Data Analytics · The Lit — Research Paper Browser

Loading the database summary…

Journal types

Journals

Year range

—

Disruptiveness index (−1 to 1)

−1.00 1.00

Filters the Team-science figures below — the papers with a computed disruption index.

Exclude non-research items (editorial board, book reviews, corrigenda/errata, announcements, indices)

—

Corpus overview takes the journal as the unit of analysis — publication volume, co-authorship, citations and team science across journals and years.
Author spotlight takes the author as the unit of analysis — add several researchers and compare their output, collaboration and impact.
Paper comparison takes the paper as the unit of analysis — add one or more papers and compare their citations, disruptiveness (D_i) and more, side by side.
All three give rich insights into research productivity, with several ways to slice and compare.

Publication volume over time

Papers published per year

Citations over time

Citations received per year

Average co-authors per paper

Mean number of authors, by year

Co-authorship distribution

Share of papers by number of authors, within scope

Citation impact by journal

Average citations per paper, within scope

Most-cited papers in scope

Citation counts are the highest of the Crossref / OpenAlex / Semantic Scholar tallies — the same count the main browser shows. Titles link to their DOI.

🚧 Beta — work in progress. Everything from here down is an experimental analysis under active development. The intention is not to replicate the Nature study itself, but to apply its measures to our own database to gain further insights into the literature it covers — so expect these figures to evolve and sharpen as the underlying citation graph grows.

Team science — do small teams disrupt? Reproducing the key measures of Wu, Wang & Evans, “Large teams develop and small teams disrupt science and technology” (Nature 570, 2019) over the slice of The Lit's citation graph mapped so far. Every figure below honours the journal, journal-type, year, disruptiveness-range and non-research filters at the top of the page (an editorial editor/area selection narrows to that journal only — see the note in “How we measure disruption”), and sharpens as more of the reference graph is built.

Loading the citation-graph analysis…

How we measure disruption

The disruption index D (Funk & Owen-Smith 2017) — read straight off the citation network, not the citation count.

Distribution of disruption

Every scored paper's D, from developing (D<0) to disrupting (D>0). Most work sits near zero — the paper's Fig. 1b.

Disruption and impact by team size — the headline result

The signature result: as teams grow, disruption (percentile) falls while citations rise.

How teams search the literature

Small teams reach for older, less-cited work (deep search); large teams build on recent, popular work.

Who lands in the extremes

Over- / under-representation by team size among the most-disruptive and the most-cited work (1.0 = as expected if spread evenly).

Most disruptive papers in scope

Highest D — work later cited on its own, apart from its antecedents. Team size is how many authors are credited on the paper (2 authors = a two-author paper).

Most developing papers in scope

Lowest D — work cited alongside the prior art it consolidated and extended. Team size is how many authors are credited on the paper (2 authors = a two-author paper).

Most disruptive & most developmental authors

Each author's D_j — the mean of D over every paper they wrote or co-wrote in scope.

Most disruptive

Most developmental

The Lit — Data AnalyticsSummary statistics for the research paper database