4 .2 Designing Experiments

1. 4.2 Designing Experiments

2. Observational Study • Observes Individuals and measures variables of interest but does not attempt to influence the responses

3. Experiment • Deliberately imposes some treatment on individuals in order to observe their responses To see how nature responds to change, we must impose the change

4. The individuals on which the experiment is done are the experimental units. When the units are humans, they are called subjects. A specific experimental condition applied to the units is called a treatment.

5. Explanatory Variables in an experiment are often called factors. Many experiments study the joint effects of several factors often called a level.

6. Comparative Experiments • Treatment Observation OR • Observation 1 Treatment Observation 2

7. We want to match treatment groups in a systematic way. • Use a control group - a placebo • If you have an Asian woman in the control group, we need an Asian woman in the experimental group.

8. Control of the effects of lurking variables is the first principle of statistical design of experiment (using a control group) • The use of chance to divide experimental units into groups is called randomization.

9. Randomization is the second major principle of statistical design of experiments

10. Completely Randomized Experiments • All the experimental units are allocated at random among all the treatments

11. Randomization produces groups of experimental units that should be similar in all respects before the treatments are applied.

12. Comparative design ensures that influences other than experimental treatments operate equally on all groups.

13. The Logic of Experimental Design • Therefore, differences in the response variable must be due to the effects of the treatments. That is, the treatments not only are associated with the observed differences in the response, but must actually cause them.

14. An observed effect too large to attribute plausibly to chance is called statistically significant. • Experiments with many subjects are better able to detect difference among the effects of the treatments than similar experiments with fewer subjects.

15. 3rd Principle of Experimental Design Called Replication: • Repeat each treatment on a large enough number of experimental units or subjects to allow the systematic effects of the treatment to be seen.

16. Basic Principles of Statistical Design of Experiments • Control of the effects of lurking variables on the response, most simply by comparing several treatments • Randomization, the use of impersonal chance to assign subjects to treatments • Replication of the experiment on many subjects to reduce chance variation in the results

17. Double-Blind Experiment • Neither the subject nor the people who have contact with them know which treatment a subject received.

18. Other Experimental Designs • Block Design: A block is a group of experimental units or subjects that are similar in ways that are expected to affect the response to the treatments. The random assignment of units to treatments is carried out separately within each block.

19. Comparing effectiveness of TV ads women SUBJECTS men Ad 1 Ad 2 Ad 3 Group 1 Group 2 Group 3 Random Assignment Compare Results Group 1 Group 2 Group 3 Ad 1 Ad 2 Ad 3 Random Assignment Compare Results

20. Matched Pairs Design • A simple block design that compares just two treatments. Each block consists of two units, these units are assigned at random. Order of the treatment is also randomized because order may effect or influence the subject’s response.