# Mill's methods - logic

Which Mill's method (not considering their limitations) would you use if you need to rule out the identity of a murderer based on the murders he already committed?

The only possible answers are method of agreement or method of difference. We are having a huge debate, some people are sure that it would be the method of agreement, others the method of difference. Each are giving reasonable explanations...but it really depends on how you build the tables. What happened if you have 3 murderers with the same characteristics? We need more opinions.

As I mentioned, the question is:

Rule out hypotheses about the identity of a serial killer based on the characteristics of each of the murders that he has already carried out. This is an example of causal inference following the method:

A) agreement

B) difference

I see it as difference to be able to rule out the identities...but I need to see it from different points of view. Feel free to share yours.

Thank you!

• Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking.
– Community Bot
Commented Jun 14, 2023 at 15:59

Long comment

In order to use so-called Mill's Methods, IMO the issue is how to "model" the problem.

Assume that we have a certain set of known murders, meaning that we know the respective murderer.

We may symbolize them with:

A occurs together with a b

B occurs together with c d

C occurs together with e f.

And we have a unknown murder: X occurs together with y z.

The assumption is that the unknown murderer X must be one of A,B,C.

We may summarize the three known murder in one single line of the table:

A B C occur together with a b c d e f