How Do Clinical Psychologists Do Research?

This page will provide a brief summary of important concepts related to clinical psychological research studies (covered in more detail in Pomerantz, 2020, pp. 137–143):

• experimental method: Most scientists, including psychologists, employ the scientific method when embarking on a research project, routinely following the same steps: deciding on the research question, observing events, and developing a hypothesis (which includes identifying an independent and a dependent variable). Once a hypothesis has been formed, it is empirically tested; interpretations of the results are reported in the study. This process is described in more detail on page 137 of your textbook.
• randomized controlled trial (or randomized clinical trial/RCT): An RCT is a study that randomly assigns participants to either an experimental or a control group. For example, researchers could want to examine whether a particular therapy for depression works. Patients who meet the diagnostic criteria for major depression would be randomly assigned to one of two groups; the experimental group would receive the therapeutic intervention, while the control group would not. Then, the groups are compared based on their baseline status (before the intervention) and on their status after receiving the therapy. RCTs are sometimes considered the gold standard for research, although some criticize their inability to replicate real-world problems.
• quasi-experiments: Some clinical psychological research questions don't lend themselves to traditional experimental design. (Typically, this happens when the researchers face many difficult-to-control variables.) In this type of scenario, a quasi-experimental design might be used. This design may not be as robust or as internally valid as a "true" experiment, but the data are important, especially when context makes a true experiment impossible.
• between-groups design: In studies using this design, comparisons are made between groups that get assigned to completely separate conditions and that receive different interventions. An experimental group receives the intervention, while the control group does not. 
• within-groups design: In studies using this design, subjects within a single group are exposed to a particular intervention, with their (often changing) results being gathered over time.
• analogue design: An analogue design approximates certain study elements when the "real thing" isn't available. It might be used when the target population isn't available or there aren't enough members of the target population to yield an adequate sample; approximations are accepted instead.
• correlational methods: Sometimes neither an experiment nor a quasi-experiment is possible, so correlational methods are used. There is no independent or dependent variable; rather, the associations between variables are observed, and inferences drawn from there. We cannot establish causality with a correlational study, but if a true experiment is impossible, correlational research still allows us to examine relationships between variables.
• case study: A case study is simply an in-depth examination of a single case or situation that can lead to more rigorous research.
• meta-analysis: A meta-analysis statistically examines and draws conclusions about a number of other studies. It's a way of combining multiple studies that all meet some criterion, which can help researchers draw firm conclusions from systematically exploring them.
• cross-sectional versus longitudinal design: A cross-sectional design examines a group of participants at a single point in time, whereas a longitudinal study follows the group over a long period of time.