6.5 Summary
- An experiment changes the state of a system to produce an organized response, in purpose to answer a specific question. The response of interest must arise from the treatment imposed by the experimenter. If not it is an observational study.
- An observational study investigates pre-existing states without attempting to affect the response.
- An observational study reveals correlations between variables whereas a well-conducted experiment can provide evidence for causation.
- Studies based on non-hypothetical research questions can only provide descriptive knowledge. Hypotheses can provide a basis for explanatory knowledge. The prerequisite for this is that the hypothesis proposes a mechanism for the observed phenomenon and not just a phenomenological relationship.
- It is sometimes possible that the response results from variations in variables other than those manipulated in the experiment, so-called background variables. Precautions are needed to verify that such variables are not causing the observed effects. In such cases it is common to use control groups or other reference data.
Get Experiment!: Planning, Implementing and Interpreting now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
