Significance Testing Principles

  • A cartoon to be used for discussing the nature of conclusions for a significance test. The cartoon was used in the November 2016 CAUSE Cartoon Caption Contest. The winning caption was submitted by Andrea Boito from Penn State University, Altoona, while the drawing was created by John Landers using an idea from Dennis Pearl. Two honorable mentions that rose to the top of the judging in the November competition included a repackaging of the classic refrain "If you torture data enough it will confess," written by Caleb Ohrn, a student at Akron University and "Did you check to see if the conditions were met? Ignore them at your own peril!" written by an anonymous author.
    View Resource
    0
    No votes yet
  • A cartoon to be used for discussing statistical hypothesis testing and the effect of outliers. The cartoon was used in the December 2016 CAUSE Cartoon Caption Contest. The winning caption was submitted by Robert Garrett, a student at Miami University, while the drawing was created by John Landers using an idea from Dennis Pearl. A second winning caption "The sadistic ANOVA problem made most students feel headed for an F test," written by Larry Lesser from University of Texas at El Paso is well-suited to stimulate a discussion of the F test in ANOVA and about general student anxiety about statistics (see "Cartoon: The Exam II")
    View Resource
    0
    No votes yet
  • A cartoon to be used for discussing the F test in ANOVA and for discussing general student anxiety about statistics. The cartoon was used in the December 2016 CAUSE Cartoon Caption Contest. The winning caption was submitted by Larry Lesser at The University of Texas at El Paso, while the drawing was created by John Landers using an idea from Dennis Pearl. A second winning caption "Mark was pleased to note that he was a significant outlier. Little did he know it was a two-sided test..." written by Robert Garrett, a student at Miami University is well-suited to stimulate a discussion of statistical hypothesis testing and the effect of outliers (see "Cartoon: The Exam I")
    View Resource
    0
    No votes yet
  • A cartoon to help students learn not to "accept" the null hypothesis. The cartoon was drawn by British cartoonist John Landers (www.landers.co.uk) based on an idea in 2016 from Dennis Pearl from Penn State University.
    View Resource
    4
    Average: 4 (1 vote)
  • A cartoon to illustrate the idea that the standard 5% significance level is an arbitrary cut-off. The cartoon was used in the February, 2017 CAUSE cartoon caption contest and the winning caption was submitted by Mickey Dunlap from University of Georgia. The cartoon was drawn by British cartoonist John Landers (www.landers.co.uk) based on an idea from Dennis Pearl from Penn State University.
    View Resource
    0
    No votes yet
  • A 5-panel gif animation than can be used in discussing setting up null and alternative hypotheses and the concept of a type I error. The idea for the animation was provided by Dr. Karen Banks from Indiana University and received second place in the 2015 A-mu-sing contest. The idea for the final panel of the gif (regarding Type I error) came from the participants at a breakout session on the use of cartoons and songs in teaching statistics at the 2015 U.S. Conference On Teaching Statistics. The cartoons are drawn by British cartoonist John Landers (first of two animations arising from the USCOTS session).
    View Resource
    0
    No votes yet
  • A 5-panel gif animation than can be used in discussing setting up null and alternative hypotheses and the concept of a type I error. The idea for the animation was provided by Dr. Karen Banks from Indiana University and received second place in the 2015 A-mu-sing contest. The idea for the final panel of the gif (regarding Type I error) came from the participants at a breakout session on the use of cartoons and songs in teaching statistics at the 2015 U.S. Conference On Teaching Statistics. The cartoons are drawn by British cartoonist John Landers (second of two animations arising from the USCOTS session).
    View Resource
    0
    No votes yet
  • A song about the various cautions that go with interpreting the P-value especially the large sample caution, a strict reliance on the 5% significance level, and errors in interpreting results as proof of a hypothesis. The song and musical arrangements in the video were written by Michael Greenacre of Universitat Pompeu Fabra, Barcelona and may be sung to the tune of Irving Berlin's "There's no business like show business." The lyrics were sung by Gurdeep Stephens in the recording. The video took second place in the 2015 A-mu-sing contest.
    View Resource
    0
    No votes yet
  • A joke for use with discussions about the relationship between sample size and power or in discussing the large sample caution in significance testing.
    View Resource
    0
    No votes yet
  • "Scaffolding" is a poem by Scottish poet Eveline Pye from Glasgow Caledonin University. The poem was originally published in the September 2011 issue of the bimonthly magazine Significance, in an article about Eveline Pye's statistical poetry. "Scaffolding" might be used in course discussions of the importance of checking assumptions in the application of statistical methods or of the value of statistical sleuthing in discovering hidden relationships.
    View Resource
    5
    Average: 5 (1 vote)

Pages