Russian White List of Peer-Reviewed Journals (journalrank / «Белый список»)

Snapshot of the official registry of peer-reviewed academic journals used to evaluate publication activity in Russia — the «Белый список» (White List), operated by the Russian Centre for Scientific Information (РЦНИ / RCSI) on behalf of the Interdepartmental Working Group of the Ministry of Science and Higher Education.

One row per journal, 100 columns: journal level in the White List (2023 / 2025 / 2026 editions), membership and last-verification dates in 36 international indexes (Web of Science, Scopus, RSCI, DOAJ, ERIH PLUS, MEDLINE, MathSciNet, zbMATH, DBLP, OpenAlex, Crossref, Wikidata, …), inclusion / exclusion dates, and the decision protocol notes.

Files

File	Description
journalrank_2026-09-19.csv	Normalized dataset — comma-separated, UTF-8 (no BOM), ISO dates, 31,576 rows × 100 fields
fields_dictionary.csv	Field-by-field dictionary (Russian from the official description file + English translation)
raw/journalrank_19.09.2026.csv	Byte-verbatim source export (TAB-separated, UTF-8 BOM) — kept for provenance
raw/journalrank_csv_description.csv	Byte-verbatim official field description file
data_provenance.json	Source URIs, SHA-256 checksums, retrieval timestamps
normalize_report.json	Machine-readable report produced by the normalization script

Columns

• Identity — title, issns (pipe-separated), links
• White List level — level_2023, level_2025, level_2026, plus derived level_current
• Index membership — 36 Yes/No columns (wos_cc, scopus, rsci, doaj, erih, medline, msn, zbm, dblp, oax, cref, wikidata, …)
• Identifiers in those indexes — srcid, doaj_id, oax_id, wikidata_id, coden, …
• Last-verification dates — *_update columns (ISO-8601 after normalization)
• Registry status — date_accepted, date_discontinued, state, notice

The level columns are deltas, not full snapshots

Each level_YYYY column is filled only where that edition differs from the previous one:

Column	Filled (of 31,576)
level_2023	29,551
level_2025	3,120
level_2026	136
level_current (derived)	31,576

level_current = level_2026 → level_2025 → level_2023, i.e. the most recent known level. Use level_current unless you specifically need the historical editions.

Dates

Source date cells are inconsistent; normalization converts them to ISO-8601 at the precision the source carries:

• dd.mm.yyyy → yyyy-mm-dd (the vast majority; 354,101 cells converted in total)
• m.yyyy → yyyy-mm — present in socabs_update (672 rows) and wpsa_update (538 rows)
• yyyy → yyyy — present in ajg_update (1,554 rows; AJG edition year) and petro_update (201 rows)

Read all date columns as strings.

Registry status

state is empty for active journals, otherwise warning (1,848) or discontinued (515). The registry's web UI lists 31,061 records — exactly this file's 31,576 rows minus the 515 discontinued ones, so this snapshot also preserves journals that were previously removed.

Usage

import pandas as pd

# dtype=str keeps "Yes"/"No" and empty cells intact (empty != missing)
df = pd.read_csv("journalrank_2026-09-19.csv", dtype=str, keep_default_na=False)

active = df[df["state"] == ""]
top = df[df["level_current"] == "1"]
in_wos = df[df["wos_cc"] == "Yes"]

from datasets import load_dataset

ds = load_dataset("fjsijijr928/journalrank", split="train")

Provenance & refresh

• Source: CSV export on https://journalrank.rcsi.science/ru/record-sources/ («Скачать список журналов» → CSV), operator РЦНИ / RCSI, https://www.rcsi.science/activity/belyy-spisok/
• Export filename: journalrank_19.09.2026.csv; retrieved 2026-10-06; latest registry decision reflected: protocol ДС/39-пр of 2026-03-25
• SHA-256 raw export: 446ae9cb4dd8e49a3417f3bc41c32608b8160b3f346eb022b7ebe28099ca79be
• SHA-256 normalized: 0cd2b54a12c8a94c43058178bcf99f890270f9236270536abcec461dfa390016
• The registry is updated continuously; there is also an open JSON API (https://journalrank.rcsi.science/ru/api-docs/docs/) for incremental refreshes. Treat this dataset as a dated snapshot, not a live mirror.

Reproduce the normalization (the script ships in this repo under scripts/):

python3 scripts/normalize_journalrank.py \
  --raw raw/journalrank_19.09.2026.csv \
  --description raw/journalrank_csv_description.csv \
  --out journalrank_2026-09-19.csv \
  --report normalize_report.json

License / License terms

The source publishes the CSV export «на условиях открытой лицензии» (under an open license). Per the source's terms of use (https://journalrank.rcsi.science/ru/info/):

• copying requires a link to the source (hyperlink for online media),
• the information is open-access state property and may not be used for commercial purposes.

Attribution: «Белый список» научных журналов / Russian White List of Peer-Reviewed Journals, РЦНИ (RCSI) — https://journalrank.rcsi.science/ru/

Russian summary

Снимок официального реестра рецензируемых научных журналов — «Белого списка» (оператор — РЦНИ). 31 576 журналов, 100 полей: уровень журнала (издания 2023/2025/2026 + производное level_current), признаки присутствия и даты проверки в 36 международных базах, даты включения/исключения, записи протоколов МРГ.

Нормализация: TAB → запятая, UTF-8 без BOM, даты ISO-8601, переименование dnb_id (idn) → dnb_id. Исходные файлы сохранены байт-в-байт в raw/, контрольные суммы — в data_provenance.json.

Колонки level_2025 и level_2026 — дельты относительно предыдущего издания; для большинства задач используйте level_current. Условия использования: указание источника + некоммерческое использование (по правилам сайта-источника).