Digital inequality

Definition and demarcation

The digital inequality thesis claims that people of higher status benefit to a greater extent from the availability of digital technologies. In general, people of higher status use digital technologies to a greater extent over time, have greater user competence and are more likely to access political, scientific and health-related information – i.e. content that is assumed to be beneficial (Zillien/Hargittai 2009). It is therefore assumed that advancing digitalisation goes hand in hand with growing rather than shrinking social inequalities.

History

Back in July 1995, the US National Telecommunications Telecommunications Administration (NTIA) published a report entitled “Falling through the Net: A Survey of the ‘Have Nots’ in Rural and Urban America”, which noted the unequal distribution of telephone, computer and modem access. Among other things, it was shown that those with a higher level of formal education were more likely to have access to the new medium, which was seen as an indication of growing inequality (NTIA 1995). The follow-up report “Falling through the Net II” (NTIA 1998) published in 1998 popularised the term “digital divide” for these internet-driven inequalities, which was later translated into German as “digitale Spaltung” or “digitale Kluft”. The first representative surveys on this topic for Germany include the ARD-ZDF online study carried out from 1997 and the study published by Initiative D21 from 2001, both of which are still published today. Around the turn of the millennium, the field of digital divide research became established in the social sciences.

Many of the early studies on the digital divide refer to the theory of the growing knowledge gap, which states that the spread of media information increases rather than reduces social knowledge differences (Tichenor et al. 1970: 159f.). If this knowledge gap hypothesis, which originally related to print media, is transferred to the Internet, even more extensive increases in social inequalities can be assumed, as the use of digital technologies is, on the whole, more cognitively, technically and economically demanding and also far more diverse than is the case for newspapers, radio or television (Bonfadelli 2002: 72f., DiMaggio et al. 2004).

Application and examples

Research on digital inequality can be divided chronologically into three overlapping phases: the phases of access, usage and impact research (Marr/Zillien 2018). Empirical digital divide research begins primarily with descriptive analyses of internet diffusion, which can be subsumed under the label of access research.

The key question in access research is the extent to which diffusion rates vary across different social groups. The percentage of users and non-users, people with and without Internet access, online and offline users are compared in a cross-section or over time, focussing on certain individual characteristics, whereby the same pattern always emerges on the whole: Higher-educated, younger, higher-income, men, city dwellers and working people are more likely to have an internet connection. However, the binary concept of the digital divide implicit in access research has been widely criticised. According to the criticism, this is undifferentiated and abbreviated, as the distinction between users on the one hand and non-users on the other is based on the assumption that either all users use the internet in the same way or that differences in use are irrelevant. The dichotomous distinction between online and offline users therefore does not take into account that there can also be significant differences between users.

In contrast to access research, usage research therefore attempts to capture inequalities in Internet use in a more differentiated way and to assume that the digital divide is a multidimensional phenomenon, which is also more adequately described by the term digital inequality. Usage research focusses on differences in terms of the technology used, usage skills or the Internet content used. To summarise, it is generally true that the quality of Internet technology, the extent of digital skills and the information orientation of Internet use correlate positively with the socio-economic status of an online user. In this context, however, it is not uncontroversial which Internet content potentially consolidates or even improves a person’s social status. In other words, it is certainly debatable which forms of internet use have a resource-enhancing effect. In addition, usage research does not usually systematically examine the short and long-term consequences of varying forms of internet usage.

Impact research on the digital divide therefore follows the idea that the focus of the analysis should not be on the differences in access and use of the internet per se, but on the resulting effects. This is based on the assumption that digital divide research must leave the level of pure description and move on to analysing the potential effects of different internet usage practices in order to prove its relevance to the social sciences and society. Consequently, impact research starts with the distribution of specific resources – such as information, social capital or opportunities for participation – and then examines the consequences of the access and usage gaps for the distribution of these resources.

Criticism and problems

Although large sections of the population today have digital competences and basic skills in the use of digital technologies, the problem of digital inequality remains unresolved. Digital inequalities can possibly be mitigated by improving digital equipment and user skills, but by no means eliminated. Moreover, with every technological advancement, supposedly overcome barriers are overcome again. This means that digital inequality will not disappear from the scientific or political agenda any time soon. Quite the opposite: advancing digitalisation is no longer just fuelling social inequalities, but rather consolidating and, as it were, automating their creation.

Research

Current research distinguishes at least five dimensions of digital inequality (van Dijk 2020; Helsper 2021): (1) In economic terms, internet-generated advantages in the area of gainful employment and consumer behaviour are stated. (2) In social terms, it is assumed that the digital development and expansion of networks contributes to an increase in social capital. (3) From a political perspective, the advantages of digital information and participation are assumed. (4) From a cultural perspective, advantages are assumed with regard to the digital reception or creation of cultural products. (5) And from an individual perspective, favourable aspects in terms of personal development or health information, which ultimately increase personal well-being, are stated. At the same time, for example, those groups that use the internet less in general and in the specific case of health information – older people, those with a lower level of education, those on low incomes – are exposed to greater health risks, which a French study describes as a double divide (Renahy et al. 2008: 9).

This corresponds to the interpretation pattern of the “Innovativeness/Needs Paradox” (Rogers 2003: 295ff.): Accordingly, those members of society who need the benefits of an innovation the most are usually among the last to adopt it. And conversely, those who are the first to integrate an innovation into their everyday lives are the least dependent on it: “This paradoxical relationship between innovativeness and the need for benefits of an innovation tends to result in a wider socioeconomic gap between the higher and lower socioeconomic individuals in a social system” (Rogers 2003: 295). As a result of the spread of many technological innovations, social inequalities increase, which corresponds with the basic assumption of the thesis on digital inequality.

Against this background, a relatively new area of digital inequality research is concerned with the reinforcement of social inequalities through the use of algorithms. The use of algorithms does not lead to a reduction in prejudice and discrimination. Rather, algorithmic calculation endows supposedly neutral decisions with a gesture of objectivity, but at the same time leads to discriminatory and inequality-enhancing effects (Eubanks 2018). While those with a higher status improve their social position through the use of algorithms, the non-privileged are put at an absolute disadvantage: They are increasingly confronted with misinformation, are also discriminated against online due to their income or skin colour or are denied loans solely because of their gender (O’Neil 2017). Existing inequalities are therefore being exacerbated and in some cases even accelerated by technology.

---