Comment Re:WTF? (Score 1) 172
In social sciences reproducibility is possible. For instance in epidemiology databases are crunched using well known statistical methods to determine correlations, then further science is applied to determine is these correlations might be causative. If a second party cannot do an equal statistical analysis and get similar results then the results are not valid. If a second party can go through the process of collecting the data and find systematic errors, then the results is not valid. This is in fact a big problem with education research. When subject to the process of real science, much if not most of the research has been shown to not meet those standards.
So social science research can be scientific, but there is a second issue. We expect research to be predictive. It is said that field such as astronomy are as unscientific as social science. But in astronomy there is an element of application. The results are used to predict other finding which then can be confirmed. This is the element that makes fields such as physiology less scientific.