CHOOSING A RESEARCH DESIGN
What Kind of Experiment Should You Do?!?
This WEBsite will serve as a tutorial on how to choose a research design. This tutorial will focus on experimental designs (as opposed to observational studies-studies that involve observing an already existing situation). To begin with, let's define what an experiment is. An experiment is a cohort (group of subjects) study in which the investigator manipulates the predictor variable-otherwise called the treatment, program or intervention-and then observes the outcome. this tutorial will focuse on experiments with two groups, one who recieve the treatment, and a comparison group that does not. We have briefly discussed the structure of a generic experimental study . What we haven't discussed, is that there are varying types of experimental designs . These types or degrees depend on how your treatment and comparison groups are selected. So, depending on how these groups are selected, you may have more or less ability to establish a causal inference. What is meant by this is that various different types of designs have respectively, greater or lesser ability to determine ig your treatment or intervention is in fact the cause of the difference in your outcome measurement between your two groups.
Now, let's go through some basic steps on how to choose a research design appropriate for the issue you want to address.
The First Question is: Are You Concerned with establishing a causal inference?
OR
Some Other Thoughts and Caveats
*Please note that this tutorial does not address monetary or feasibility
considerations of research design. Although a certain type of experiment
may be the best for what you want study, it may not be feasible, nor may
it be affordable. These issues can be addressed once you decide on what
your goal/manpower/budget is for your study*.
*This tutorial is meant to be an introduction to research design. Therefore,
the designs mentiones are of the simplest form. research design can become
quite complicated, and the designs can become quite intricate. So I hope
you find this tutorial to be helpful, and a good introduction to research
design, and that it sparks your interest in research design.
****There are study designs that involve only one group (a group to which/whom you address your issue), these groups will not be dealt with in this tutorial. Good Luck! .
Please Email me with any information
or comments about this web page
4/9/97
