Common Issues in Interpretation I: Box-and-Whisker Plots
Not only are there common issues in experiments, but there are many common ways that people misinterpret statistical results. Box-and-whisker plots (also known as box plots) are a great tool for summarizing data, but can be a source of confusion for new analysts. To refamiliarize yourself with boxplots, You can read our summary of them in the Chapter 5 exercise, "Practice working with eGulf," or on various websites, such as Wikipedia (https://en.wikipedia.org/wiki/Box_plot).
Following an experiment regarding how computer keyboard layouts affect people's typing speed, the principal investigator, Kurt Devorak, compared the hand sizes among respondents in the treatment group and control group to make sure his sample was balanced. He plotted the results in two box-and-whiskers plots (illustrated in the R workspace). Which of the following conclusions cannot be determined from these boxplots (or almost any boxplot)?
This exercise is part of the course
Causal Inference with R - Experiments
Exercise instructions
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
boxplot(Treatment,Control,names=c("Treatment","Control"),main="Hand Sizes (inches) across Experimental Groups")