How do you know whether your findings are valid?

Validity is key during the analysis and interpretation of scientific results. Without validity, results cannot be generalised and therefore any research becomes less significant. Essentially the term ‘valid’ originates from the Latin ‘vallidus’ meaning strong. But how can your findings be ‘strong’ in order to falsify your null or alternative hypothesis and still provide valid findings?

Validity: ‘Evidence that a study allows correct inferences about the question it was aimed to answer or that a test measures what it set out to measure conceptually’

(Discovering Statistics using SPSS, third edition, Field)

So for your findings to be valid they must be accurate and appropriate, whilst referring to the question you originally aimed to answer. They must represent what you tested and they must be strong in the sense that the content validity is high; clearly showing that what you have tested represents your field of study.

Validity can be expressed in different ways Internal/Criterion validity – what goes on within a study (whether the researcher tested what was meant to be tested) and External/Ecological validity – what happens outside of the study (how well the findings can be generalised to other situations and people).

To maximise validity, any potential confounding or extraneous variables need to be minimised. So to ensure your findings are valid, the research itself needs to be as tightly controlled as possible to avoid other variables other than those being measured affecting the results. If findings do not solely show the effect of one Independent variable on a Dependent variable, the research’s validity is reduced.  

‘Any research can be affected by different kinds of factors which, while extraneous to the concerns of the research, can invalidate the findings’

(Seliger & Shohamy, 1989)

To ensure validity, instruments and methods used during research must first be reliable. Without reliability, results cannot be accurate and therefore the study cannot measure what it was intended to measure, reducing the validity of the study itself. If data is not reliable, through inaccurate data collection/experimental methods, the results automatically cannot be valid.    

Data such as qualitative data can be reliable because of the depth and detail involved, but does this lend itself to validity? Qualitative data can be valid as it is true to life, but perhaps only valid for one person or a small minority? It’s very difficult to generalise Qualitative data to the larger population, but does this mean that data is still valid? Or are the findings therefore invalid as a whole?

To know whether your findings are valid, surely you must question all aspects of research as to how methods could invalidate findings. Only when these factors have been eradicated to the best of the researchers’ ability can your findings be valid.