Have you come across the term triangulation while working on your research paper? You might have a rough idea of what it means, but you’re not entirely sure?
Then sit back and relax.
In this video, I will explain briefly but precisely what triangulation in research is all about. Additionally, I’ll provide you with all four types of triangulation and how you can implement this technique.
This way, you can elevate your qualitative research design to the next level and make your research methodologically robust.
Triangulation (Word Origin)
You can easily derive the meaning of triangulation from Latin. “Tri” means three and “angulus” means angle. So, triangulation involves “measuring in a triangle,” a concept that originates from land surveying.
However, outside of land measurement and geometry, empirical social research has adopted this term. And that’s what we’re focusing on now.
Triangulation in Research
When we talk about triangulation, we are on a methodological level. It’s about how a specific research design can provide as much insight as possible using one or more methods.
While it is commonly associated with qualitative research, it can also be applied in quantitative and mixed methods research.
To avoid confusing you, let’s look at the definition of “triangulation” in the research context from our colleague Flick (2008, p.12):
“Triangulation involves taking different perspectives on a subject under investigation or more generally: when answering research questions. These perspectives can be realized in different methods applied and/or different chosen theoretical approaches, both of which are related to each other.”
The goal of triangulation is to gain deeper insights than would be possible with just a single method or a single theoretical perspective.
Using the metaphor of land surveying, the position of an object can be determined more accurately when viewed from at least two different angles.
4 Types of Triangulation
To help you apply triangulation in your scientific work, here are the four most prominent types (Denzin, 1970; Flick, 2011).
#1 Method Triangulation
This form is probably the most commonly used. Denzin, the father of triangulation, even distinguishes between within-method and between-method triangulation.
Within-method triangulation could involve using two different interview guides to loosen the constraints of methodological decisions when creating the research design.
Between-method triangulation would involve adding a second method as described earlier. In our example, you could distribute an additional online questionnaire to the employees or evaluate the user data of the system.
#2 Data Triangulation
With this approach, you need different data sources. The method can remain the same, as can the phenomenon you are investigating.
To vary the data sources, you can change time, location, and people. There are almost endless possibilities, as you can already triangulate within each of these dimensions.
Time
Let’s take an example. Suppose your method is limited to expert interviews. You conduct interviews in a company and want to accompany the introduction of a new logistics system. You could triangulate within the dimension of “time.”
You select your expert, such as the head of the logistics department. Provided the company agrees, you could conduct an interview with the expert at two or three different points in time.
Here you would gain wonderful insights, for example, into the time before the introduction, the introduction process, and the experiences after the system has been used in the company for some time.
Location
In the same scenario, you could also triangulate the location. You could find two other companies where the system is also being implemented. Then you interview the heads of the logistics departments in these companies.
This way, you can make comparisons and examine the “phenomenon,” whatever you are investigating during the system adaptation, from different perspectives.
Data Subjects
Additionally, and I always recommend this, you can triangulate the data subjects. In addition to the logistics manager, you could include a warehouse specialist and a mid-level manager.
Of course, you can triangulate all three dimensions, but this also increases the effort. Consider which type of data triangulation would provide the most value for answering your research question.
Investigator Triangulation
This type involves briefly switching sides. Two or more researchers can prevent subjective distortions or a so-called “bias” on the part of the researcher.
In our example of expert interviews, at least two interviewers would have to be used. It would not be enough for you to conduct interview 1 and your fellow student interview 2. You would have to do it together, take notes independently, and then compare your evaluations.
This type of triangulation is only really feasible if you work in a group.
Theoretical Triangulation
The last form of triangulation is quite exciting but also not easy for novices like you and me to implement. Before analyzing your data, you must be aware of your theoretical background.
This means which theory you use to understand the data or the phenomenon. Different theories offer different perspectives. In theoretical triangulation, you would apply several different theoretical frameworks to the data and view the phenomenon from different angles.
For example, you could develop a codebook for analyzing your interviews based on a behaviorist theory. Then, analyze your transcript again, this time with a codebook developed using a different sociological or psychological theory. Your imagination is the limit here.
Of course, always provided you argue well.
Triangulation in Research: Validation vs. Balance
To fully understand the concept of triangulation, it’s worth looking at the debate that has been carried out in the research literature over the past few decades.
It was Denzin (1978) who originally proposed triangulation as a strategy for validating research results. His idea was to use an additional method to ensure the accuracy of an analysis. But that this additional method is conducted on a much smaller scale.
This approach, however, has been repeatedly criticized (e.g., Mayring, 2001), leading more and more researchers to argue that different methodological approaches or theoretical perspectives should better be considered equal.
It is also important to understand that the research design in qualitative methods does not always have to be strictly predefined.
Most textbooks suggest a certain approach, with steps you should take, important quality criteria, and so on.
But in qualitative research, it is always possible to deviate from a blueprint if certain circumstances in your research require it.
Triangulation, too, should be understood as an open concept rather than something that needs to follow a strict guideline.
The Difference Between Triangulation and Mixed Methods
If you’re familiar with my article on mixed methods, you might wonder what the big difference is, since different methods are combined there too.
Mixed methods and triangulation are indeed two related concepts within empirical social research. They share similarities, such as the combination of different methods.
But: Mixed methods represent an independent research strategy that explicitly combines quantitative and qualitative methods to benefit from the strengths of both approaches.
Triangulation, on the other hand, is a much broader concept, which not only involves the combination of methods (although it can) but also includes theoretical perspectives and other subjective viewpoints.
Moreover, unlike in mixed methods, you can do what is called “within method” triangulation, which could be a combination of two different qualitative methods.
References
Denzin, N. K. (1978). Triangulation: A Case for Methodological Evaluation and Combination. Sociological Methods, 339-357.
Flick U. (2008). Managing quality in qualitative research. London, England: Sage.
Flick, U. (2011) Introducing Research Methodology: A Beginner’s Guide to Doing a Research Project. Los Angeles: Sage
Mayring, P. (2001). Combination and Integration of Qualitative and Quantitative Analysis. Forum Qualitative Sozialforschung Forum: Qualitative Social Research, 2(1).