What are the disadvantages of a quasi-experimental design?

• A. Potential confounding variables — lack internal validity; May have population bias
• B. High internal validity
• C. Always randomized
• D. No confounding

Answer: (A)

Quasi-experimental designs are used when random assignment isn’t possible, so groups may start out unequal. The main drawback is that without randomization you can’t confidently rule out confounding factors—preexisting differences between groups or other variables that change alongside the intervention. These confounds threaten internal validity, making it harder to say that any observed effect is due to the intervention itself. There’s also a risk of population bias, meaning the sample may not represent the broader population because groups aren’t created randomly. This combination—confounding variables, weaker internal validity, and potential sampling bias—best captures why quasi-experimental designs have disadvantages.

High internal validity would require randomization, which quasi-experiments lack. They are not always randomized, and no-confounding is unlikely given the nonrandom allocation, which is why those descriptors don’t fit.