Confirmation Bias and the Open Access Advantage: Some Methodological Suggestions for the Davis Citation Study