Agregadores de dados altmétricos: analisando o altmetric.com e o webometric analyst

Objective: reports the experience of using altmetric data aggregators Altmetric.com and Webometric Analyst, in a study on the presence of articles from Brazilian scientific journals on Information Science on social media platforms. Methods: the study was characterized as descriptive, of qualitative and quantitative nature, which analyzed the aggregators Altmetric.com and Webometric Analyst. The corpus used to test them consisted of 13 Brazilian journals in Information Science, encompassing 2,724 scientific articles published in the period from 2011 to 2018. Result: from the use of the two aggregators to collect altmetric data about articles from Brazilian journals in Information Science, it was primarily verified that different methodologies and tools generate different results, with discrepancies and specific similarities. It was also noticed that the DOI numbers of some articles were incorrect, that is, the link of these DOIs pointed to an error page in CrossRef, assuming that was the reason for the small number of positive results in the Altmetric.com via API. Although the Webometric Analyst has shown greater coverage of altmetric data for articles from Brazilian journals in Information Science, possibly due to the search strategy used, in this case author, title and year, some inconsistencies were noticed in Mendeley, such as an article with different versions or with different publication dates, showing divergent counts of readers for the same article. Conclusions: it is considered that the tools used for generating altmetric data should be understood as facilitators of data collection on social media platforms, however one must be aware of the scope and limitations of each of these tools, as well as of the collection chosen for testing and capture. As the nature of the platforms is uneven, they can be employed as to complement one another.