Understanding Triangulation in Research: Methods and Validity
Understanding Triangulation in Research
It is assumed that whatever technique is used, it points to the same reality. It does not change the reality itself; what changes, at most, is the focus. However, all approaches must tend towards a convergence of results. If they do not converge, this is one of the main problems: triangulation has been poorly implemented in at least some of the techniques.
Triangulation has been established as a methodology and strategy for assessing and monitoring research already completed, more to enhance an understanding of social reality rather than as part of a design intended to increase knowledge about social reality. The term ‘triangulation’ was developed by Campbell and Fiske in the field of attitude tests. In principle, it involves measuring several different features of a collection of individuals using various methods, so that each measurement is independent. Thus, it generates a correlation matrix comprised of the traits measured. The measures of the same trait should correlate more closely with each other than with measures of different traits that involve different methods. The multitrait-multimethod matrix seeks to ensure that only relative validity could provide the use of a single method. Its application to the relationship between quantity and quality is questionable, since the latter is beyond the scope of measurement.
Denzin’s formulation of triangulation goes beyond a specific field of action, so it can be one of the major steps in methodological development towards a joint qualitative-quantitative approach. For Denzin, triangulation is a way to examine the validity of the conclusions that can be assumed with respect to the data; therefore, it is part of the hypothetical consistency between data.
Denzin’s Sources of Triangulation
According to Denzin, possible sources of triangulation are:
- Data
- Observations
- Theories
- Methodological triangulation, or using multiple methods in investigating the same phenomenon, with the intention of balancing the biases of each method.
The logic of triangulation is that the researcher will be more certain of their results if they use different methods, given that each has its advantages and specific biases.
Criticisms of Denzin’s Proposal
Denzin’s proposal has been criticized from three points of view:
- Determining Convergence: Determining whether there has been a convergence is a delicate exercise, especially when it comes to the convergence of qualitative and quantitative data. The expectation of convergence of the results within the validity is established under the hypothesis of observing different aspects, such as from the principle of the double quantitative and qualitative aspect of reality. Then, convergence could be taken as an expression of that, instead of a double aspect, it is one.
- Handling Discrepancies: Even if Denzin acknowledges the possibility that the results of various methods do not match, he just analyzes its consequences. It follows that everything seems destined to fall on the results. What to do when methods do not match, which he ignores? Bryman and Patton described the tendency of researchers is to rely more on qualitative evidence when this happens. The immediate experience that offers quality seems to be decisive, but this is not an answer.
- Emphasis on Externalities: Triangulation places strong emphasis on the externalities of the methods. In his view, it is the comparison between methods which generates the validity of the observation, bypassing the mediation of the observer. The very separation between different types of triangulation, such as observers and methodology, again leaves aside the methods that may or may not coincide in a synthesis that is the researcher.
The researcher is the one who metabolizes methods, and it is their practice that makes them compatible, as practical methods are developed by specific subjects.
A Fourth Critique
There is a fourth critique that views triangulation as part of an external, objective reality that can be accessed from the use of various instruments or perspectives. Thus, more complex interpretations of triangulation, like Erzberger and Prein, define it as obtaining a three-dimensional representation of the object, combining two complementary two-dimensional images. Indeed, the very concept of validity involves the acceptance of an observed social reality that itself is built on observation, as is established from the constructivist approach, is far from their budgets.