Custom Search

Tuesday, November 9, 2010

Reliability and Validity Concepts

Reliability and Validity Concepts:
The most prized qualities of psychological research are reliability (reliability the extent to which a given finding will be consistently reproduced on other occasions) and validity (validity the extent to which a given study investigates what it purports to investigate). Put simply, reliability relates to our confidence that a given finding can be replicated – and is not just a ‘freak’ or chance occurrence. Reliability in psychological research has much the same meaning in relation to a car. A reliable car is one that nearly always works. A reliable finding can nearly always be reproduced. Validity relates to our confidence that a given finding shows what we believe it to show. A valid car is a genuine car (a car that does what it is meant to do). A valid finding is a genuine finding (a finding that is what it purports to be – i.e. one that enhances your understanding in the manner indicated). Imagine we carry out a study in which we send someone to an exotic overseas location on a number of occasions to see whether this makes them think more favourably of their partner. Let us assume that it does. If it does so repeatedly, then it is a reliable finding. However, the study does not necessarily show that absence makes the heart grow fonder, but may instead demonstrate that exotic holidays make people feel better about their lives in a general way – including about their partners. The finding is therefore almost certainly not valid in a specific sense. Disputed validity is one of the most common and thorny problems in psychological research. Disputes arise when findings which purport to show one thing are reinterpreted to suggest that they actually show something completely different. The fact that the validity of research is often questioned is no bad thing. Indeed, this form of analysis and debate is central to psychological research and to one’s skill as a researcher.

No comments: