“few” is really unscientific. everyone knows for an experiment to yield reliable results you need a representative sample size. and then repeat the experiment to see if it gives the same result. only then you can be confident.