Unit 3 summary
You should be able to
- Conduct a statistical hypothesis test of an observed result from an experiment that uses random assignment
- Write an appropriate null hypothesis for an experiment that uses random assignment
- Use Monte Carlo simulation in TinkerPlots™ to find the expected experimental variation from an experimental study, if the null hypothesis were true (randomization test)
- Calculate a p-value
- Formulate a conclusion
- Evaluate the internal validity of a study using the following criteria:
- Temporal precedence
- Correlation of cause and effect
- No palusibel alternative explanations
You should understand
- How to interpret the mean of a dummy variable
- How to interpret the difference in means between two groups
- The logic behind a randomization test, including:
- Why we take the observed outcomes as fixed
- Why we randomly reassign the observed outcomes to experimental groups
- Why we sample without replacement
- What the distribution of results represents
- What a p-value represents
- What internal validity means
- The role of random assignment in drawing cause-and-effect conclusions (internal validity)
TinkerPlots™ skills
- Create a split dot plot
- Create a “shuffler” sampler
- Use the
Ruler tool to find the difference in means
- Use the
Reference line and Divider tools to count the values in a distribution that are as or more extreme than a given value
Vocab
- Categorical variable
- Numeric variable
- Dummy variable/Dummy coding
- Experiment
- Random assignment
- Experimental variation
- p-value
- Confounding variable
- Internal validity/cause-and-effect
- Probabilistically equivalent