# Significance Level

• ### Cartoon: The Boyfriend

A cartoon suitable for a course website that makes use of a boxplot to display an outlier and also uses the term "statistically significant" in its punch line. The cartoon is number 539 (February, 2009) from the webcomic series at xkcd.com created by Randall Munroe. Free to use in the classroom and on course web sites under a creative commons attribution-non-commercial 2.5 license.

• ### Quote: Ramseyer on significance

Old statisticians never die they just become nonsignificant. Quote by Gary Ramseyer (1934 - 2012) listed as joke #56 on http://www.ilstu.edu/~gcramsey/Gallery.html Gary C. Ramseyer's First Internet Gallery of Statistics Jokes
• ### Cartoon: Meaningless Statistics

A cartoon that can be used in teaching about summary statistics. 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.
• ### Reading and Interpreting Tables and Graphs Involving Rates and Percentages

This survey assesses statistical literacy. The survey focuses on the general use of informal statistics in everyday situations: reading and interpreting tables and graphs involving rates and percentages.
• ### The Probability of Penalizing the Innocent Due to Bad Test Results

This is an example of "growing" a decision tree to analyze two possible outcomes. The tree's branches examine the two possible conditions of employee drug use with corresponding probabilities. This example looks at the final outcome probabilities of being correctly and incorrectly identified versus testing accuracy.
• ### Project #3 Hypothesis Testing Project

The assignment begins with creating a summary of and tables for the data, then walks the student through the steps of creating a hypothesis testing report. It uses the data set ncbirth200.xls, which is a random sample of 200 births from the data set ncbirth1450.xls.
• ### **Java Applets for Power and Sample Size

This collection of free, interactive Java applets provides a graphical interface for studying the power of the most commonly encountered experimental designs. Intended to be useful in planning statistical studies, these applets cover confidence intervals for means or proportions, one and two sample hypothesis tests for means or proportions, linear regression, balanced ANOVA designs, and tests of multiple correlation, Chi-square, and Poisson. Each applet opens in its own window with sliders, which are convertible to number-entry fields, for manipulating associated parameters. Controlling for the other parameters, users can change sample size, standard deviation, type I error (alpha) and effect size one at a time to see how each affects power. Conversely, users can manipulate the power for the test to determine the necessary sample size or margin of error. Additional features include a graph option by which the program plots a dependent variable (i.e. power) over a range of parameter values; the graph is automatically updated as the parameters are changed. Each dialog window also offers a Help menu which provides instructions for using the applet. The applets can be used over the Internet or downloaded onto the user's own computer.
• ### Hypothesis Testing for a Proportion and Small Samples

This site explains small sample hypothesis testing for a normal population and hypothesis testing for a population proportion. Includes examples and exercises.
• ### Analysis Tool: Significance of Difference Between Correlation Coefficients

Using the Fisher r-to-z transformation, this page will calculate a value of z that can be applied to assess the significance of the difference between two correlation coefficients, r_a and r_b, found in two independent samples. If r_a is greater than r_b, the resulting value of z will have a positive sign; if r_a is smaller than r_b, the sign of z will be negative.

• ### Analysis Tool: 2x2 Contingency Table (Version 1)

For a table of frequency data cross-classified according to two categorical variables, X and Y, each of which has two levels or subcategories, this page will calculate the Phi coefficient of association; perform a chi-square test of association, if the sample size is not too small; and perform the Fisher exact probability test, if the sample size is not too large. For intermediate values of n, the chi-square and Fisher tests will both be performed.