Chapter 17 ISCAM Ch. 4 Intro
CHAPTER 4: COMPARISONS WITH QUANTITATIVE VARIABLES.
This chapter parallels the previous one in many ways. We will continue to consider studies where the goal is to compare a response variable between two groups. The difference here is that these studies will involve a quantitative response variable rather than a categorical one. The methods that we employ to analyze these data will therefore be different, but you will find that the basic concepts and principles that you learned in Chapters 1-3 still apply. These include the principle of starting with numerical and graphical summaries to explore the data, the concept of statistical significance in determining whether the difference in the distribution of the response variable between the two groups is larger than we would reasonably expect from randomness alone, and the importance of considering how the data were collected in determining the scope of conclusions that can be drawn from the study.
Section 1: Comparing groups β Quantitative response
Section 2: Comparing two population means
-
Investigation 4.2: The elephants in the room β Comparing groups
-
Investigation 4.2 cont: The elephants in the room cont. β t procedures
-
Investigation 4.3: Left-handedness and life expectancy β Factors influencing significance
Section 3: Comparing two treatment means
-
Investigation 4.4: Lingering effects of sleep deprivation β Randomization tests
-
Investigation 4.5: Lingering effects of sleep deprivation (cont.) β Two-sample t-tests
-
Investigation 4.6: Ice cream serving sizes β Two-sample t-confidence interval
-
Investigation 4.7: Cloud seeding β Strategies for non-normal data
Section 4: Matched Pairs Designs
-
Investigation 4.8: Speed it up β Independent vs. paired design, technology
-
Investigation 4.9: Speed it up (cont.) β Inference (simulation, paired t-test)
-
Investigation 4.10: Comparison shopping β Application
-
Investigation 4.11: Smoke alarms β McNemarβs test (paired categorical data)
Examples
-
Example 4.1: Age Discrimination? β Randomization test
-
Example 4.2: Speed Limit Changes β Two-sample t-procedures
-
Example 4.3: Distracted Driving? (cont.) β Paired t-procedures
-
Example 4.4: Comparison Shopping in the Future β Power for paired vs. independent samples
Chapter 4 Summary
