Confirmation Bias (DP IB Psychology): Revision Note

Confirmation bias

• A cognitive bias is a faulty or distorted way of perceiving or understanding the world

  • A cognitive bias is a kind of heuristic

• Confirmation bias is the tendency to overlook or ignore information which does not align/agree with preconceived ideas about a person/event/situation/group and focus only information which supports existing views and attitudes

  • E.g., if I suspect that my husband is being unfaithful I will look for evidence of this in his behaviour and ignore examples which do not support my suspicions

Confirmation bias in research

• Researchers may:

  • Selectively record results supporting their hypothesis

  • Ignore findings that challenge it

• This leads to low validity and reinforces stereotypes

Research which supports confirmation bias

Pavkov & Lewis (1989)

Aim:

• To investigate whether race/ethnicity and confirmation bias determine decisions made by clinicians in terms of a diagnosis of schizophrenia.

Participants:

• Patients from four mental health hospitals in Chicago, USA

• The sample comprised two thirds male, two thirds aged 18-34 years

• They were from neighbourhoods that represented both Black-dominant and White-dominant populations

Procedure:

• The researchers interviewed the participants while they were in hospital

  • One of the interviews was diagnostic (i.e., to determine the nature of the patient's mental illness) and was conducted by an expert who had not been told the aim of the research

  • The researchers conducted the other interview, which focused on social-psychological measures, such as how socially integrated the patient was and how aware they were of their condition

Results:

• The researchers found that Black patients were more likely to be given a diagnosis of schizophrenia than White patients

  • This was particularly the case in hospitals which were located in Black-dominant neighbourhoods

  • This happened even when the misdiagnosed patients had previously been diagnosed with a different mental illness, i.e., not schizophrenia

Conclusion:

• Patients may be being misdiagnosed simply due to the colour of their skin; this is evidence of confirmation bias

• Clinicians may use stereotypes rather than objective evidence

Evaluation of confirmation bias

Strengths

• Having an understanding of confirmation bias should enable people to avoid the mistake of labelling or stereotyping others who are not from one's own cultural (or other) group

• Interviews provide rich, detailed data which has strong explanatory power, and gives insight into the experiences of the participants

Limitations

• There are a range of extraneous variables involved in the above research

  • Using four separate hospitals housing a wide variety of staff, patients, equipment, location issues, etc., means that it is difficult to draw meaningful conclusions from the findings

• The participants may show social desirability bias, which would impair the validity of the findings

Link to concepts

Causality

• Measuring confirmation bias is difficult:

  • It is a subjective, ill-defined variable

  • Apparent evidence of confirmation bias may actually reflect other factors

• Pavkov & Lewis' (1989) findings suggest racial confirmation bias in the diagnosis of schizophrenia

• However, alternative explanations include:

  • clinician fatigue

  • limited time/resources

  • understaffing

• Misdiagnoses were based on only one interview, which may be an unrealistic expectation for accurate diagnosis

Bias

• The study may suffer from sampling bias:

  • Participants were drawn from four hospitals in Chicago only

  • This limits representativeness and generalisability

• To fully assess how confirmation bias affects diagnosis, the study should be replicated in other states/countries with more diverse samples