epsy341small.gif (20859 bytes)
Experimental  Research

The major feature that distinguishes experimental research from other types of research is that the researcher manipulates the independent variable.  There are a number of group designs in experimental research. Some of these qualify as experimental research, others do not.

The statistics by themselves have no meaning. They only take on meaning within the design of your study. If we just examine stats, bread can be deadly.

The term validity is used three ways in research...

  1. In the sampling unit, we learn about external validity (generalizability).
  2. In the survey unit, we learn about instrument validity.
  3. In this unit, we learn about internal validity and external validity. Internal validity means that the differences that we were found between groups on the dependent variable in an experiment were directly related to what the researcher did to the independent variable, and not due to some other unintended variable (confounding variable). Simply stated, the question addressed by internal validity is "Was the study done well?" Once the researcher is satisfied that the study was done well and the independent variable caused the dependent variable (internal validity), then the research examines external validity (under what conditions [ecological] and with whom [population] can these results be replicated [Will I get the same results with a different group of people or under different circumstances?]). If a study is not internally valid, then considering external validity is a moot point (If the independent did not cause the dependent, then there is no point in applying the results [generalizing the results] to other situations.). Interestingly, as one tightens a study to control for treats to internal validity, one decreases the generalizability of the study (to whom and under what conditions one can generalize the results).

There are several common threats to internal validity in experimental research. They are described in our text.  I have review each below (this material is also included in the PowerPoint presentation for this unit):

When planning a study, it is important to consider the threats to interval validity as we finalize the study design. After we complete our study, we should reconsider each of the threats to internal validity as we review our data and draw conclusions.

Del Siegle, Ph.D.
Neag School of Education - University of Connecticut