In a survey, the researcher records information from a collection of units such as animals, plants or plots of land but does not actively influence them. In an experiment, the researcher actively makes changes to some of these units by adjusting characteristics before recording information from them. The adjusted characteristics are called factors and only a well-designed experiment can lead to conclusions about whether these factors causally affect the measurements from the units.
The introduction describes some general principles for the design of experiments that should be followed to ensure that the resulting data can be meaningfully interpreted, and discusses practical issues about conducting experiments.