Formal and substantial Internet information skills: The role of socio-demographic differences on the possession of different components of digital literacy

The literature about digital inequality has pointed out the role of so–called “digital skills” in contributing to a full exploitation of the opportunities of the Web for individuals. Research has started to measure the differences in online skills on a socio–demographic base, finding relevant disparities. Since different components of digital skills have been described in theory, it is not clear which of them are influenced by specific social variables and which are not. This study goes a step further in the analysis of “digital skills,” concentrating on two different components of them: “formal information skills” and “substantial information skills.” Complex search tasks were assigned to a quota sample made up of young people in northern Italy, divided by gender and education level. The results show that when other important variables associated with digital skills (age, experience with the Web, availability of hardware and software) are kept constant differences in gender and education have an influence on substantial information skills, but not on formal ones. This provides evidence for the assumption that a substantial part of digital skills represent a socially relevant factor for digital inequality, even when education and the spread of new media have standardized the level of formal skills. Introduction: Inequalities in Internet use What kind of ability is it which quickly and correctly finds different kinds of information on the Internet when it is needed? Is it a manual–technical ability whose measurement is to be considered by the same standards as videogame playing ability? Or is it an ability which can influence the cultural and social life of an individual? In the sociological tradition, differences in using media and accessing information among individuals has very rarely been considered a relevant source of inequality. Even in the specific field of mass media sociology (with the relevant exception of the knowledge gap hypothesis [1]), the abilities to use and exploit media have not been linked directly to social inequality. The advent of digital media has changed this situation and opened a wide debate about inequalities derived from different possibilities of exploiting new informational opportunities. The use of the Internet differently from that of mass media (like radio and TV) is based very much on personal resources. The activities of information selection, evaluation, filtering, recombination — which were once carried out by media professionals — are shifting back to the individual. As such, individuals have more freedom but also find themselves alone in building their own agendas, content and strategies relative to information consumption. The digitalization process has been the means for getting media consumption out of the Fordist era, offering possibilities of personalization and customization already present in other areas of consumption. However, in the case of consumption of information, differentiation can have larger and less predictable social consequences impacting more profoundly on the cultural, professional and social lives of individuals. Initially worries about the “digital divide” were focused on the growing disadvantages experienced by those without a connection to the Internet. However, since 2000 much literature has been devoted to enlarging the concept of the “digital divide” into a broader one of “digital inequality” (DiMaggio, et al., 2004; Hargittai, 2002, 2004a; van Dijk and Hacker, 2003; van Dijk, 2005). Apart from the problems based on a sheer lack of connection, notions of “digital inequality” examine the effects of differences in the use of the Internet among people who already have physical access to the computers and digital networks. As new media spreads and becomes more pervasive in our daily lives, the real digital divide seems to be based on what people do with media, more than on technologies and connectivity available to them. As van Dijk [2] effectively states, it seems that the problem usually called “digital divide” is “deepening where it has stopped widening”, that is when the number of people without access to the Internet diminish, a new and more profound divide can be created because of different levels of ability and opportunities to exploit a very complex medium with diverse and abundant content. Many authors have proposed various descriptions of the causes of “digital inequality”. Particularly relevant among the components of digital differences are: autonomy of use (DiMaggio, et al., 2004); motivation (van Dijk, 2005); support from social networks (DiMaggio, et al., 2004); and, types and impact of use (Bonfadelli, 2002; van Dijk, 2005; Liff and Shepherd, 2005). This paper examines digital skills, described by a number of other researchers (van Dijk and Hacker, 2003; van Dijk, 2005; Warschauer, 2002; Hargittai, 2002, 2004a, 2004b; Hargittai and Shafer, 2006; Steyaert, 2000; Norris, 2001). To date, there have been few studies which try to empirically observe “digital skills”. Van Dijk (2005) notes that “despite a growing literature on Web–site usability issues (much of it from library science and social informatics), we know little about how and why skill is related to personal characteristics”. Most of the works focus on psychological differences, pointing out the importance of personal characteristics such as field dependency/independency (Wang, et al., 2000; Palmquist and Kim, 2000; Kim and Allen, 2001; Juvina and van Oostendorp, 2004); gender (Ford, et al., 2001); experience (Lazonder, et al., 2000); and, locus of control, motivation and interest (Juvina and van Oostendorp, 2006). Even scarcer are the studies which take socio–demographic characteristics into consideration. Hargittai (2002) reported the first research of this kind based on a random sample of residents in a socially heterogeneous New Jersey county. The subjects were tested while they tried to locate several kinds of material online. Hargittai defined the concept of skills as the ability to complete a task and the amount of time spent for completing it. The author found large differences among the subjects, influenced by age, education and prior experience with the technology. Hargittai did not find significant differences based on gender. But which of the different components of digital skills are we talking about? Here I argue that social variables have a different influence on digital skills depending on which component of them we test. The research reported here makes a first important distinction: on the basis of van Dijk’s model (van Dijk, 2005), I try to observe “formal” and “substantial information skills” separately. In this paper research results attempt to differentiate and separately measure formal and substantial information skills in retrieving information on the Internet. This is done in order to understand if the two components of skills are influenced differently by social characteristics such as gender and education capital while maintaining other variables constant. The results suggest that formal and substantial skills have very different social implications. Formal and substantial information skills Many authors have tried to define what can be assumed by the term “digital skills”. This very complex concept is formed by different problems, and is connected to what is believed to be a real new form of literacy (Warschauer, 2002). Many scholars who analyzed the concept of digital skills distinguish between two main components: “technical/operational skills”, which are needed to operate computers and basic software, and “informational skills”, those needed to navigate, select, evaluate and re–use information. The main feature of this second component, “information skills”, is to be formed not only by skills directly connected to the use of new media, but also by skills which were present in traditional literacy and become even more important in the era of digital media. Van Dijk (2005) proposes what is probably the most interesting model so far to distinguish between different parts of digital skills. • Operational skills • Informational skills o formal o substantial • Strategic skills The first component, that of “operational skills”, regards the ability needed to operate computers and Internet connections with their basic applications. Its standardization is well represented by the seven modules of the European Computer Driving License (ECDL). Individuals in many western countries demonstrate considerable differences in operational skills [3]. However, differences in this component seem to be resolved in the long term with formal education (parallel to literacy) and especially with the natural spread of technologies in the population. This makes differences in operational skills more a contingent rather than a structural social problem. The third component, strategic skills, concerns the abilities to determine the general goals of media use in a universe of information overload and disorganization. Strategic skills are “defined as the capacities to use computer and network sources as the means for particular goals and for the general goal of improving one’s position in society”. [4] As van Dijk argues, such a goal–oriented attitude is more important with new media and in a “network society” than it was in the past. We focus in this paper on the second component: “information skills”. Van Dijk defines them as the skills to search, select and process information in computer and network sources. As strategic skills are very hard to measure, most research on Internet skills implicitly or explicitly focuses on information skills. In fact, they can be observed when subjects need to solve complex information retrieval activities, when operational skills are constant and when the task is given [5]. Van Dijk proposes an interesting sub–distinction of information skills. “Formal information skills are the abilities to understand and work with the formal characteristics of a particular medium ... Substantial information skills are the abilities to find, select, process and evaluate information in specific sources following particular questions.” [6] In other words, formal information skills are related to the forms in which contents are organized inside new media, such as the Internet. For example, these competences are needed to move rapidly and efficiently about hyperlinks, to know and make proper use of features such as bookmarks, to understand the structure of an URL and to be able to locate the home page starting from a complex address. However, more profound skills are needed to use the structure with proper content. Substantial information skills are needed for this deeper digital literacy component. They are needed to search, select, process and evaluate information as it is presented on the Internet and new media. Examples of activities which need substantial information skills are: evaluate the nature of a site (commercial, personal home page, governmental, lobby), evaluate the reliability of the contents of a site, be able to select the most useful links in a search engine page results, simplify complex research activities, produce good queries for search engines [7] In the practice of navigation, formal and substantial information skills are interrelated. For example, an individual can be good at evaluating the content of a site (substantial skill) if he knows how to efficiently use links and combo menus in order to easily navigate it and process its contents (formal skills). However, these two components of skills are also conceptually very different. In the analysis of Internet information retrieval skills (as said before, mostly focused on information skills) identified differences were not divided between these two levels. We know that social characteristics such as age, gender, education can have an influence on information retrieval skills on the Internet (Hargittai, 2002, 2004a, 2004b; Dietrichson, 2001). But we do not know what specific kind of skills can be most influenced by these social variables. By knowing this, it will be possible to understand where the social relevance of the differences of digital skills lies and to better focus policies and solutions. The goal of this study is to investigate if there is a different influence by gender and education capital on formal and substantial information skills, when other important variables (age, experience, conditions of access) are kept constant. We do this by separately measuring the performances of people in tasks which are related to one or the other component. My hypothesis is that gender and education influence Web information retrieval performances, even among people with the same physical access conditions and the same age and length of experience in using the Internet. In particular, among users with these characteristics, gender and education capital influence performance in information retrieval only for substantial information skills and not formal information skills (as van Dijk, 2005, defines them). This hypothesis is derived from preliminary and non–scientific observations of information retrieval activities. The hypothesis can also be inferred theoretically by van Dijk’s model: the higher the level in the hierarchy of digital skills, the less abilities depend on the mere familiarity with the media and are linked to more profound skills. Gender and education are two important sociological variables in the field of new media use. The second has been considered by many scholars as a relevant predictor of selective use, good acceptance and storage of information, both in “old media” (Tichenor, et al., 1970) and in new media (van Dijk, 2000, 2005, 2006; DiMaggio, et al., 2004; Hargittai, 2002, 2004a, 2004b; Bonfadelli, 2002). As for gender, it is also an important variable in the field of new media studies. Some research (Hargittai and Shafer, 2006; Liff and Shepherd, 2005) considered it more relevant with respect to self–evaluation of skills (it is clear that females tend to underrate themselves) than to real performances. Hargittai and Shafer (2006) found that females are equal to males when it comes to actual performance. Will this finding be confirmed after a separate analysis of formal and substantial information skills? To separately test the two components, I have used an exploratory methodology. Navigation performances inside closed hypertexts (especially inside a single site) have been used as indicators of formal information skills. In closed hypertext navigation tasks, users only have to locate information in a given information structure; they follow links, use combo menus, maps and internal search engines. In this activity, formal information skills are extremely prominent since finding an information is only an activity of location. Navigation performances involving the entire Web have been used as indicators of substantial information skills. To retrieve information in the open Web, not only formal information skills are needed but mostly substantial information skills (especially in the first parts of the session). To be successful, users have to create appropriate queries on search engines, select results efficiently on the basis of their quality, and know where to look for resources. Although in open Web navigation also some formal information skills are needed, performances in these types of activities can be used to measure the impact of substantial information skills in comparison with performances in closed hypertext tasks, where mostly formal skills are needed. If my hypothesis is true, we expect to find different effects of gender and education capital in closed hypertext and open Web tasks. In the former differences should be negligible while in the latter differences should be relevant. This research is significant to our understanding of digital inequalities. It is possible that with the spread of the Internet and new media to the majority of the population operational and formal information skills will be standardized by use and by more formal training in schools. If the hypothesis is true, it means that even among people of the same age, with the same technical tools and experience with the Internet, social characteristics will influence those skills needed for a complete exploitation of the Internet’s potential. By understanding if substantial information skills continue to be influenced by social characteristics, even when formal information skills are constant, we can more specifically examine which skills are relevant to inequalities in the long term. Hence, these results are relevant to educational planning and for policies aimed at reducing digital inequalities. If my hypothesis is true, schools must devote a special effort to substantial information skills instead of operational skills.