The uniform formatting of citations/references is a basic requirement for and a distinct feature of scientific journals. In order to fully achieve this in Tokovi osiguranja, references of all papers are routinely checked and if needed corrected using a software tool called RefFormatter.

RefFormatter automatically matches references with their original bibliographic descriptions, as given in reliable, the so-called normative databases, such as CrossRef, Medline, etc. During this and subsequent operations, references are parsed into their component parts (author, title, etc.) to be corrected, amended with the missing elements, and reformatted to meet the citation style in use by the Journal.

The references that cannot be found by RefFormatter in any of the normative databases are left to the authors to format them semi-automatically using Reformater’s functionalities.

The discrepancy between citations and references is a strong indication of the doubtful quality of a paper. The full correspondence between in-text citations in an article and citations given in the reference lists is ensured in Tokovi osiguranja in a two-step, semi-automatic process using SCIndeks Assistant.

In the first step, SCIndeks Assistant activates a software tool called CiteMatcher, which processes the full texts of submitted articles in order to locate the citations that are not given in the reference list, and conversely, the references given in the reference list but not in the body of the article. Based on the findings, CiteMatcher generates a report in the form of an error table.

In the second step, SCIndeks Assistant delivers the table to the authors and the editorial staff, who correct the errors, either by editing the body of the article, or the reference list.

Citing retracted or corrected articles, as well as papers published in fake or predatory journals, has become a serious burden to modern scholarly publishing. Listing such references in an article directly questions the legitimacy of the citing article as a whole.

In order to protect our authors from unintentional and unconscious citing of illegitimate references, Tokovi osiguranja uses a special tool called Mind the Trap. Its purpose is to alert authors of the presence of illegitimate references in their manuscripts.

Mind the Trap processes the literature section of each manuscript and marks visibly the references listed in its Blacklist. The Blacklist includes 10,000+ illegitimate (retracted) papers and more than 30,000 fake, predatory, and suspected journals. The Mind the Trap Blacklist is updated annually.

The Mind the Trap results are attached to the manuscript to be used at some point of the reviewing process. The decision whether to discard illegitimate references from the reference list and the body of the article, or to retain them, is left to the authors of the manuscript.

All papers published in Tokovi osiguranja are routinely submitted to plagiarism detection through CrossCheck/iThenticate within SCIndeks Assistant. iThenticate is the only anti-plagiarism tool regarded to be sufficiently efficient for use in journal publishing. In iThenticate, plagiarized text is detected by matching the screened article with a huge corpus of international scholarly literature exclusively provided by CrossRef. In addition, the corpus comprises research content available on the Internet.

Based on the results of detection, the Editorial Board decides whether a submitted article with a substantial plagiarism score should be rejected or, as it happens in some cases, returned to the authors for correction, e.g. to add the missing citations. In making decisions, the Editorial Board follows the standards set by COPE. When using iThenticate and resolving special cases, it relies on support by CEON/CEES, as the SCIndeks Assistant provider and the CrossRef contractor.

Papers published in Tokovi osiguranja are automatically added to iThenticate’s document corpus in order to protect them from being plagiarized.

In Tokovi osiguranja, reviewers are expected to make well-founded decisions about the acceptance or rejection of manuscripts and offer useful suggestions aimed at improving the reviewed manuscripts that they recommend for publication.

This valuable practice is encouraged by enabling authors to evaluate the usefulness of reviews. In addition to the assessment by the Editorial Board, in Tokovi osiguranja, reviewers are also rated by authors using the software tool Reviewing Reviewers. By responding a standard online questionnaire within SCIndeks Assistant, authors evaluate reviewers in terms of what parts/aspects of their papers, and to what extent, have been improved thanks to the reviewers’ interventions and suggestions. The procedure of reviewing reviewers by authors is double-blind.

The same tool quantifies and accumulates results for all reviewers and compiles an annual report that can guide the Editorial Board in instructing and selecting reviewers.