Validity of Research Results

Validity is the extent to which a question or scale is measuring the concept, attribute or property it says it is.

For example, if one is measuring an attribute of a product such as Brand Recognition, how do we know that the question (or questions) used to measure brand recognition are valid? Usually, this is a result of what we know of the "perceived meaning" of the question from previous times it has been used. Validity can also be optimised by careful pretesting of alternative questions designed to measure the same concept.

In the academic sphere, validity can take many forms. Construct validity for example refers to the ability of a measure to relate meaningfully to other similar measures used before. In commercial market research, many companies rely on face validity – the extent to which the respondent knows what is being measured and it seems sensible to them. Validity can be differentiated from reliability, another property of good research.

In practice, validity can also refer to the success of the project in retrieving valid results. There are many sources of error that can reduce the validity of a project including poor sample selection and resultant bias, simple coding errors, mis-understanding of management and research questions by the researchers and misundertanding of the investigative questions by the respondents. Other errors include asking leading questions, unconscious non-verbal prompts on right answers, vindictive respondents, or inappropriate methodologies used to analyse the raw data.

Competent market researcher emphasise validity of results in their research conduct, even over presentation and colourful reports, which can sometimes obfuscate the actual validity of a market research survey.
blog comments powered by Disqus
Crimes in Design Webinar
Subscribe to our Monthly Newsletter