Positivism is also applied to politics, or, as some folks other than I tend to call it,
political science. Positivists, in this field, are those who make statements based on observations, and form hypotheses which are then tested using observational techniques (since the
Geneva Convention forbids experimentation in this area). They are distinct from, say,
Normative folks, who practice politics by describing what
should be rather than what
is.
Personally, I think that to call political science a science is a dangerous trivialization of the field, and introduces all sorts of baggage which is not useful in most cases and downright harmful in others. Some political scientists I know argue that political science is an 'observational hard science' in much the same way astronomy is, since experimentation is impractical to the point of impossibility, and only observation can be used to form and test hypotheses. However, I believe this to be a fundamentally flawed comparison for one central reason: astronomy is based on logical primitives which are readily verified through practical experimentation - Newtonian physics, gravity, constants such as the speed of light, and even relativity. The relationship of these testable 'primitives' of theory to the observations made in astronomy can be directly proven through application of simple scaling. No-one has ever managed to convince me, at least, that there are behavior patterns which humans follow in individual cases (or groups follow, or whatever) which are testable which can be scaled up to explain the interaction of groups such as nations, states, decisionmakers, etc. etc. There is a branch of political thought which attempts to explain international and intergroup dynamics using the experimentally-verified psychology of the human mind; however, their models tend to fail fairly spectacularly when used in attempts at prediction.