Abstract

Studies of Internet use continue to show a gap between those with and without access to the Internet and its resources. However, recent work indicates that this is not a straightforward divide about access; instead there are many variants of access, use, and presence online. This paper examines these variants, bringing together primarily US and European studies on the digital divide to identify social facilitators and inhibitors to online use that can inform policy and practice. This paper reviews research on who is and is not online, where access is gained, and what promotes or inhibits use of the Internet. This leads to identification of a number of interrelated social and technical factors that underlie access differences, including technical and social infrastructures, social networks, and content.

1. Introduction

While statistics from surveys in the early 2000’s have suggested that differences in access across traditional categories of age, race, gender, and socioeconomic status are disappearing, recent surveys are less optimistic that the digital divide is – or ever will – close. Many question the measure of access as the primary indicator of “electronic equity” ([46] p. 80) since differences continue to be present across demographics in time spent online, activities engaged in online, bandwidth used, and perceived usefulness of Internet technology ([35][36] [27][23][7]). Persistent differences in interest between users and non-users suggest that current divides are not passing phenomena, but instead represent different ways of approaching and participating online ([29][7]).

The many patterns of digital use belie the notion of a digital divide ([38]), with its connotations of all or nothing access to the benefits of contemporary information infrastructures, and its implications of barriers to access that can be swept away with a single intervention. Recent work addresses inclusion and exclusion as a digital spectrum rather than a divide (e.g., [29][43][47][56]). But even a spectrum suggests access and use that fits neatly on a line from low- to high-end use, with connotations of inferior and superior participation in today’s western, technologically-oriented world. What is emerging is a multi-patterned, multi-faceted, and fluid picture of use ([3][29][55][7]). Ideas of literacy and fluency become more appropriate in this case. This paper re-examines research on Internet access and use to identify social facilitators and inhibitors that, instead of contributing to a divide or spectrum, influence individuals’ digital literacy and fluency.

2. Current State of the Digital Divide

Early studies of computer and Internet users revealed a population that was predominantly young, white, male, of high socio-economic status, living in developed countries, and English speaking. As more people of all descriptions come online, there has been hope that the divide was closing. Major studies in the U.S., Canada, and Europe have shown that since the early 1990s, there has been considerable growth in the percentage of formerly under-represented groups online including women, non-whites, older users, low income earners, non-college graduates, and non-native English speakers (e.g., [27][40][11]).

Yet, overall growth is not equally distributed. A 2005 report on the expansion of Internet use in the EU by the Commission of the European Communities (CEC [7]) finds that while the gender gap seems to be closing, and more unemployed and self-employed are coming online, “Internet penetration among housepersons, especially women, older citizens, retired people and in rural areas is clearly lagging behind” (p. 8). Similarly, a Pew Internet and American Life study by Lenhart and colleagues found that, in 2002, “56% of [the 42% of Americas who are] non-Internet users do not think they will ever go online. These people are generally the poorer, older segment of the not-online population, and are more likely to be white, female, retired and living in rural areas” ([29], p. 4). Van Dijk and Hacker ([55]), drawing from U.S. and Netherlands data, also paint a pessimistic picture about closing the digital divide. They find increasing relative differences across demographics. Households with higher incomes, a head of household with higher education, lower age of head of household, and white ethnicity (U.S. data only) show greater possession of computers with the gap widening in percentage terms from 1984 to the late 1990s (1998 or 2000 depending on the studies available).

Reports such as these point to persistent patterns of Internet use that indicate more than specific barriers preventing willing participants from getting online. This is well described in the CEC report [7] which suggests the unlikely disappearance of the digital divide. This report identifies three patterns of Internet uptake, and describes how these are evident in traditional categories of the digital divide. The first pattern suggests the digital divide may be a temporary issue, with groups catching up to forerunners in the medium term. At present this looks like the case for the gender divide, and for older populations, particularly as the latter increasingly include digitally aware age cohorts. However, recent figures from a September 2005 Pew survey, which show a much higher number of Americans online (72% overall; [48]), also shows gender differences again—75% of U.S. men online, compared to 69% of women—so perhaps the gender gap is not after all a temporary divide, but instead a recurring one.

The second pattern noted by CEC suggests the digital divide is an issue of ever evolving delays, with groups catching up in the very long term, but lagging behind with every new innovation. The CEC report finds this to be likely the case for low income and less educated groups, and possibly the case for some countries newly entering the EU, and for rural areas and users. This is also likely the pattern for differences across other countries. Data on the percentage of the population online show high levels (over 50%) for countries such as Australia, Canada, Hong Kong, Japan, New Zealand, Singapore, South Korea, Sweden, Taiwan and the US, and mid to low levels for Aruba (34%), Bahrain (21%), Chile (20%), Israel (17%), Argentina (10%), Kuwait (9%), Lebanon (8%), South Africa (7%), and Brazil (7%) (2001, from [42]). Elsewhere in the world numbers are far lower, even in European countries where 1% or less are online in the Ukraine (in 2001), Bosnia (in 2001), and Albania (2000) ([42]). The lags are likely to remain; even as each country increases its own percentages, catching up is going to be hard to do.

The third pattern suggests the digital divide is a continuing issue of delay and exclusion, with considerable delay between social groups, and some groups never catching up. CEC suggests this is possibly the case for low income, less educated groups, and looks like it is the case for some countries, and for rural areas and users. Other reports show that this may also be the case for the disabled. Adults and children with disabilities are proportionally under-represented online. In 2002, 38% of Americans with disabilities were online compared to 58% of all Americans; 28% of these non-users reported their disability made it “difficult or impossible for them to go online” ([29]); and Livingstone and Bober ([43]) found UK children and young adults with disabilities used the Internet less frequently than others. As Rice [62] has pointed out, delayed use is impoverished use since individuals have not gained the experience and skills from being online. Thus, delay is an issue of concern.

These three patterns, like much of the rhetoric of the digital divide, refers to not being online as a failing in keeping up, whether temporary or permanent, the rhetoric is of lag and delay, all predicated on a benchmark of practices of early adopters – white, young, male, affluent, English speaking, in developed countries. Continuing a long standing western, pro-innovation bias (Rogers, 1995), the standard is the adoption of the new rather than alternate measures of practice, norms and culture. The rhetoric of delay, and the unspoken assumption that being online is e-inclusion, are too simplistic and taken-for-granted for the complexity of the international, multi-cultural, multi-use and multi-purpose phenomenon that is the Internet.

When a standard is based on a moving target – early adopters – it is particularly susceptible to the rhetoric of delay because no matter how fast one adopts, the benchmark is moving just as fast. However, since this means that every user will in some way lag behind the earliest adopter, it becomes a meaningless distinction. Developed countries already see that those who own computers and use them may still lag in terms of use of the latest technologies. Thus, it is more appropriate to move from issues of ownership, adoption and use of a specific technology to digital literacy and fluency, looking to see if users have created a meaningful place for technology in their lives. To explore a more nuanced view of inclusion and exclusion, we need to look at use, and at the social facilitators and inhibitors to online fluency that exist even when individuals have achieved access to a computer and a connection to the Internet.

2.1. Usage Differences

Disparities in gross measures of access are only part of the data about being online. Differences across age, gender, race, and socio-economic indicators are evident in ways the Internet is used. Howard, Rainie and Jones ([36]), for example, found that on any given day, among those with access, more of the men, whites, higher income earners, more educated and more experienced users are likely to be online. Among UK children, boys spend more time online than girls, have been online longer, have more online skills, and higher levels of self-efficacy ([43]). Another recent study also finds that men have higher perceptions of their abilities online than do women [63]. The following examines some of these differences in activities and use by gender, race, income, occupation and experience online.

While some of these differences may be ways that one demographic lags behind another, other may be a matter of choice or preference. The listings below show differences, but, at present, these should not be taken to be inferior or superior types of use.

Gender: Several studies find that men are more likely than women to use the Internet for news, product, financial, sports and hobby information, online trading, banking and gaming ([36][37]); women (and seniors) use it more for health information ([39]). While both men and women use email extensively, women report a stronger liking for email than men (78% of women say they look forward to reading email vs. 62% of men; [49]), using it for relationship building ([50]), and somewhat more for connecting with family ([21][49]).

Race: A 2003 report from Pew ([30]) states “there are few instances where online whites, blacks, and English-speaking Hispanics report equal participation levels” (p. 77). Non-whites trail in use of email, but use more IM and chat rooms than whites; are less active in information activities, but more active in hobby and entertainment activities (e.g., downloading music, playing games, looking for sports information). This continues trends from 2001, when a smaller proportion of Black and Hispanic Internet users emailed, searched for news, conducted searches for product/service information or made online purchases ([39]).

Income: Different kinds of use are evident by household income (see Table 1). The higher the income, the more likely users are to engage in email/IM, searching for information, buying online, and banking and trading. Lower income households are more likely to be playing games, doing school work, searching for jobs, and chatting online.

Table 1: Online Activities by Household Family Income, 2001. Percentage of Internet users age 3 or older; income endpoints only; trends are consistent across categories of income ([39], p. 34)

Income < 15,000 >75,000

Increase with increasing income

E-mail/IM 72.0 89.1

News/weather/sports 53.5 67.0

Product/service info. 54.9 73.5

Health info. 29.5 38.9

Government info. 28.1 35.1

Product/service purchases 26.1 49.1

Online banking 12.8 23.0

Trade stocks, bonds, funds 3.2 13.8

Approximately the same across income

Radio/TV/movies 20.0 19.8

Online education 4.0 4.0

Decrease with increasing income

Playing games 47.0 37.5

School assignments 37.1 24.6

Job search 23.0 14.6

Chat rooms/listservs 23.0 16.5

Phone calls 6.7 5.1

Age: Fewer older users engage in each of the different kinds of online activities, although email is a major use by all age groups. Like women, those over 55 are more likely to look for health information. The major activities of children and young adults are schoolwork, email, online games, radio/movies, and using chatrooms.

Experience: Are the differences by gender, race, income and age the profile of disadvantaged groups, or of groups learning their way online and using what makes sense to them at their time of life? Considerable difference in use is found related to years of experience online. Experienced users spend considerably more time online: 6 hours a week for those with one year of experience, and over 16 hours a week for those with over 4 years online (U.S. figures: [37], see also [34][33]). These Netizens ([36]) are online more often, involved in more online activities than others, and online more for professional reasons. These experienced users have acquired a fluency with online life; they incorporate the Internet into all aspects of their lives, with greater comfort spending and managing their money online, and communicating online to support social relationships. They make less of a separation between being online and not being online, easily ‘toggling’ between on and offline life ([10]).

Yet, Netizens only represent 16% of the adult US Internet users (as of 2000). Along with Netizens, Howard, Rainie and Jones ([35][36]) find: Utilitarians (28% of adult Internet users) who approach the Internet functionally as a tool; Experimenters (26%) who are on their way to becoming netizens, moving beyond fun activities to information retrieval tasks; and Newcomers (30%) who have arrived online more recently, using it primarily for fun activities, and generally from one place (either work or home).

Since 30% of the users in each gender, race, age, and income groups can be expected to be Newcomers, learning their way online, many of the ‘lag’ effects may instead be learning effects because it is women, non-whites, and low income groups who have shown the greatest increase in online presence. For age, both old and young are new users, each also learning their way online, but with different preferences for activities. Thus, the patterns noted by the CEC report must co-exist at any time with particular user’s experiential and life stage. Although beyond the scope of this paper, it is important to tease out learning effects from persistent lag effects and apportion to each their impact on e-inclusion and e-exclusion.

Intermittent and non-users. While some are coming online now on their way to becoming Netizens – including the many new, young users who begin to use computers almost as soon as they can reach a keyboard – others are only partial or occasional users, with no interest, inclination, or support to make further use of the Internet. In 2002, a Pew study ([29]) looked in more detail at the then 42% of Americans who said they did not use the Internet. Of these, 69% (24% of Americans) were Truly Disconnected, not using the Internet nor having others near them using it. Others were intermittent users: some currently disconnected (Net Dropouts, 17% of non-users), and some connected (Intermittent Users, estimated as 27-44% of current users). Perhaps of most interest was a set of non-users called Net Evaders (20% of non-users) who had access to the Internet through others. This latter group reveals an important social aspect to Internet use – access through one’s social network (discussed below).

3. Social Facilitators and Inhibitors

Although data on use by each demographic is very useful, effects of each are intertwined and so even these data are too simplistic for explaining Internet and digital divide issues. For example, household income greatly affects children’s use of the Internet, but so does being in school; women’s use and the presence of computers in the household are affected by whether there are children present. The evidence also suggests that there are social facilitators and inhibitors at work in forming individual’s attitudes and practices with Internet technology. The following discusses facilitators and inhibitors under the headings of infrastructure, culture, individual differences, and social networks. Readers will, quite rightly, notice overlap in factors in these categories as can be expected for the non-linear phenomenon of digital literacy.

3.1. Infrastructures and Access

Digital divide issues play out in differences in where individuals gain access to technology, and where societal infrastructures have provided the means for access. Network infrastructure is important for all users, but access to local computer technologies is particularly important for those who do not own computers, for intermittent users and for those who use the Internet through others. In the US, non-users are not that far from access: “60% of non-users know of a place in their community where Internet access is publicly available, while 76% of Internet users know of public access sites. Most of those who know of local access points say those access points are easy to reach. The most frequently identified location of public access is a library.” ([29], p. 3-4).

Children’s access provides an apt example of the impact of public access to computers. A U.S. National School Board Foundation study in 2000 of where children used the Internet [31], showed low income and African-American children overwhelmingly more likely to use the Internet from school than home: Household income: <$40,000 : 68% from school and 46% from home; >%75,000: 57% from school and 86% from home; Race: White children: 56% from school and 73% from home; African-American children: 71% from school and 35% at home.

Schools then are an important access point for those traditionally on the exclusion side of the digital divide. But schools need to recognize and take on that role. In their study of UK children, Livingstone and Bober’s [43] found both a lack of formal instruction about the Internet and a lack of critical skills for evaluating Internet content. They point to the schools as an important location for increasing all students’ knowledge about computing and the Internet and raising information literacy skills.

Combining concerns about information literacy with findings about access by income and race suggests an even greater need for school-based learning as a means to address digital divide issues. Although Livingstone and Bober note that the issue is now one of “quality of use (as assessed by time use, skills and range of online activities)” rather than of technology access (p.12), household lack of means or interest can bring the site of access into play in forging quality use.

Network Infrastructures. Disparities across geography carry with them variation in local infrastructural support such as electricity, broadband, and wireless capabilities, typically privileging urban over rural users, and developed over developing countries ([60][7][8][58]). It is often taken-for-granted that access via broadband or other high speed connection is available, and thus, that applications can be written that require such transfer. For example, a 2005 study in the U.S. by National Center for Academic Transformation ([32]), examining the use of technologies for education of underserved students in higher education, found that making accommodations for the low bandwidth access students had was the second most important issue after access: a quiz feature at one university did not work for dial-up users; and another university had to make a special effort to optimize 400 graphic, audio and video files used in a particular course. These effort appeared to be worthwhile. Results showed the online components that were added to 15 courses (3 fully online programs, 12 blended with off and online components) greatly helped the underserved students. Extra access to help for review of materials, and the added flexibility of working online, helped working, adult learners because it gave them choices about when and where they would learn.

Broadband. While geographical constraints are often cited as the reason broadband is not widely used, other factors such as Internet use and government policy play a large role (see [61] for a discussion of U.S. policy). Use of broadband is another indicator of the divide between users: broadband adopters are much more intense users that those continuing to use dial-up systems. A 2005 Pew survey ([23]) found that the 53% of Americans who go online from home using broadband (33% of the American adult population) engaged in 2.8 activities per day compared to 1.7 per day for dial-up users (indicating broadband users are 65% more active than dial-up users); similarly, the 54% of Canadians who use broadband are more intense users, spending 55% more time online than dial-up users, and viewing twice as many web pages (Nua [40]). Those not using broadband show a disinclination toward adopting this technology. While intense U.S. Internet users are adopting broadband and want broadband when they don’t have it, 58% of dial-up users do not want broadband (February 2004 data, [23]). This puts into question their ability to access new online content engineered to be broadcast and received via high speed access, and assumed to be easily accessible to all online users.

3.2. Culture

Content and Representation Online. The discussion so far has highlighted issues relating to users – getting them online with computers and Internet connections, getting them to use what is online. But where is the concern about what is online? While many rightly concern themselves with issues of access to resources, others are turning their attention to the content of the Internet. The numbers are clear that certain countries, regions, and demographies are not participating fully in populating and seeding the Internet. Where is their representation online? Fewer members of a population online means not only fewer able to download resources, but fewer also placing resources online, and fewer others to communicate with online from these cultures, regions, and countries.

One gross measure of this is the amount of content available in different languages online. Estimates put the number of English speakers online at between 32% (2005) and 44% (2001) (Internet World Statistics [25], 30.6%, March, 2006; Global Reach [15], 35.8%, 2004; Bowen [4] 44%, citing [12]), with later figures lower than earlier ones as the number of non-English speakers increases rapidly (with particular increases in Chinese speakers, [26]). Yet, estimates suggest 70% of Internet content is in English (Global Reach [15], 68.4%; Bowen [4] 70% for 2001). This shows approximately twice the content for English speakers than for others, even though users prefer to access the web in their own language ([54]). (For more on language diversity and the Internet, see [45].)

A growing area of interest looks to online content for explanations of use and non-use, as well as trying to remediate the lack of representation online. Not only language, but also resources that speak to particular cultures, ages and communities are important for making the move online worthwhile and relevant to potential adopters and users (e.g., [27][56][57][53][9][6]). Many grassroots and non-profit community networking initiatives strive to do just that, bringing the Internet and its resources to the local community in a way that is relevant and representative of its needs. (e.g., [2][6][13][17]).

Technology. Differences in culture also affect norms for adoption and use of different technologies (phones, cell phones, personal computers, laptops, palmtops), and thus the critical mass of others with whom individuals can share knowledge and communicate via particular media. Without the appropriate device (computer, cell phone) and network infrastructure, individuals lack the means to participate in the circle of social and informational activity around them – contact via phone or email, access to government, commercial, and other kinds of information. Different kinds of work also have made for certain preferences in media use. For example, some have found a preference among farmers for radio broadcasts for information delivery ([17]). Government adoption of the Internet for information delivery must be weighed against readiness of particular sectors to receive information this way. For example, during the 2001 foot and mouth disease crisis, the UK government used the Internet to post information for farmers, but one estimate put the number of farmers with computers at that time at only 25% ([17]). Public access could not close the gap in this case since farmers with infected farms were not allowed to travel, and others were greatly discouraged from moving about the countryside or meeting in public places.

Rural. Rural users are further impacted by the lack of organization on their behalf. Many cities have put efforts toward Internet infrastructures, either strategically as is the case in Philadelphia, Pennsylvania (although Philadelphia had to navigate legal challenges from internet service provider Verizon Communications), or by grassroots growth as in Seattle, Washington, and the “patchwork of smaller wireless networks, often funded by local councils” across the UK ([16] [24]). “Muni nets” are begin established in many cities around the world, including London; New York; Taipei; Calgary; Adelaide; St. Louis, Missouri; and Leiden, Netherlands [16].

There is some good news for rural areas in this respect. In the U.S., where opposition to city sponsored wi-fi has restricted practices such as Philadelphia’s move to establish a wireless (or ‘unwired’) city, internet service providers see no profit opportunity in rural initiatives and thus there is no opposition to regional government implementations. Although not (yet) widespread, in 2005, Wired News ([59]) reported a ‘Wi-Fi cloud’ over rural Oregon’s Morrow County, permitting farmers and businesses to access Internet services within a 700 square mile area. While city systems generally require users to pay for access, the Morrow County service is free to the general public, financed through contracts with the city and county agencies, and large farming business.

3.3. Individual Differences

Differences are evident among users and reluctant or non-users in their attitudes to the Internet. Research suggests awareness of the Internet is a key separator between users and non-users ([27]); for those aware of the Internet, perceptions of its usefulness further separate users and non-users including net dropouts, with the latter not considering or finding the Internet to be of interest or use to them. Lenhart et al ([29]) report that 52% of non-users say ‘I don’t want it’ or ‘I don’t need it’ as major reasons for non-use; 43% give worry about pornography, credit card theft, and fraud as major reasons; 30% say it is too expensive; 29% don’t have time; 27% find the Internet too complicated; and 11% say they don’t have a computer. European surveys show a similar profile (German Online Nonusers Survey, cited in [55]): half those not connected to the Internet report they did not want to be: 91% of non-users reported they don’t need a PC for job or privately; 80% that they don’t know any home use; 77% say ‘no time or liking’; 47% PCs too expensive; 41% PC means less social contact; 38% don’t think they can handle a PC; 28% say ‘PC’s damage health’; 23% ‘I reject computers’; 19% ‘PC for job sufficient’.

Lenhart et al ([29]) find that individuals who “are socially content—who trust others, have lots of people to draw on for support, and who believe that others are generally fair …feel they have control over their lives, …read newspapers, watch TV, and use cell phones and other technologies are more likely to use the Internet than those who don’t.” (p. 4). This profile reinforces the conclusion drawn by Nie [33] that connectivity goes to the already connected; those with more social resources are better positioned to take on the new technology.

3.4. Social Networks

Social networks are implicated in a number of ways relating to getting and staying online, gaining access to resources, and learning how to become Internet literate and fluent. Coming online can be a challenging activity for new users. They can be embarrassed by their lack of computer and typing skills ([29]), unfamiliar with conventions of online communication [5], unsure of the size and make-up of the audience online ([51]). Having family and close friends can provide both social and technical help for getting online. Similarly, access at work or school can help bring skills into the house to be shared with others (see below, [39]). The following sections explore how use by close friends and relatives is an important driving force for adoption and fluent use, but also an important access point for non-users. Recent statistics indicate that 74% of non-users have family members and close friends who go online ([29]).

Children. Children are a key reason that both computers and Internet access come into the household ([31][52][37]). A 2000 National School Board Federation study ([31]) found 36% of parents report children’s education as a primary reason for computer purchase (27% buy it for business reasons), and 39% as the reason for Internet access at home.

Income greatly affects use by children: younger users in high income households use the internet in much higher proportions than those in low income households. Among US children (10-17 years) in highest income level, 87.5% use the Internet overall, and 82.5% use it at home, whereas in the lowest income level, 45.7% use the Internet overall, and only 21.4% at home. Being at school greatly affects use by 18-24 year olds: 85% of those in school or college use the Internet, compared to 51.5% of those not in school ([39]; for more on children’s use, see [43]).

As noted above, school is a good setting for students to learn about the Internet. It is also a good way for knowledge and skills to reach the household. Children who are learning to use computers at school bring new computing skills into the home. Livingstone and Bober ([43]) report from their UK study of children’s use of the Internet, that “Children usually consider themselves more expert than their parents, gaining in social status within the family as a result. Among daily or weekly internet users, 19% of parents describe themselves as beginners compared with only 7% of children, and only 16% of parents consider themselves advanced [users] compared with 32% of children” (p.14). Studies also find that women’s access to the Internet is markedly higher when there are children in the household (70% versus 57%; [37]).

Occupation. Where an occupation involves computer use at work, it significantly enhances the likelihood of computer use at home, and thus in the household for children and other household members ([39][29]see Table 2). NTIA reports that in 2001, 77% of households in which someone uses a computer or the Internet at work own a computer or use the Internet from home, compared to 35% for other households; this holds across other demographic factors (income, education, race, age of adults). “Use at work not only acquaints someone with the utility of the technology, it also provides an opportunity to climb a sometimes frustrating learning curve in an environment with technical support. This acquired knowledge can then be taken home and shared with other members of a household.” ([39], p. 62-3).

Table 2: Internet / E-mail Use at Work by Occupation (Percent of employed persons age 25 and over in 2001; [39], p. 62)

Managerial/professional specialty 20.4

Technical/sales/administrative support 21.5

Precision production, craft, repair 8.2

Farming, forestry, fishing 10.0

Service 6.2

Operators, fabricators, laborers 3.6

Proxy Use. Lenhart et al ([29]) found that 20% of non-users – the Net Evaders – are not so distant from the Internet; they live with someone who uses the Internet at home; 74% of non-users say they have family members and close friends who go online. In studying Internet use in a farming community, Hagar [17] found that family members not in the household often served as online searchers for non-users (e.g., sons or daughters living on their own farms), and farm women were main users on behalf of the family. Once online, users feel more confident when they know who they are talking to, and know the relationships among groups of communicators [5]. Online learners report bringing the Internet to their family, as well as maintaining distant friendships because of their new found skills and confidence [22].

Family, children in school, and friends bring Internet skills into the household, giving the non-user access to the resources of the Internet, either by proxy, as they ask others to look things up for them, or directly as they learn to use the computer that is now situated in the household. Moreover, once online, access to others via the Internet through email, online groups, discussion lists, etc. accelerate access to further resources. The Internet brings others into the household – distant family, friends, or new online acquaintances – who can be resources for information and further help in acquiring skills.

Critical Mass. Adopting the Internet, particularly for communication purposes, is easier and more advantageous when there are known others to communicate with online. As a group rather than individual adoption issue, it is quite likely that pockets of non-use will exist solely because of group network effects rather than any other kind of demographic characteristic. However, demographics may exacerbate or extend the time it takes for adoption to begin and reach critical mass. This effect is also likely to be greater for adoption among those who do not already know each other (and/or fear interaction with strangers). Weakly tied communicators require an external authority to implement and support a communication infrastructure for their use because their own weak ties do not extend to commit to such an interaction. Strongly tied communicators – friends, family – are much more likely to be able to influence each other to adopt a new means of communication, as well as to call on each other for help in getting started ([18][19]).

3.5. New Literacies

Beyond classifications by traditional socioeconomic measures are differences in fluency with the new literacies of online communication. This includes not only facility with the primary language in which resources are presented (e.g., with English, the predominant language of the Internet; [29]), but also fluency in online language and communication norms (e.g., emoticons, SMS text), and the growing and emerging variety of online communication applications, technologies and programming environments (e.g., blogs, wikis, podcasts, html, RSS, java).

Multiple Innovations. Going online means adopting multiple innovations: not just technical innovations of computers, Internet connections, and Internet service providers, but also complex social and organizational practices that create online venues for socializing, self-expression, learning, sharing knowledge, and working together ([28][44][20][14]). Users differ in their comfort with exposing their thoughts and information on the web (and also with understanding the extent of that exposure), which is, in turn, related to their familiarity and identity with the online environment, including knowing the norms of the local online group or community, and how well they know other members of the community ([5]).

As online communication becomes more integrated into family communication, the more users can begin to go online in a way they perceive as safe. Instead of engaging with the whole of the Internet, they need only begin with family and friends. Similarly, as use becomes commonplace at work or in educational settings – where a critical mass of others is there to communicate with, and where support is available for use – the easier it is to become fluent with the new language.

4. Summary

The recent data on Internet use confirms what writers have been saying for the last five years – that notions of a divide, with a singular barrier of ‘access’ separating users from non-users is inadequate for furthering discussion of inclusion in digital society. New to the discussion is acceptance that differences in access and use will persist, and are rooted as much in social as technological factors. Also new is the notion that individual access is not the full picture, but that technical and social infrastructures – computer networks, computer terminals, social networks, schools, and libraries – play a major role for users by facilitating access and learning about the Internet. Wider scale social networks also matter in providing content of interest to potential and actual users, and the critical mass of others for and with whom to create and exchange these resources. This review suggests several interrelated layerings to the social issues encapsulated in statistical differences in access by race, gender, socio-economic status, and region that are important for future discussion:

Infrastructure. Private, institutional, and/or government agency support for the physical devices (e.g., computers) and networking capabilities (telephone lines for dial-up, broadband services); reliability and availability of electricity, broadband, public access terminals, computers and network access at work, school, and home.

Content. Materials of interest to readers of the Internet (information repositories, online courses, open access bulletin boards, commercial sites); materials in the language of the reader; content relevant culturally (e.g., relating to countries – France, Korea, China; or ethnicity, e.g., African American, Hawaiian); socially (e.g., gender, ethnicity, age appropriate); locally (e.g., regional, civic, local information); and to individual life stage and life course [1], user group or community, group lifecycle [22], and local concerns.

Social Networks. Others in one’s social network with formal requirements to be online (for work, school); others to communicate and engage with online; others to help with getting online and/or getting resources from the Internet for them, and for help in being online. Support in acquiring and using the technology (e.g., providing computers, teaching computer basics, troubleshooting computer and connectivity problems). A critical mass of (relevant) content; of people to start and sustain interactive forums, online groups; and of acceptance of results of online innovations, e.g., trust in online information, trust in credit card use online, acceptance of online degrees.

Literacy. Competence with technology hardware and software, applications and online environments; text, graphics and information literacies; and literacy in first and second languages. Fluency in use so that on and offline become seamless parts of a whole.

5. References

[1] Anderson, B., and Tracey, K. (2002). “Digital living: The impact (or Otherwise) of the Internet in everyday British life”, in B. Wellman and C. Haythornthwaite (Eds.), The Internet In Everyday Life. Oxford, UK: Blackwell. pp. 139-163.

[2] Bishop, A. P., Bazzell, I., Mehra, B., and Smith, C. (2001). “Afya: Social and digital technologies that reach across the digital divide”, First Monday, 6(4). http://www.firstmonday.org/issues/issue6_4/bishop/index.html.

[3] Boase, J., Horrigan, J. B., Wellman, B. and Rainie, L. (2006). The Strength of Internet Ties. Pew Internet and American Life Project. Retrieved May 28, 2006 from: www.pewinternet.org/ pdfs/PIP_Internet_ties.pdf

[4] Bowen, T. S. (November 21, 2001). English could snowball on Net. Technology Research News. (http://www.trnmag.com /Stories/2001/112101/ English_could_snowball_on_Net_112101.html

[5] Bregman, A. and Haythornthwaite, C. (2003). “Radicals of presentation: Visibility, relation, and co-presence in persistent conversation.” New Media and Society, 5(1), 117-140

[6] Brock, A. (2005). “ ‘A belief in humanity is a belief in colored men:’ Using culture to span the digital divide”, JCMC, 11(1), article 17. http://jcmc.indiana.edu/vol11 /issue1/brock.html

[7] Commission of the European Communities (2005). eInclusion revisited: The local dimension of the information society. http://europa.eu.int/comm/employment_social/news/2005/feb/eincllocal_en.pdf

[8] Davidson, T., Sooryamoorthy, R. and Shrum, W. (2002). “Kerala connections: Will the Internet affect science in developing areas?” In B. Wellman and C. Haythornthwaite (Eds.), The Internet In Everyday Life (pp. 496-519). Oxford, UK: Blackwell.

[9] Ess, C. and Sudweeks, F. (2005). “Culture and computer-mediated communication: Toward new understandings.” JCMC, 11(1), article 9. http://jcmc.indiana.edu/vol11/ issue1/ess.html

[10] Fallows, D. (2004). The Internet and Daily Life. http://www.pewinternet.org/ pdfs/PIP_Internet_and_Daily_Life.pdf

[11] Fallows, D. (2005). How Women and Men Use the Internet. http://www.pewinternet.org/ pdfs/PIP_Women_and_Men_online.pdf

[12] Gandal, N. and Shapiro, C. (2001). The Effect of Native Language on Internet usage. Paper presented at the Telecommunications Policy Research Conference, October 27-29, 2001, Alexandria, VA.

[13] Gaver, W.W., Dunne, A., Pacenti, E. (1999). “Design: Cultural probes.”Interactions, 6(1), 21-29.

[14] Girard, M. and Stark, D. (forthcoming). “Socio-technologies of assembly: Sense-making and demonstration in rebuilding lower Manhattan.” In D. Lazer and V. Mayer-Schoengerger (Eds.), Governance and information: The Rewiring of governing and deliberation in the 21st century. Oxford: Oxford University Press.

[15] Global Reach (Sept. 2004). Global Internet Statistics (by Language). Retrieved May 31, 2006 from: http://global-reach.biz/globstats/index.php3, and http://www.glreach.com/globstats/refs.php3

[16] Guardian Unlimited (March 3, 2005) “All wired up.” Retrieved Dec. 22, 2005 from http://technology.guardian.co.uk/online/story/0,,1428626,00.html

[17] Hagar, C. (2005). The Farming Community in Crisis. Unpublished doctoral dissertation, University of Illinois at Urbana Champaign.

[18] Haythornthwaite, C. (2002). “Strong, weak and latent ties and the impact of new media.” The Information Society, 18(5), 385-401.

[19] Haythornthwaite, C. (2005). “Social networks and Internet connectivity effects.” Information, Communication and Society, 8(2), 125-147.

[20] Haythornthwaite, C. (forthcoming). “Articulating divides in distributed knowledge practice.” Information, Communication & Society.

[21] Haythornthwaite, C. and Kazmer, M. M. (2002). “Bringing the Internet home: Adult distance learners and their Internet, Home and Work worlds”, in B. Wellman and C. Haythornthwaite (Eds.), The Internet in Everyday Life. Oxford: Blackwell. pp. 431-463.

[22] Haythornthwaite, C., Kazmer, M. M., Robins, J. and Shoemaker, S. (2000). “Community development among distance learners: Temporal and technological dimensions.” Journal of Computer-Mediated Communication, 6(1). http://www.ascusc.org/jcmc/vol6/issue1/ haythornthwaite.html.

[23] Horrigan, J. (2005). Broadband Adoption at Home in the United States: Growing But Slowing. 33rd Annual Telecommunications Policy Research Conference. Retrieved Dec. 22, 2005 from http://www.pewinternet.org/ pdfs/PIP_Broadband.TPRC_Sept05.pdf

[24] Intel (no date). Most Unwired City Survey. Retrieved Dec. 22, 2005 from http://www.intel.com/ personal/wireless/unwiredcities/

[25] Internet World Statistics (March, 2006). Top Ten Languages Used in the Web. Retrieved May 31, 2006 from: http://www.internetworldstats. com/stats7.htm.

[26] Johnson, B. (2005). English grip on internet being eroded. Guardian Unlimited, Technology Blog, Tuesday Aug. 16, 2005. Retrieved Dec. 22, 2005. http://blogs.guardian.co.uk/technology/archives/2005/08/16/english_grip_on_internet_being_eroded.html

[27] Katz, J. E. and Rice, R. E. (2002). Social consequences of Internet use: Access, involvement and expression. Cambridge, MA: MIT Press.

[28] Kling, R. (1999). “Can the ‘next-generation’ Internet effectively support ‘ordinary citizens’?” The Information Society, 15(1), 57-64.

[29] Lenhart, A., Horrigan, J., Rainie, L., Allen, K., Boyce, A., Madden, M. and O’Grady, E. (2003). The Ever-Shifting Internet Population: A New Look at Internet Access and the Digital Divide. Retreived Dec. 23, 2005 from: http://www.pewinternet.org/ pdfs/PIP_Shifting_Net_Pop_Report.pdf

[30] Madden, M. (Dec. 22, 2003). America’s Online Pursuits. The Changing Picture of Who’s Online And What They Do. Pew Internet and American Life Project. Retrieved May 28, 2006 from: http://www.pewinternet.org/ pdfs/PIP_Online_Pursuits_Final.PDF

[31] National School Board Federation (March 28, 2000). Safe & Smart: Research and Guidelines for Children’s Use of the Internet. Retrieved January 14, 2006 fromt: http://www.nsbf.org/safe-smart/full-report.htm

[32] National Center for Academic Transformation (NCAT) (2005). “Information Technology: A problem or solution for underserved students.” Retrieved Dec. 22, 2005 from: www.thencat.org/Newsletters/Oct05.htm#1

[33] Nie N. H. (2001). “Sociability, Interpersonal Relations, and the Internet: Reconciling Conflicting Findings.” American Behavioral Scientist, 45(3), 420-435.

[34] Nie, N. H. and Erbring, L. (Feb. 17, 2000). Internet and society: A preliminary report. Stanford Institute for the Quantitative Study of Society (SIQSS), Stanford University, and InterSurvey Inc. Available at http://www.stanford.edu/group/siqss/

[35] Howard, P. and Jones, S. (2003). (Eds.) Society Online: The Internet in Context. Sage.

[36] Howard, P., Rainie, L. and Jones, S. (2002). “Days and nights on the Internet”, in B. Wellman and C. Haythornthwaite (Eds.), The Internet In Everyday Life. Oxford, UK: Blackwell. pp. 45-73.

[37] Cole, J.I., Suman, M., Schramm, P., van Bel, D., Lunn, B., Maguire, P., Hanson, K., Singh, R., and Aquino, J-S. (2000). The UCLA Internet report: “Surveying the digital future”. Retrieved May 31, 2006 from: http://www.digitalcenter.org/ pdf/InternetReportYearOne.pdf

[38] NTIA (2000). Falling through the net: Toward digital inclusion. National Telecommunications and Information Administration, US Department of Commerce. Available online at: http://www.ntia.doc.gov/ntiahome/digitaldivide/.

[39] NTIA (2002). A Nation Online: How Americans are Expanding Their Use of the Internet. US Department of Commerce.

[40] Nua (2001). US No Longer Dominates the Net. Available online at: http://www.nua.ie/surveys/ index.cgi?f=VS&art_id=905356771&rel=true

[41] Nua (2003). Forrester Research: One in five european seniors online. Retrieved Jan. 19, 2006 from http://www.nua.com/surveys/ index.cgi?f=VS&art_id=905358750&rel=true.

[42] Nua (2005). How Many Online. http://www.nua.com/surveys/ how_many_online/europe.html

[43] Livingstone, S. and Bober, M. (2005). UK Children Go Online: Final Report of Key Project Findings. Economic and Social Research Council: UK. Retrieved January 16, 2006 from: http://news.bbc.co.uk/1/shared/bsp/hi/pdfs/28_04_05_childrenonline.pdf

[44] Orlikowski, W.J. (2002). “Knowing in practice: Enacting a collective capability in distributed organizing.” Organization Science, 13(3), 249-273.

[45] Paolillo, J. Pimienta, D., Prado, D. et al. (2005). Measuring Linguistic Diversity on the Internet. Montreal, Canada: UNESCO. Retrieved May 26, 2006 from: http://unesdoc.unesco.org/ images/0014/001421/142186e.pdf

[46] Patterson, R. and Wilson, E. J. III (2000). “New IT and Social Inequity: Resetting the research and policy agenda.” The Information Society, 16, 77-86.

[47] Pinkett, R. (2000). Bridging the Digital Divide: Sociocultural Constructionism and an Asset-Based Approach To Community Technology and Community Building. Paper presented at the 81^st Annual Meeting of the American Educational Research Association (AERA), New Orleans, LA, April 24-28, 2000. Retrieved, Dec. 28, 2005 from: http://llk.media.mit.edu/papers/archive/aera2000.pdf

[48] Pew Internet and American Life Project (Dec. 8, 2005). Data Update. Retrieved Dec. 22, 2005 from: http://www.pewinternet.org/ pipcomments.asp?m=12&y=2005

[49] Rainie, L., Fox, S., Horrigan, J., Lenhart, A. and Spooner, T. (May 10, 2000). Tracking Online Life: How Women Use the Internet to Cultivate Relationships with Family and Friends. Retrieved Dec. 22, 2005 from: http://www.pewinternet.org/pdfs/Report1.pdf

[50] Boneva, B. and Kraut, R. (2002). “Email, Gender and Personal Relationships”, in B. Wellman and C. Haythornthwaite (Eds.), The Internet In Everyday Life. Oxford, UK: Blackwell. pp. 372-403.

[51] Sproull, L. and Kiesler, S. (1991). Connections: New ways of working in the networked organization. Cambridge, MA: The MIT Press.

[52] Statistics Canada (2000). “Plugging in: Household Internet use.” The Daily. Monday, December 4, 2000. Available online at: http://www.statcan.ca/Daily/English/ 001204/d001204a.htm

[53] Stanley, L. D. (2003). “Beyond access: Psychosocial barriers to computer literacy.” The Information Society, 19 (5), 407-416.

[54] Today Translations (2005). Language Usage on the Internet. Retrieved Jan. 19, 2006 from: http://www.todaytranslations.com/index.asp-Q-Page-E-Language-Usage-on-the-Internet—99144709.

[55] van Dijk, J. and Hacker, K. (2003). The digital divide as a complex and dynamic phenomenon. The Information Society, 19(4), 315–326.

[56] Warschauer, M. (2002). “Reconceptionalizing the digital divide.” First Monday, 7 (7), 1–15. http://firstmonday.org/issues/issue7_7/warschauer/index.html

[57] Warschauer, M. (2003a). Technology and Social Inclusion. Cambridge, MA: MIT Press.

[58] Warschauer, M. (2003b). “Dissecting the “Digital Divide”: A case study in Egypt.” The Information Society, 19, 297-304.

[59] Wired News (October 17, 2005). Wi-Fi Cloud Covers Rural Oregon. Retrieved Dec. 22, 2005 from http://www.wired.com/news/wireless/0,1382,69234,00.html

[60] Nicholas, K. (2003). “Geo-policy barriers and rural internet access: the regulatory role in constructing the digital divide.” The Information Society 19 (4), 287-295.

[61] Wilhelm, A.G. (2003). “Leveraging sunken investments in communications infrastructure: A policy perspective from the United States.” The Information Society, 19(4), 279-286.

[62] Rice, R.E. (2002). “Primary issues in Internet use: Access, civic and community involvement, and social interaction and expression”, in L. Lievrouw and S. Livingstone (Eds.), Handbook of New Media. London, UK: Sage. pp. 105-129.

[63] Hargittai, E. and Shafer, S. (2006). “Differences in actual and perceived online skills: The role of gender.” Social Science Quarterly 87 (2), 432-448.