Replication Rate by Journal Rank
Do papers published in more prestigious journals replicate more or less often? Each original paper is matched to its journal’s SJR overall rank and classified as “replicated” if at least 75% of its effect replications succeeded under the selected criterion. Metric shown: SJR overall rank, at publication year.
Rate = success / (success + failure + reversal); inconclusive and unrecorded outcomes excluded, where success means the outcome recorded in the database. Unit: replication effect (n = 6,931 — originals whose journal is matched in the SCImago rank dataset). How this is defined.
4,518 papers plotted · 400 excluded (no publication-year SJR overall rank) · 838 excluded (inconclusive outcome)
Replication rate by SJR quartile
All papers grouped by their journal’s SCImago best-quartile in the publication year (Q1 = most prestigious). Whiskers are 95% intervals from a journal-cluster bootstrap (1,000 resamples of journals with replacement). Most replicated papers come from Q1 journals, so the higher quartiles have few papers and wide intervals.
Replication rate by SJR overall rank quintile
All papers with a publication-year SJR overall rank, split into five equal-count bins (note: rank 1 is the most prestigious, so bins run best → worst left to right). Whiskers are cluster-bootstrap 95% intervals.
Per-journal replication rate vs SJR overall rank
One point per journal with at least 15 plotted papers. Point size reflects the number of papers; whiskers are 95% Wilson CIs. Lower rank numbers are more prestigious journals.
Rank metrics. All four come from the SCImago Journal Rank (SJR). SJR percentile is 100 × (1 − (rank − 1) / N) over that year’s N ranked journals, so it is comparable across years even though the raw rank is not (SCImago ranked ~17,000 journals in 1999, growing to ~32,000 by 2024). SJR quartile is the journal’s best subject-category quartile (Q1–Q4). SJR overall rank is the raw position that year (1 = highest SJR). Journal h-index is SCImago’s cumulative journal h-index (a snapshot, not year-specific). Percentile, quartile, and rank are taken per publication year; papers before 1999 use the 1999 ranking.
Matching. Journal names in the replications database are joined to the SCImago data by ISSN with a normalized-name fallback (~93% of matched journals). Books, preprints, and conference proceedings are not ranked and are excluded.
Attribution. SJR data: SCImago, (n.d.). SJR — SCImago Journal & Country Rank [Portal], retrieved from scimagojr.com (used non-commercially).
Data: replications_database_2026_08_01_120356.csv; rank metrics generated 2026-08-03.