evaluation

• generalisability?

  • good: 12 hospitals in 5 areas of US of different funding and size, also 8 patients with mixed occupations.
  • bad: only hospitals in US and study 3 only 1 person was used to collect data (young women). bias?
• reliability?

  • good: standardised procedures (empty, hollow, thud) and fact 12 hospitals were visited.
  • bad: ethical concerns, can it be conducted again today with same procedure?
• applications?

  • supports self fulfilling prophecy, effect of labels means decrease in likelihood of treatment.
  • DSM process of diagnosis is flawed
  • calls out psychiatric hospitals on diagnosis process.
  • should more toward a treatment of therapy for mental health rather than being drug-dependent.
• validity?

  • good: (eco) increased mundane realism, natural field setting produces a natural response and all pp used same symptoms. (internal) quantitative results.
  • bad: (internal) cannot manipulate any variables for cause and effect. Ps had wrong diagnosis, not all hospitals have same controls so each situation is different.
• ethics?

  • good: shows how patients are truly treated and nurses aren't aware of being monitored so no socially desirable answers seen. nurses are responsible for own behaviour rather than pp so any psyc harm isn't technically rosenhans doing.
  • bad: no consent or debrief is seen, psyc harm is still induced and patients deceived (but necessary?)