Meta-Ethnography: Noblit and Hare’s Seven Phases and eMERGe Reporting

Meta-ethnography is a seven-phase method, developed by George Noblit and R. Dwight Hare in 1988, for synthesizing findings across qualitative studies through interpretation rather than aggregation. Instead of pooling numeric effect estimates the way a meta-analysis does, a meta-ethnography treats each included study’s own interpretations as data, translates those interpretations into one another, and builds a new, higher-order account that goes beyond what any single study said on its own. It is the oldest and, by most published counts, still the most widely used method of qualitative evidence synthesis. This guide covers Noblit and Hare’s seven phases in the order they are meant to be applied, the analytical concepts that make the method distinctive (reciprocal translation, refutational synthesis, line-of-argument synthesis), and the eMERGe reporting guidance that now sets the expected standard for writing one up.

Meta-ethnography is not a systematic review, a meta-analysis, or an integrative review

Meta-ethnography sits in a different family from the quantitative and mixed-design synthesis methods most research-administration guidance defaults to. A systematic review and meta-analysis typically restrict inclusion to a single comparable study design and statistically pool results against a shared risk-of-bias framework — see CASRAI’s side-by-side comparison of review types for the definitional distinctions among a narrative review, systematic review, and scoping review. Meta-ethnography instead works exclusively on qualitative, interpretive accounts — typically studies using ethnographic, phenomenological, grounded-theory, or interview-based methods — and its output is not a pooled effect size but a new interpretation, a “third-order” construct built from what the original studies’ authors already concluded.

It is also a different method from an integrative review following Whittemore and Knafl’s five stages, even though both are qualitative-synthesis-adjacent and both are used heavily in nursing and health-services research. An integrative review is defined by combining evidence from mixed methodologies — experimental, quasi-experimental, non-experimental, qualitative, and theoretical sources — in one review, precisely because that methodological breadth is the problem it is built to handle. Meta-ethnography is narrower and more specifically interpretive: it works on qualitative accounts only, and its defining move is translating each study’s own concepts and metaphors into the terms of the others, not evaluating and combining evidence across incompatible study designs. Choose an integrative review when the underlying evidence base is genuinely mixed-method; choose a meta-ethnography when the question is what a qualitative literature, taken together, actually means.

Noblit and Hare’s seven phases

Noblit and Hare set out their method in Meta-Ethnography: Synthesizing Qualitative Studies (Sage, 1988). The seven phases are iterative and overlapping in practice — reviewers routinely move back and forth between reading, comparing, and translating rather than completing each phase once and moving on — but they are meant to be entered in this order:

1. Getting started. Identify an intellectual interest or phenomenon that existing qualitative research is positioned to inform, and confirm meta-ethnography is the right method for it before committing to a full review.
2. Deciding what is relevant. Define the boundaries of the synthesis and identify and select the qualitative study accounts to include — not necessarily an exhaustive systematic search, though contemporary practice (and eMERGe, below) expects a transparent, reproducible search and selection process.
3. Reading the studies. Read the included accounts repeatedly, noting the metaphors, concepts, and interpretations each study’s authors used to explain their own findings — the “second-order constructs,” in the method’s terminology.
4. Determining how the studies are related. List the key concepts and metaphors from each study and juxtapose them against one another to see which studies address similar phenomena and how their interpretations line up or diverge.
5. Translating the studies into one another. Systematically compare each study’s concepts against the others’, treating each study as if it were being interpreted through the lens of the rest — the analytical step that gives the method its name.
6. Synthesising translations. Compare the translations produced in phase five to build a synthesis that goes beyond a list of individual translations — a genuinely new, higher-order interpretation, not just a summary of what phase five produced.
7. Expressing the synthesis. Communicate the synthesis in a form suited to its intended audience — academic, policy, or practice — which may mean a written narrative, a conceptual model, or another format depending on who the synthesis is for.

The analytical concepts that distinguish meta-ethnography

Two ideas do most of the interpretive work inside phases five and six, and a third-order/second-order/first-order distinction runs through the whole method:

• First-, second-, and third-order constructs. First-order constructs are research participants’ own words and accounts, as reported in the primary studies. Second-order constructs are the primary studies’ authors’ own interpretations of that data — their themes and concepts. Third-order constructs are what the meta-ethnographer produces: a new interpretation built from translating and synthesising the second-order constructs across studies, not simply restating them.
• Reciprocal translation. Used where studies address directly comparable phenomena and their findings can, in effect, be added to one another — each study’s concepts are translated into the terms of the others and vice versa, iteratively, until a stable set of translated concepts emerges.
• Refutational synthesis. Used where studies stand in real tension or contradiction rather than reinforcing one another. Rather than discarding the disagreement, refutational synthesis treats the refutation itself as something to interpret and explain — why studies of the same phenomenon reached different conclusions is itself a finding.
• Line-of-argument synthesis. The highest-order move: building a “whole” out of a set of translated parts, functioning as inference and theory development that goes beyond any individual study or translation — the closest analogue to what phase six, synthesising translations, is meant to produce.

The eMERGe reporting guidance

Poor and inconsistent reporting of meta-ethnographies was a long-standing, documented problem: reviewers using the method often did not report their process transparently enough for readers to judge how the seven phases had actually been carried out, which limited how much confidence journals, policymakers, and practitioners could place in the findings. eMERGe — developed by France and colleagues, funded by the UK National Institute for Health Research’s Health Services and Delivery Research programme (project 13/114/60), and published in January 2019 simultaneously in the Journal of Advanced Nursing, BMC Medical Research Methodology, Psycho-Oncology, and Review of Education — is the first reporting guideline built specifically for meta-ethnography, rather than a generic qualitative-synthesis checklist applied to it after the fact.

The guidance was developed through a mixed-methods process that began in June 2015: a methodological systematic review of existing meta-ethnography conduct and reporting guidance; an audit of published meta-ethnographies plus interviews with end-users to identify good and poor reporting practice in the wild; an international, multidisciplinary consensus-building process (a workshop plus two eDelphi studies); and, finally, drafting the guidance itself with explanatory notes. The result is 19 reporting criteria, organised under Noblit and Hare’s own seven phases rather than as a separate structure — the guidance’s developers found that phases four through six (determining how studies are related, translating them into one another, and synthesising translations) were where reporting quality was weakest and needed the most explicit clarification. Citation: France, E.F., et al. (2019). “Improving reporting of meta-ethnography: The eMERGe reporting guidance.” Journal of Advanced Nursing, 75(5), 1126–1139, doi:10.1111/jan.13809.

eMERGe is a reporting guideline, not a conduct manual — it does not replace Noblit and Hare’s original method, and it does not mandate a single way of doing meta-ethnography. What it mandates is transparency about which choices were made at each of the seven phases, so that a reader (or a journal reviewer, or a guideline developer relying on the synthesis) can judge the rigor of the work rather than take it on faith.

Frequently asked questions

Is meta-ethnography a type of systematic review?

Not in the conventional sense. It can be conducted with a systematic, reproducible search strategy, and increasingly is, but its core method — translating studies’ interpretations into one another to build a new interpretation — is fundamentally different from a systematic review’s aggregative, protocol-driven approach to combining comparable quantitative findings. Meta-ethnography sits within the broader family of qualitative evidence synthesis methods, alongside approaches such as thematic synthesis and framework synthesis.

How many studies does a meta-ethnography need?

Noblit and Hare’s original method was developed and demonstrated on a relatively small number of studies, and meta-ethnography has traditionally been considered best suited to a modest, manageable set of qualitative accounts precisely because the translation process is interpretive and labor-intensive rather than mechanical. Applying the method to a very large number of studies is possible but raises well-documented practical challenges in keeping the translation process rigorous and traceable.

What is the difference between reciprocal and refutational synthesis?

Reciprocal translation is used when studies broadly agree or address directly comparable phenomena, so their findings can be translated into a shared vocabulary and combined. Refutational synthesis is used when studies genuinely conflict, and treats explaining that conflict — rather than resolving or averaging it away — as part of the synthesis itself.

Is following eMERGe mandatory to publish a meta-ethnography?

No single reporting guideline is universally mandatory the way, say, a specific ethics approval is, but many journals publishing qualitative synthesis now expect authors to reference a recognised reporting guideline, and eMERGe is the one built specifically for this method. Following it, or explaining a deliberate departure from it, is standard practice for a methodologically credible meta-ethnography today.