<p>This study investigates the co-authorship patterns in questionable journals within Russian academia, focusing on the social sciences from 2014 to 2017. Large-scale publication data from the Scopus database is analyzed to determine whether articles in questionable journals can be distinguished from those in legitimate journals based on collaboration patterns. The analysis reveals significant differences, particularly in the prevalence of institutional and national co-authorships, as well as the number of authors per paper. Papers with five or more authors are more likely to appear in questionable journals, suggesting a possible link to unethical practices such as the sale of authorship slots. The geographical distribution of authors is also examined, revealing that suspicious collaborations often involve participants from geographically distant regions or countries facing similar publication pressures, such as Kazakhstan. The findings highlight the impact of research evaluation systems and publication pressure on the rise of dubious co-authorship practices, driven by the requirement for international publications. This research underscores the importance of scrutinizing co-authorship data used in broader academic studies, as questionable journals can introduce significant biases, particularly in data from developing countries.</p>