In 2016 we began the Research In Focus series, highlighting new and interesting findings from NIDILRR-funded studies, presented in a reader-friendly format. As a follow up, we offer our occasional series Research In Brief, where we break down some of the terms you might find in peer-reviewed studies.
In this month’s Research in Brief, we explore different types of quantitative research designs (more on research methods). A quantitative study is any research study where measurements are taken to compare groups of people, look at changes over time, or look at the effects of a treatment. Different types of quantitative studies have different advantages and disadvantages. For example:
- In a cross-sectional study, researchers measure several variables in the participants at the same time, and then test to see if two or more variables are related to each other. For example, a researcher might ask a group of participants to report on their experiences with depression, pain, and sleep quality in a survey. The researcher may then test whether people who report higher levels of depression tend to report worse sleep quality or higher pain levels. Cross-sectional research is relatively easy to conduct, and researchers can measure a large number of variables simultaneously. However, cross-sectional studies cannot tell us the cause of an association, only that an association exists. In the above example, we cannot tell if getting poorer sleep causes depression, or if feeling depressed causes poor sleep; only that the two are related.
- In a longitudinal study, researchers collect measurements from the same group of participants at 2 or more time points, and compare the measurements to see if they have changed over time. The researchers may also measure one variable at the first time point, and then measure a second variable later. For example, researchers might measure sight-reading skills in a group of 3-year-old children, and then measure their school performance and test scores over the next 15 years to find out whether early reading skills predict later academic performance. Longitudinal studies can help clarify the predictive relations between variables, but they cannot tell us for sure whether one variable causes another. For example, children who score higher on early reading tests may have more involved parents, which might explain both their higher early reading scores and their better academic performance later in childhood. Also, sometimes participants drop out of the study before all the measurements are taken.
- A quasi-experiment is a study where measurements are taken before and after people receive a treatment. However, the participants are not randomly assigned to the treatment. For example, researchers might use a quasi-experiment to compare a group of patients in a hospital that is using a new exercise program with a group of patients receiving standard care. Quasi-experiments are useful when a new program cannot be randomly assigned, but it is difficult to draw firm conclusions about the effectiveness of the program.
- An experimental study is a study in which participants are randomly divided into “experimental” and “control” groups. The participants are assigned to a group based on the random flip of a coin, or a random number, so that the groups will be as similar as possible before the study. Then, the participants in the experimental group receive the new intervention, while the participants in the control group receive something different. When the experiment is testing a clinical treatment, it might be called a Randomized Controlled Trial (RCT). During experimental studies, researchers usually take measurements from the participants in both the experimental and control groups at the beginning and the end of the study. They will then test whether the participants show bigger changes in the experimental group than in the control group. An experiment is the best way to test whether a new treatment has a measurable impact on participants. However, often it is not possible or ethical to randomly assign people to experimental or control groups; for example, it is not ethical to randomly assign some people to get a disability and others to not get a disability. So in those cases, researchers might use a different type of study, like a quasi-experiment to compare people with and without the disability.
To Learn More:
For examples of cross-sectional studies highlighted in our Research In Focus series:
- People with Childhood Disabilities May Be at Higher Risk for Chronic Diseases as Young Adults
- Strong Relationships May Help Adolescents Adjusting to a Traumatic Brain Injury
- Recovery Is Possible for People with Serious Mental Illnesses
NIDILRR has funded the longest running longitudinal study of people with spinal cord injury, the Health, Employment, and Longevity Project, headed by James Krause, PhD. Now in its 47th year, the study has followed cohorts of individuals with SCI, surveying them every three to five years.
For examples of quasi-experimental studies highlighted in Research In Focus:
- Digital Books with Dynamic Text May Show Promise to Help Young Children with Autism Spectrum Disorders Learn Sight Words
- A New Tablet-Based System Shows Promise in Helping Teens with Autism Spectrum Disorder Stay on Task at School
- A New Online Program May Help People with TBI Build Emotion Regulation Skills
For examples of experimental studies highlighted in Research In Focus: