Validity and reliability

Validity means ‘having some foundation based on truth’ whilst the term reliability means ‘able to be trusted, predictable or dependable’. (Collins English Dictionary). The terms mean very different things in a quantitative or qualitative research context. 

In a quantitative context, working within a scientific or positivist model of thinking, research results are objective measurements of phenomena and can be repeated and reproduced by other researchers – a notion of stability (i.e. reliability). The term validity refers to notions of accuracy – the truthfulness of the results.  The use of both reliability and validity refers to ideas of generalisation (applicable to larger populations or larger/wider situations), standardisation and robustness – hard objective data.

The context of qualitative research is totally different from quantitative research in that the data is semantic rather than numerical.  ‘Quantitative researchers seek causal determination, prediction and generalisation of findings whilst qualitative researchers seek instead illumination, understanding and extrapolation to similar situations’ (1).

 

The influence of post-modernism on research theory and practice has made the possibility of finding an absolute reliable truth a fallacy.   This applies to both quantitative and qualitative research.  Different practitioners and end users of research have different models of theory most of which are implicit rather than explicit.  Thus ‘data’ (numerical or semantic) is selected, interpreted and reported differently depending on the over-arching theory e.g. how advertising works (persuasion vs. engagement), whether attitude precedes behaviour or vice versa, whether psycho-analytical concepts are helpful in understanding human behaviour or not etc.

 

The pragmatic practitioner and user of qualitative research believes that valuable and robust research findings can be delivered through good working practices, transparency and clarity of theory and presentation from the start to end of the project. 

 

In the final analysis, a research study (quantitative or qualitative) either has or does not have credibility in the context of previous research and information, combined with the experience of those involved in the market, the category, the brand, the customers, the problem and so on.  Perfect validity is paradoxically impossible to achieve because human behaviour and motivations are so complex (2).

Reference:

1. Golafshani, N (2003) Understanding Reliability and Validity in Qualitative Research.  The Qualitative Report, Vol 8 No. 4

2. Denzin, N and Lincoln, Y (2000) Handbook of Qualitative Research. Sage