Research design is essentially the framework used to conduct a study. While there are several types, two stand out as the most prominent and commonly used: experimental and observational.
Experimental research is when you manipulate one or more independent variables to observe their effect on a dependent variable, yet while controlling all other factors. In this type of research, researchers do intervene in the environment. Since researchers have some control over it, random assignment is thus used to minimize bias. (We'll explore this concept in greater detail later.)
Usually, there are would be two groups: a control group and an experimental group. For instance, to test the new drug efficiency, participants would be randomly assigned to either a group receiving the drug or a placebo (a fake drug). This would allow researchers to draw conclusions, whether one factor causes another. Experimental research is commonly referred to as a “scientific study” or simply an “experiment.”
On the other hand, observational research design studies phenomena as they naturally occur, without any manipulation from the researchers. In this type, researchers only observe or record data, without any intervention for any of the variables. These studies show if there is a relationship between variables, yet observational research doesn’t establish a cause-and-effect relationship. An example is to, for instance, observe the eating habits of people in a restaurant to study dietary patterns.
Observational research further has three types: cohort studies, case–control studies, and cross-sectional studies.
The first type, cohort studies, is a type of longitudinal study, which means that it tracks a group of individuals (a cohort) over a specific period of time, those who share a particular characteristic or experience.
These studies observe how exposure to certain factors may affect outcomes over time. Observations like those may span over months or even years. An important aspect is that participants, as mentioned, should share a specific trait or behavior. An example is to follow a group of smokers over 10 years to observe rates of lung disease compared to non-smokers.
The second type is case-control study. Such type involves two groups: a case group and a control group. The case group has a specific attribute that the control group does not have. Researchers then compare the two groups to see if the case group exhibits a particular characteristic more than the control group. Such study is also called retrospective case study. This is because those studies usually looks back in time to examine data that has already occurred.
Essentially, researchers are trying to analyze past records or information to observe any patterns; such patterns may show that there are a relationship between risk factors (things that might increase the likelihood of a condition) and outcomes (the condition or result being studied). Taking from smokers vs non-smokers example, researchers may try to compare a group of smokers (case group) with non-smokers (control group) to assess whether smokers have higher rates of lung disease. (Note that the case group was specifically selected because they already possess the condition the researchers are investigating; in this case, smoking.)
The third and final type is cross-sectional studies, where researchers are trying to examine a group of people (called a population) at a single point of time. It is the researchers who are trying to a get a snapshot to gather specific information about the population during this specific point. This contrasts with longitudinal studies, where data is collected over time. You can see the huge difference between cross-sectional studies and longitudinal ones. Contrary to longitudinal studies, in cross-sectional studies, the data is only collected once, without following the participants over time. However, similar to cohort studies, they compare different groups of people, e.g., smokers vs. non-smokers, yet they do not establish cause-and-effect.
Cross-sectional studies more observational in nature, meaning researchers are basically trying to observe and record data without manipulating anything. Going back to our smoking example, a cross-sectional study might attempt to compare the lung health of smokers and non-smokers at a single point in time. One possible way is that researchers can publish a survey to a group of people; they would ask, for instance, whether the person is smoking and would also test their lung function.