Final answer:
Random sampling improves external validity in descriptive and correlational research, while random assignment enhances internal validity in experiments.
Step-by-step explanation:
In descriptive and correlational research, the practice of random sampling helps improve external validity by ensuring that participants adequately represent the population of interest. In contrast, in experiments, the practice of random assignment helps improve internal validity by ensuring that participants are fairly distributed between groups. This process reduces the likelihood of systematic differences that could bias the results and allows researchers to more confidently attribute differences in outcomes between groups to the manipulations of the independent variable rather than to preexisting differences among participants.