So far, I`ve talked a lot about observations and how we can improve them, but we often observe to learn something about cause-and-effect relationships, what depends on what? I will conclude our discussion on observations by presenting a framework that will allow us to understand how scientists draw conclusions based on their observations, experiments and simulations. The logic of observational conclusions was described by John Stuart Mill, a philosopher who is best known for writing about freedom, but who also contributed much to what we think of science. Mill was interested in how we could use observations and experiments to determine the causes or what depends on what. He introduced a number of methods of argumentation on empirical data that we now call Mills` methods. If there is such a variable, the Mills agreement method identifies this as cost. So let`s look at a simple example with only four dependent variables and one independent variable, whether a person has fallen ill or not. In this case, Mills` accord method would label tofu as the probable cause of the disease. It is the only food that all the people who got sick ate. Now let`s look at Mill`s method of disagreement. He is a close relative. We create the same table with dependent and independent variables, but in this case, look for a model where a difference in the dependent variable only appears if there is a difference in a variable. .