Nothing is perfect. There are always reasons that the results/findings of a study might be wrong or misleading. These reasons are called threats to validity of the work (also sometimes called limitations, but don’t confuse this use of the term with, say, the limits of what tool will or won’t do; it’s all about limitations to the conclusions that can be drawn from the results/findings).
As rebutter, your job is to identify and list the key threats to the validity of the research.
Here are some important tips to consider when rebutting in this course:
The rebutter should demonstrate that he/she has read and understood the paper.
There are several very common threats to the generalizability of findings in CS education research. The rebutter should always discuss these ones (if applicable).
The human subjects studied might not be representative of the target population (e.g., all learners).
The software projects studied might not be representative of the target population (e.g., all software projects).
The tasks studied might not be representative of the target population (e.g., all programming tasks)
However, note that the rebutter is expected to look for and discuss threats beyond the above three.