Hi Shirom,

If I understand you, the best practice is to do quality control of 5-10% of your survey questions (two observers). The quality control person should be a different person than the data collector.

All the best

I don't think so. The more the observers the less the bias

If the observer is a junior, a pilot study should be done, and the measurements and decisions of the junior should be compared with the results of a senior. A statistical test should be done and the differences should not be significant. The degree of agreement between the two observers would also be assessed. So if there are no big differences between the two observers, in this case the junior can start collecting the data.