>Low statistical power is an obvious symptom of weak research.
No it isn't. It is a simple one but it isn't obvious. Low statistical power is just one of many indicators of research quality. Qualitative research with sample sizes in the single digits can be done well and research with large statistical power can be done poorly. The measure of quality in research is much more conceptual and cannot be reduced to rule based measures without failing completely.
Statistical power includes effect size, so "qualitative" research (with large, easily measurable effects) can have high statistical power with only a few samples.
It shouldn't even be an indicator. The less certain you are about your hypothesis set, the less statistical power you should seek on your research. That way you can optimally filter the set and reject most falsehoods faster and cheaper.
But it is currently a useful indicator. This is a symptom of a problem by itself.
No it isn't. It is a simple one but it isn't obvious. Low statistical power is just one of many indicators of research quality. Qualitative research with sample sizes in the single digits can be done well and research with large statistical power can be done poorly. The measure of quality in research is much more conceptual and cannot be reduced to rule based measures without failing completely.