Observational Study — Definition, Formula & Examples
An observational study is a type of statistical study in which the researcher collects data by observing subjects without assigning treatments or manipulating any variables. The researcher simply records what naturally occurs.
In an observational study, the investigator measures or surveys members of a sample without imposing an intervention. Because subjects are not randomly assigned to treatment and control groups, any association found between variables cannot be used to establish a cause-and-effect relationship; confounding variables may account for the observed association.
How It Works
To conduct an observational study, you identify the population of interest, select a sample, and then measure the variables you care about — without intervening. For example, you might survey 500 adults about their sleep habits and their GPA, then look for an association. Because you did not control who sleeps more or less, lurking variables (like work schedule or stress level) could explain any pattern you find. That is why observational studies can reveal correlations but cannot, on their own, prove causation. Researchers choose observational studies when experiments would be unethical, impractical, or too expensive — such as studying the long-term effects of smoking on lung health.
Example
Problem: A school nurse wants to know whether students who eat breakfast score higher on a math test. She surveys 200 students, asking whether they ate breakfast that morning, then records each student's test score. The 120 students who ate breakfast averaged 82 points; the 80 who skipped breakfast averaged 74 points. Can she conclude that eating breakfast causes higher scores?
Identify the study type: The nurse did not assign students to eat or skip breakfast — she only observed their existing behavior. This is an observational study.
Note the association: The breakfast group scored an average of 8 points higher than the no-breakfast group.
Consider confounding variables: Students who eat breakfast may also get more sleep, have more family support, or feel less anxious. Any of these lurking variables could partly explain the score difference.
State the conclusion: The nurse can say there is an association between eating breakfast and higher test scores, but she cannot claim that breakfast causes higher scores because no treatments were randomly assigned.
Answer: There is an observed association (breakfast eaters scored 8 points higher on average), but causation cannot be established from this observational study.
Why It Matters
Observational studies appear throughout AP Statistics (standard S-IC.3) and are tested on nearly every AP exam in the study-design section. Medical researchers, epidemiologists, and social scientists rely on observational studies when randomized experiments are impossible — for instance, studying the long-term health effects of pollution exposure. Understanding their limitations helps you critically evaluate news headlines that confuse correlation with causation.
Common Mistakes
Mistake: Claiming cause and effect from an observational study
Correction: Without random assignment, confounding variables cannot be ruled out. State that the study shows an association or correlation, not that one variable causes changes in the other.
Mistake: Confusing a survey with an experiment just because data is collected
Correction: The defining feature of an experiment is that the researcher imposes a treatment. If subjects are only measured or surveyed without intervention, it is observational — regardless of how structured the data collection is.