Final answer:
An experiment on a single individual can yield useful, albeit not broadly generalizable, conclusions, typically serving as a basis for more comprehensive research. Large sample sizes are preferred for broader generalizability, and not all data must confirm a hypothesis to be useful, as disconfirming evidence can also advance scientific understanding.
Step-by-step explanation:
An experiment conducted on a sample of just one person can still produce useful conclusions, and it would be true that not all experiments need large sample sizes to be impactful. However, studies on a single individual, often called case studies, are usually more exploratory or hypothesis-generating in nature rather than conclusive.
They allow for in-depth investigation of a particular case which can be very useful, especially in fields like psychology and medicine. That said, conclusions drawn from single-subject research are not broadly generalizable due to the small sample size.
In the context of scientific experimentation, it is also true that larger sample sizes are generally better because they tend to provide more reliable data that can be generalized to a larger population. Nonetheless, an experiment can indeed be done on humans; such experiments are subject to rigorous ethical review to protect participants.
Furthermore, experimentation is not the only valid type of scientific investigation, as observational studies, surveys, and other methods also provide valuable insights. It is false that data which does not support a hypothesis is not useful. Disconfirming evidence is essential in science as it helps refine or reject hypotheses and advance understanding.
Experiments should be designed to test predictions derived from hypotheses, where the researcher asks, 'If my hypothesis is true, what would I expect to observe?' and compares the expectations against actual observations.