Chapter 28: Triangulation

What is triangulation?

Triangulation is the combination or blending of more than one participant group, researcher, theory and/or method in the same research. Its purpose is to understand the phenomenon under study¹ by determining consistency, or ‘truth’.¹ Triangulation can be used to demonstrate the rigour, validity and credibility of research findings.² While the purpose of triangulation is not to confirm results, but rather to understand differences, it can be difficult to explain inconsistent results when discussing the research undertaken.

There are four main types of triangulation²:

• Theoretical triangulation is the use of more than one theory to guide the research process. For example, a researcher might analyse data on family violence by applying feminist and critical theory, and they might also apply structural functionalist theory (see Section 1) when examining family violence as part of a complex system. By applying different theories, the data is able to be interrogated through theoretical lenses, which can lead to deeper understanding of the findings and greater nuance than a single theory might support.
• Researcher triangulation is the use of multiple (two or more) researchers to collect and / or analyse data. The researchers may have different disciplinary backgrounds and experiences, and will also bring their professional and personal interpretations to the data. For example, research approaches to consumer and community involvement (or patient and public involvement) might advocate for patients to be involved in the analysis of data, to include patient perspectives in the interpretation of the data. In a study developing a ‘BroSupPORT’ portal and examining issues facing men with prostate cancer,³ researchers found that health professionals were not sure that a Patient Reported Outcome comparator tool would be helpful in prompting health-seeking behaviour, but participants with prostate cancer welcomed such a tool. Focusing a patient lens on data in this study was important because it was able to highlight differences between perspectives of health professionals and patient participants. If only health professionals had been consulted the tool would not have been considered helpful and would have been ruled out as an option for the portal.
• Methodological triangulation is the use of multiple (two or more) methods to collect and analyse data. The data collection methods might include focus groups, interviews, photovoice, observations, field notes and more. In essence, it is bringing together the various methods used to collect data and can provide a more nuanced explanation of results. Methodological triangulation can include quantitative methods to support or harmonise results. Using quantitative and qualitative methods together enables the research to answer the questions of ‘what’ and ‘why’ (see Chapter 11: Mixed Methods). The BroSupPort portal study³ is a good example of methodological triangulation because it used a combination of workshops, interviews and focus groups to collect data.
• Data triangulation uses more than one data source and / or method of analysis to interrogate the data. Data sources may include interviews with people in a range of roles in an organisation, rather than only those in one particular role. Data analyses might include data from both inductive and deductive perspectives. Data triangulation might also include different data sources, such as qualitative (e.g. interviews) and quantitative (e.g. surveys). In the BroSupPORT portal study³ data were gathered at workshops, focus groups and interviews. Surveys, mind maps, River of Life activities and problem trees (in printed form), along with field notes taken at each workshop, were used to collect data. A range of techniques was used to analyse the data including, but not limited to, descriptive content analysis.

Table 28.1 provides examples of the four main types of triangulation. Other types of triangulation, such as ‘time’ and ‘space’³, are not covered in this chapter because they are used less often.

Table 28.1: Examples of triangulation

How to conduct triangulation

How triangulation is conducted depends on the type of triangulation.

• Theoretical triangulation requires an introduction to each theory and can be written as a literature review. The theories are described and then compared, to elicit inferences that will form the basis of data interpretation. For example, a feminist theory will inform data collection in such a way that girls and women (and women’s marginalised groups) will be deliberately sought out and included in the research study. Analysis would include a focus on gender identity, patriarchal oppression, diversity of culture and background, and would seek to demonstrate women’s points of view through a feminist lens. If, for example, a study is about women patients, the data collection and analysis would focus on how or whether women are represented in the data, and how women are medically treated by healthcare practitioners. Women’s own perspectives would be sought and analysed, to understand their perspectives.
• Researcher triangulation is often described in the type of data being analysed, and can often be read in the researcher’s statement of positionality or in the reflexivity section of a journal paper or report⁹. Some forms of thematic analysis (not reflexive thematic analysis) requires more than one investigator to read, re-read, code and re-code interviews or focus groups. When it is not a requirement of the method of analysis, triangulation should still be considered, in order to address concerns about the rigour, validity and credibility of findings of a single researcher. Including more than one researcher and participant can leads to greater divergence and the potential for nuanced findings.
• Methodological triangulation is used often in the literature. A decision is made about how to conduct the research, on the basis of the research question or aim. Often in mixed methods research, a qualitative component seeks to answer the question, ‘Why?’ and the quantitative component seeks to test a hypothesis or answer the question, ‘What?’. However, many qualitative methods might be included, such as interviews, focus groups, newspaper clippings, to answer the research question(s). When using methodological triangulation, the researcher is looking to expand their understanding of the findings. For example, if a survey and interviews are the mixed methods used in a study, the researcher would seek to compare and contrast the findings of both methods, to gain a comprehensive understanding of the phenomenon, and then would describe how the findings support or diverge in answering the research question(s). Thus, a study exploring barriers and enablers in the implementation of the 6-PACK falls prevention program¹⁰ incorporated a cluster randomised control trial, economic and program evaluations, and surveys and focus groups. The findings were triangulated and results suggested that regular, practical face-to-face education and training for nurses were key to successful falls prevention program implementation in acute hospitals, as were provision of equipment; audit, reminders and feedback; leadership and champions; and the provision of falls data .
• Data triangulation involves using and analysing more than one participant group. It is often considered an aspect of methodological triangulation because different methods usually involve more than one source of data. Data collection needs to be well-defined and conducted. Once the data from all participant groups has been examined, the findings are compared and contrasted to assist in answering the research question(s).

It’s important to remember that triangulation can involve more than one type of triangulation, and this is often the case with mixed-methods research. For example, in mixed-methods research, methodological, investigator and data triangulation may be used to demonstrate the full findings of the research. While Table 28.1 has listed each type separately, examining some of the example papers will show that there is more than one type of triangulation in the studies. Strict adherence to only one triangulation type can make researching the phenomenon more difficult.

Advantages and challenges of triangulation

Comparing and contrasting theories, data sources, methods and data analyses can ensure strong reliability and validity in research results. However, this can also be time-consuming and resource-intensive. Attention needs to be paid to the nuances of the research, to provide holistic explanations. There are times when triangulation may not be considered necessary, and this also needs to be understood when addressing the research question. For example, if the purpose of the research is to develop a new theory, there may be no need to include more than one method, data point or theoretical foundation.

Summary

Triangulation is the use of more than one data source, investigator, theory or method in the same research. There are four main triangulation types: each provides a means for examining the research from different perspectives and for ensuring the rigour, validity and credibility of findings.

References

1. Patton MQ. Enhancing the quality and credibility of qualitative analysis. Health Sciences Research. 1999:34, 1189-1208.
2. Denzin NK. Sociological Methods: A Source Book (2nd ed). Mcgraw-Hill: 1978.
3. Shemesh B et al. Codesigning a patient support portal with health professionals and men with prostate cancer: an action research study. Health Expect. 2022:25, 1319-1331. doi/10.1111/hex.13444
4. Cohen L, Manion L, Morrison K. Research Methods in Education. Routledge; 2017.
5. Yeh A, Trang P. Avicii’s S.O.S.: a psychobiographical approach and corpus-based discourse analysis on suicidal ideation. Psychology of Language and Communication. 2022;26(1):207-241. doi.10.2478/plc-2022-0010
6. Lundell S, Pesola U et al. Groping around in the dark for adequate COPD management: a qualitative study on experiences in long-term care. BMC Health Serv Res. 2020;20:1025. doi.10.1186/s12913-020-05875-2
7. McCrone L & Kingsbury M. Combining worlds: a mixed method for understanding learning spaces. Int J Qual Methods. 2023;22. doi:10.1177/16094069231173781
8. Johnson M, O’Hara R et al  Multiple triangulation and collaborative research using qualitative methods to explore decision making in pre-hospital emergency care. BMC Med Res Methodol. 2017;17(11). doi:10.1186/s12874-017-0290-z
9. Llewellyn-Beardsley J et al “Nothing’s changed, baby”: how the mental health narratives of people with multiple and complex needs disrupt the recovery framework. SSM – Ment Health. 2023;3(100221). doi:10.1016/j.ssmmh.2023.100221
10. Ayton D et al. Barriers and enablers to the implementation of the 6-PACK falls prevention program:  pre-implementation study in hospitals participating in a cluster randomised controlled trial. PLOS ONE. 2017;12. doi:10.1371/journal.pone.0171932