3.10 Unit 3 summary

3.10.1 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

3.10.2 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)

3.10.3 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

3.10.4 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