Wkład naukowców w ranking IDUB (Inicjatywa Doskonałości – Uczelnia Badawcza). Polscy badacze w Google Scholar

Thesis/Objective – Google Scholar is a tool that is widely used not only to search the scientific literature, but also to obtain information on researchers’ scientometric measures. In this article, we will verify whether, based on GS data, users with the highest measures will be identified as associated with the best universities in Poland, called IDUBs. Methodology – Stepwise logistic regression models with cross-validation were used to find variables significantly influencing the  correct  automatic  classification. Findings and  conclusions – The best models in terms of predictive quality were obtained using the h-index, the type of university, the annual number of publications and the year of the first publication as predictors. Student’s t-tests showed statistically significant differences in the mean values of the h-index, the i10 index and the number of publications (p<0.001, p<0.001 and p=0.013, respectively) between researchers from the best 10 universities in Poland (associated as IDUBs) and scientists from other  academies.  The  scholars  characterized  by  high  scientometric  measures were affiliated to IDUB schools – this relationship is observed within the scope of universities, not technical or medical schools. Due to the free and open nature of the GS, the data obtained from it are heterogeneous and often incomplete, making automatic processing and analysis difficult. These complications are particularly evident when aggregated rather than individual data being analysed. Despite these limitations, the results obtained make it possible to cope with the rapid growth of scientometric data and may lead to the creation of new measures for assessing the scientific output of scientists.