Abstract
This article discusses the heterogeneity in children’s appropriation and use of the internet that make up contemporary digital divides. Based on a survey of Portuguese children in mandatory education (8- to 17-year-olds), it relies on multivariate statistical procedures to build a topological mapping of internet use patterns. Variations in digital practices and parental mediation are analysed in relation to social backgrounds and demographic traits. Four clusters of users were thus identified: ‘self-reliant cybernauts’, ‘nurtured cybernauts’, ‘nurtured beginners’ and ‘unguided rookies’. This article aims to contribute to deepening the debate on digital divides and digital diversity within the sociology of childhood.
Introduction
The expansion of information and communication technologies (ICT) and their individual appropriation is a mass phenomenon in contemporary western societies. Within the household and at school, children appear to be leaders in this innovation process, which has been transforming and reshaping daily scenarios and relationships. However, this remarkable diffusion of ICT does not automatically mean universal democratization of access or even uniformity of usage within the system. It is thus worthwhile exploring questions such as:
How do internet uses vary in children and reflect the structural diversity of childhood?
How are these variations linked to social characteristics, such as age, gender, social class, or influenced by patterns of parental mediation?
Against this backdrop, our aim was to build a coherent picture of digital divides and diversity in children’s use of the internet (places, protagonists and the social contexts to which they are linked) by analysing empirical data from a survey conducted in 2008 of Portuguese school children (aged between 8 and 17 years old), living in contrasting areas of the country. Our main argument is that consistent and sociologically significant aggregations of internet use practices, levels of skill, family support and types of media environment can be identified by clustering these variables according to their varying degrees of interrelationship. We intend to show that social inequality persists even within the so-called ‘digital generation’ and that ‘technological fixes’ are insufficient to democratize the benefits of an information society.
Looking beyond the traditional dichotomy of users vs non-users, this research contributes to ongoing theoretical discussions on contexts of internet use and their role in contemporary childhood, in a particular (and understudied) national setting: Portugal. This article also makes an innovative contribution in methodological terms, since it relies on a multivariate analysis that combines a wide range of variables concerning internet uses, social background and family context.
The article is structured in five steps: first, we offer a brief review of the literature on new media at home and digital divides among children; then, we give a succinct presentation of the data and research design; next, we describe and justify the main methodological procedures and statistical operations, followed by a summary of the main results, namely a topological mapping of patterns of internet use by children and identification of sociologically grounded groups; we follow this with a theoretically and empirically based discussion on digital diversity; finally, we address the main conclusions of this study.
New media and technologies at home: A digital generation?
Statistical data from Eurostat demonstrate the widespread diffusion of new technologies within Europe: internet users amounted to 67 percent in the EU27 in 2009, ranging from 37 percent in Romania to 90–91 percent in the Netherlands and Sweden. Children and young people (6–17 years old) lead in internet use (75 percent). The growth trend among children is remarkable: overall users (aged 6–17) increased from 68 percent in 2005 to 75 percent in 2008, due to the rising access rate of younger cohorts (6–13 years old) (McQuillan and d’Haenens, 2009). Households made up of families with children are those that convert most rapidly and intensely to technological innovation and bring new media into their homes (Hasebrink et al., 2008). Internet diffusion and access in Portugal, though not very high among adults (48 percent), is almost universal among children: 95 percent among the under-15s in 2009 (Eurostat).
One of the most recent and striking domains of social change is precisely this sudden emergence of a networked technological family (Kennedy et al., 2008) and new childhood cultures associated with intense exposure to and use of new media. As Livingstone (2002: 1) states ‘the home is being transformed into the site of a multimedia culture’.
At the same time, new values concerning childhood and children and families’ educational mobilization in all social milieus have strongly influenced the social demand for ICT. Contemporary parents celebrate the educational potential of the internet (to support school work and promote academic success), investing part of the family budget in purchasing new equipment and internet connection services (Almeida and Vieira, 2006).
In Europe, research on children’s access and uses of the internet emerged in 2000 and has grown steadily ever since. The majority of these studies focus on teenagers (14- to 15-year-olds) and deal with topics such as daily internet access and use, online activities and interests, skills and networking, learning at school, new web surfing opportunities and risks (Danoso et al., 2009).
Digital divides are one of the most interesting theoretical issues concerning children and the internet that has arisen from international research. The dichotomy between ‘haves and have nots’ was an issue typical of the 1990s, which was later revealed to be too simplistic, partial and inaccurate (Livingstone and Bober, 2005), since there is evidence that equal and massive access to the internet does not mean universal and uniform usage (Hargittai, 2010). However, access issues should not be simply dismissed: it cannot be taken for granted that all families or children have internet access at home. The geography of ICT dissemination reveals inequalities between countries in Europe (Hasebrink et al., 2007), between regions in the same country (Tsatsou et al., 2009) and between groups of children. On the one hand, despite a triumphalist rhetoric (Tapscott, 1998), a naturalized and proficient access to the internet is still a privilege of wealthier families and children, including only restricted segments of the lower classes (Lee, 2008). Despite official efforts to promote digital inclusion and massive public school provision of ICT, the home cannot be ignored as a locus of inequality (re)production.
The diversity of access modes goes beyond the sharp contrast between ‘haves and have nots’ and reveals other ways of reinforcing social inequalities: type of computer (laptop or desktop) and internet connection (broadband vs dial-up connection); flat-rate vs pre-paid connection, services used and number and location of home computers (common areas vs bedrooms). These are all factors contributing to a multidimensional reality that does not fit into a single dichotomous formula. Other authors (Buckingham, 2008; Sigalès and Mominó, 2009) call our attention to the reinforcement of an ‘in-school’ and ‘out-of-school’ divide. Distance is fast growing between informal, spontaneous, self-directed, motivated online cultures that value the child as an active and autonomous learner, as opposed to a formal school culture, subject to hierarchy, that regards the child as a passive and dependent learner. ‘Home-style learning’ is clearly gaining ground over ‘school-style learning’ (Buckingham, 2008).
Contemporary research has shown that inequality does not just play a role in access to ICT, but rather the gap is shifting towards the inside of the system itself, through different modes and abilities of internet use. More diversified, sophisticated, creative and proficient uses of the internet are concentrated in groups of children whose parents have the highest educational and digital literacy levels (Buckingham, 2008; Livingstone and Helsper, 2007). A ‘second-level digital divide’ is at play (Hargittai, 2002).
Children’s appropriation of the internet is thus far from homogeneous or uniform. Social distinctions shape ‘digital diversity’, a concept that appears to be more appropriate to describe the wide range of modes of accessing, using and giving sense to the internet (McQuillan and d’Haenens, 2009). Although the issue of digital divides and diversity almost immediately suggests the usefulness of developing a typology of internet users, few studies have attempted it. Three of these are particularly relevant, since the types reveal specific modes of use and are then connected to external, contextual variables. However, one proposes a discontinued and fragmented universe of distinct types while the two others a linear scale from lower to higher levels of inclusion.
Focusing on ICT uses in the classroom, Holloway and Valentine (2004) highlighted the importance of ‘communities of practice’ and divided UK secondary school cyberkids into four groups, characterized by a diversified repertoire of social and technical skills: the ‘technoboys’, the ‘lads’, the ‘computer competent girls’ and the ‘luddettes’. This clearly shows the impact of gender stereotypes on modes of computer appropriation.
Instead of a compartmentalized construction of mutually exclusive types, Livingstone and Helsper (2007) put forward ‘a continuum of digital inclusion’ based on the results of a national survey of 9- to 19-year-olds in the UK, using two main criteria: the frequency of use (non-users, low users, weekly and daily users); and the breadth of use, i.e. the range of online opportunities (basic users, moderate users, broad users and ‘all-round’ users). Older children and middle-class children take advantage of more opportunities than younger and less affluent ones.
Following the same path, Kalmus et al. (2009) propose a general ‘ladder of online opportunities’, in which children’s abilities evolve in stages, mainly according to age: school-favoured uses; activities related to popular uses of communication and entertainment; resource-bound uses, mainly entertainment opportunities; and advanced uses, such as interactive and creative activities.
This research strives to contribute to these theoretical debates and enrich an emergent field of study. We aim to describe and explain children’s diversity in internet usage, by bringing to the discussion up-to-date and innovative data on digital divides. Young children (not their gate-keepers) were surveyed directly. Diversity and patterns of use are presented not in a linear continuum or as isolated categories, but in a topological map of internet use patterns, built upon a wide array of interconnected variables and grounded in differentiated social contexts.
Data and research design
This research is based on an ongoing project concerning children and the internet in Portugal, funded by the Gulbenkian Foundation. The following analysis relies on a survey launched in 2008, applied to 3049 children (aged 8–17) in the final years of each level of compulsory education (fourth, sixth and ninth grades) at Portuguese public and private schools located in contrasting areas of the country. Self-completed by the children, the questionnaire focused on computers and internet use, at home and at school. The breakdown of the sample by sociodemographic variables is shown in Table 1.
Sociodemographic characteristics of the survey sample
As we can see, computer use and internet access are almost widespread among Portuguese children, even though a small group is still excluded from technological innovation at home (9 percent do not have a family computer; 21 percent do not have an internet connection). But does this mean that all the other ‘included’ children use ICT and enjoy its benefits in a similar way? Or, conversely, can different profiles of computer and internet use be discernible according to a specific set of conditions?
In order to answer these questions, we identified a group of variables that ought to play a significant role in influencing internet use by children and set out to explore how they acted together in order to create different sociologically significant patterns, which were correlated with social and demographical variables such as age, sex and educational and social backgrounds. 1 In view of this objective and considering the nature of the variables included in this analysis (nominal and ordinal), the most suitable statistical method was to perform a multivariate procedure, through multiple correspondence analysis. The raw data concerning the 10 variables mentioned in Table 2 underwent an extensive reconstruction process. 2
Dimensions, recoded variables and categories used in the analysis
The strategy used to identify children’s internet use patterns covered the four dimensions displayed in Table 2: digital skills and ICT use experience, diversity and intensity of electronic practices, children’s media environment and the existence of rules on internet use at home.
In order to determine the degree of children’s digital skills we posed a set of questions regarding the type of activities they were able to carry out with computers, 3 the time elapsed since they had begun to use the internet and how they had learned to use computers and the internet.
We assessed the diversity and intensity of children’s electronic practices using five different variables. The first was the typology of young internet users, derived from a set of 16 questions on internet activities, later recoded into four different variables concerning the intensity (low to high) of internet use practices: communication, 4 education, 5 entertainment 6 and other. 7 We conducted a cluster analysis 8 of these four variables in order to identify homogeneous profiles of internet users among the children surveyed. Each profile results from a specific combination of intensity of use of each of the four types of aggregated internet activities, defining a typology of young internet users composed of: ‘all-round cybernauts’, who are omnivorous, intense practitioners of all internet activities; ‘diligent students’, who concentrate almost exclusively on educational activities; ‘committed gamers’, who focus on entertainment and playing online games; and ‘incipient users’, singled out by the fact that they do not perform any activity intensively (Almeida et al., 2011).
Web-based communication 9 encompasses three categories: communication with peers only; communication with peers and parents; and no web communication at all. Web navigation strategies 10 were clustered into four categories: suggested navigation (webpages visited are mainly suggested by friends or family); autonomous navigation (no suggestions followed, predominance of web surfing); cumulative navigation (a combination of the two former strategies); and, finally, hesitant navigation (no clear navigation strategy discernible in very inexperienced users). The last two variables concern the use of online social networks and the number of distinct types of webpages usually visited. 11
The remaining variables used in the model regard the children’s media environment at home (number of resources available to them 12 ) and the existence of rules on internet use set by parents.
A multiple correspondence analysis was performed on this set of 10 variables resulting in the identification of four clearly different modes of internet use by children and their respective constraints and conditions, organized by two fundamental dimensions. 13 The first dimension concerns the scale of digital skills attained and the intensity of use, the second concerns the degree of parental mediation and control of children’s internet use. This second dimension is deducted from the variables pertaining to parental control and nurturing (equivalent to the ‘parental warmth’ described by Valcke et al., 2010: 456): existence of internet rules, the family as source of digital learning, online communication with parents and web navigation guided by suggestions from family members. Within each dimension, a clear-cut distinction emerged between users according to their degree of computer and internet use skills, intensity and diversity of practices performed (Table 3, Figure 1).
Multiple correspondence analysis: Patterns of internet use and parental mediation
A topological map of internet use by children
The intersection between positive and negative poles of these two dimensions configures the Cartesian Plan shown in Figure 1, with four different quadrants identifiable. It maps the relationship between the different categories of the variables included in the analysis, as measured by their degree of proximity to each other.
The first quadrant (upper-left, ‘incipient use without parental mediation’) aggregates the least experienced users (amateurs when dealing with computers and with less than one year of internet use), with a rather narrow use of the internet (incipient users), limited to a hesitant navigation of the World Wide Web, and who do not use the internet to communicate with others. They rely solely on help from their friends. Their media environment is less rich than other children’s and their parents have no influence over their dealings with the internet, probably because they are non-users.
The second quadrant (below-left, ‘moderate use with parental mediation’) encompasses slightly more experienced children (although still amateurs, they have been using the internet for the past year or two), characterized by a moderate use of the internet, focused mainly on educational activities (diligent students). They learned how to use the internet with their family’s help, at school or on an IT course. Family supervision is present, through the rules concerning internet use and suggested navigation strategies. Communication with others, via online social networks or instant messaging, is absent.
The third and fourth quadrants (on the right, intensive use) share similarly high levels of proficiency with computers (professionals and experts) and the internet (all-round cybernauts). They have been using ICT for a considerable amount of time (more than three years) and show a remarkable diversity of visited websites (more than eight types), including online social networks, as well as a rather similar media environment (they own all the resources included in the questionnaire).
The divide between these two quadrants stems from parental mediation. Children included in the third quadrant (below-right, ‘intensive use with parental mediation’) learned how to use the internet with the help of their parents, have rules concerning internet use, communicate with peers and family members via the internet, and both follow suggestions regarding which pages to visit and carry out autonomous navigation. Conversely, children located in the fourth quadrant (upper-right, labelled ‘intensive use without parental mediation’) learned how to use the internet by themselves, have no parental rules on internet use, communicate only with peers and apply autonomous navigation strategies exclusively.
Finally, the multiple correspondence analysis object scores were subjected to a cluster analysis, 14 locating each respondent within a specific pattern, through which four clusters emerged. The ‘self-reliant cybernauts’ cluster corresponds to the pattern of ‘intensive use without parental mediation’ and is the largest, including about one third of the sample (1005 individuals); the second cluster, labelled ‘nurtured cybernauts’, corresponds to intensive use with parental mediation (777 respondents, 25 percent). Less intensive uses of the internet are concentrated in smaller clusters: ‘nurtured beginners’ correspond to moderate use with parental mediation (710 children, 24 percent) and ‘unguided rookies’ to incipient use without parental mediation (521 individuals, 17 percent). The next section demonstrates how social context variables play a decisive role in digital diversity, by analysing the composition of these clusters when crossed with the classical sociodemographic variables.
The methodology used represents an innovative contribution to studies in this field, in the sense that it goes beyond existing typologies based on a limited set of variables, by relying on a comprehensive multivariate analysis that combines internet use with social background and family context.
The massive variable reconstruction conducted in order to carry out this multivariate analysis (due to the oversimplified question and answer options of the questionnaire applied to a sample with substantial age differences) may constrain its reproducibility in other contexts. Yet, we are confident that if direct questions covering the same research issues (internet practices, intensity of use and parental mediation strategies) were asked by other teams in different research contexts, some relevant comparative data could emerge.
Results: Contexts and clusters of internet use
The distribution of these clusters by sociodemographic variables demonstrates the association between the diverse patterns of computer and internet use by children and their sociodemographic background (see Table 4).
Clusters by sociodemographic variables (in percentages)
p < .001; Cramer’s V = .097. ** p < .001; Cramer’s V = .362. *** p < .001; Cramer’s V = .166. **** p < .001; Cramer’s V = .151.
Significant differences were found between boys and girls regarding patterns of internet use and parental mediation. Girls are mostly present in the ‘nurtured’ clusters, suggesting that they are more prone to be subjected to parental influence than boys (prevailing in the self-reliant cybernauts), who are trained from an early age to be self-reliant and to become more autonomous.
Age is also a significant factor for understanding the differences between patterns of computer and internet use. The data presented seem to imply an evolutionary path that follows the succession of age cohorts: from incipient and hesitant practices to more intense, diverse and reliant uses of ICT (see, for instance, the progression of ‘self-reliant’ cybernauts in the three age cohorts). However, this path is not exclusive: a relevant portion of young internet users remain ‘nurtured cybernauts’ and, more residually, in the other two groups of less proficient users.
Children’s patterns of internet use and parental mediation also vary according to the educational backgrounds of their families. 15 Parental influence tends to increase alongside their level of education. Three patterns can be detected. The less educated parents (elementary and lower intermediate degrees) play a modest and detached role in their children’s internet use, who are mainly either ‘unguided rookies’ or ‘self-reliant cybernauts’. Over two-thirds of children with parents with intermediate levels of qualification are skilled cybernauts, most of them free from parental mediation. Finally, parents with higher education adopt a much more interventionist approach vis-a-vis their children’s electronic media activities, scoring the highest proportions of ‘nurtured’ users.
Social class has an impact similar to educational background, due to the high degree of correlation between the two variables. Less qualified workers (industrial and pluriactive workers) follow their children’s ICT use less closely (higher proportions of ‘unguided rookies’ and ‘self-reliant cybernauts’), certainly due to a lack of ability and skills associated with a shorter educational path. The children of the intermediate classes (clerical and sales workers), coming from an intermediate educational background, are in a half-way position concerning these patterns of internet use and parental influence (nurtured users make up almost 50 percent). Social classes with higher educational credentials and revenue (professionals and technicians and, to a lesser extent, employers and managers) tend to exercise parental guidance over their children’s activities on the internet. ‘Nurtured users’ reach higher proportions among these social strata, although some differences are noticeable between the two categories. Professionals and technicians’ children are more evenly distributed throughout the usage profiles (except ‘unguided rookies’) because their children are marginally younger (augmenting the ‘nurtured beginners’ share), due to an unintended overrepresentation in the sample; ‘employers and managers’ is a less homogeneous category in terms of educational credentials and includes some intermediate educational strata, thus favouring an increase in the ‘self-reliant cybernauts’ profile.
Discussion
The data presented support the notion that the digital divide is much more complex than existing theories, mentioned earlier, account for. Even with widespread access to computers and the internet at home, remarkably different patterns of internet use emerge. Diversity between children exists and is not randomly distributed – it is associated not only with core sociodemographic variables (age, gender, parents’ educational background and social class) but also with parenting styles. Technology is not introduced neutrally into children’s daily life: unequal social contexts shape its appropriation.
Our typology of four clusters, which are more than mutually exclusive types (Holloway and Valentine, 2004) or a linear continuum of use patterns (Livingstone and Helsper, 2007), gives a more detailed and comprehensive view of digital diversity, organized by experience and intensity of usage, on the one hand, and by parental mediation, on the other.
Experienced and intensive users engage in diverse internet uses, perform a vast array of activities (all-round cybernauts) and combine these with autonomous navigation strategies, instant messaging and online social networks. Less experienced users are more inclined (or limited by their circumstances) to narrow their internet experience to one type of use (education or games) or even to remain as incipient internet users. This status as less experienced user somehow limits all other digital practices, implying a need to choose between activities because the restricted skills of the user are not adequate to enable proficient use of all of them, similar to the ‘taste for the necessary’ asserted by Robinson (2009). As stated in the literature (Hargittai, 2010; Lee, 2008; Livingstone and Helsper, 2007) and demonstrated in this present study, users’ skills and capabilities are not just a product of individual attributes (such as age or gender). They are shaped by the child’s family’s position in the social structure (i.e. parental educational and professional resources), which is in turn closely connected to parental mediation styles at home.
In better-off families, the number and diversity of computers and the availability of continuous and wireless internet connections at home encourage children to become intense, experienced and prolific internet users. The parents, themselves, are digitally skilled and thus act as their children’s tutors (Vekiri, 2010: 947). Adopting different strategies (Edwards and Alldred, 2000; Livingstone and Helsper, 2008; Valcke et al., 2010), they may help and guide their children’s use of computers and the internet, suggesting websites and activities, correcting mistakes and solving technical problems and even exercising some mild surveillance of electronic activities in order to detect misconduct. The dichotomy of ‘digital natives’ vs ‘digital immigrants’ (Prensky, 2001) that opposes generations, children vs adults, does not seem to apply here: by sharing common interests and experiences, the internet probably contributes to closing the generational gap and creating a new technological family culture.
Conversely, in lower class families, with poorer media environments, parents are non-users or extremely incipient users; they can hardly provide technological support or supervision to their children – except to set rules regarding schedules or durations of use. In these families, technological innovation reinforces the generation and knowledge gap between children and adults (Clark, 2009), and the dichotomy between ‘digital natives’ and ‘digital immigrants’ gains currency. In contemporary societies, networked families (Kennedy et al., 2008), in which generations share and benefit from technological innovation, coexist with families where technology reinforces a pre-existing generational gap based on literacy.
However, unlike previous studies, our analysis shows that the ‘digital class divide’ is not linear. Within affluent families, different patterns emerge: although material access to technology is given in both cases, highly educated parents tend to nurture (which can be a way of constraint) the internet use of their children whereas less qualified but well-off parents are less intervening and their children more self-reliant. Equally, children from less privileged backgrounds can overcome social hurdles, becoming highly proficient users, through self-taught efforts and peer group support, even though the risk of remaining an incipient user is higher and the degree of return brought about by internet use is significantly lower.
Returning to the individual attributes, age is a major predictor of internet use. Digital skills, use intensity and navigational abilities rise in such a dramatic manner alongside children’s ageing and socialization processes that other powerful explicative variables such as class or educational background are somewhat relegated to a minor role. Growing up means climbing a ladder, from incipient and hesitant practices to more intense, diverse and reliant uses of ICT. Younger children are typically more influenced and supervised more closely by parents. Older children are more autonomous, surfing the web on their own, mostly following the influence of peers, which mirrors their transition from family to peer group socialization in other spheres of life. This is why younger children (8–10 years old) are particularly visible among ‘nurtured beginners’ or ‘cybernauts’, while older ones prevail among the ‘self-reliant cybernauts’.
Although not as significant as in other research (see, for instance, Livingstone et al., 2005), gender also plays a part in differentiating internet use and in distinguishing degrees of parental influence. Even if intensity of use shows a rather balanced distribution between girls and boys (with around 60 percent intensive users in both groups), reinforcing the notion that diversity does not play a major role at the technological entrance gate (democratization of access), but rather that the distinct modes of appropriation are structurally embedded in unequal social contexts. Gender stereotypes and socialization practices also shape digital diversity. Very much engaged in their school project (Baudelot and Establet, 1992), attentive to family relations and respectful of educator authoritative control (Kirwill et al., 2009; Livingstone and Helsper, 2008), girls have a greater propensity to engage in online educational activities, to communicate with peers and family and to follow their parents’ advice and suggestions. Conversely, boys, growing up with more permissive socialization norms, tend to overemphasize their digital skills and knowledge (disregarding parental aid and advice) and to either engage in a diversified internet use (‘all-round cybernauts’) or turn into heavy game players. They are the majority among the ‘unguided rookies’. Apart from computer skills, where boys are slightly more proficient than girls, there are no other relevant differences in intensity or quality of internet use with respect to gender.
Finally, a word on the national setting of this study: a digital divide does exist in Portugal, between adults, whose internet use rate is lower than the European average, and children, who are much closer to their European counterparts. Our results may potentially be extrapolated to other contexts, since sociodemographic factors transcend national frontiers. Thus, it is probable that these four clusters can be found in other countries, albeit with different sizes, according to specific socioeconomic structures.
Conclusions
Media appropriation strongly shapes contemporary childhood, but this does not mean a homogeneous experience for all children. Digital diversity is evident and it is deeply shaped by family social background, parental mediation or individual variables such as age and gender. As in other domains, it seems more appropriate to consider the coexistence of a plurality of ‘childhoods’ instead of a universal and single category of ‘childhood’ in society (Prout and James, 1997; Sirota, 1998). The topological map and the clusters presented in this article to demonstrate the digital diversity among children, based on a wide range of variables and focusing on internet use, should contribute to this debate.
Childhood deserves conceptual autonomy and to be studied in its own right. However, within this paradigm, giving children a voice should not imply detaching them from their family (Brannen and O’Brien, 1995), since it is the strategic micro-setting where macro inequalities are (re)produced and where generational relations are forged (Alanen and Mayall, 2001).
As demonstrated in this study, children’s location in one of the four clusters is closely associated to their parents’ educational and professional background, which in turn favours distinct forms of parental mediation. In well-off families, educated parents are themselves internet experts and active tutors and supervisors of their children’s digital literacy, creativity and proficiency. In lower class families, children are the leaders of new media appropriation, while parents are kept out of the system, reinforcing the notion of a generational and knowledge gap. Dichotomies opposing ‘immigrants’ vs ‘natives’, adults vs children (Prensky, 2001), or perspectives announcing the rise of a young ‘revolutionary’ ‘digital generation’ in a globalized world (Tapscott, 1998) appear to oversimplify a significantly more complex and diverse social reality. There is no ‘technological fix’, such as providing universal access to computers and the internet, that will single-handedly eliminate the weight of social inequality on abilities to make use and reap the benefits of digital resources.
Age and, to a lesser extent, gender also affect diversity. On a par with social space, time is a key variable in describing and explaining digital differentiation between children. However, in addition to the synchronic portrait of these different types in the present (as presented in this article), a longitudinal approach following different cohorts of children as they grow up could bring new insights about modes and speeds of progression between clusters along the time curve. The emergence and diversity of individual and group digital trajectories is certainly a promising area of research for the near future. The discursive dimension of internet use and activities, i.e. their meaning and representations for the children involved, is missing in this first stage of research, which explicitly adopted an extensive and quantitative perspective. This is to be complemented with the qualitative and in-depth approach planned for the second stage of the research project.
Footnotes
Acknowledgements
We are grateful to the Calouste Gulbenkian Foundation for funding this project and to the two anonymous referees for the extremely useful comments.