What If Conditions Aren't Satisfied?
The pipeline data table contains a large number of observations so that we could use the t-test despite the indications of a very skewed population. What are our options if we don't meet the conditions for the t-test?
If the population is reasonably symmetric, but normality is questionable, we could use the Wilcoxon Signed Rank test. This is an option presented in the Test Means command, as illustrated in Figure 9.4. Wilcoxon's test is identified as a non-parametric test, which just indicates that its reliability does not depend on the parent population following any specific probability distribution. Rather than testing the mean, the test statistic computes the difference between each observation and the ...
Get Practical Data Analysis with JMP® now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
