Examples
Worked examples
- Is an instance
Robert Slavin's own 1986 and 1995 papers used best evidence synthesis to review educational interventions where included studies varied too much in design and outcome measures for a single pooled statistic to be meaningful, but where a purely narrative review risked treating weak and strong studies as equally informative.
- Is an instance
New Zealand's Ministry of Education runs an ongoing 'Iterative Best Evidence Synthesis' programme, which applies the method at a larger scale to synthesize evidence on education interventions for policy use.
Counter-examples
Looks similar, but isn't
- Not an instance
A meta-analysis that computes a single pooled effect size across every eligible study, weighting each purely by sample size or inverse variance, is not a best evidence synthesis -- BES explicitly weights studies by methodological quality and applicability, not just precision.
- Not an instance
A traditional narrative review that discusses relevant studies without a stated, reproducible standard for what counts as strong evidence is not a best evidence synthesis -- BES requires an explicit, documented internal- and external-validity standard applied consistently across the included literature.
Editorial commentary
Best evidence synthesis (BES) is a review method that sits deliberately between a traditional narrative literature review and a formal meta-analysis. It was introduced by education researcher Robert Slavin as an ‘intelligent alternative’ to both: unlike a narrative review, it applies an explicit, reproducible search and quality standard; unlike a meta-analysis, it does not require pooling every eligible study into a single statistical estimate when the underlying studies are too heterogeneous in design, population, or outcome measure for that pooling to be meaningful.
The internal-validity / external-validity standard
The defining feature of BES is its quality criterion: a study counts as strong evidence only if it is high in BOTH internal validity (the study design minimizes bias, so its result can be trusted for the sample studied) AND external validity (the sample and setting are similar enough to the population the review is meant to inform that the result actually generalizes). A tightly controlled study on a narrow, unrepresentative sample and a large real-world study with weak controls are both treated as limited evidence under this standard — neither automatically outweighs the other, and the synthesis discusses why. This is a deliberate contrast with a meta-analysis, where a well-powered but narrowly-generalizable trial can still dominate a pooled estimate simply because of its precision.
How the synthesis itself is written
BES still requires a systematic, documented literature search, similar in rigor to a systematic review‘s search strategy. What differs is the synthesis step: instead of a single pooled statistic, the reviewer writes a structured narrative that explicitly weighs each study’s quality and applicability, explains disagreements between studies rather than averaging over them, and states the overall conclusion in terms of the strength and consistency of the best available evidence — not a single combined effect size with a confidence interval.
Where it is used
BES originated in, and remains most closely associated with, education research, where interventions are frequently tested across highly heterogeneous school settings, populations, and outcome measures that make a single pooled meta-analytic estimate hard to interpret meaningfully. New Zealand’s Ministry of Education has run an ongoing, large-scale ‘Iterative Best Evidence Synthesis’ programme applying the method to inform education policy, and the approach has also been used in other social-science and evaluation contexts facing the same heterogeneity problem.
Machine-readable encodings
Use in your systems
<role vocab="credit"
vocab-identifier="https://casrai.org/dictionary/"
vocab-term="Best Evidence Synthesis (BES)"
vocab-term-identifier="https://casrai.org/dictionary/term/best-evidence-synthesis-bes" />{
"@context": "https://schema.org",
"@type": "DefinedTerm",
"@id": "https://casrai.org/dictionary/term/best-evidence-synthesis-bes",
"name": "Best Evidence Synthesis (BES)",
"identifier": "https://casrai.org/dictionary/term/best-evidence-synthesis-bes",
"description": "Best evidence synthesis (BES) is a review method, originated by Robert Slavin as an explicit alternative to both traditional narrative reviews and formal meta-analysis, that combines a systematic, exhaustive literature search with a narrative interpretation focused specifically on the studies with the strongest internal AND external validity -- rather than either pooling every study statistically (as meta-analysis does) or summarizing the literature without explicit selection criteria (as a narrative review does). A study qualifies as 'best evidence' under this method when its design minimizes bias (internal validity) and its sample and setting generalize to the population the review is meant to inform (external validity); studies that are rigorous but narrow, or broad but methodologically weak, are treated as lower-quality evidence and weighted accordingly in the narrative synthesis rather than excluded outright or averaged in on equal footing.",
"inDefinedTermSet": "https://casrai.org/dictionary/domain/research-outputs#set",
"url": "https://casrai.org/dictionary/term/best-evidence-synthesis-bes",
"sameAs": [],
"license": "https://creativecommons.org/licenses/by/4.0/",
"publisher": {
"@id": "https://casrai.org/#organization"
},
"author": {
"@id": "https://casrai.org/#editorial-team"
},
"datePublished": "2026-08-31T23:35:38",
"dateModified": "2026-08-31T23:35:38",
"inLanguage": "en-GB",
"isAccessibleForFree": true
}
