5 Statistics you (probably) missed: Non-parametrics and interpretation

 

This chapter covers

  • The history and purpose of statistical tests
  • Evaluating and using non-parametric alternatives to common parametric tests
  • Using the chi-square test for categorical comparisons
  • Mitigating the likelihood of false-positive and false-negative results
  • Using statistics to ensure accurate findings
The number of ways you can misunderstand statistics is infinite. The number of ways you can understand it is finite.
—Dr. Lawrence Tatum

Parametric tests cannot be used in every situation. Despite statistics coursework covering the same limited topics, the tests we covered in chapter 4 are not the only options available to you as an analyst. When you cannot expect to produce reliable results with a t-test, ANOVA, or any other method we have covered so far, you have a wide range of non-parametric tests available for evaluating your data and making inferences about the broader population.

To provide some context for the underlying logic of parametric statistical tests, we will first briefly cover the development of these tests. Once you understand their intended purpose, you will be prepared to answer challenging stakeholder questions, communicate the limitations of a test, and think critically about an extensive range of questions you could answer in your organization and interviews as you seek to grow your career.

5.1 The landscape of statistics

5.1.1 The evolution of statistical methods

5.1.2 Choosing your approaches responsibly

5.2 Non-parametric statistics

5.2.1 Comparisons between groups on continuous or ordinal data

5.2.2 Exercises

5.2.3 Comparing categorical data

5.2.4 Recap

5.3 Responsible interpretation

5.3.1 Statistical errors

5.3.2 P-hacking

Summary