If the causes of behavior(in experimental terms,the independent variables)are doubtful,then the behavior(the dependent variable)may or may not be related to the cause.
The resolution of this profound dilemma is not easy;however,the issue has been reviewed in two influential books;Webb,Campbell,Schwartz,and Sechrest,Unobtrusive measures:Nonreactive Research in the Social Sciences(1966),and Cook and Cambell,Quasi-Experimentation:Design & Analysis Issues for Field Settings(1979)