The Rosenhan experiment aimed to test the validity of psychiatric diagnosis and determine the role of bias and stigma in classifying abnormal behaviors. It sparked public debate regarding the methods and findings of the study. In Rosenhan’s field experiment, the simulated symptoms of the pseudopatients were the independent variable (IV), while the psychiatrists’ admission and diagnostic label of the pseudopatients were the dependent variables (DV). However, confounding variables still existed, and Rosenhan attempted to control them to eliminate the method error. For instance, Rosenhan’s experiment was conducted in 12 different hospitals located in five states, ranging from old and poorly staffed to private and research-oriented. In doing so, Rosenhan extended the geographical location and socio-economic groups covered in the experiment, thus, reducing method error and allowing the results to be more generalized.
Another example of the attempt to control extraneous variables in the study is choosing confederates from different backgrounds. According to Rosenhan, among three female and five male pseudopatients were “three psychologists, а pediatrician, а psychiatrist, а painter, … а housewife,” and a psychology graduate student. Furthermore, all the participants employed pseudonyms, and those involved in the mental health care sector alleged another occupation. Therefore, additional attention and bias from the hospital staff were eliminated to improve the experiment’s precision. Nevertheless, a number of other extraneous variables should be accounted for, such as staff’s education, pseudopatients’ ethnicity, and cultural background, as well as the level of nervousness in simulated patients due to the experiment’s novelty.