Evidence Synthesis Professional Competency and Career Roadmap

Knowing a method is not the same as being ready to own the decision

By Dr. Esmaeel Saeedy Robat , Founder of MetaSyn Academy · Published meta-analyst (Nature Human Behaviour, 2026)

Evidence checked: 12 August 2026

This roadmap helps you make a more disciplined claim about your own evidence-synthesis capability. Instead of producing a percentage or telling you that you are “job ready,” it asks what responsibility you want to take on, what you have actually done, what evidence supports that claim, and what you should develop next.

Evidence synthesis has well-developed standards, but there is no single psychometrically validated competency scale that applies to every reviewer, information specialist, statistician, methodologist, educator, consultant and HTA professional. Existing frameworks are useful precisely because they are more bounded. Townsend and colleagues developed six competency domains for librarians involved in systematic reviews and explicitly allowed different levels of involvement according to role and institutional context.1 CABI’s Evidence Synthesis Skills Framework is broader in occupational scope and was designed to support career planning and training-needs analysis, but it is positioned primarily within the agrifoods sector.2

That leaves a practical problem. Researchers still need a way to ask: What can I do now? What can I do only with supervision? Which responsibility am I prepared to own independently? Where should I involve a specialist? A useful developmental resource can support that reflection without pretending to be a licensing examination.

If you are still deciding which professional responsibility fits your background, start with Careers in Evidence Synthesis: Roles, Skills and Professional Pathways . This resource begins one step later: after you have a direction in mind, it helps you examine the evidence behind your own capability claims.

Why this is a self-audit, not a competency test

Self-assessment can support reflection and identify perceived learning needs, but it should not be treated as an objective measure of performance. A systematic review by Davis and colleagues found limited agreement between physicians’ self-assessments and external measures of competence, with the least skilled practitioners often being among the least accurate self-assessors.3 A broader meta-analysis of self-assessed knowledge in education and workplace training likewise found that self-ratings related more strongly to affective constructs such as motivation and satisfaction than to measured cognitive learning.4

The implication is not that reflection is useless. It is that the question needs to change. “How good am I at meta-analysis?” invites a vague confidence judgment. “Which meta-analytic decisions have I made independently, what output records those decisions, and who has reviewed that work?” is much harder to answer casually.

Four levels describe responsibility, not prestige

Miller’s classic assessment pyramid distinguishes knowledge from demonstrated performance.5 Townsend and colleagues adapted that logic for systematic-review librarians, but their framework did not prescribe required levels and did not provide a validated test for assigning them.1 This roadmap therefore uses simpler workplace language. The labels are transparent design anchors for reflection, not psychometric categories.

How the roadmap interprets a gap

The tool deliberately does not know what a “methodologist” or “consultant” must score. No defensible universal target exists. Instead, you define the responsibility you intend to take on within each domain. The roadmap then compares that intention with your current self-reported level.

If you select independent responsibility for searching but record only foundational awareness, the tool identifies a development priority. If you intend to lead a complex quantitative synthesis but currently practise under supervision, it can flag the gap and make specialist collaboration a reasonable next action. That is not a verdict about your career. It is a structured comparison between two claims you made yourself.

Evidence should match the claim you are making

A course certificate can demonstrate that structured learning occurred. It does not by itself show that the learner can resolve an unfamiliar methodological problem independently. A publication demonstrates participation in a research output, but a byline alone does not reveal which tasks each author performed. CRediT helps make contributions more explicit by describing 14 contributor roles, including methodology, formal analysis, data curation, software, validation and supervision.6

That is why this resource asks you to record evidence next to the capability claim. The categories are not a validated hierarchy. They are prompts that make the claim more inspectable: training, supervised work, independent artifacts, or leadership and quality-assurance evidence. A Level 3 or Level 4 self-rating without any inspectable evidence should trigger reflection rather than confidence.

When the audit reveals a methodological gap, go back to the method

This roadmap identifies development needs. It does not reteach the underlying review methods. If your profile exposes a specific gap, move back into the Academy’s methodological resources and work on that task directly.

AI literacy belongs in the profile, but prompting is not the competency

Current cross-organization guidance treats responsible AI use in evidence synthesis as a question of suitability, evaluation, transparency, oversight and accountability rather than simple access to a tool. Cochrane’s RAISE work also emphasizes that authors remain accountable for final content and that AI use should not compromise methodological rigor or research integrity.78

For this roadmap, basic AI literacy therefore means being able to ask whether automation is suitable for a task, understand relevant limitations, verify consequential outputs, protect confidential information and document use appropriately. Building custom models or automated evidence pipelines is specialist capability. Knowing a collection of prompts is not, by itself, evidence of either.

Use the profile as a cycle, not a verdict

Competence is not permanently acquired at one date. Methods change. Reporting standards change. Software changes. New review designs create unfamiliar decisions. AI changes the distribution of work. Professional development therefore works better as a cycle: choose a responsibility, inspect your current evidence, identify one meaningful gap, develop or collaborate, obtain feedback, and then review the claim again.

A useful roadmap should sometimes tell you not to become the specialist

One of the most important outputs from a professional audit is not “learn everything.” Evidence synthesis is collaborative. Advanced information retrieval, complex statistical modelling, qualitative synthesis, HTA methods, measurement science and automated evidence systems can each require deep specialist capability. A generalist can understand why that expertise matters without claiming to possess it.

If a domain is central to the project but peripheral to your intended professional role, the right development action may be collaborate with a specialist. That is not a failed competency. It is often evidence of professional judgment.

The same principle protects the roadmap from becoming another credential checklist. You do not need Level 4 everywhere. You need a defensible match between the responsibility you intend to own, the capability you can demonstrate, and the expertise available around you.