In experimental design, is it necessary to do randomization? If yes, why?

Data Science, AI and ML
,
#1

In experimental design, is it necessary to do randomization? If yes, why?

#2

If you use randomization in your experiments, you guard against bias . For example, selection bias (where some groups are underrepresented) is eliminated and accidental bias (where chance imbalances happen) is minimized. You can also run a variety of statistical tests on your data (to test your hypotheses) if your sample is random.
details about randomization:https://www.statisticshowto.com/randomization-experimental-design/

#3

There are two reasons.

The first and most obvious reason is to avoid bias. An experimenter good select experiments units to maximise the desired outcome of an experiment.

The second reason is more subtitle. When we randomise properly provided we have an adequate sample size the distribution of experimental error will be approximately normally distributed allowing statistical analysis of the results.