Rosenhan's Study (1973)

To see if the sane could be distinguished from the insane using the DSM classification system and, if they can be differentiaed, how sanity can be indentified.

• Sent 8 pseudo-patients to 12 different hosptials, there were 3 women and 5 men.
• All gave false names and occupations, but all other information was true, other than symptoms.
• All said they could hear an unfamilier voice of the same sex saying 'empty', 'hollow' and 'thud'. 
• As soon as they were admitted they stopped all abnoraml behaviour.
• They took part in ward activities, spoke to staff and fellow patients as they would normally and respond to instructions from staff.
• Some wrote down their observations about the ward, the patients and the staff.
• If asked how they felt, they said they felt fine and no longer had any symptoms.
• In some hospitals the pseudo-patients approached members of staff with requestions such as 'pardon me, could you tell me when I am likely to be discharged?'

• All pseudo-patients were admitted, and none were detected as being sane.
• All but one of them had a diagnosis of schizophrenia in remission.
• They stayed in hospital for between 7 and 52 days, with the average being 19 days.
• In 3 hospitals, 35 out of 118 patients were suspicious about the insanity of the pseudo-patients.

Rosenhan concluded that staff in psychiatric hospitals were unable to distinguish those who were sane from those who were insane, and that DSM was therefore not a valid measurement of mental illness at that time.

Whilst the pseudo-patients' observations wpuld try to be objective, some subjectivity and the emotions of pseudo-patients could have influenced these observations.

There are ethical problems, as the hospital staff were decieved about the patients' symptoms. Nor did they know they were in a study so were unable to give consent. However Rosenhan didn't name any staff or hospitals so there was no risk of identification.  

All doctors, either of physical or mental illness, ted to play safe and go for the most serious or common diagnosis before ruling it out, rather than immediately trying to see if the patient is faking it. Psychiatrists have to be especially careful, as if they release someone with a mental disorder there can be serious consequences (for them or for others). 

The pseudo-patients insisted on being admitted to the hospital, which is an important symptom in itself, so the voicesweren't the only symptom they presented. Nor was their behaviour totally normal once they were admitted, as they didn't say that they were in fact normal and inststed on being released (which is what most normal people would do).

Psychiatrists point out that DSM has been revised since Rosenhan's study, and such results are less likely. Another study now is therefore unlikely to replicate he results.

Spitzer argues that the diagnosis of schizophrenia in remission was in fact due to how the pseudo-patients behaved and not to the fact that the psychiatrists couldn't tell they were normal, as it is a very rare dagnosis for real patients. Therefore, he argues, the psychiatrists recognised there was something different about the pseudo-patients.

The study was carried out in actual psychiatric hospitals, using real staff who were unaware of the study, so it has ecological validity.

As a range of hospitals were used from around the country (America), including old and new and using different methods of funding, the results could be generalised to other psychiatric hospitals at the time.

The number of days the pseudo-patients stayed in hospital is an objective measurement, and the fact that the pseudo-patients could see what life was like from a patient's perspective adds validity.