Resumen
This study aimed to characterize the results of research earnings management through real activities (GRAR) as well as possible gaps for future researches. For that, a bibliometric research was carried out, with a descriptive purpose, a documental procedure and a quantitative approach, and it was applied the technique of content analysis and the analysis of social networks through UCINET software. Scientific articles were collected, all of them published in international accounting journals of the area available in Scopus database. The seminal study of Roychowdhury (2006) on GRAR was the base of which a sample of 50 publications was obtained. The results show that the year of 2014 was the one with the highest number of publications on GRAR with a greater significance from the Journal of Business Finance & Accounting, although the production related to the theme has not evolved in a period of six consecutive years. It was also observed that among those 50 publications, 27 were conducted in the United States, and 22 of them presented no theory in theoretical construction. Nevertheless, it was observed that the Agency Theory was the most frequent used theory among the other ones in publications. Regarding the application of GRAR metric, it was found that only 11 publications used the isolated model of actual activities (Roychowdhury, 2006). In contrast, 39 publications applied together GRAR and models based on accruals. In view of these results, it is assumed that the theme and the model proposed by Roychowdhury (2006) have not yet been disseminated by the researchers, and so little has been consolidated in the literature of the area since the related publications are not constant. However, there was a considerable evolution in the period of 2014 and 2015, demonstrating an interest in the application of this metric, which in turn tends to be legitimized by researchers of the area, as well as other metrics of quality of previous accounting information. Finally, this research contributes to the development, expansion and identification of research-related gaps in national and international literature, providing insights for a new research, which may use variables and theories from a different perspective.