Resource Library

June 22, 2010 Activity webinar presented by Paul Roback, St. Olaf College and hosted by Leigh Slauson, Capital University. This webinar describes an in-class activity, motivated by Case Study 1.1.1 in The Statistical Sleuth, in which students compose haiku poems about statistics. Their poems are used to introduce two-sample t-tests and randomization tests. In addition, the in-class experiment leads to good discussion about experimental design issues, where students compare our design to the actual experiment described in Amabile et al.(1985) "Motivation and Creativity: Effects of Motivational Orientation on Creative Writers", Journal of Personality and Social Psychology 48(2): 393-399. I use this activity on the first day of our second course in applied statistics (Statistical Modeling), but it could easily be used in an introductory course as well. Examples of haiku poems which have resulted from this activity can be found under CAUSEweb > Resources > Fun > Poem (direct link), or at www.causeweb.org/cwis/SPT--FullRecord.php?ResourceId=1883.

You can't fix by analysis what you bungled by design. is a quote by American education researchers Richard J. Light, Judith D. Singer, and John B. Willett. The quote is found in the preface of their 1990 book "By Design: planning research on higher education".

This 6 minute 39 second video can be used to teach the difference between correlation and causation. For example, that a relationship between X and Y might be explained by X causing Y, Y causing X, or a third factor that drives them both. The video is episode #109 (Nov 10, 2009) in the Psych Files podcast series produced and hosted by Michael A. Britt, Ph.D. at www.thepsychfiles.com. Video is free to use in the classroom or on course websites under a non-commercial ShareAlike creative commons license.

A joke to teach the meaning of "significance" written in 2011 by University of Texas at El Paso professor of Mathematical Sciences, Lawrence Mark Lesser (1964-).

There is no such thing as no chance is a quote by American businessman and founder of the Ford Motor Company, Henry Ford (1863-1947). The quote is from a speech given to the 1930 class of students at Edison Laboratory in Menlo Park, New Jersey. It is referred to on page 14 of the 1930 book "American Florist".

A song lyric by Dennis Pearl of The Ohio State University written as a parody of the 1960 tune "Hit the Road Jack" by Percy Mayfield; made popular by Ray Charles in his 1961 recording. What to say in class before song: There are times when the mode may be preferred to the mean - especially if the concept of interest is tied to understanding the most likely situation. You might remember that Ray Charles used to sing a song about this... In a class where Bayesian and Maximum Likelihood methodology has been introduced you might add the following after the first sentence "For example when you assume a uniform non-informative prior for a parameter, then the m.l.e. coincides with the mode of the posterior distribution - and the mean of the posterior distribution may not be a good estimate." Tip for Teaching: The song takes up a bit too much class time for delivering its message. Thus, for in-class use, it is recommended to play only the first verse or three. Musical accompaniment realization and male vocals are by Joshua Lintz, female vocals are by Mariana Sandoval from University of Texas at El Paso.

This applet performs the Student's t test on two sets of data, and reports the average and variance for both sets of data, the t score, degrees of freedom, and one and two tailed P values.

This java applet can be used to determine whether or not the means in two sample populations are significantly different.

This applet demonstrates the concept of power. Users select the hypothesized mean, the alternative mean, the sample size, and the number of samples. The applet shows the hypothesized histogram and the alternative histogram. Users then select either the level of significance and set alpha or the rejection region and set the test statistic. The applet then shows the p-value (in red) and power (in green). User can also determine the direction of the test by clicking the inequality sign.

This page of Statistical Java describes 11 different probability distributions including the Binomial, Poisson, Negative Binomial, Geometric, T, Chi-squared, Gamma, Weibull, Log-Normal, Beta, and F. Each distribution has its own applet in which users can manipulate the parameters to see how the distribution changes. The parameters are described on the main page as well as situations that would use each distribution. The equations of the distributions are not given. To select between the different applets you can click on Statistical Theory, Probability Distributions and then the Main Page. At the bottom of this page you can make your applet selection. This page was formerly located at http://www.stat.vt.edu/~sundar/java/applets/