Resource Library

A cartoon to be used for discussing the selection of the best explanatory variable in a regression model. The cartoon was used in the March 2017 CAUSE Cartoon Caption Contest. The winning caption was submitted by Michael Posner, from Villanova University. The drawing was created by British cartoonist John Landers based on an idea from Dennis Pearl of Penn State University. A second winning entry, by Michele Balik-Meisner, a student at North Carolina State University, may be found at www.causeweb.org/cause/resources/fun/cartoons/variable-wheel-i Three honorable mentions that rose to the top of the judging in the March competition included "No no no! You randomize AFTER you select your research topic!" by Mickey Dunlap from University of Georgia; "This isn't what I meant by random variable!" by Larry Lesser from The University of Texas at El Paso; and "We find this method of finding 'significant' predictors to be quicker than using stepwise regression and it is even slightly more reproducible." by Greg Snow from Brigham Young University.

This case study starts by the simple comparison of the prices of houses with and without fireplaces and extends the analysis to examine other characteristics of the houses with fireplace that may affect the price as well. The intent is to show the danger of using simple group comparisons to answer a question that involves many variables. The lesson shows the R code for doing this analysis; however, the data and the model could be used with another statistical software.

This paper comes from researchers at the NASA Langley Research Center and College of William & Mary.  

"The experience of retinex image processing has prompted us to reconsider fundamental aspects of imaging and image processing. Foremost is the idea that a good visual representation requires a non-linear transformation of the recorded (approximately linear) image data. Further, this transformation appears to converge on a specific distribution. Here we investigate the connection between numerical and visual phenomena. Specifically the questions explored are: (1) Is there a well-defined consistent statistical character associated with good visual representations? (2) Does there exist an ideal visual image? And (3) what are its statistical properties?"

This resource was prepared to give the practicing engineer a clear understanding of probability and statistics with special consideration to problems frequently encountered in aerospace engineering. It is conceived to be both a desktop reference and a refresher for aerospace engineers in government and industry. It could also be used as a supplement to standard texts for in-house training courses on the subject. 

This page calculates the Poisson distribution that most closely fits an observed frequency distribution, as determined by the method of least squares. Users enter observed frequencies, and the page returns the fitted Poisson frequencies, the mean and variance of the observed distribution and the fitted Poisson distribution, and R-squared.

This page will calculate the lower and upper limits of the 95% confidence interval for a proportion, according to two methods described by Robert Newcombe, both derived from a procedure outlined by E. B. Wilson in 1927. The first method uses the Wilson procedure without a correction for continuity; the second uses the Wilson procedure with a correction for continuity.

Calculates the z-ratio and associated one-tail and two-tail probabilities for the difference between two correlated proportions, such as might be found in the case where the proportions are based on the same sample of subjects or on matched samples.

To assess the significance of any particular instance of r, enter the values of N[>6] and r into the designated cells, then click the 'Calculate' button. Application of this formula to any particular observed sample value of r will accordingly test the null hypothesis that the observed value comes from a population in which rho=0.

This page will perform a t-test for the significance of the difference between the observed mean of a sample and a hypothetical mean of the population from which the sample is randomly drawn. The user will be asked to specify the sample size as the page opens.

This calculator performs the following for a contingency table up to 5x5: chi-square analysis; Cramer's V; two asymmetrical versions of lambda; the Goodman-Kruskal index of predictive association; other measures relevant to categorical prediction. Key Word: Categorical Analysis.