1. Agresti/Franklin Statistics, 1 of 56  Section 4.3 What Are Good Ways and Poor Ways to Experiment?

2. Agresti/Franklin Statistics, 2 of 56 An Experiment Assign each subject (called an experimental unit ) to an experimental condition, called a treatment Observe the outcome on the response variable Investigate the association – how the treatment affects the response

3. Agresti/Franklin Statistics, 3 of 56 Elements of a Good Experiment Primary treatment of interest Secondary treatment for comparison Comparing the primary treatment results to the secondary treatment results help to analyze the effectiveness of the primary treatment

4. Agresti/Franklin Statistics, 4 of 56 Control Group Subjects assigned to the secondary treatment are called the control group The secondary treatment could be a placebo or it could be an actual treatment

5. Agresti/Franklin Statistics, 5 of 56 Randomization in an Experiment It is important to randomly assign subjects to the primary treatment and to the secondary (control) treatment Goals of randomization: Prevent bias Balance the groups on variables that you know affect the response Balance the groups on lurking variables that may be unknown to you

6. Agresti/Franklin Statistics, 6 of 56 Blinding the Study Subjects should not know which group they have been assigned to – the primary treatment group or the control group Data collectors and experimenters should also be blind to treatment information

7. Agresti/Franklin Statistics, 7 of 56 Example: A Study to Assess Antidepressants for Quitting Smoking Design: 429 men and women Subjects had smoked 15 cigarettes or more per day for the previous year Subjects were highly motivated to quit

8. Agresti/Franklin Statistics, 8 of 56 Example: A Study to Assess Antidepressants for Quitting Smoking Subjects were randomly assigned to one of two groups: One group took an antidepressant daily Second group did not take the antidepressant (this group is called the placebo group)

9. Agresti/Franklin Statistics, 9 of 56 Example: A Study to Assess Antidepressants for Quitting Smoking The study ran for one year At the end of the year, the study observed whether each subject had successfully abstained from smoking or had relapsed

10. Agresti/Franklin Statistics, 10 of 56 Example: A Study to Assess Antidepressants for Quitting Smoking Results after 1 year: Treatment Group: 55.1% were not smoking Placebo Group: 42.3% were not smoking Results after 18 months : Antidepressant Group: 47.7% not smoking Placebo Group: 37.7% not smoking Results after 2 years : Antidepressant Group: 41.6% not smoking Placebo Group: 40% not smoking

11. Agresti/Franklin Statistics, 11 of 56 Example: A Study to Assess Antidepressants for Quitting Smoking Question to Think About: Are the differences between the two groups statistically significant or are these differences due to ordinary variation?

12. Agresti/Franklin Statistics, 12 of 56  Section 4.4 What Are Other Ways to Conduct Experimental and Observational Studies?

13. Agresti/Franklin Statistics, 13 of 56 Multifactor Experiments Multifactor Experiments: have more than one categorical explanatory variable (called a factor).

14. Agresti/Franklin Statistics, 14 of 56 Example: Do Antidepressants and/or Nicotine Patches Help Smokers Quit?

15. Agresti/Franklin Statistics, 15 of 56 Matched-Pairs Design Each subject serves as a block Both treatments are observed for each subject

16. Agresti/Franklin Statistics, 16 of 56 Example: A Study to Compare an Oral Drug with a Placebo for Treating Migraine Headaches Subject Drug Placebo 1ReliefNo Relief First matched pair 2Relief 3No Relief

17. Agresti/Franklin Statistics, 17 of 56 Blocks and Block Designs Block: collection of experimental units that have the same (or similar) values on a key variable Block Design: identifies blocks before the start of the experiment and assigns subjects to treatments with in those blocks

18. Agresti/Franklin Statistics, 18 of 56 Experiments vs Observational Studies An Experiment can measure cause and effect An observational study can yield useful information when an experiment is not practical An observational study is a practical way of answering questions that do not involve trying to establish causality

19. Agresti/Franklin Statistics, 19 of 56 Observational Studies A well-designed and informative observational study can give the researcher very useful data. Sample surveys that select subjects randomly are good examples of observational studies.

20. Agresti/Franklin Statistics, 20 of 56 Random Sampling Schemes Simple Random Sample: every possible sample has the same chance of selection

21. Agresti/Franklin Statistics, 21 of 56 Random Sampling Schemes Cluster Random Sample: Divide the population into a large number of clusters Select a sample random sample of the clusters Use the subjects in those clusters as the sample

22. Agresti/Franklin Statistics, 22 of 56 Random Sampling Schemes Stratified Random Sample: Divide the population into separate groups, called strata Select a simple random sample from each strata

23. Agresti/Franklin Statistics, 23 of 56 Observational Studies Well-designed observational studies use random sampling schemes

24. Agresti/Franklin Statistics, 24 of 56 Retrospective and Prospective Studies Retrospective study: looks into the past Prospective study: follows its subjects into the future

25. Agresti/Franklin Statistics, 25 of 56 Case-Control Study A case-control study is an observational study in which subjects who have a response outcome of interest (the cases) and subjects who have the other response outcome (the controls) are compared on an explanatory variable

26. Agresti/Franklin Statistics, 26 of 56 Example: Case-Control Study Response outcome of interest: Lung cancer The cases have lung cancer The controls did not have lung cancer The two groups were compared on the explanatory variable : Whether the subject had been a smoker