What Is an Observational Study? | Guide & Examples
An observational study is used to answer a research question based purely on what the researcher observes. There is no interference or manipulation of the research subjects, and no control and treatment groups.
Observational studies are generally used in hard science, medical, and social science fields. This is often due to ethical or practical concerns that prevent the researcher from conducting a traditional experiment. However, the lack of control and treatment groups means that forming inferences is difficult, and there is a risk of confounding variables impacting your analysis.
Types of observation
There are many types of observation, and it can be challenging to tell the difference between them. Here are some of the most common types to help you choose the best one for your observational study.
|Naturalistic observation||The researcher observes how the participants respond to their environment in “real-life” settings but does not influence their behavior in any way||Observing monkeys in a zoo enclosure|
|Participant observation||Also occurs in “real-life” settings, but here, the researcher immerses themselves in the participant group over a period of time||Spending a few months in a hospital with patients suffering from a particular illness|
|Systematic observation||Utilizing coding and a strict observational schedule, researchers observe participants in order to count how often a particular phenomenon occurs||Counting the number of times children laugh in a classroom|
|Covert observation||Hinges on the fact that the participants do not know they are being observed||Observing interactions in public spaces, like bus rides or parks|
|Quantitative observation||Involves counting or numerical data||Observations related to age, weight, or height|
|Qualitative observation||Involves “five senses”: sight, sound, smell, taste, or hearing||Observations related to colors, sounds, or music|
|Case study||Investigates a person or group of people over time, with the idea that close investigation can later be generalized to other people or groups||Observing a child or group of children over the course of their time in elementary school|
|Archival research||Utilizes primary sources from libraries, archives, or other repositories to investigate a research question||Analyzing US Census data or telephone records|
Types of observational studies
There are three main types of observational studies: cohort studies, case–control studies, and cross-sectional studies.
Cohort studies are more longitudinal in nature, as they follow a group of participants over a period of time. Members of the cohort are selected because of a shared characteristic, such as smoking, and they are often observed over a period of years.
Case–control studies bring together two groups, a case study group and a control group. The case study group has a particular attribute while the control group does not. The two groups are then compared, to see if the case group exhibits a particular characteristic more than the control group.
For example, if you compared smokers (the case study group) with non-smokers (the control group), you could observe whether the smokers had more instances of lung disease than the non-smokers.
This often involves narrowing previously collected data to one point in time to test the prevalence of a theory—for example, analyzing how many people were diagnosed with lung disease in March of a given year. It can also be a one-time observation, such as spending one day in the lung disease wing of a hospital.
Observational study example
Observational studies are usually quite straightforward to design and conduct. Sometimes all you need is a notebook and pen! As you design your study, you can follow these steps.
Step 1: Identify your research topic and objectives
The first step is to determine what you’re interested in observing and why. Observational studies are a great fit if you are unable to do an experiment for ethical or practical reasons, or if your research topic hinges on natural behaviors.
Step 2: Choose your observation type and technique
In terms of technique, there are a few things to consider:
- Are you determining what you want to observe beforehand, or going in open-minded?
- Is there another research method that would make sense in tandem with an observational study?
- Does it make a difference to your analysis if your participants know you are there?
- If yes, make sure you conduct a covert observation.
- If not, think about whether observing from afar or actively participating in your observation is a better fit.
- How can you preempt confounding variables that could impact your analysis?
Overall, it is crucial to stay organized. Devise a shorthand for your notes, or perhaps design templates that you can fill in. Since these observations occur in real time, you won’t get a second chance with the same data.
Step 3: Set up your observational study
Before conducting your observations, there are a few things to attend to:
- Plan ahead: If you’re interested in day cares, you’ll need to call a few in your area to plan a visit. They may not all allow observation, or consent from parents may be needed, so give yourself enough time to set everything up.
- Determine your note-taking method: Observational studies often rely on note-taking because other methods, like video or audio recording, run the risk of changing participant behavior.
- Get informed consent from your participants (or their parents) if you want to record: Ultimately, even though it may make your analysis easier, the challenges posed by recording participants often make pen-and-paper a better choice.
Step 4: Conduct your observation
After you’ve chosen a type of observation, decided on your technique, and chosen a time and place, it’s time to conduct your observation.
When conducting observational studies, be very careful of confounding or “lurking” variables. In the example above, you observed children as they were dropped off, gauging whether or not they were upset. However, there are a variety of other factors that could be at play here (e.g., illness).
Step 5: Analyze your data
After you finish your observation, immediately record your initial thoughts and impressions, as well as follow-up questions or any issues you perceived during the observation. If you audio- or video-recorded your observations, you can transcribe them.
- If you conducted your observations in a more open-ended way, an inductive approach allows your data to determine your themes.
- If you had specific hypotheses prior to conducting your observations, a deductive approach analyzes whether your data confirm those themes or ideas you had previously.
Step 6: Discuss avenues for future research
Observational studies are generally exploratory in nature, and they often aren’t strong enough to yield standalone conclusions due to their very high susceptibility to observer bias and confounding variables. For this reason, observational studies can only show association, not causation.
If you are excited about the preliminary conclusions you’ve drawn and wish to proceed with your topic, you may need to change to a different research method, such as an experiment.
Advantages and disadvantages of observational studies
- Observational studies can provide information about difficult-to-analyze topics in a low-cost, efficient manner.
- They allow you to study subjects that cannot be randomized safely, efficiently, or ethically.
- They are often quite straightforward to conduct, since you just observe participant behavior as it happens or utilize preexisting data.
- They’re often invaluable in informing later, larger-scale clinical trials or experiments.
- Observational studies struggle to stand on their own as a reliable research method. There is a high risk of observer bias and undetected confounding variables.
- They lack conclusive results, typically are not externally valid or generalizable, and can usually only form a basis for further research.
- They cannot make statements about the safety or efficacy of the intervention or treatment they study, only observe reactions to it. Therefore, they offer less satisfying results than other methods.
Observational study vs. experiment
The key difference between observational studies and experiments is that a properly conducted observational study will never attempt to influence responses, while experimental designs by definition have some sort of treatment condition applied to a portion of participants.
However, there may be times when it’s impossible, dangerous, or impractical to influence the behavior of your participants. This can be the case in medical studies, where it is unethical or cruel to withhold potentially life-saving intervention, or in longitudinal analyses where you don’t have the ability to follow your group over the course of their lifetime.
An observational study may be the right fit for your research if random assignment of participants to control and treatment groups is impossible or highly difficult. However, the issues observational studies raise in terms of validity, confounding variables, and conclusiveness can mean that an experiment is more reliable.
If you’re able to randomize your participants safely and your research question is definitely causal in nature, consider using an experiment.
Frequently asked questions
- How do you define an observational study?
An observational study is a great choice for you if your research question is based purely on observations. If there are ethical, logistical, or practical concerns that prevent you from conducting a traditional experiment, an observational study may be a good choice. In an observational study, there is no interference or manipulation of the research subjects, as well as no control or treatment groups.
- What is the difference between an observational study and an experiment?
- What is a quasi-experiment?
- What’s the difference between exploratory and explanatory research?
- What is experimental design?
Experimental design means planning a set of procedures to investigate a relationship between variables. To design a controlled experiment, you need:
- A testable hypothesis
- At least one independent variable that can be precisely manipulated
- At least one dependent variable that can be precisely measured
When designing the experiment, you decide:
- How you will manipulate the variable(s)
- How you will control for any potential confounding variables
- How many subjects or samples will be included in the study
- How subjects will be assigned to treatment levels
Experimental design is essential to the internal and external validity of your experiment.