Triangulation of data analysis techniques

Abstract

In psychology, as in other disciplines, the concepts of validity and reliability are considered essential to give an accurate interpretation of results. While in quantative research the idea is well established, in qualitative research, validity and reliability take on a different dimension. Researchers like Miles and Huberman (1994) and Silverman (2000, 2001), have shown how these issues are addressed in qualitative research. In this paper I am proposing that the same corpus of data, in this case the transcripts of focus group discussions, can be analysed using more than one data analysis technique. I refer to this idea as ‘triangulation of data analysis techniques’ and argue that such triangulation increases the reliability of the results. If the results obtained through a particular data analysis technique, for example thematic analysis, are congruent with the results obtained by analysing the same transcripts using a different technique, for example correspondence analysis, it is reasonable to argue that the analysis and interpretation of the data is valid.