Test Reliability and validity in psychometric assessments

QUESTION 1: What best describes reliability in a psychometric test?

That an empathy test can also measure the mass of an object
That a questionnaire designed to assess one construct is not useful to measure a different one
That metrics eliminate the need for training
That questions must be ambiguous to capture variations

That the metrics do not require technical validation
That questions are unambiguous and scales accurately reflect performance hierarchies
That speculation replaces research
That the test produces extreme fluctuations

Constant time consistency and technical validation of the assessed psychological factor
Random results and frequent changes
Measuring mass instead of length
Avoid studies supported by mathematical bases