Abstract
Parametric tests are designed for idealized data. In contrast, nonparametric tests are designed for real data: skewed, lumpy, having a few warts, outliers, and gaps scattered about. Nonparametric methods are workhorses of modern science, which should be part of every scientist's competence. Beyond that, they are very valuable for learning data literacy because they encourage the student to gain a tangible “feel” for the data they are examining. Carrying out nonparametric tests may involve reordering the datapoints into ranks, pairing them up across groups, flipping coins to determine outcomes, or shuffling subjects or samples among different groups.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
