Search results

This paper raises for discussion and debate the question of generalisability in qualitative research. It is our view that, in the current national climate of the need for ‘evidence’ for policy and strategy success, many people claim to be more swayed by ‘numbers’ than words, and generally feel that quantitative or statistical research is the only kind that allows confidence in the generalisability of the results. This view is equally as evident in the field of adult learning research, where much of the headline data is in the form of numerical statistics based on sampling regimes. The outputs of this type of research are often designed for use in policy development and planning environments. The validity of these regimes and outputs is seldom questioned, while qualitative studies are often given limited credence because of assumptions about their worthwhileness based on premises from another paradigm. The downside of the quantitative view is that numbers simply cannot provide the evidence for the vast array of important questions that need answering. However, the question of what evidence is the right kind to support a particular point should be the ‘horses for courses’ one - of choosing the methodology that will gather the best data for answering the specific research question in front of us. In order to provide some more tangible issues to further the debate and discussion, after a literature review on the subject that brings together what (little) there is written on the subject, [we then] draw on examples of research conducted over the last 15 years, finally posing some areas in which the generalisation from qualitative research might be studied further with a view to establishing some parameters for its formalisation.