When I say … triangulation

Abstract

If you ever happen to get lost between Breenaun and Leenane in the west of Ireland, don't despair – there is an easy way out of your dilemma. Look to your right for the peak of Devilsmother and straight across the water towards the crags of Ben Gorm. Then look at your map and you should be able to pinpoint exactly where you are and where you need to go. Losing one's way is an occupational hazard for the hiker. The process described above is one of triangulation, in which you establish your location in relation to recognisable land points on a map; from this has developed the concept of triangulation in research. Triangulation may be defined as ‘the use of two or more methods of data collection in the study of some aspect of human behaviour’.1 Triangulation involves obtaining different perspectives on the same issue and thus should both increase researchers' confidence in their findings and allow a fuller understanding of the richness and depth of medical education research findings. For example, you might carry out a questionnaire survey of medical students to establish their level of satisfaction with their training, followed by a series of focus group discussions; if you get the same results, your confidence in your findings will be strong and you will probably have obtained a broader understanding of students' feelings from the focus groups. As this example shows, triangulation can be used in quantitative and qualitative research studies. There are different types of triangulation. These include time triangulation, space triangulation, combined levels of triangulation, theoretical triangulation and methodological triangulation.1 In the context of medical education research, triangulation most often refers to methodological triangulation; however, it is best to make this explicit when you use the term. The same holds true for the occasions on which you may mean another form of triangulation (such as theoretical triangulation, in which researchers cite different theories to explain a phenomenon). Methodological triangulation aims to explicitly alleviate the problem of ‘method-boundedness’, in which, for reasons of tradition, favouritism or simply limited experience with different research methods, researchers use only a single method to look at the object of study. Triangulation is not a panacea, however, and it has been criticised on a number of accounts. Firstly, if the data on which you base your analyses are not correct, you can triangulate all you like but you will be no nearer to the truth at the end of it. Even if two or more methods produce similar findings, you should be wary of jumping to an immediate conclusion that the findings must be correct: it is possible that both sets of findings are wrong for different reasons (http://www.referenceworld.com/sage/socialscience/triangulation.pdf). A second problem concerns the wrongful assumption that data drawn from different methodologies can be directly compared or crunched together to reach a definitive conclusion. What is the difference between triangulation and multi-method research? Purists might say that you should use the term ‘triangulation’ only when you are checking the validity of findings ‘by cross-checking them with another method’ (http://www.referenceworld.com/sage/socialscience/triangulation.pdf), but this may be too restrictive a view. Triangulation has occasionally been criticised as overused and occasionally misused; one way to avoid falling into this trap is to describe in simple terms exactly what you and your co-researchers did.2 If all of this is too confusing and you are still on a hillside in the west of Ireland, there is always your mobile phone, or the Drisheen Meerabo – but that is another story.