Confidence and RISC: How Russian papers indexed in the national citation database Russian Index of Science Citation (RISC) characterize universities and research institutes

The paper analyses Russian Index of Science Citation (RISC), a national citation database. We continue our previous study (Moskaleva et al., 2018) and focus on difference between bibliometric indicators calculated on, so to say, “the best” journals, so called RISC Core, and those which take into account all Russian journals available. Such a difference may show focuses of insitutional actors on different document types, publication strategies etc.

💡 Research Summary

The paper provides a comprehensive examination of the Russian Index of Science Citation (RISC), a national citation database that aggregates Russian scholarly output across journals, conference proceedings, books, patents, dissertations and other research artefacts. Building on a previous study (Moskaleva et al., 2018), the authors focus on the distinction between bibliometric indicators derived from the “RISC Core” – a curated set of high‑quality Russian journals that are indexed in the Web of Science Core Collection (WoS CC), Scopus, and the Russian Science Citation Index (RSCI) – and those calculated from the entire RISC corpus, which includes all Russian‑affiliated publications regardless of venue quality.

RISC was launched in 2005 on the eLIBRARY.RU platform and, since 2010, has incorporated metadata from Scopus and WoS for Russian‑affiliated papers through a partnership with Elsevier. As of August 2018, the database contains more than 31 million documents and over 360 million references, with a journal list exceeding 60 000 titles; roughly 6 000 of these are fully indexed, while another 1 000 belong to the Core (WoS CC + Scopus + RSCI). Figure 2 in the paper demonstrates a pronounced skew: the Core accounts for only 23 % of all papers but attracts 93 % of citations, mirroring Garfield’s classic observation of citation concentration in the Science Citation Index.

To assess how the Core versus the whole database influences institutional evaluation, the authors extracted publication data for 96 Russian universities and 337 research institutes (predominantly Russian Academy of Sciences entities) covering the period 2012‑2016. The RISC reports provide counts of total publications, publications indexed in RISC only, publications in the Core, and sub‑counts for WoS CC, Scopus, and RSCI journals. The analysis proceeds along several dimensions:

1. Self‑upload practices – Many institutions augment their RISC profiles by uploading local conference proceedings and minor article collections via the “Science Index” service. Figure 3 shows that universities with less than 10 % of their output in the Core tend to upload a substantial proportion of their documents themselves, inflating apparent productivity without improving citation impact.

2. Disciplinary specialization – The authors compute the share of STM (Science, Technology, Medicine) papers for each institution and compare it with the share of WoS CC/Scopus papers. Figure 5 reveals a strong positive relationship: institutions with a high STM proportion are far more likely to have a sizable fraction of their output in international databases. No university with less than 50 % STM output exceeds a 20 % share of WoS/Scopus papers. Conversely, SSH (Social Sciences and Humanities)‑focused universities rely heavily on national journals and have a low Core presence.

3. Research institutes vs. universities – Institutes, being more discipline‑specific, display a clearer dichotomy (Figure 6). SSH‑oriented institutes publish predominantly outside the Core, whereas STM‑oriented institutes have a high Core share.

4. Contribution of RSCI vs. international databases – Figure 7 presents regression analyses of Core composition. For STM institutions, the slope for WoS/Scopus contributions is 1.43 (R² = 0.90), indicating that each additional Core paper is more likely to stem from an international venue. For SSH institutions, the slope for RSCI contributions is 0.89 (R² = 0.86), showing a heavier reliance on the national index.

The empirical findings lead to two principal policy implications. First, adopting the RISC Core as the primary bibliometric basis curtails manipulative practices such as mass uploading of low‑impact conference materials, because the Core’s citation‑richness makes it resistant to artificial inflation. Second, while the Core adequately captures the impact of STM research, SSH scholars depend on high‑quality national journals (RSCI). Therefore, continued quality assurance and international visibility for RSCI titles are essential to ensure fair assessment of humanities and social‑science output.

In conclusion, the study demonstrates that RISC Core provides a robust, quality‑oriented metric for evaluating Russian universities and research institutes. It reflects the majority of citation impact while representing a modest fraction of total publications, thereby offering a more reliable gauge of scholarly performance than the unfiltered RISC corpus. The authors recommend broader institutional adoption of Core‑based metrics, increased transparency in Core journal selection, and targeted strategies to integrate SSH research more fully into internationally recognized citation networks.