Resource Library

This activity focuses on basic ideas of linear regression. It covers creating scatterplots from data, describing the association between two variables, and correlation as a measure of linear association. After this activity students will have the knowledge to create output that yields R-square, the slope and intercept, as well as their interpretations. This activity also covers some of the basics about residual analysis and the fit of the linear regression model in certain settings. The corresponding data set for this activity, 'BAC data', can be found at the following web address: http://www.causeweb.org/repository/ACT/BAC.txt

If you can't measure it, I'm not interested. A quote by Canadian educator and management theorist Laurence Johnston Peter (1919 - 1990) from "Peter's People" in "Human Behavior" (August, 1976; page 9). The quote also appears in "Statistically Speaking: A dictionary of quotations" compiled by Carl Gaither and Alma Cavazos-Gaither.

This joke can be used to motivate class discussions on the assumptions underlying drawing conclusions from data (especially the assumption of stationarity). The joke is a revision of a story in "The Angel's Dictionary: A modern tribute to Ambrose Bierce" by Edmund Volkart - also quoted in "Statistically Speaking: A dictionary of Quotations" by Carl Gaither and Alma Cavazos-Gaither (page 62). The revision (to make the story suitable for classroom use) was written by Dennis Pearl, The Ohio State University.

A cartoon that might be used in teaching about data quality issues. Cartoon by John Landers (www.landers.co.uk) based on an idea from Dennis Pearl (The Ohio State University). Free to use in the classroom and on course web sites.

A cartoon to teach about the concept of independent versus dependent variables. Cartoon by John Landers (www.landers.co.uk) based on an idea from Dennis Pearl (The Ohio State University). Free to use in the classroom and on course web sites.

January 26, 2010 webinar presented by Alicia Gram, Smith College, and hosted by Leigh Slauson, Capital University. This webinar describes an activity that uses data collected from an experiment looking at the relationship between two categorical variables: whether a cotton plant was exposed to spider mites; and did the plant contract Wilt disease? The activity uses randomization to explore whether there is a difference between the occurrence of the disease with and without the mites. The webinar includes a discussion of the learning goals of the activity, followed by an implementation of the activity then suggestions for assessment. The implementation first uses a physical simulation, then a simulation using technology. (Extra materials, including Fathom instructions for the simulation, available for download free of charge).

This presentation is a part of a series of lessons on the Analysis of Categorical Data. This lecture covers the following: Mantel-Haenszel estimator of common odds ratio, confounding in logistic regression, univariate/multivariate analysis, bias vs. variance, and simulations.

This course covers methodology, major software tools and applications in data mining. By introducing principal ideas in statistical learning, the course will help students to understand conceptual underpinnings of methods in data mining. It focuses more on usage of existing software packages (mainly in R) than developing the algorithms by the students. The topics include statistical learning; resampling methods; linear regression; variable selection; regression shrinkage; dimension reduction; non-linear methods; logistic regression, discriminant analysis; nearest-neighbors; decision trees; bagging; boosting; support vector machines; principal components analysis; clustering. Perfect for students and teachers wanting to learn/acquire materials for this topic.

The emphasis in this course will be understanding statistical testing and estimation in the context of "omics" data so that you can appropriately design and analyze a high-throughput study. Since the measurement technologies are evolving rapidly, important objectives of the course are for students to gain a basic understanding of statistical principles and familiarity with flexible software tools so that you can continue to assess and use new statistical methodology as it is developed for new types of data.

By the end of the course, you should be able to tailor the analysis of your data to your needs while maintaining statistical validity.  You should come out of the course with insight so that you can assess the validity of new statistical methodologies as they are introduced as well as understand appropriate statistical analyses for data types not discussed in the class. 

Perfect for students and teachers wanting to learn/acquire materials for this topic.

This is a graduate level survey course that stresses the concepts of statistical design and analysis in biomedical research, with special emphasis on clinical trials. Perfect for students and teachers wanting to learn/acquire materials for this topic.