Simply put falsifiability is the capacity for some proposition, statement, theory or hypothesis to be proven wrong.
Yes. And that’s only possible if you can test the hypothesis or theory, WNK.
Falsifiability can be determine by the ability of hypothesis to perform such test, by providing instruction in the hypothesis how to perform the FUTURE tests (eg how to set up experiment in a controlled environment (thus, in lab), where you can find (observe or detect) and gather evidences, measure it and quantify it, compare and verify it; these would be the test results of your observation).
If the tests have already been performed, right or wrong, then the hypothesis is already falsifiable, because you would test results and the results would contain data to your observation (your data would be your recordings, eg measurements, quantities, comparisons, etc).
The actual Testing stage and Analysing stage of Scientific Method would be occur during observation.
The Falsifiability stage should occur before the Testing stage, they should come in the form of INSTRUCTION of (A) how to set up the experiments, or (B) how, where and when you would find the evidences.
The formulation of the hypothesis should contain explanation of preliminary observation, the theoretical and proof-based model in the form of mathematical equations, formulas and constants, that are used part of predictive model, and instruction on to test the explanation and predictions through observation via evidence finding/gathering or experiments.
If you have proposal on how to perform the tests, then that’s indication that the hypothesis is falsifiable.
A concept or model isn’t falsifiable if you have no instruction on how you would test the explanation and predictions.
When Michael Behe wrote his papers on Irreducible Complexity (IC, 1993) and later wrote his book on Darwin’s Black Box (1996), he offered no instruction on how to test IC, so he has no evidences and therefore no verifiable data. All Behe has, as his data come from computer simulations.
Anyone can make computer simulations that favor whatever model, hence, these data (from his computer simulations) are not permissible, nor qualified as scientific evidence.
This is why the scientific community at large, and even in his fellow biochemist professors he worked with at the biochemistry department, rejected Behe’s IC as nothing more than pseudoscience. Irreducible Complexity doesn’t even qualified as hypothesis, let alone as accepted and verified scientific theory.
Irreducible Complexity is example of what unfalsifiable concept or model are.