MétaCan
Menu
Back to cohort
Record W4200389440 · doi:10.1101/2021.12.06.21267383

Data and code availability statements in systematic reviews of interventions were often missing or inaccurate: a content analysis

2021· preprint· en· W4200389440 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenuemedRxiv · 2021
Typepreprint
Languageen
FieldDecision Sciences
TopicMeta-analysis and systematic reviews
Canadian institutionsOttawa HospitalUniversity of Ottawa
FundersNational Health and Medical Research CouncilMedical Research CouncilAustralian GovernmentUniversity of Ottawa
KeywordsComputer scienceCitationSystematic reviewScopusCode reviewInformation retrievalDownloadSample (material)Raw dataData fileData scienceWorld Wide WebSoftwareMEDLINESoftware qualityDatabaseSoftware development

Abstract

fetched live from OpenAlex

Objectives: To estimate the frequency of data and code availability statements in a random sample of systematic reviews with meta-analysis of aggregate data, summarise the content of the statements and investigate how often data and code files were shared. Methods: We searched for systematic reviews with meta-analysis of aggregate data on the effects of a health, social, behavioural or educational intervention that were indexed in PubMed, Education Collection via ProQuest, Scopus via Elsevier, and Social Sciences Citation Index and Science Citation Index Expanded via Web of Science during a four-week period (between November 2nd and December 2nd, 2020). Records were randomly sorted and screened independently by two authors until our target sample of 300 systematic reviews was reached. Two authors independently recorded whether a data or code availability statement (or both) appeared in each review and coded the content of the statements using an inductive approach. Results: Of the 300 included systematic reviews with meta-analysis, 86 (29%) had a data availability statement and seven (2%) had both a data and code availability statement. In 12/93 (13%) data availability statements, authors stated that data files were available for download from the journal website or a data repository, which we verified as being true. While 39/93 (42%) authors stated data were available upon request, 37/93 (40%) implied that sharing of data files was not necessary or applicable to them, most often because "all data appear in the article" or "no datasets were generated or analysed". Discussion: Data and code availability statements appear infrequently in systematic review manuscripts. Authors who do provide a data availability statement often incorrectly imply that data sharing is not applicable to systematic reviews. Our results suggest the need for various interventions to increase data and code sharing by systematic reviewers.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Direct model labels (unvalidated)

Per-model category and study-design labels from the labeling rounds. They are machine output, unvalidated, and the disagreement between models ships as data. No study design here is MEDLINE-validated yet.

Model armCategoriesStudy designConfidence
gemmaMetaresearch
Domain: Reproducibility · Genre: Empirical
About the Canadian research system: no · About a Canadian topic: no
Observationallow
gptMetaresearchOpen science
Domain: Reporting · Genre: Empirical
About the Canadian research system: no · About a Canadian topic: no
Observationalmedium
models splitAgreement compares identical category sets and study designs across arms.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.278
metaresearch head score (Gemma)0.181
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch, Meta-epidemiology (narrow), Meta-epidemiology (broad), Scholarly communication, Insufficient payload (model declined to judge)
Consensus categoriesMetaresearch
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.335
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.2780.181
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0160.004
Bibliometrics0.0010.003
Science and technology studies0.0000.000
Scholarly communication0.0010.000
Open science0.0050.004
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0080.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.932
GPT teacher head0.619
Teacher spread0.313 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it