Triangulation

Triangulation is a research strategy that uses multiple methods, data sources, investigators, or theories to study the same phenomenon, with the goal of enhancing the credibility and comprehensiveness of the findings. The term is borrowed from surveying and navigation, where the location of a point is determined by measuring it from two or more known positions. In research, triangulation operates on the principle that no single method or data source can fully capture the complexity of a phenomenon, and that convergence of findings from different approaches provides stronger evidence than any single approach alone. There are four main types: data triangulation (using multiple data sources), methodological triangulation (using multiple research methods), investigator triangulation (involving multiple researchers), and theoretical triangulation (applying multiple theoretical perspectives).

Every research method has inherent strengths and limitations. Surveys can measure prevalence across large populations but may miss the nuances of individual experience. Interviews capture rich detail but are limited to small samples. Observational data reflects actual behavior but may miss the reasons behind it. By combining approaches, triangulation compensates for the weaknesses of any single method while capitalizing on its strengths. When findings from different methods converge, confidence in the conclusions increases. When they diverge, the differences themselves are informative, pointing to aspects of the phenomenon that require further exploration or to methodological factors that influence results. Triangulation is central to establishing trustworthiness in qualitative research and to strengthening the validity of mixed methods studies.

Data triangulation involves collecting data from different sources, time points, or settings. For example, a study of patient safety might combine incident reports, patient surveys, and observational field notes. Methodological triangulation combines different methods, either within one approach (such as two different questionnaires) or across approaches, most often qualitative and quantitative, such as using survey data to establish the prevalence of a phenomenon and interviews to explore its meaning and context. Investigator triangulation involves having multiple researchers independently code data, analyze results, or interpret findings, reducing the influence of any single researcher's perspective. Theoretical triangulation examines data through different theoretical lenses, such as analyzing health behavior through both rational choice theory and social ecological theory. In practice, the researcher specifies the purpose of triangulation (convergence, testing whether different approaches corroborate the same finding, or a more complete picture, in which each approach captures a different facet of the phenomenon), collects data through the multiple approaches, analyzes each dataset according to its own method, and then integrates the findings through comparison and synthesis.

Use triangulation when studying complex phenomena that cannot be fully captured by a single method, when the validity of findings is critical for policy or practice decisions, when seeking to enhance the trustworthiness of qualitative research, or when different stakeholders may have different perspectives on the same issue. Data triangulation is appropriate when the phenomenon manifests differently across contexts or over time. Methodological triangulation is valuable in mixed methods research where the research questions require both breadth and depth. Investigator triangulation is important in team based research and when researcher bias is a concern. Theoretical triangulation is useful when competing theories offer different explanations for the same phenomenon.

A common mistake is treating triangulation as automatic validation, assuming that if multiple methods are used, the findings must be correct. Convergence strengthens evidence but does not prove it; all methods could share the same systematic bias. Another error is using triangulation superficially, collecting data through multiple methods but analyzing them separately without genuine integration or comparison. Researchers sometimes confuse triangulation with mixed methods research; triangulation is one possible purpose of mixed methods, but not all mixed methods studies employ triangulation. Ignoring divergent findings across methods, rather than exploring and explaining them, wastes one of triangulation's most valuable contributions. Finally, adding methods purely to claim triangulation, without a clear rationale for how each method contributes to understanding the phenomenon, adds complexity without analytic value.

A mixed methods study evaluates the implementation of a falls prevention program in residential aged care facilities. The quantitative component tracks fall rates before and after program implementation across 15 facilities (data triangulation across sites), finding a 23% reduction. Semi structured interviews with 30 care staff explore barriers and facilitators (methodological triangulation), in which staff described ownership of the program as central to sustained adherence. An independent researcher recodes a subset of the interview transcripts (investigator triangulation), achieving 91% agreement on the major themes. The researchers interpret the findings through both implementation science and organizational culture theories (theoretical triangulation), noting that implementation science explains the process variables while organizational culture theory better accounts for the variation in outcomes across facilities.