An important consideration when conducting research and evaluation is whether you can rely on your data as being a true portrayal of the views or position of your population of interest. It is important to be sure that the findings being presented have not been unintentionally skewed because of how the questions were asked, or something unique about the respondents that you may have overlooked.
Triangulation is about using a range of techniques to provide insight into a topic that is particularly useful as a way of cross-checking the research findings.
It is particularly useful as there is often never one ideal way to conduct research; there are many ways to find information, and each method has its advantages and disadvantages.
Triangulation is also referred to as a mixed methods approach to research. It enables you to validate data through cross-verification from more than two sources. By using triangulation you are testing the consistency of the findings and increasing your chance of controlling any threats to the validity of the results.
There are four main reasons for triangulation, as outlined on the Better Evaluation website:
- Widening and deepening your understanding of the findings.
- Disproving a hypothesis using more than one approach.
- Confirming a hypothesis provided by one approach, by using another.
- Explaining more fully–using multiple techniques–the meaning behind the data.
I am currently conducting an evaluation of the Small Town Transformation initiative for Regional Arts Victoria. The research I will conduct to inform my evaluation will comprise lots of data from six regional towns, lots of viewpoints, a range of perspectives and both qualitative and quantitative data. I intend to use a range of methods to collect this data, including:
focus groups
in-depth interviews
surveys
observations
secondary research
stakeholder interviews and
data monitoring
Using each of these approaches I may be asking the same question of different respondents, or even the same question of the same respondents but in more than one way. By exploring the topic using different techniques I can be sure that my results will not hold bias. For example, I will check that what I learn in qualitative discussions supports my survey findings, and that these results have been supported in results from other similar studies from secondary research sources. I will cross check what I observe in the small towns with what people tell me through both qualitative and quantitative research, to be sure that what I observe is consistent with what others perceive is going on. I will not just rely on evidence from one town; I will compare and contrast both qualitative and quantitative findings across all six towns, to look for consistencies, or else reasons for inconsistencies. I have set up community panels of residents in each of the six towns and I intend to re-visit them over time to allow me to compare and contrast their views with other similar participants.
This hopefully explains the importance of triangulation. If you commission research or evaluation be sure that your consultant has an opportunity to look for ways in which the findings can be validated using more than one method, to be sure you can rely on the results.
I intend to embrace the triangle as I go forth and evaluate!