What Is Triangulation, and What Is It Actually Supposed to Accomplish?

What does triangulation actually add to a research study?

Triangulation is often invoked as a sign of methodological rigor: collect data from several sources, use more than one method, compare the findings, and the study becomes more credible.

That description captures part of the idea, but it can also encourage a mechanical interpretation. Researchers may add interviews to observations, ask several stakeholder groups the same questions, or involve another researcher and then simply state that the findings were “triangulated.” What the triangulation was intended to accomplish sometimes remains unclear.

The more useful question is therefore not simply whether a study uses triangulation. It is what is being triangulated, why those particular perspectives were chosen, and what the researcher expects to learn from their relationship.

Triangulation examines a phenomenon from more than one relevant vantage point

Triangulation is more than collecting the same answer twice

Why is it called triangulation?

The term draws on the broader idea of determining or understanding something from multiple reference points. In research methodology, the metaphor has been extended to examining a phenomenon using different sources, methods, investigators, or theoretical perspectives.

In qualitative research, triangulation has often been described as a strategy for developing a more comprehensive understanding of a phenomenon and for examining the convergence of information from different sources. Methodological discussions also describe triangulation as potentially useful for increasing confidence in findings, exploring complexity, and identifying biases associated with relying on a single approach.

The geographical metaphor should not be taken too literally, however. Research triangulation does not normally produce one mathematically exact “true position.” Social phenomena can legitimately appear different depending on who experiences them, where they occur, how they are measured, and which theoretical lens is used.

There is more than one kind of triangulation

A widely cited classification associated with Norman Denzin distinguishes data, investigator, theory, and methodological triangulation. These categories remain useful for understanding what researchers mean when they say that a study is triangulated, although terminology and implementation can vary across methodological traditions.

These forms should not be treated as boxes that must all be checked. A study may use one type because it serves the research question and have no reason to use the others.

Triangulation can examine convergence

One traditional purpose of triangulation is to determine whether evidence obtained in different ways points toward a similar interpretation. If independent and appropriate sources converge, researchers may have greater confidence that a finding is not merely an artifact of one source or procedure.

Suppose students report that instructor feedback is routinely delayed, faculty members independently describe difficulty meeting feedback deadlines, and appropriately interpreted administrative records show long intervals between submission and feedback release. The convergence of these different forms of evidence may strengthen the conclusion that delayed feedback is a recurring implementation issue.

Convergence is most informative when the evidence is genuinely relevant and sufficiently independent. Three reports derived from the same administrative database are not three independent confirmations merely because they appear in different documents.

Triangulation can provide complementarity

Different sources and methods often illuminate different aspects of a phenomenon rather than reproducing the same information.

Observation might show that instructors rarely use a particular educational technology during class. Interviews could reveal why: perhaps instructors consider it useful but lack reliable classroom connectivity. Institutional documents might establish that the technology is officially encouraged but not required.

The three forms of evidence do not answer the same question. Together, however, they may produce a more comprehensive account of implementation than any one source could provide alone.

This is why deciding whether to collect evidence from more than one source should begin with what each source can contribute, not with a requirement that all sources repeat the same measurement.

Triangulation can reveal disagreement

A common but limiting assumption is that successful triangulation means all findings agree. Sometimes disagreement is exactly what makes the triangulation informative.

Imagine administrators reporting that a new research policy is well understood across a university, while faculty interviews reveal widespread uncertainty. Rather than treating the faculty accounts as a failure to confirm the administrative view, the researcher might investigate the discrepancy itself. The disagreement could reveal differences between policy intention, communication, interpretation, and implementation.

Discordant findings can arise because sources occupy different positions, because the phenomenon varies across settings or time, because measures capture different constructs, or because one or more sources contain error. Triangulation creates an opportunity to investigate these possibilities. It does not provide an automatic formula for deciding which account wins.

Triangulation can expose method-specific blind spots

Every data collection method makes some aspects of a phenomenon easier to see and others harder to see. Interviews can provide detailed accounts of participants' interpretations but rely on what participants can and will articulate. Observation can document behavior in context but may provide limited access to intentions or private experiences. Administrative records can capture recorded events while omitting the meanings surrounding them.

Combining methods may therefore help researchers identify findings that depend heavily on one method's characteristics. This is one reason data triangulation and methodological triangulation should be distinguished: varying the source is not the same as varying the way evidence is generated.

Triangulation does not remove bias

Researchers sometimes write as though triangulation “eliminates bias.” That is too strong. Multiple sources or methods can help expose some method-specific assumptions or weaknesses, but they can also share biases.

Two self-report instruments may both be affected by social desirability. Several stakeholder groups may share the same institutional assumptions. Two analysts may approach data from similar theoretical positions. Different methods may also depend on the same underlying operational definition.

Triangulation can therefore contribute to credibility or validity when thoughtfully designed, but it does not make evidence bias-free.

Triangulation is not simply another name for mixed methods

Triangulation can occur entirely within qualitative research. A qualitative study might compare interviews, observations, and documents or obtain accounts from several participant groups. Investigator or theory triangulation likewise does not require quantitative data.

Mixed-methods research, by contrast, involves substantive qualitative and quantitative components and their purposeful integration. Triangulation may be one rationale for combining components in a mixed-methods design, but the concepts are not synonymous.

Likewise, having both numerical and textual material does not by itself establish mixed-methods research.

More triangulation is not automatically better

A study does not become progressively stronger every time another source, method, investigator, or theoretical perspective is added. Each addition creates methodological and practical consequences.

An unnecessary second method may increase participant burden. Several theoretical perspectives can produce conceptual clutter if their relationship is poorly developed. Multiple analysts can generate additional interpretations without improving rigor if there is no clear process for using those interpretations.

The relevant question is whether the additional vantage point addresses a meaningful limitation, provides necessary evidence, tests an important interpretation, or expands understanding of the phenomenon. The assumption that more sources or methods automatically make evidence stronger should therefore be resisted.

Plan what you will do when the evidence does not agree

Triangulation is easiest to describe when everything converges. The methodological work becomes more interesting when it does not.

Before data collection, consider what different patterns might mean. If two sources disagree, are they supposed to measure exactly the same phenomenon? Do they refer to the same period and setting? Is one reporting perception while another records behavior? Could different participants legitimately experience the phenomenon differently?

A useful triangulation strategy therefore specifies not only what will be compared but how convergence, complementarity, and divergence will be interpreted. Otherwise, researchers risk celebrating agreement while treating inconvenient disagreement as noise.

Triangulation can reveal the gap between policy and practice

The purpose of triangulation here is not to make three sources vote on which account is true. It is to understand the phenomenon more adequately by examining it from positions that have different access to the policy and its implementation.

Common misconceptions about triangulation

State what you are triangulating and why

If you plan to claim triangulation in a proposal or manuscript, you should be able to explain more than the fact that several forms of evidence were collected. Identify the dimension that varies and what examining that variation contributes to the study.

Most importantly, decide what you will do with disagreement before it appears. Triangulation is more defensible when conflicting evidence is treated as something to understand rather than something to remove until the findings look consistent.

Before claiming that your study uses triangulation

Frequently asked questions about triangulation

Triangulation should create insight, not merely repetition

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.