The word experimental research has a range of definitions. In the strict sense, experimental research is what we call a true experiment.
This is an experiment where the researcher manipulates one variable, and control/randomizes the rest of the variables. It has a control group, the subjects have been randomly assigned between the groups, and the researcher only tests one effect at a time. It is also important to know what variable(s) you want to test and measure.
A very wide definition of experimental research, or a quasi-experiment, is research where the scientist actively influences something to observe the consequences. Most experiments tend to fall in between the strict and the wide definition.
Experimental research design is centrally concerned with constructing research that is high in causal (internal) validity. Randomized experimental designs provide the highest levels of causal validity. Quasi‐experimental designs have a number of potential threats to their causal validity. Yet, new quasi‐experimental designs adopted from fields outside of criminology offer levels of causal validity that rival experimental designs.
The design of research is fraught with complicated and crucial decisions. Researchers must decide which research questions to address, which theoretical perspective will guide the research, how to measure key constructs reliably and accurately, who or what to sample and observe, how many people/places/things need to be sampled in order to achieve adequate statistical power, and which data analytic techniques will be employed. These issues are germane to research of all types (exploratory, explanatory, descriptive, evaluation research). However, the term “research design” typically does not refer to the issues discussed above.
The term “experimental research design” is centrally concerned with constructing research that is high in causal (or internal) validity. Causal validity concerns the accuracy of statements regarding cause and effect relationships. For example, does variable 1 cause variation in variable 2? Or does variable 2 cause variation in variable 1? Or does variable 3 cause variation in both variables 1 and 2? And what is the magnitude of the causal relationships among the variables? Thus, research design as used herein is a concern of explanatory and evaluation research but generally does not apply to exploratory or descriptive research.
Criteria for Establishing Causal Inferences
The three classic criteria necessary to support a causal inference, according to the philosopher John Stuart Mill, are:
(1) Association (correlation),
(2) Temporal order, and
The criterion of association requires that there is a systematic relationship between the cause and effect variables. This criterion is by far the easiest to determine. The second criterion of temporal order is a bit more complicated. The temporal order criterion requires that the cause, or more precisely variation in the cause variable, must occur before the observed variation in the effect variable.
The third criterion of no spuriousness is by far the most difficult to achieve. This criterion requires that the observed relationship between the cause and the effect variables must not be due to other omitted or unmeasured third variables. Using the relationship between delinquent peers and offending as an example, this criterion requires that this relationship cannot be due to homophily or any other potential explanation. Because there are usually many, many potentially relevant third variables and many of these third variables are unobserved, the criterion of no spuriousness can be quite difficult to achieve.