reliability

• what is reliability?

  • A measure of consistency  
  • If asked about ASSESING RELIABILITY IN EXAM talk about test-retest reliability or inter observer reliability  
• test-retest reliability

  • if a person completes a depression questionnaire on Monday they should get the same results on Friday. When the 2 scores are correlated, if the co-efficient is over +0.8 the questionnaire has strong reliability for assessing depression  
• interobserver reliability

  • When 2 or more observers agree on recording behaviour is a consistent way 
  • Behavioural categories are used in observations which are subjective so when. the separate observers correlate their observations there should be similarity 
  • Ainsworth's strange situation found 98% agreement across observers on the behavioural categories observed which makes finding more meaningful 
• test-retest reliability ao3

  • A limitation or test-retest reliability is demand characteristics as the same ppts are used which could make it a memory test as they're recalling what they previously answered 
  • Another limitation is its time consuming if the second test takes place over a year later 
• improving reliability: interviews

  • Training the interviewer reduces bias  
  • Also, by having the same researcher oi conduct all interviews so there's no difference in tone, facia; expressions etc  
• Improving reliability: procedures

  • Means there is more control over extraneous variables, so they don't become cofounding variables 
• improving reliability: operationalising behaviours

  • Operationalising behavioural categories (making them measurable) to reduce subjectivity and increase objectivity