Social media repertoires: Social structure and platform use

Abstract

We analyze how the size and composition of social media repertoires is associated with key sociodemographic variables: age, gender, socioeconomic status, education, and occupation. Specifically, we ask what is the association between these variables and: (a) social media use as a whole, (b) the number of platforms people include in their social media repertoires, (c) the platforms included in such repertoires, and (d) the most prevalent repertoires? To answer these questions, we analyze data from an in-person survey (N = 700) about use of media and communication technologies conducted in 2016 in Argentina by a polling firm. Our findings indicate that: (a) the odds of using social media are higher among younger people, women, those with higher socioeconomic and educational levels, and those employed; (b) whereas an increase in age is associated with a decrease in the size of the repertoire, higher educational attainment is associated with an increase in the number of platforms included in the repertoire; (c) age, gender, education, and occupation are significantly associated with the inclusion of different platforms in the repertoire; and (d) some of these variables are significant for the uptake of different repertoires, but not others. We interpret these findings drawing upon scholarship about digital inequalities and social media repertoires, and we assess and reflect on their implications for research on the digital divide.

Keywords:

• Digital divide
• Latin America
• platforms
• repertoires
• social media

Notes

1 Research on media repertoires (Hasebrink and Hepp 2017; Klein, Walter, and Schimank 2018; Swart, Peters, and Broersma 2017) is grounded in studies of traditional or mass media, such as television (Hasebrink and Popp 2006), and its emergence is related to the multitudinous options afforded by cable television (Ferguson and Perse 1993).

2 Use of multiple media can have larger societal implications, for instance, in the spheres of news consumption and political participation. Media diversity tends to increase content diversity (Dubois and Blank 2018), with important implications for exposure to ideologically varied news. As individuals resort to multiple media or platforms to inform themselves, their information repertoires expand, and political participation can find new avenues (Diehl, Barnidge, and Gil de Zúñiga 2019).

3 Although the survey included individuals ranging from 18 to 75 years old, there is one case with age reported at 77.

4 The outcome “number of platforms” was constructed, which means that respondents were not specifically asked about the number of social media platforms they used concomitantly. For internal validity purposes, future research should incorporate such a question—and eventually compare how constructed measures vary with relation to self-reported counts.

5 The fact that we model the use of each platform separately might present a limitation to this study since it implies comparing the significance of coefficients across models that were designed for platforms with different levels of uptake as well as different frequencies of use. We decided to offset this limitation with an analysis of the concurrent uptake of different platforms, as per the multinomial logistic regression model whose results are presented in Table 7.

6 We decide to exclude from this ranking the platform Google Plus despite its reported percentage of users reaching almost 45%, since it was not included in qualitative interviews conducted in Argentina as part of the same study (see Boczkowski, Matassi, and Mitchelstein 2018). Furthermore, respondents could potentially mistake it for “use of Google,” as in the search engine.

7 Full tables with predicted probabilities for variables set at different values available upon request.

8 We believe that age not showing statistical significance for the uptake of Snapchat might be a product of a relatively small number of overall Snapchat users (N = 54).