1. Identifying the question

• What is it that I want to know?
• Is it possible to find out what I want to know?
• Is the data obtainable? (birth weight – socio economic, drugs, alcohol etc.)
• Is it ethical to obtain such data? (Tuskegee)
• If not, is there a reasonable substitute?
• Are my assumptions reasonable?
• Is my approach reasonable?

II. Designing a Study

• Survey
• Identify the population of interest
• Obtain a representative sample of that population
  • Simple Random Sampling
  • Stratified Sampling (M-F, Age groups)
  • Systematic Sampling (class roster, census list)
  • Multi-Stage Sampling (Neilson ratings)
• Sources of Bias
  • Voluntary Response (Sherry Hite M&M or Ann Landers)
  • Non-response bias (day phone)
  • Response bias (people lie)
  • Undercoverage
• Observational Studies
• Used when a designed experiment is not ethical
• Subjects studied over a period of time in natural setting
• Case/Control – Control must match (Love Canal)
• Record Variables of interest
• Confounding is a major issue
• Designing an Experiment
• Researcher has control over the subjects or units in the study
• An intervention takes place that otherwise would not occur
• Randomization used to assign treatments
• Strongest case for causality
• Classic example, randomized clinical trial
• EDA – Exploratory Data Analysis
• A data set exists

• Explore for relationships, trends, differences
• Prelude to a study
• Data summary for conclusion

III. Collecting Data

• Identify variables
• Identify types of variables
• Discrete – Countable (number of days the temp. was above 70)
• Continuous – uncountable (gas gauge)
• Identify Limits of measurement or observation

4. Analyze the data
5. Make Conclusions and Discuss Limitations

• What are the answers to the original hypotheses?
• What are the limitations of the study?
• What conclusions does the study not make?
• What new questions arise from this study?