Types of Study Design on the MCAT
By Andrew Dombrowski
The Psychology and Sociology section of the MCAT has become well-known for including questions about experimental design. Moreover, experimentally-based passages occur in all three science sections. For this reason, gaining a solid understanding of experimental design can allow you both to succeed on questions that directly ask about it and to more efficiently approach and comprehend passages that describe experimental findings.
Familiarizing yourself with some of the main terms involved in research (independent vs. dependent variables, confounding/mediating/moderating variables, longitudinal vs. cross-sectional research, etc.) is important, but the best way to take your understanding of these terms to the next level is by exploring why researchers make the choices they do, not just memorizing terms.
In this blog post, we’ll look at various types of study designs and why they’re appropriate in certain contexts.
One major distinction is between observational and experimental research designs. Experimental designs may be the most familiar from science laboratory courses. To put it simply, in an experiment, you do something and observe the result. The “something” can vary widely, with real-world examples ranging from trying out new cell culture conditions to create a model for a disease process to administering patients a new kind of drug.
To make sense of the results, though, you need to have a point of comparison, which is provided by positive and negative controls. Students in high-school and college-level lab courses often find controls to be tedious, largely because the experiments that are done in such courses are relatively simple and have intuitively obvious outcomes. To help solidify the concepts of positive and negative controls, let’s work through some real-world examples where they’re needed.
A negative control is a group that does not receive the treatment. A classic example of when negative controls are important is when an experiment tests a new drug. Imagine that you’re testing a new anti-flu drug. If you administer your new drug to patients and observe them get better, that doesn’t really tell you anything, because the vast majority of people with the flu will get better anyway. In this context, you need to figure out whether your drug helps people recover faster, and for that, you need a negative control group.
A positive control is a group that receives a treatment with a known effect. A common reason for including positive controls is to make sure that your experimental setup is working correctly. Let’s say you do a cell culture experiment and observe nothing. Is your result meaningful, or is it possible that you just screwed up the cell culture protocol (which is realistic, since cell culture can be a pretty finicky process)? A positive control helps you tell the difference.
If patients are randomly assigned to receive the treatment or a control, a study is a randomized controlled trial (RCT). RCTs are considered to provide high-quality evidence, and multiple RCTs can be synthesized in systematic reviews and meta-analyses, which are considered to provide the highest level of evidence regarding treatments.
In experiments involving patients, it is useful to “blind” them to whether they receive the real medication or a placebo, to help control for the placebo effect. Likewise, it is useful to “blind” examiners to the status of the patients they examine, in order to avoid confirmation bias. The latter type of blinding also occurs in lab-based studies; for example, the person evaluating the growth of cells in a culture plate should not know whether the cells received a real treatment or a control treatment. Studies with one type of blinding are known as single-blinded studies, and studies with both types are known as double-blinded studies.
You might ask why a study would ever use a single-blinded design if a double-blinded setup is better. Simply, it might not be possible. One example might be clinical research assessing the impacts of support groups on the quality of life of patients with a certain disease. There’s no way to administer a “placebo” support group—patients will know whether they’re attending a support group or not. However, you could still blind the clinicians caring for them to their status.
Observational studies, as the name suggests, involve analyzing something that is happening regardless of whether scientists are interested in it. There are several reasons why a researcher might conduct an observational study. Some common examples include assessing complex population-level phenomena like the impact of diet on heart disease and cancer, or analyzing the spread of an infectious disease among the community. In essence, there are contexts where performing an experiment would be impossible and/or unethical.
Cross-sectional studies involve taking a snapshot of a population of interest at a moment in time. For instance, a study might record the demographic and personal information about individuals, as well as health metrics such as blood pressure, fasting glucose, and cholesterol levels, to see whether any associations emerge. In contrast, a longitudinal study would take the same basic approach, but incorporate multiple measurements to enable the research to track trends over time.
Case-control studies take individuals with a certain condition (“cases”) and match them to similar individuals without that condition (“controls”), in an attempt to identify factors likely to contribute to the development of a condition. This is particularly useful for illnesses that are not yet well understood. They can suggest useful avenues of further research, but are not considered to provide conclusive evidence.
Finally, case studies are descriptions of specific patients with remarkable or interesting conditions. They aren’t generally considered to be conclusive, but can be quite useful when it comes to alerting clinicians to emerging threats or raising awareness of rare but serious diseases.
There’s still more to study when it comes to interpreting research for the MCAT, but hopefully this review has helped clarify certain types of study design. A major lesson here is to always ask why researchers do things in a certain way, because you will absorb the material better this way than if you just memorize definitions.
Just getting started in your MCAT prep? Consider our free practice bundle. This bundle includes a half-length diagnostic, access to our first full-length practice exam, a sample QBank quiz, and a demo of our course. You can sign up here. Interested in learning more about the Psych/Soc section of the MCAT? We offer a Content Review book as well as a Strategy and Practice book on this section. You can view this and the rest of our best-selling MCAT books here.
We wish you the best in your studies!