Statistics

1. Simple Summaries are easier then long lists.

1.1. Dotplot

1.2. Five-Number Summary

1.3. Upper & Lower Quartiles

1.4. Median

2. Always include the base rate or baseline risk when discussion a change in the rate or risk of something.

3. A Sample of a few hundred, or a few thousand can give reasonably accurate information about a population of many millions.

3.1. Population

3.2. Random Sample

3.3. Sample Survey

3.4. Margin of Error

4. Cause-and-Effect conlusions cannot generally be made on the basis of an observational study.

4.1. Observational Study

4.2. Variable

4.3. Confounding Variable

5. A statistically significant finding does not always have practical importance. The implied direction of cause and effect may be wrong. Any study should be viewed in the context of society at the time it was done.

6. An unrepresentative sample, no matter how big, tells you almost nothing about the population.

6.1. Nonresponse bias

6.2. Self-Selected Sample

6.3. Volunteer Sample

7. Unlike Observational studies, cause-and-effect conclusions can generally be made on the basis of randomized experiments.

7.1. Randomized Experiment

7.2. Treatment

7.3. Random Assignment

7.4. Placebo

7.5. Statistically Significant and Practical Importance