How difficult do you think it is to reduce the individual and social bias that affects scientific experiments?

Question

An indication of the results we want to achieve?

Answer 1

For the physical sciences it is really quite simple. The use of controlled experiments, the publication of work and the reproduction of that work by others is very efficient at filtering out inadvertent bias and even deliberate bias or even outright fraud.

In the case of social sciences where controlled experiments are difficult it can be quite hard. The model used by medical researchers is probably as good as can be done. Double blind tests eliminate some degree of bias. In addition large populations may be required to get good statistics. Even so it can be very hard to draw valid conclusions. That is even more a problem if good statistics cannot be obtained.

In general you need design experiments with care to control for inadvertent bias and you need to be ethical to not succumb to cheating. Keep in mind you will get caught.

Answer 2

Your question's too vaugue. There are many types of experiments and issues to research. Some are more controllable than others.