Chapter 16

Evaluating Experiments


Bullet Understanding the difference between an experiment and an observational study

Bullet Critiquing an experiment’s manner of design

Bullet Evaluating the results of an experiment correctly

Experiments are one of the major vehicles for data collection. To assess whether the results of a hypothesis test, confidence interval, or regression analysis are credible and believable, you must first examine the collection of the data. I have a saying: “Garbage in equals garbage out.” It means that no matter how sophisticated your graphs, charts, and data analysis are, they mean nothing if you base them on biased and inaccurate data.

In this chapter, you determine whether an experiment is well done and the impact the process has on the results.

Distinguishing Experiments from Observational Studies

Observational studies and experiments are vastly different in terms of the work that goes into them (their design and execution) and what comes out of them (the conclusions you can make). You need to be able to distinguish one type of study from another.

An observational study is just as it sounds: a study where the researcher merely observes the subjects and records ...

Get Statistics Workbook For Dummies with Online Practice, 2nd Edition 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.