided by David nichols at SPSS. An SPSS
program for generating data is at CorrGen2.html,
and it should not be difficult to modify it for other software, especially if you don't
want an exact pattern of correlations.
If you only want two variables with a particular
correlation between them, then the preceding program is overkill. A simpler program is
available at CorrGen.html
Religious Fundamentalism
- This is an example based on Sethi and Seligman (1994) on the relationship between the
level of religious fundamentalism and personal optimism. It illustrates a one-way analysis
of variance, the Bonferroni multiple comparison procedure, the averaging of correlation
coefficients, and multiple regression.
Testosterone and Antisocial Behavior
- These are data from Dabbs and Morris (1991) on the relationship between antisocial
behavior and testosterone levels for 4462 adult males. This presents an interesting
example of chi-square and the possibility of loglinear models.
Reading Comprehension and the SAT
- These are data from Katz, et al. (1990) on the relationship between a reading
comprehension items and SAT scores when the student does not see the passage on which the
reading comprehension items are based. The illustrate the use of correlation/regression
and (weakly) the t test. This example is used extensively in the 4th edition
of the Fundamentals book.
Explanatory Style and Athletic Performance
- These are data from Seligman, et al. (1990) on the relationship between Explanatory
Style (Optimism-Pessimism) and performance following failure in an athletic event
(Swimming). It illustrates an independent group t test and simple correlation
and regression.
Treatment of Posttraumatic Stress Disorder--Anova Example
- These are data from Foa et al, 1991 on the treatment of Posttraumatic Stress Disorder in
Rape Victims. It illustrates a one-way analysis of variance, multiple comparison
procedures, and eta-squared and omega-squared.
Return to
Dave Howell's Statistical Home Page
University of Vermont Home Page
Last revised 6/25/98