Rosenhan, D.L. (1973) On being sane in insane places

All the information about Rosenhan's Experiment on sane people in insane places and the 'stickiness of labels'

HideShow resource information

Rosenhan Intro

Background Info: Since the 1900s, metal ilness has been subject to the medical model that assumes the characteristics of a mental illness can be grouped and identified together. However, the symptoms of a mental illness are not as clean cut as the medical model suggests. They are subjective and this increases the likelihood of incorrect diagnosis.

Aim: The study aimed to test the hypothesis that the classification system (DMS) used by psychiatrists to diagnose mental illness was reliable, and how useful this diagnosis can be.

Hypothesis: Psychiatrists cannot reliably tell the difference between people who are sane and those who are insane.

Procedure: Rosenhan did two different pieces of research.

1 of 9

Rosenhan Study 1

Study 1

Method: Field Experiment

Aim: To see if sane individuals can be diagnosed and admitted to a psychiatric hospital.

Participants: 8 pseudopatients (5 males, 3 females) who had no history of mental illness.

Procedure: Each pseudopatient contacted a hospital for an appointment and reported they were hearing voices saying words such as ‘empty’, ‘hollow’ and ‘thud’ (independent variable).

- They changed their names and occupations (if it was related to psychology).

- Apart from this participants provided accurate details of their lives.

- If patients were admitted and diagnosed as insane (dependent variable) then they stopped reporting hearing voices and behaved normally.

- Each patient would remain in hospital until they were discharged.

- While in the hospital patients recorded their experiences, including interactions with staff and patients.

2 of 9

Rosenhan Study 1 results


·      All the pseudopatients were admitted- diagnosed with schizophrenia or remission

·      Pseudopatients spent between 7 and 52 days in hospital

·      When the were discharged they had a diagnosis of ‘in remission’ i.e. they were still considered schizophrenic

·      All their behaviour while in hospital was influenced by their diagnosis e.g. normal behaviour like queuing for lunch was seen as oral acquisitive behaviour

·      For almost 90% of the time staff left patients alone, spending time in their offices. Psychiatrists were rarely seen entering the ward.

·      When pseudopatients stopped members of staff and asked them a question, only 4% stopped to talk.

·      There was a lack of personal space, and lack of personal choice.

3 of 9

Rosenhan Study 2

Study 2

Method: Participant Observation

Aim: To see if the insane can be diagnosed as sane

Participants: Staff in the psychiatric hospital

Procedure: The staff in the hospital were aware of Rosenhan’s first study. They were informed that over 3 months  one or more pseudopatients would try to get admitted. Each member of staff was asked to rate the likelihood that the patients who sort to be admitted were pseudopatients. In reality, no pseudopatients would attempt to be admitted.

4 of 9

Rosenhan Study 2 Results


Over the three month period

  • 193 patients were admitted
  • 19 were thought to be pseudopatients by a psychiatrist and one other member of staff
  • 23 were thought to be pseudopatients by at least one psychiatrist
  • 41 were thought to be pseudopatients by at least two members of staff
5 of 9

Rosenhan Conclusions


  • In study 1 there were more type 2 errors made (diagnosing a healthy person as sick)
  • In study 2 there were more type 1 errors made (diagnosing a sick person as healthy)
  • Rosenhan did not criticise the staff and attributed nay failures to the environment, and the medical model of diagnosis they were working with.
  • It is not possible to distinguish the sane from the insane in psychiatric hospitals.
  • The labels provided by diagnosis influence perceptions of behaviour making it more likely that normal behaviour is interpreted as insane.
6 of 9

Rosenhan Evaluation Strengths


  • It is a very creative design, testing the reliability of diagnosis
  • Using pseudopatients meant that Rosenhan could gain insight into the experience of being in a psychiatric hospital
  • The study is as ecologically valid as it is possible to be
  • The variety of hospitals used in the study means that the study can be generalised to different institutions
  • The study uses both quantitative and qualitative data to inform results
  • It highlighted the difficulty in applying the medical model of diagnosis to mental health, and has practical applications in raising awareness of the impact of being diagnosed as mentally ill, and improving practice. This makes the study very useful.
7 of 9

Rosenhan Evaluation Weaknesses


  • It could be argued that as the pseudopatients were not really insane then their experiences would be different to the usual patient
  • There was a backlash from the psychiatric community after the study was published. The pseudopatients were pretending to be unwell, and if there pretence was effective then we should expect them to be admitted and diagnosed with some form of mental illness.
  • The sample of pseudopatients was very small, and was biased towards those who have an understanding of psychological issues (6/8 had psychological backgrounds). This may have had an impact on their behaviour in the wards.
  • It could be argued that Rosenhan only selected qualitative data that supported the quantitative data collected.
8 of 9

Rosenhan Key Words

Key Words

Stickiness of Labels: Labelling someone as mentally ill can have a detrimental effect upon the person. Such labels are very sticky and difficult to get rid of. People are rarely seen as cured from mental illness, and there is much prejudice in the world today surrounding mental illness. It is, therefore, very important that diagnoses are accurate.

9 of 9


No comments have yet been made

Similar Psychology resources:

See all Psychology resources »