4 Sampling Variation and the Bootstrap Test
In Unit 3, we discovered that, even under the null hypothesis of no group differences, group means from randomized studies vary because of experimental variation. That is, variation in the result occurs because of random assignment. Recall in Unit 2, that the chance variation was a function of the sampling process; different samples drawn from the population (model) produced different results. When results vary because of the sampling process, the chance variation is referred to as sampling variation.
Outline and Goals of Unit 4
The following schematic outlines the course readings, in-class activities, and assignments for Unit 4.
|4.1 📖 Sampling variation|
|4.2 🔨 Speed skating|
|4.3 🔨 Crazy in love (part 1)|
|4.4 📖 External validity evidence and random sampling|
|4.5 🔨 Sample size exploration|
|4.6 📖 Validity evidence and inferences|
|4.7 🔨 Comparing study designs|
|4.8 📖 Observational studies and the bootstrap test|
|4.9 🔨 Murderous nurse|
|4.10 🔨 Summative activity: Movie sequels|
In the readings, course activities, and assignments in Unit 4, you will explore the process of modeling sampling variation to be able to evaluate observed differences between groups. You will learn about the bootstrap test (a Monte Carlo method for evaluating whether an observed result in compatible with sampling variation from a hypothesized model) and how to carry out this test using TinkerPlots™. You will also learn why random sampling helps provide validity evidence for generalizing results to the population (external validity evidence). Lastly, you will learn how to evaluate group differences from observational studies.