Statistical Home Page

David C. Howell


GreenBlueBar.gif GreenBlueBar.gif

Collections of Web Pages

Statistical Methods for Psychology


Book Cover This website is intended to support Statistical Methods in Psychology, 8th edition by David C. Howell. (If you are using the 6th or 7th edition, simply change the URLs by changing "methods8" to "methods6" or "methods7" in the address bar.) The website is intended to serve a number of functions. It will provide data that you can use for examples and exercises, answers to exercises (some of them), and other material that you may find useful. In addition, I have tried to incorporate important material on topics that I don't discuss in the text, and helpful hints about where to find useful information on the web. 

You can find the web site at Methods Book Link

Fundamental Statistics for the Behavioral Sciences


This website supports Fundamental Statistics for the Behavioral Sciences, 8th edition by David C. Howell, was published in 2013. These pages are intended to provide data that are referenced in examples and exercises, answers to exercises (some of them), a set of Java applets that illustrate concepts in the book, guides to the use of SPSS and two freely downloadable packages that do much of what SPSS will do. I also have other material that you may find useful. In addition, I have tried to incorporate important material on topics that I don't discuss in the text, and hints about where to find useful information on the web.

You can find the web site at Fundamentals Book Link

Additional Material on Many Topics


Over the years I have written many pages on a variety of statistical topics. The material referred to here is generally not discussed in depth in either of my books, but is material that is certainly useful. Some of this material is in draft form, and it may be revised from time to time. Please remember where you found it, and always give credit if you use it. I would appreciate suggestions for further additions and/or clarification.

A page listing each of the topics covered can be found at this link. S ee also the pages related to the Methods text under the heading of "Supplementary Material.

Programming with R


The programming language R has become a very common language for the analysis of statistical data. I have collected a great many programs and examples, some of which are concern learning R, some are programs for specific analyses, and some are examples to highlight points. I seem to have scattered these progrmas across several web sites, but I am attempting to bring them together under one roof.

You can find the web site at Programs in R.html

Randomization Tests and Bootstrapping


Most discussions of statistical tests concern parametric tests, where we attempt to draw conclusions about the parameters of the population(s) from which the data were draw. Another, very important, set of procedures are often known as "resampling procedures," which include randomization tests and bootstrapping. I cover those extensively in a large set of pages. The pages begin here.

Randomization tests differ from parametric tests in almost every respect.

  • There is no requirement that we have random samples from one or more populations—in fact we usually have not sampled randomly.
  • We rarely think in terms of the populations from which the data came, and there is no need to assume anything about normality or homoscedasticity.
  • Our null hypothesis has nothing to do with parameters, but is phrased rather vaguely, as, for example, the hypothesis that the treatment has no effect on performance.
  • Under the null hypothesis, the score that is associated with a participant is independent of the treatment that person received.
  • We do calculate some sort of test statistic, however we do not compare that statistic to tabled distributions.
  • Instead, we compare it to the results we obtain when we repeatedly randomize the data across the groups, and calculate the corresponding statistic for each randomization.
  • Even more than parametric tests, randomization tests emphasize the importance of random assignment of participants to treatments.

  • Useful Internet
    Sites


    There are many more important Internet sites related to statistical analyses than anyone can reasonably list. I have picked out some of my favorites, but an Internet search will find you many more. To find my favorite sites, click here. If you have any good ones to suggest, please let me know.