Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one
Contents

Academy · Research Data, Analytics & AI · MetaLab Meta-Analysis Tools

MetaLab: Two Distinct Meta-Analysis Tools Sharing the Same Name

“MetaLab” refers to two unrelated research tools: a crowdsourced, continuously updated meta-analysis database for developmental psychology, and a separate MATLAB-based toolbox designed for meta-analysis of dense basic-science data. Confusion between the two may direct a researcher to an inappropriate tool.

Pubrica AcademyWritten by a research data analyst Reviewed bya meta-analysis methodologist Published10 October 2026 Last checked10 October 2026 Reading time7 minutes Sources10 peer-reviewed

MetaLab is a name shared by two distinct pieces of meta-analysis software, built for different fields, by different research teams, for different kinds of underlying data. The first is a web-based, crowdsourced, continuously updated database of meta-analyses in developmental psychology and language acquisition research, built to let researchers rerun and extend existing meta-analyses as new studies are published. The second is a MATLAB-coded toolbox built specifically for meta-analyzing the kind of dense, heterogeneous quantitative data produced in basic cell and molecular biology research, where studies rarely use the same experimental model or measurement units. Neither tool is more “correct” than the other; a researcher should establish which one matches the relevant field and data before choosing either.

Learning objectives

  • The two distinct tools that share the name MetaLab, and which fields each one serves
  • Why basic-science meta-analysis needed a purpose-built MATLAB toolbox rather than standard meta-analysis software
  • How to determine from the data which MetaLab, if either, suits a given use case

Common misconceptions

  • “MetaLab is a single piece of software, so any description of 'what MetaLab does' applies regardless of context.” The crowdsourced MetaLab built for developmental psychology (metalab.stanford.edu) and the MATLAB-coded meta-analysis toolbox built by Mikolajewicz and Komarova for basic life-science research are two separate, unrelated tools published by different research groups for different kinds of data, and conflating them may direct a researcher toward software that does not fit their field.

MetaLab as a crowdsourced developmental-psychology database

The better-known MetaLab in cognitive and developmental science is a web platform, hosted at metalab.stanford.edu, that stores standardized, continuously updatable meta-analyses of findings in infant and child development, with a strong concentration in language acquisition research. Bergmann and colleagues describe the platform's purpose in their 2018 paper in Child Development as addressing a specific replicability problem in the field: individual developmental studies often have small samples because testing infants and young children is logistically difficult, so any single study's effect size estimate is imprecise, and a static, one-off meta-analysis quickly goes out of date as new studies are published1. MetaLab's database structure allows a meta-analysis to be rerun automatically as new effect sizes are contributed by the research community, rather than requiring a fresh literature search and full re-analysis each time, and it is explicitly designed to be used by researchers planning new studies to run power analyses against the cumulative effect size estimate rather than an outdated single-paper number1.

MetaLab as a MATLAB toolbox for basic-science meta-analysis

A separate and unrelated tool, also named MetaLab, is a meta-analytic toolbox coded in MATLAB R2016b, published by Mikolajewicz and Komarova in their 2019 paper in Frontiers in Physiology2. This MetaLab implements a specific methodology the authors developed for basic life-science research, where individual studies rarely share a common experimental model, measurement unit, or outcome scale, in contrast to clinical research, where outcomes are often standardized enough for conventional meta-analysis software to pool directly2. The toolbox is distributed through a public code repository for academic and statistical use, and the paper demonstrates its application with worked examples including intracellular ATP concentrations in osteoblasts and ATP release kinetics in mechanically stimulated mammalian cells, illustrating the kind of dense, heterogeneous quantitative data it is built to handle2.

Rationale for a dedicated meta-analysis approach to basic-science data

Mikolajewicz and Komarova's paper explains the gap their MATLAB toolbox addresses: basic life-science research produces a dense volume of quantitative data, but the field has made comparatively little use of rigorous quantitative synthesis across studies, unlike clinical research where systematic review and meta-analysis are standard practice2. Part of the reason is methodological: basic-science studies routinely differ in experimental model (cell line, animal model, in vitro versus in vivo), measurement technique, and units, which conventional meta-analysis software built around standardized clinical effect sizes is not designed to harmonize. Their published methodology provides both a theoretical framework for handling this kind of heterogeneity and the computational tools, packaged as the MATLAB MetaLab toolbox, to carry out a quick review or a full systematic review followed by meta-analysis on this kind of data, with the stated goal of producing pooled estimates of physical outcomes and their variability that can support hypothesis generation, study design, and computational modeling2.

Position of each MetaLab within the meta-analysis ecosystem of its field

Neither MetaLab exists in isolation from its field's wider methodological infrastructure. The developmental-psychology MetaLab is closely linked to the ManyBabies Consortium, a large-scale collaborative replication initiative in infant research that runs coordinated multi-lab studies partly to generate the kind of large, standardized effect-size data the MetaLab database is built to store and continuously update3. Lewis and colleagues' 2020 review in the Annual Review of Developmental Psychology describes this ecosystem directly, framing MetaLab's crowdsourced database and ManyBabies' coordinated replications as complementary responses to the same underlying problem (small, imprecise individual effect sizes in infant research), approached from the synthesis side and the data-generation side respectively3. The basic-science MetaLab, by contrast, sits alongside more general-purpose statistical meta-analysis tools such as the metafor package for R, a widely used library for fitting meta-analytic models that is not specific to any one field, but which does not include the specific heterogeneity-handling methodology Mikolajewicz and Komarova built for basic-science data with inconsistent experimental models and units2.

Practical introduction to each tool

For the developmental-psychology MetaLab, the usual first step is to browse the hosted database by topic area, such as infant-directed speech preference or statistical word segmentation, to determine whether a relevant meta-analysis already exists, and then either to query its underlying data for a power analysis or, where a research question is not yet covered, to contribute new effect sizes through the platform's documented submission process so that the relevant meta-analysis can be extended rather than redone from scratch1. For the basic-science MATLAB toolbox, the first steps are to confirm access to MATLAB R2016b or a compatible later version, to obtain the toolbox from its public code repository as described in Mikolajewicz and Komarova's paper, and to work through the paper's own worked examples, such as the intracellular ATP concentration dataset, before applying the same workflow to a new dataset, since the paper is written explicitly as a practical guide alongside the software itself rather than a purely theoretical methods paper2.

The replicability problem that motivated the developmental-psychology MetaLab

The specific replicability concern Bergmann and colleagues describe merits explicit statement, since it is the clearest justification for building a continuously updatable database rather than a series of one-off published meta-analyses. Infant and young-child studies are expensive and slow to run relative to studies with adult participants, which keeps individual sample sizes small, typically a few dozen infants per study, and small samples produce individual effect-size estimates with wide confidence intervals and a correspondingly higher chance that any single published study's result is a poor estimate of the true underlying effect4. A conventional, one-off meta-analysis pools the literature available at one point in time and then becomes outdated as new studies accumulate, requiring a fresh full re-analysis to stay current, which few research teams have the incentive or resources to repeat on a regular schedule. Lewis and colleagues' later synthesis of this literature argues that MetaLab's crowdsourced, continuously updatable structure directly addresses this problem of obsolescence, giving the field a living reference point for statistical power calculations that stays current as the underlying evidence base grows, rather than freezing the field's best estimate at whatever point a single research team last published a formal meta-analysis3.

Determining which tool, if either, suits a given dataset

The practical question for a researcher who encounters the name MetaLab is which field and data type they are working with, since the two tools do not overlap in function. A developmental psychology, cognitive science, or language-acquisition researcher looking to locate or contribute to an existing meta-analysis, or to run a power analysis against a cumulative effect size, should look to the Stanford-hosted crowdsourced platform1. A basic life-science researcher working with heterogeneous experimental data (different models, measurement techniques or units across the studies being synthesized) and needing a method and toolbox built specifically for that kind of harmonization, should instead look to Mikolajewicz and Komarova's MATLAB-based MetaLab and its accompanying methodology paper2. Neither tool is a general-purpose substitute for standard meta-analysis software such as RevMan or metafor in R for conventional clinical effect-size pooling, which typically rely on established random-effects models and heterogeneity statistics such as DerSimonian and Laird's method and Higgins and Thompson's I² statistic678; both MetaLab tools were built to solve a specific gap in their own field, not to replace this general-purpose toolkit. Cumming's accessible guide to effect sizes, confidence intervals, and meta-analysis is a useful starting reference for researchers less familiar with these underlying statistical concepts, regardless of the software ultimately used9.

Key points

Key points

MetaLab is not a single tool: the name refers to a crowdsourced developmental-psychology meta-analysis database (metalab.stanford.edu) and, separately, to a MATLAB-coded toolbox for meta-analysis of heterogeneous basic-science data. The field and data type should be confirmed before either is adopted.

Practical application

Practical application

  1. The field and data type should be identified before a search for MetaLab is undertaken. Researchers in developmental psychology or language acquisition are likely to require the crowdsourced database; those in basic cell or molecular biology with heterogeneous experimental data are likely to require the MATLAB toolbox.
  2. The methodology paper for the selected MetaLab should be consulted. Bergmann et al. (2018) describes the crowdsourced platform's purpose and structure; Mikolajewicz and Komarova (2019) describes the MATLAB toolbox's methodology and worked examples.
  3. Neither MetaLab should be assumed to replace standard meta-analysis software for conventional clinical data. Both tools were built for specific gaps in their own fields; conventional clinical effect-size pooling is generally still better served by established software built for that purpose.

Frequently asked questions

No. They are independent projects built by different research groups for different fields; the shared name appears to be coincidental rather than indicating any relationship between the platforms.

No. It is structured specifically around standardized effect sizes in developmental and cognitive science studies and is not built to harmonize the heterogeneous experimental models and measurement units typical of basic life-science research.

The toolbox is distributed through a public code repository described in Mikolajewicz and Komarova's 2019 paper for academic and statistical use; the repository's current license terms should be checked directly before use, as these can change over time.

Because it is coded in MATLAB R2016b, running it requires access to MATLAB or a compatible environment; this is a meaningful practical consideration compared with the web-based, no-installation-required crowdsourced platform.

References

  1. Bergmann C, Tsuji S, Piccinini PE, Lewis ML, Braginsky M, Cristia A, Frank MC. Promoting Replicability in Developmental Research Through Meta-analyses: Insights From Language Acquisition Research. Child Dev. 2018;89(6):1996-2009. read 10 October 2026. View on Child Development
  2. Mikolajewicz N, Komarova SV. Meta-Analytic Methodology for Basic Research: A Practical Guide. Front Physiol. 2019;10:203. read 10 October 2026. View on Frontiers in Physiology
  3. Lewis ML, Braginsky M, Tsuji S, et al. A Quantitative Synthesis of Early Language Acquisition Using Meta-Analysis. Annu Rev Dev Psychol. 2020;2:57-77. read 10 October 2026. View on Annual Review of Developmental Psychology
  4. Frank MC, Bergelson E, Bergmann C, et al. A Collaborative Approach to Infant Research: Promoting Reproducibility, Best Practices, and Theory-Building. Infancy. 2017;22(4):421-435. read 10 October 2026. View on Infancy (ManyBabies founding paper)
  5. Viechtbauer W. Conducting Meta-Analyses in R with the metafor Package. J Stat Softw. 2010;36(3):1-48. read 10 October 2026. View on Journal of Statistical Software
  6. Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med. 2002;21(11):1539-1558. read 10 October 2026. View on Statistics in Medicine
  7. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188. read 10 October 2026. View on Controlled Clinical Trials
  8. The Cochrane Collaboration. Review Manager (RevMan), Version 5.4. read 10 October 2026. View RevMan software page
  9. Cumming G. Understanding the New Statistics: Effect Sizes, Confidence Intervals, and Meta-Analysis. New York: Routledge; 2012. read 10 October 2026. View on Routledge
  10. Harrer M, Cuijpers P, Furukawa TA, Ebert DD. Doing Meta-Analysis with R: A Hands-On Guide. Boca Raton: Chapman & Hall/CRC Press; 2021. read 10 October 2026. View on Taylor & Francis

Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.

How to cite this article

Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.

Pubrica Academy. (2026, October 10). MetaLab: Two Distinct Meta-Analysis Tools Sharing the Same Name. Pubrica. https://pubrica.com/academy/concepts-definitions/what-is-metalab-software-in-matlab/

This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.

Which MetaLab Fits My Project?

Answer these before deciding which MetaLab, if either, is the right tool.

[ ] Field: developmental/cognitive/language-acquisition psychology, or basic life science (cell/molecular biology)?
[ ] Data type: standardized psychological effect sizes, or heterogeneous experimental measurements across models/units?
[ ] Goal: locate/extend an existing meta-analysis and run a power analysis, or harmonize dense heterogeneous basic-science data?
[ ] Environment: need a no-installation web platform, or have access to MATLAB?
[ ] Read the matching methodology paper (Bergmann et al. 2018 or Mikolajewicz & Komarova 2019) before starting

Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.

How Pubrica can help

Pubrica’s data and meta-analysis team assists researchers in selecting and applying the meta-analysis methodology and tooling appropriate to their field and data structure, rather than defaulting to the first tool encountered.

Find something else

Search the Pubrica Academy

Access is free and requires no account. Enter a topic, or open the full library to filter by category, sub-topic or resource type.

Selection of the MetaLab tool appropriate to the research field

A scoping assessment is free and takes two business days. It provides a written assessment of what the work requires and which specialist would undertake it.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.