Login Through Gosuslugi? Factors of Public Attitude Towards E-government Services on Social Media

Abstract

The focus of the work is related to the public perception of government practices within the framework of digitalization policy. Electronic practices of interaction with the government have long been widespread among most Russians. This is confirmed by both public opinion polls and Russia’s high positions in the world rankings of e-government development. In this context, a lot of studies have been conducted to examine human adaptation to new digital services, as well as the reasons for the successful introduction and distribution of public e-services. However, citizens’ motivations to refuse certain forms of electronic services are still poorly explained. In this regard, despite the generally positive picture of the perception of digital public services, some types of emotional assessments and reactions deserve additional consideration. There is a reason to believe that perception is driven not by the service itself, but by its inherent characteristics, which have become the main variables for our analysis: perceived mandatory usage, the sensitivity of the personal data (especially medical and biometric data) and the digital quality (digital nature) associated with the absence of an analog alternative. The database for the analysis contained comments on posts regarding the performance of the national e-government platform Gosuslugi published on the social networking site Vkontakte. The comments were subjected to double-independent encoding. With thematic modeling as a common method of machine text analysis, we confirmed the effect of the selected characteristics on the use of words and the subject of posts. If the digital service had at least one of the properties, the texts contained words with pronounced negative connotations. The second stage of the study was regression analysis, where the types of user reactions were determined as dependent variables: politicization and conspiracy sentiments, as well as the presence of a negative attitude in the text of the publication. The results showed that the main factor for each of the responses is the perceived mandatory usage. At the same time, the initial assumptions that this factor can breed the effect of digital quality and sensitivity have only partially been confirmed. We hope that our proposed approach, based on a combination of content analysis, thematic modeling, and regression analysis, would contribute to the available survey tools in such a difficult field of research as the study of citizens’ attitudes to government institutions and practices.

Acknowledgments. This research is supported by the Basic Research Programme, HSE University.