Nothing
inst/benchmark/results_replication_enriched.md
In rtransparency: Identifies Indicators of Transparency
Replication detector validation (replication-enriched sample)
Package version 0.9.9. 250 open-access PMC articles selected for external-
validation / replication language (PubMed [tiab] search) and hand-labeled for
the replication indicator (a replication or external/independent validation
reported as PERFORMED). The general 2023 sample has too few replication
positives for a stable sensitivity estimate; this enriched sample provides one.
Positives: 111. Sensitivity 92.8, Specificity 34.5, PPV 53.1.
Specificity and PPV here are not representative of unselected literature: the
sample is deliberately rich in validation language, which is the detector's
hardest discrimination, so it concentrates false positives (internal splits
and reviews that discuss validation). The 2023 1000-article sample gives the
representative specificity (98.5). Sensitivity, estimated on the large positive
set, is the stable quantity this benchmark contributes.
Try the rtransparency package in your browser
Any scripts or data that you put into this service are public.
rtransparency documentation built on July 1, 2026, 9:07 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Replication detector validation (replication-enriched sample)
Package version 0.9.9. 250 open-access PMC articles selected for external- validation / replication language (PubMed [tiab] search) and hand-labeled for the replication indicator (a replication or external/independent validation reported as PERFORMED). The general 2023 sample has too few replication positives for a stable sensitivity estimate; this enriched sample provides one.
Positives: 111. Sensitivity 92.8, Specificity 34.5, PPV 53.1.
Specificity and PPV here are not representative of unselected literature: the sample is deliberately rich in validation language, which is the detector's hardest discrimination, so it concentrates false positives (internal splits and reviews that discuss validation). The 2023 1000-article sample gives the representative specificity (98.5). Sensitivity, estimated on the large positive set, is the stable quantity this benchmark contributes.
Try the rtransparency package in your browser
Any scripts or data that you put into this service are public.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.