Validity asks whether the investigation actually tests the intended relationship. If two important variables change together, causation becomes difficult to interpret. Reproducibility asks whether others can obtain compatible results using the described method. More repeats do not repair a fundamentally invalid design.
Errors should be anticipated. Random variation causes readings to scatter; systematic error shifts readings in a consistent direction. Risk controls reduce likelihood/severity of harm. Ethical requirements protect people, animals and environments. Work involving Country/Place, cultural heritage or community knowledge may require permission and respectful protocols; scientific curiosity does not override those responsibilities.