The Dual Digital Divide The Information Highway In Canada The Dual Digital Divide The Information Highway in Canada (c) copyright 2000 PIAC Principal Author and Co-investigator Andrew Reddick (PIAC) Co-authors and Co-investigators Christian Boucher (Ekos Research Associates Inc.) Manon Groseilliers (Ekos Research Associates Inc.) With funding from: Human Resources Development Canada Industry Canada The Public Interest Advocacy Centre Suite 1204, 1 Nicholas Street, Ottawa, Ontario K1N 7B7 Canada Canadian Cataloguing and Publication Data Reddick, A.; Boucher, C.; Groseilliers, M. The Dual Digital Divide - The Information Highway in Canada ISBN 1-895-060-31-1 Table of Contents EXECUTIVE SUMMARY 5 THE DUAL DIGITAL DIVIDE 5 TYPOLOGY OF NON-USERS 7 SUMMARY OF FINDINGS 8 INTRODUCTION AND METHODOLOGY 10 Introduction: 10 The Internet in Canada 11 Dual Digital Divide 12 Methodology 14 ACCESS AND THE CANADIAN INFORMATION HIGHWAY - POLICY CONTEXT 15 General Policy Framework 15 Community Access Program 16 Community Learning Networks 17 OVERVIEW OF ACCESS AND THE CANADIAN INFORMATION HIGHWAY 19 Canadian Communications Household 19 Access and the Internet 22 General Non-User Attitudes and Values 24 Diversity of Choice and Diversity of Life 29 TYPOLOGY OF THE "NON-USERS" IN THE DIGITAL DIVIDE 32 Summary Dimensions of the Digital Divide 32 Who is lagging behind in the Digital Divide and Why? 41 Population Array of Non-Users 44 SOCIAL CONTEXT 50 The Immutability of the Dual Digital Divide? 50 Literacy 51 Incomes, Cost and Affordability 53 Diversity of Content 56 Changing Patterns in Consumption? 59 U.S. Digital Divide 60 CONCLUSION 63 REFERENCES 68 Executive Summary The purpose of this study was to analyze Canadian households at the margins in terms of access to the Information Highway, generally considered to be the Internet. These households have been described as the 'have-nots' or 'non-users'. This study provides a better understanding of the attitudes, practices and the needs of those who are not connected to the Internet and the circumstances under which some of these non-users are likely to go online. Analysis was based on a number of research questions. These included: How do we account for some members of social sub-groups being online and not others?; What are the most important factors or variables that affect decisions by people whether or not to be online?; What are the different types of non-user groups?; How important is cost/affordability as a barrier or obstacle to access?; What policy options are available to address connectivity problems? The findings of the report were based on several methodologies. A review was conducted of reports, articles, books and other documents from sources which included Industry Canada, Human Resources Development Canada, the Public Interest Advocacy Centre and statistical analysis by Statistics Canada and the National Telecommunications and Information Agency on the issue of the digital divide. In addition, a review and detailed multivariate analysis was undertaken on the results of the 1997-1998 study the Information Highway and the Canadian Communications Household to create a typology of non-users (Ekos Research Associates Inc.). Analysis in this study is correlated with the strategies, goals and, the access and content development models of the federal departments Industry Canada and Human Resources Development Canada (HRDC). These two departments operate major programs that, together, have the goals of facilitating access, individual and community development, and content development on the Internet. The study also identifies the implications and opportunities for other federal departments (e.g., Heritage, Treasury Board Secretariat) and other levels of government to address issues involving access and content. The Dual Digital Divide Canadians have experienced the burgeoning growth of the Information Highway and the Internet over the past few years. However, the levels of awareness and the use of these new technologies and services are highly polarized along social class and generational lines, creating a digital divide. From 1997 through 1999, higher income households were three times more likely to than lower income households to have home access. By 1999, about two thirds of upper income households had access from home as compared to about one in four low income Canadians. Contrary to the widespread enthusiasm about the Internet, a significant number of Canadians will remain unconnected for many years to come. However, the concept of a digital divide is much more complex than a simple differentiation between those who are connected and those that are not. In addition to the usually understood division of users and non-users (the first divide), non-users are not a homogenous group, but clearly form into two broad groups (the second divide). Together, users and the segmented non-user groups form a Dual Digital Divide. Of those not connected, in the first group (near users) are those who have varying degrees of interest in being connected, but are unable to due to a number of barriers, of which the most important are cost/affordability and literacy. The second group (distant users) consists of those who have little or no interest at all in the Internet and online services, or perceive no value in the Internet to meet their every day economic or social needs. For this latter group, if faced with the need to access the Web in the future, it can be expected that barriers, such as availability of service, cost, literacy and capacity to use, etc., will become predominant factors of concern. Evidence suggests that it is highly unlikely that the Dual Digital Divide will be overcome. In future, if the Internet becomes as essential as local telephone service, then depending on market conditions (availability, cost), a public policy response will likely be required to ensure universal access. The variable levels of connectivity, lack of interest in use, and the lack of perceived need for access are factors that cut across all socio-economic groups. However, barriers and obstacles to access are aggravated for those in the lower social classes who have less resources or skills available to overcome these. Connectivity to the Internet exhibits a pattern of a differential level of access based on social class, generation, gender and value factors (perceived need and interest). This pattern mirrors the long established differential levels of access that exist for many other products and services in society (e.g., reading materials, education, cable TV, portable phones, etc.). Trends in connectivity suggest that this polarization in access is likely to become the norm in society, at least for the foreseeable future. Data from research conducted in late 1998 and early 1999 indicates that a plateau effect is developing for Internet access, whereby at least one in five upper and upper middle income households remain unconnected. There is still growth in connectivity with lower and lower middle income households. However, with the lower social class, connectivity remains low, overall, and comparatively the digital divide has actually widened since 1996. Typology of Non-users Using multivariate regression analysis, this study created typologies of non-users. The three typologies that were developed provide insights into the attitudes, values and social circumstances that affect connectivity. These typologies have useful implications for policy to address the communication and information needs of those who will likely remain unconnected. Type 1 non-users recognize some value of the Internet in meeting some of their needs, but face the primary obstacles of cost/affordability and technical skill development. This group is considered to be 'near users'. Social literacy is also an important factor for these individuals, as well as the other two Types. Public access sites (CAP) and community learning services (CLN) have important roles to play in meeting the needs of Type 1 non-users. Public education and awareness initiatives by organizations (community, NGO's, government) providing these services would likely facilitate and improve the use of the Internet by this group of non-users. Broadly speaking, Type 1 non-users tend to cluster around younger Canadians (44 and younger), but this is not exclusive. There is an important gender gap in the non-user typology (overall 61% women, 39% men). Women are over-represented in Type 1 by a ratio of 2 to 1. Type 2 non-users similarly face technical and cost barriers, but as great a challenge is the perceived or very real lack of personal or social benefit and value of Internet service. Improved content (economic, social and cultural), designed and developed to better meet the existing needs of these non-users, may facilitate greater online activities. Technical skills training, content relevant to social needs and the availability of public access sites are also important components of an access strategy for this group. There is a tendency (again not exclusive) for the Type 2 non-user to be in the pre-retirement cohort (45-55 yrs). At the federal government level, major roles exist in access and content activities for HRDC (CLN), Industry Canada (CAP), Heritage (social, cultural content), and Treasury Board Secretariat (TBS) (access and government services). Type 3 non-users are also relatively far removed from online activities, particularly in the near future. They are not likely to see or derive any personal benefit from access, and are not likely to have the resources or social skills and interests to benefit from current Internet access. Broadly speaking, seniors and retired Canadians (55+ yrs) tend to make up much of Type 3 non-users. For both Type 2 and Type 3 non-users, strategic needs assessments (social needs) combined with content development initiatives may increase the propensity for Net use in the longer term. The proportion of women is slightly higher in Type 2 and Type 3 non-user than men. It is important to note that all the non-user types are clearly dominated by women. From a public policy perspective, the reality is that the majority of non-users will need to be provided with communication and information services in a variety of traditional and standard formats or means, and accessible for those with different skill and income levels, for some time to come. The federal departments with mandates in these primarily social policy issues include HRDC, TBS and Heritage. Summary of Findings General: 1. For non-users, occasional and regular Internet users, core policy and program objectives of government will need to be concerned with such issues as: access; technological and social literacy; social capacity and application; indigenous social and cultural content development; and the continued availability of multiple means of access and information distribution, with comparable quality and service levels. Cost: 1. Cost/affordability was found to be the most important barrier or obstacle to connectivity. 2. Cost and technological literacy were found to be important factors in the likelihood of use in the future. For about half of all sub-groups the ability to use the Internet (technical proficiency and understanding) was important. 3. Affordability or the cost of Internet service was an important criterion for at least two thirds of those in the lower and upper middle, and upper income segments, and for just over half of those in the lower income demographic. Literacy: 4. Literacy as a barrier extends beyond reading, writing and technical literacy. It also involves social literacy or social capacity which involves individuals' abilities to understand and use information in ways that are beneficial and meaningful to their everyday lives. Government initiatives need to incorporate this broader meaning of literacy. 5. Access to the Internet will not, on its own, overcome the social and economic inequalities and cleavages in communities or society at large. However, as the Internet joins the mix of mainstream communication technologies, programs that address technological access and social proficiency will be necessary to create opportunities for people to maintain a competent level of participation in society. In the alternative, non-users risk being further disadvantaged over the long term. Public Access: 6. Those in the lower income groups are more likely than higher income households to have recently accessed the Internet from a public location, a friends house, Internet Café or some other community location. At the same time, individuals from all social classes have made use of public access locations, whether to learn about the Internet, how to use it, or because home access was not possible. 7. There is a need for ongoing government support for the integration of community access and networking initiatives. These community-based networking services create opportunities to effectively address some of the core issues of public access, training and content development. 8. Looking ahead, strategic analysis and planning will be needed on such questions as: the appropriate number and locations of public access sites; the development of community networking groups; local management responsibility; quality of service standards; services available to the public; content development; the terms of access for different services; and government funding obligations and responsibilities. 9. There is a need to improve communication strategies, at the local and national levels, to increase the awareness by non-users of the availability of access sites and, of information resources on the Internet that may meet their specific needs and interests. Content: 10. While there is an increasing abundance of commercial content on the Net, a major weakness in Canada is the lack of a diversity of quality indigenous, socially and culturally relevant content. Diverse socially and culturally relevant content (local, regional, national) is necessary for individuals to meet their daily social, economic and cultural needs. 11. There is a need for more public involvement and support in the development, cataloguing in a Canadian context (standards for access), and dissemination of citizenship and other socially relevant content. Diversity of Access: 12. In an analysis of preference by the public (Net users and non-users) of the use of different technologies to gain access to government for information, the public clearly indicated that a diversity of means of access is required and will be in the future (i.e., in-person, mail, Internet, phone, fax). The Internet is not displacing these, but becoming another component of the overall mix of means of access. Introduction and Methodology Introduction: The purpose of this study was to analyze Canadian households at the margins in terms of access to the Information Highway, largely in the context of the Internet. These households have been described as the 'have-nots' or 'non-users'. Preliminary research on Internet access and use has suggested that non-users at the margins were largely comprised of those in the lower social class and included those living in rural locations in Canada. The majority of the other non-users from the middle or upper social strata were considered likely to come online as public awareness and general use of the Internet by others increased over time. This study is intended to provide a better understanding of the access attitudes, practices and needs of those who are not connected, and in particular those in the lower social class segments. There are several research questions used for this analysis. How do we account for some members of a social sub-group (e.g., with shared demographics such as age, income, education, etc.) being online and others not? What are the most important factors or variables that affect decisions by people whether or not to be online? Are there different types of non-user groups, and if so, what are they? What attitudes, practices or other attributes help us to understand the differences that explain why some non-users are likely to go online, while others will not? Are there particular barriers arising from these which could be addressed through government policy or programs or by the market in some way to facilitate greater online access? How important is cost/affordability as a barrier to online access? How important is this variable in relation to other factors? Is cost mainly a problem for low income Canadian households not online? How important is this variable for higher social class segments? How important is cost/affordability for those already online? Analysis addressing these issues is largely based on the baseline research of the Information Highway and Canadian Communication Household study conducted from 1997 through 1998. For comparative purposes, and to help establish a broader sociological context, other research materials have been used in this study. These research materials include information from Statistics Canada, reports for the National Telecommunications Information Agency (NTIA) in the United States and other documents from government and non-government sources in Canada. Analysis in this study is correlated with the strategies, goals and, the access and content development programs of the federal departments Industry Canada and Human Resources Development Canada (HRDC). These two departments operate major programs that involve facilitating access and content development on the Internet. The Internet in Canada Over the past few years, Canadians have experienced the burgeoning growth of the Information Highway. The Information Highway involves a diverse array of information and communication technologies and related applications. While for many the Internet is considered the central new communications application, new information, entertainment, financial and economic applications are also being made available through wire-based, terrestrial wireless and satellite technologies. At the same time, the patterns of awareness and, in particular, the use of the new technologies and services indicate that a significant polarization exists along social class and generational lines. Results from the baseline study the Information Highway and the Canadian Communications Household found that the use of home computers and the Internet were significantly higher among upper income households. This research also found that youth and better educated Canadians tended to have the highest levels of access and use. These differences in access are not, however, unique to the Internet. Beyond, the Internet and computers, differential levels of access also exist in other communications technologies. These include standard services, such as cable television, and newer services such as cellular or PCS phone, telephone banking and satellite television (Ekos, 1998). Though consisting of many new technologies, the Internet has emerged as the backbone of the Information Highway. From 1996 through 1998, overall, access to the Internet from home by Canadians showed fairly rapid growth from twenty eight percent to thirty three percent of households. At the same time, most of this growth occurred with higher social class segments. Figure 1 Source: Ekos Research Associates Inc., 1998. As shown in Figure One, higher income households were about three times more likely than lower income households to have home access. Households in middle income groups similarly had differential levels of access, increasing with affluence. These results suggest that a polarization or differential level of access, already evident with other communication technologies and other services in Canada, such as health or education, could also become the norm with the Internet. Preliminary research from the 1999 version of the continuing Information Highway and Canadian Communication Household study indicates that while, overall, access to the Internet from home has increased significantly, most of this growth has occurred in upper middle and upper income groups. Lower middle and lower income households still lag in access with about the same proportional differences (23% access from home for lowest income level and 65% access from home for upper income segment) (Ekos, 2000). Dual Digital Divide When thinking about this digital divide, it is important to note that the notion of a divide is considerably more complex than simply some people being connected and others not. It also extends beyond the ideas of the availability of numerous community locations being available to facilitate access, when home access is not viable, or such issues as the technical ability or literacy necessary to use this form of communications. A more accurate or appropriate description of the emerging pattern of access and use of the Internet is that of a Dual Digital Divide. In addition to the division of users and non-users (the first divide), non-users clearly form into two broad groups, 1) those who have varying degrees of interest in being connected (near users), but are unable to due to a number of barriers, and 2) those who have little or no interest at all in the Internet and online services (the second divide). In this latter group, faced with the need to access the Web at some point in the future, it can be expected that barriers, such as availability of service, cost, literacy and capacity to use, etc., will become predominant factors of concern.1 The lack of connectivity and the lack of interest in use, or the perceived lack of need for access are factors that cut across all socio-economic groups. There are many possible explanations for these attitudes, including: a genuine disinterest; a lack of perceived value; inability to use technology; a lack of awareness of the range of services and information available and how these may meet individual needs, among others. At the same time, there are as many individuals who would like to be connected and use the Internet but cannot due to such barriers as cost and lack of skill, knowledge, etc.. Again evidence of these barriers can be found with many who are not connected across all the socio-economic groups. However, barriers to access are aggravated for those with less resources or skills available to overcome these. These barriers could be aggravated by factors such as employment status, education, regional disadvantages (e.g., economically deprivation), generational factors, among others. Overall however, these barriers tend to be most significant for those in the lower social class. Recent research by Statistics Canada also points to the development of a digital divide in Canada. The report, Canadians Connected, found that while Internet penetration rates have increased across income quartiles, education, age groups and geographic location, at the same time, the growth in the 'haves' has been coming from the higher as opposed to lower income groups. Through 1996 and 1997 this gap was widening. The report also noted that there was a trend of higher overall rates of growth in access in the less connected sub-groups; a pattern if it holds which could lessen the divide over time. 2 This differentiation moves beyond a simplistic notion of a divide between 'have' and 'have-nots', conveying a pejorative view of first and second class information society citizens. It may be true that those not connected are already, or will be, disadvantaged in some way. And this is arguably the case with many who could realize a benefit through access (e.g., skills development, employment opportunities, literacy, etc.) but are unable to because of a real barrier to access. However, it is also necessary to give the public its due, and recognize that one does not 'fail the test' by not being online! Of the approximately fifty percent of Canadians that are not connected, about half see no value, use or purpose in the Internet for them. Whether or not this group has more wisdom than the rest of us about access, they should not be construed or treated a priori as second class citizens, or simply categorized with the others not online as being socially or economically disadvantaged by not having access. Moreover, as a matter of policy, these individuals should not be disadvantaged with the best quality of information and services only being available on the Web, and second rate content and services provided by other, second rate, means. The other important consideration in the debate about access and the digital divide is the essential or optional nature of Internet service. The Internet, as yet, is still an optional service as opposed to an essential service, such as basic local telephone service, though for many, particularly the elites and knowledge workers in society, this is changing. At the same time, Canada is shifting to an economy and society where communications is playing an increasingly important role. While the degree to which we become an 'information society' may be debatable, in future many Canadians will need to have access to fully benefit both economically and socially. The degree to which not having all or a majority of Canadians connected is a problem from social and economic policy perspectives increases as non-users become truly disadvantaged in some way, or if this has some deleterious affect on the delivery of public services, or perhaps, the viability of some market services. For example, if some government information and services are only available on the Internet in future, thereby making some aspects of the Internet an essential service, then the digital divide takes on a new importance, both at the individual and social levels. This study offers a closer analysis of the different non-user segments, or those at the margins. This analysis provides a deeper understanding of the attitudes and practices of these different segments and considers such questions as: what are the main barriers to access? Are there appropriate policy responses to address these barriers? What may encourage some non-users to get online? What other options for service will necessary to facilitate those who are not likely to come online, even perhaps, in the long term? Methodology Research and analysis for this report was based on statistical analysis and literature from Industry Canada, Human Resources Development Canada, Statistics Canada, the NTIA in the U.S., as well as other books, articles, documents and reports relevant to the issue. Non-user typologies were created using multi-variate analysis on the 1997-98 Information Highway and Canadian Communications Household study. This multidimensional analysis was based on three steps. Basic 'factors' were distilled from the full set of attitudes tested in the surveys. The second step identified parameters for a logistic regression model that estimated the likelihood of being an Internet user or non-user. Regression modeling was supported by a demographic profiling of users and non-users. Finally, a typology of non-users was created and an examination was conducted of the differences in the underlying structures of attitudes supported by each segment of non-user. The first Information Highway and Canadian Communications Household study was conducted in three waves from September 1997 through June 1998. The information collected from over 600 variables in this study formed one of the most comprehensive baselines on the communication attitudes and practices of Canadian households, and in particular, in the context of the Information Highway. The first wave of the study involved a telephone survey with a national sample of 3,522 Canadians aged 18 and over. This was conducted in September 1997. The second wave involved a mail-back diary questionnaire. This questionnaire was completed by 1,271 respondents who had participated in the first wave. This research was conducted between February and April 1998. The final wave, completed in June 1998, consisted of a telephone poll. This involved 1,767 respondents from the first wave and another 434 respondents who participated in the study for the first time. Access and the Canadian Information Highway - Policy Context General Policy Framework Starting in the 1980's, through policy and regulatory changes, the federal government has been restructuring Canada's communications industry to facilitate: economic development; competition; and a shift to the electronic production and delivery of economic, social and cultural services. Through the 1990's, the Government of Canada extended these initiatives to lay the groundwork for an Information Highway to facilitate the development of an information society. The general themes guiding this policy framework were to facilitate Canada's transition to a knowledge society; to be the most connected nation in the world; and to realize economic growth and competitiveness domestically and internationally. 3 Building on the 1994 Speech from the Throne that introduced the Information Highway strategy, the government, through Industry Canada, formed the Information Highway Advisory Council (IHAC). The strategic framework guiding the work of IHAC and, complementary programs and initiatives of other federal departments, involved a number of core themes and objectives relating to: universal and affordable access; lifelong learning and skills; job creation; increased competitiveness and economic growth; the development of Canadian cultural content; benefits from electronic commerce; access to government and other public services and information online, among others. The final IHAC report made a number of recommendations in the area of access to address roles for the government. These recommendations included facilitating rural access; government funding for community access initiatives, including libraries and not-for-profit community networks and sites in both rural and urban areas; government funding to facilitate the development of public and other forms of non-commercial content; and that government information and services be made available in a range of different formats to facilitate access. 4 While in varying degrees, all federal government departments are involved in some way in the achievement of these objectives, the two departments that have been most active in addressing the concerns of access, skills and individual and community use of the Information Highway are Industry Canada (IC) and Human Resources Development Canada (HRDC). HRDC operates a wide range of programs concerned with youth, literacy, skills, jobs, community and individual development. The initiative that combines many of these interests within the context of access is the Community Learning Network program. Industry Canada similarly undertakes many initiatives to address different dimensions of the Information Highway agenda, including: the Community Access Program (CAP); SchoolNet; VolNet; and Computers in the Schools, among others. However, the CAP program is most centrally concerned with public access. The CAP program is a technology-based initiative, framed with economic and industrial policy objectives, but also facilitating social policy objectives. Economic and industrial policy objectives include stimulating the use and demand for Information Highway equipment, content and services. Individual, community and national development objectives underlie this aspect of the program. Social policy objectives, while not a core mandate of Industry Canada, are facilitated and encouraged by providing facilities or technological platforms and activities to encourage literacy and learning, skills development, employment opportunities and content development. Building on the existing technical access platforms (sites or network groups) in communities, HRDC's CLN program is concerned with social policy objectives central to the department's mandate. These include: the use of technologies as tools to support and enable learning; skills development; employment opportunities; and, individual and community development. Much of the CLN program is concerned with content and services. Community Access Program Stemming from the objective of Canada to become the most connected nation in the world by the year 2000, the government introduced its' Connecting Canadians agenda in 1998. Featuring six action areas, the Canada On-Line component involves access by the public to the Internet. 5 CAP is an initiative developed to help provide Canadians with affordable access to the Internet and the skills to use it effectively. Public locations, such as schools, libraries and community centers provide local sites for individual access. Initially, CAP was planned to connect 1,500 rural and remote communities. Following under the Connectedness Agenda, CAP has been designed to establish access sites in 5,000 rural communities and up to 5,000 sites in urban communities by March 31, 2001. CAP is made available on a partnership basis. Partners can include: provincial and territorial governments; community groups; social agencies; libraries; schools; volunteer groups and the business community. The objectives of the CAP program are to: 1. Provide Canada's rural and urban communities with better and more affordable access to the Information Highway and to raise awareness about its potential for creating jobs and growth; 2. Stimulate the development of new electronic learning tools and services by and for communities; 3. Provide Internet training facilities for local entrepreneurs, employees, educators, students and others interested in improving their information and networking skills; 4. Stimulate the electronic delivery of government and other services and obtain feedback from citizens about how they would like these presented.6 Community Learning Networks The CLN program involves time-limited pilot projects involving community partners that offer multiple locations in a community that provide a wide range of learning resources. At a general level, the program is intended to raise awareness about, and participation in, community-based programs and initiatives involving opportunities in learning and skills development. Major goals include enhancing the social and economic development of communities and the lives of individuals in those communities through lifelong learning. A key goal of the program is to make learning opportunities more accessible to Canadians. This social policy-based program is geared to address the needs of those Canadians at a particular risk of being marginalized or left behind in the new information based society and economy. This includes those who have specific learning needs to enhance their employability; those who require alternative forms of learning to meet their needs and who may be marginalized or become part of the information 'have-nots". The initiative also creates opportunities for those with special needs, such as seniors and persons with disabilities. The program will spend $29 million from 1999 through 2002, with up to $100,000 per year for each project.7 The CLN project is also intended to create models for other communities and community-based organizations in the areas of learning, ongoing organizational support and maintenance, and different approaches to learning opportunities. In a broader context, the CLN program is part of HRDC's mission that includes enabling Canadians to manage transitions in their lives. An underlying goal is to reduce inequality and to promote growth through the support of both social and economic development in the workplace and the community. This includes concerns with: lifelong learning; employment access and adjustment; safe, fair and productive workplaces; equitable participation; and reasonable income security. Recognizing the diversity of circumstances and needs of Canadians, HRDC's strategy is to provide information and services using the full range of delivery/access modes including: print; telephony; CD-ROMs, television, in-person assisted services and the Internet. 8 Overview of Access and the Canadian Information Highway Canadian Communications Household While there is much interest in how Canadians will get access, that is choice of infrastructure services, there is much less about what they are getting access to and why. The majority of communication companies are not taking big risks in new services, with the exception of the Internet. Much of the new content offerings tend to be existing products or services that have already proven themselves using other technologies (e.g., business telephone services, broadcasting) or in other consumer formats, such as newspapers, reference materials, educational content, and home shopping. The products in other formats are repackaged in electronic format with the recreation of these markets online. As is the norm with new products or services, such as the Internet, communication companies have initially focused their efforts on the best market segments from which to recoup the high costs of development. These segments are business, professional and upper income consumers. The high-end consumer segment, comprising the more advanced communication households, tends to be an early adopter of a wide array of new technologies. This segment also has the greatest disposable income for content services. This also means that, at least during the initial development phases, the content and services made available are likely to reflect the demands and needs of this segment, or specialized sub-groups, such as education, as opposed to the general consumer market. Subsequently, in addition to the novelty factor, the low awareness about new services such as the Internet two to three years ago, and the low value attached to such services by the general public, is explained to some extent through this development strategy. 9 The public's awareness of the Internet has increased over this period, as has connectedness. However, use value still tends to be interest-specific as opposed to general and populous, though this too has been changing more recently with the proliferation of more general content, including electronic commerce services. There is a considerable variation in the degree to which Canadian households have different communication products and services. Most households have standard services such as telephone and television. However, clear patterns of differential levels of access exist in households with newer services, such as cellular or PCS phones, Internet, satellite TV, home computers, and even to some extent, with what many consider to be a standard service, cable television. In general, the likelihood for greater levels of connectivity are linked to such variables as income, education, gender, location and technological literacy. Other important variables include need, affordability, and perceived value. Upper social classes, those with better education and employment, are more likely to have a range of communication technologies and services. Where there tends to be more proliferation of a particular technology or service across social classes, the incidence of use or ownership also tends to be higher with those with higher income, higher education, and living in an urban area. Coupled with this are generational differences. For example, youth are high users of some technologies, e.g., the Internet, and seniors tend to lag considerably behind youth and other segments. 10 Figures Two through Five demonstrate some of the differentials in access and use of a number of technologies by social class, using income as a key indicator. Figure 2 Source: Ekos Research Associates Inc., 1998. n=3522. Figure 3 Source: Ekos Research Associates Inc., 1998. n=3522. Figure 4 Source: Ekos Research Associates Inc., 1998. n=3522. In Canada, as other western developed countries, there has been much enthusiasm about how quickly different social segments have subscribed to the Internet as compared to other modern communications technologies, such as cable television and telephone. The enthusiasm about the Internet, divorced from a serious consideration of how it is situated in our broader social and economic relations, risks the making of unfounded assumptions and expectations about the likelihood of a large majority of Canadians, regardless of social class, being ubiquitously connected from home, and when this may occur. Cable television, available since the late 1950's and early 1960's, provides a useful illustration of this point. Cable TV is generally valued as an information/entertainment communication service by a large number of Canadians. It is also a service that can be described as fairly mature in its development as compared to the Internet. Figure 5 Source: Statistics Canada, 1997. However, as demonstrated in Figure Five, after more than thirty years availability, there still exists substantial differences in cable subscriber ship based on social class and geographical location; patterns which are not unlike those of the emerging bifurcation with Internet service. While national cable service penetration averages about three quarters of households (74%), there are substantive differences between upper (83.1%) and lower income (64.7%) households; a pattern that has not changed over the last several years. The differentials in rural location are largely technologically based arising from the unavailability of service in many areas. However, regional affordability and economic disadvantage issues also account for some of these differences. In the context of other communication services, such as cable television, if Internet pricing remains stable and some value is perceived it can be expected that there will be continued growth of subscriber ship from home, particularly in the upper and middle classes. However, at the same time it can be expected that there may be a lower overall adoption of the Internet in lower income households, and with some social sub-groups, such as seniors. As well, in spite of Net hype and enthusiasm, it would be anomalous if a plateau effect in subscriber ship didn't start to occur with the Internet over the next few years, as has been the case with other modern communication technologies. Early results from the 1999 Information Highway study (Ekos) suggests that a plateau in Internet use is already occurring with the upper income ($60k+) and upper middle income ($40-60k) segments. Evidence of differential levels of access as a social norm based on class and illustrated by such variables as cost and disposable income is extensive in Canadian society. For example, household expenditures on such items as communications, education and reading materials exemplify this pattern. Table One below demonstrates these levels of access/use. Upper income households spend almost three times as much as low income households on communication services, and four times as much on reading materials and education.11 Table One: Average Household Expenditures by Income Quintile, 1997. Average Lowest Q 2nd Q 3rd Q 4th Q 5th Q Communications $924 563 753 934 1041 1331 Education $659 311 258 531 802 1395 Reading Materials $275 120 199 263 306 486 Source: Statistics Canada, Spending Patterns in Canada, 1997. Access and the Internet During the study period, a majority of lower income households reported that they did not have a home computer (66%) or Internet service from home (83%). Almost half (48%) of lower middle income households were likely to have a home computer, but only one in five (20%) were likely to have a home Internet account. Upper middle income households were more likely to have a home computer (60%) but less than one in three were likely to have home Internet (27%). The majority (79%) of upper income households were likely to have a home computer and almost half (46%) home Internet. 12 Figure 6 Source: Ekos Research Associates Inc., 1998. Accounting for the differential levels in access is complex and involves a number of variables. As noted above, the matrix of variables which can bear on whether individuals access the Internet include income, education, gender, location and technological literacy. Research has also shown that as important, if not more so for many, are other variables of need, affordability, and perceived value. The Internet will not likely be used uniformly among those who are not currently online. Even those online exhibit differential use patterns. For example, while females make up over half of non-users, those online tend to largely be light users. Males, on the other hand, tend to make up the majority of moderate and heavy users. In generational terms, Canadians over the age of 55 account for about one in three non-users (37%). Most of the heavy users of the Internet are represented by individuals under the age of 35 (66%). In comparison, about one third (30%) of the age demographic are non-users. 13 There is also a differential in access between urban and rural users. The main reason underlying this difference has less to do with the availability of service as issues relating to perceived value of the service and cost. 14 In general terms, there are some differences underlying the initial purposes of why people go online. For example, lower income users are more likely to use the Internet for leisure purposes, whereas higher income users are more likely to state education as a main reason for home Internet service. At the same time, a significant number of users with lower levels of education see a main use of the Internet as a means to improve their education and skill levels. Beyond these initial justifications, the use of different content on the Internet tends to vary more as users gain greater online experience. Figure 7 Source: Ekos Research Associates Inc., 1998. In addition to these types of demographics, which provide some useful generalizations about users and non-users and household communication patterns, more interesting questions tend to involve understanding the attitudes and perceptions by non-users about their reluctance or disinterest in being online, and the barriers faced by those non-users who wish to be online but are unable to. At the same time, it is important to note that non-users do not form into simple homogenous groups of those who are or are not interested in being online. General Non-User Attitudes and Values When queried about home Internet access, the cleavage between the two main non-user groups clearly forms around a perceived lack of need or interest on the one hand, and cost or affordability on the other. Interestingly, at this relatively early stage of public awareness about the Internet, close to one in five of all income groups may have had some interest in connectivity, but considered cost to be a barrier. At the same time, roughly the same number had little interest in this service, and about a third did not see how the Internet had any relevance to their daily information or communication needs (Figure 8). Figure 8 Source: Ekos Research Associates Inc., 1998. Preliminary data from the 1999 version of this research study, suggests that the Internet is increasingly being perceived as an important or necessary service by more Canadians. In this changing context of the perceived value of the Web, cost has now become a more important barrier for low income families and some in the middle income groups. Lack of interest was about the same. However, lack of need as a factor has dropped across segments, but respondents were more likely to identify issues such as no computer, an old computer, or access from another location as explanations for not having service from home. 15 The potential benefits and range of possible uses of the Internet are extensively discussed and described by promoters and users among the information elite in government and industry. The general public, on the other hand, in the context of their every lives and activities, did not find the lack of Internet access from home to be problematic (Figure 9). Figure 9 Source: Ekos Research Associates Inc., 1998. At the same time, during the research, deeper probing was conducted with respondents to ascertain the level of interest that existed in the Internet. This research provides a general indication that non-users again tend to divide into a group which are potential or near users, depending on a range of variables (e.g., value of content, cost, etc.), and a second group which are quite disinterested. As shown in Figure Ten, a slight majority of non-users from all segments with the exception of upper income households would not have the Internet at home if they had a choice. Figure 10 Source: Ekos Research Associates Inc., 1998. However, an expectation about the potential use and value of Internet service by many respondents was revealed when attitudes involving importance, ease of use and cost of Internet service were explored. A majority of respondents from all social classes felt that Internet access from home would be at least somewhat or very important at some point (Figure 11). Figure 11 Source: Ekos Research Associates Inc., 1998. Technological literacy and cost were found to be important factors in a consideration of the likelihood of use in the future. For about half of all sub-groups (Figure 12) the ability to use the Internet (technical proficiency and understanding) was important. Figure 12 Source: Ekos Research Associates Inc., 1998. Similarly, affordability or the cost of Internet service was an important criterion for at least two thirds of those in the lower and upper middle, and upper classes, and just over half of those in the lower income demographic (Figure 13). Figure 13 Source: Ekos Research Associates Inc., 1998. In a review of incidence of use in a previous three month period, while home access is the predominant means, those in the lower income groups are more likely than higher income households to have recently accessed the Internet from a public location, friends house, Internet Café or some other community location. At the same time, some respondents from all social classes have made recent use of public access locations, whether to learn about the Internet, how to use it, or because home access was not possible (Figure 14). Figure 14 Source: Ekos Research Associates Inc., 1998. In a comparison of the perceived relative importance of the Internet to other standard communication technologies or services, the Internet tended to be rated fairly highly (39%). The majority of those that considered the Internet to be an essential service largely represented the upper income segment of the population, where just over half (53%) of this group considering the Internet essential as compared to less than one third (28%) in the low income category. Figure 15 Ekos Research Associates, 1998. Interestingly, many respondents showed moderate support for basic Canadian and U.S. cable as essential services, with about one in five identifying these as optional services. But, in practice, most cable subscribers receive both cable services, suggesting that perceived importance and value may not always be directly linked to actual use levels. Diversity of Choice and Diversity of Life A clear theme that has emerged from the research is a pragmatism on the part of the public in how they use different technologies to meet different needs. There has been much speculation about the potential for the Internet to displace traditional means of accessing products and services. It may very well be the case that the public will increasingly face the need to use new communications technologies, such as the Internet, to obtain some products and services, for example new media. In other instances, people may have no choice in how they get access to a product or service if a marketing strategy of 'product substitution' is employed, for example the replacement of vinyl records with cd's. 16 In an analysis of preference by the public in the use of different technologies to gain access to government for information, the public clearly indicated that a diversity of means of access is required. Figure 16 Source: Ekos Research Associates, Inc., 1998. As demonstrated in Figure Sixteen, the full range of traditional access technologies, in addition to the Internet, were seen as important means for contacting the government. While there is some difference between the income groups, further research indicated that the particular technology employed was closely related to several core factors. These factors are: the needs of the person; the nature of the service being sought (e.g., general information, applying for something, providing or accessing personal or sensitive documentation, etc.); and access to technology and technological literacy. While Figure Sixteen involves accessing government information, technological preference will vary considerably for all segments for other types of information or services. Internet (email) has become an important access technology in this mix of diversity of technological means of access, particularly for those in the higher income groups. However, it is also used to a minor extent by those who do not have home Internet access. As a general trend, the Internet has become an additional method for transactions, as opposed to a replacement. Beyond government services, this pattern or strategy is also being adopted by the private sector, where Internet browsing and purchasing is an added on or complementary means of access, in addition to other traditional means, such as in-store, phone, catalogue, etc.. In addition to how the Internet is used from home as part of a basket of consumer information and communication technologies, it is also useful to compare its use in relation to other social and cultural activities. Figure 17 Source: Ekos Research Associates Inc., 1998. The Internet has had a definite impact on how individuals allot their time to different activities. About one in four Internet users were watching less television (28%) or relying less on newspapers for their news (24%) due to the Internet. Another twenty four per cent were buying less books and magazines because some of this material is available on the Net. At the same time, it should not be assumed that the Internet will simply displace how people access different media products or other social and cultural activities. As indicated in Figure 17, there is a healthy competition for people's time arising from a wide range of daily obligations, responsibilities and leisure activities. 17 The value ascribed by people to these activities extends beyond whether it can be done online or not. Convergence may integrate some of these technologies, such as Internet, television, video and music, or content, such as books, newspapers and magazines. But the social and familial contexts of some of these activities are as important, if not more so, than whether or not delivery through the Internet happens to be technically efficient. The Internet may displace the extent to which individuals devote time to other activities, but in reality it just adds another type of activity to the overall mix. The diversity of the range of activities of which people are involved in their lives also informs an understanding of why many Canadians do not perceive a sufficient value in the Internet to justify being connected from home. In this sense, people should not be faulted or 'fail the test' for not being online. This discrimination in tastes and values speaks much to the importance of the need for the development of relevant, easy to access and use content on the Internet as it develops over the next several years to truly offer a complement, as well as an alternative, to traditional social, cultural and leisure activities for many connected and still unconnected Canadians. Typology of the "Non-Users" in the Digital Divide The search for a more complete understanding of Canadians' attitudes to the digital divide is elusive. The complex and often contradictory nature of beliefs and perceptions makes it difficult to discern the basic patterns. Moreover, simple relationships between attitudes and background characteristics do not easily fit together to produce a coherent "big picture". To address these limitations we have produced a multidimensional analysis of the Canadian public. This analysis provides a more realistic and direct tool for arraying different constellations of attitudes, beliefs, behaviours and preferences about the information highway. This multidimensional analysis provides two major advantages. First, it "solves" some of the apparent contradictions in the data. For example, it will show how different demographic characteristics and experiences can come together to yield the highest levels of concern. The overall patterns help explain why certain types of Canadians think and feel the way they do about the information highway. Secondly, the multidimensional analysis provides a much more meaningful and practical tool for developing policy responses, which will suit the unique character of different types of Canadians. The Canadian public should not be viewed as a monolith. Any responses, which assume this overall similarity, are destined to miss the mark. This multidimensional analysis presented in this section is based on three steps. In the first step, the basic "factors" were distilled from the variety found in the full set of attitudes tested in the survey. The second step involved identifying parameters for a logistic regression model that estimated the likelihood of being an Internet user or non-user. Regression modeling (based on attitudinal factors) is supported by a demographic profiling of Users and Non-users. The third step is to create a typology of non-users and examine the differences (if any) in the underlying structure of attitudes supported by each segment of Internet Non-user. Summary Dimensions of the Digital Divide The survey questionnaires identify a number of broad attitudes and behaviours on issues surrounding the Information Highway. The initial transformation of the data consisted in creating sixteen scaled variables based on the original 83 variables. These scaled variables were introduced in an effort to reduce the sheer volume of information and group similar variables under one heading. The scales are a composite of the results from the relevant variables, with "positive" and "negative" responses gaining a higher and lower value on the scale respectively. A brief synopsis of the factors is presented in Exhibit 5.1.1. A description of their meaning follows, along the highlights of some of the key correlations between the dimensions and other social variables. Exhibit 5.1.1 Tele-Work LPACT 'Impact of working at home on family life' HPAST 'Impact of working at home on working hours' EDEST 'I am in control of my own economic destiny working at home' APPEA 'How appealing do you find the idea of working at home' FHOME 'How often do you think you will conduct work from home in the future' TPACT 'Impact of working at home on time pressures' FPACT 'Impact of working at home on finances' NPACT 'Impact of working at home on networks / work contacts' HWORK 'To what extent do you agree that new technology will make working at home more possible' WHOME 'How often do you conduct work from your home' Positive Imagery ATT4 'IH Image Positive-Negative' ATT3 'IH Image Helpful-Harmful' ATT7 'IH Image Liberating-Dehumanizing' ATT14 'IH Image Educational-Mindless' ATT2 'IH Image Secure-Insecure' ATT12 'IH Image Tool-Toy' Concern with Cultural Impact Q51C 'There should be mechanisms in place to ensure there is enough Canadian content on the IH' Q51D 'Government should provide additional funding through agencies to ensure strong Canadian content on the IH' COMPA 'How important is it to you to ensure the development and availability of Canadian content on the IH' Q53AM 'A good use for the IH would be to provide Canadians with information about what it means to be Canadian ' Q51A 'I am really worried that Canadian identity will suffer as we move towards IH' Technophobia FAST 'The government is moving too fast in using new technology to deliver information and services' SERV 'When dealing with the government, I would rather deal with a person even if it means slower services and greater costs' Q53H 'I personally know some people who spend so much time at home using the Internet that it has a negative impact on the quality of their family life' EDGE 'Government is on the leading edge in implementing new technologies' Q53R 'The IH is reducing the level of privacy in Canada' Competency/ Technophilia Q33D 'Rate your ability - working with computers' FAM 'How familiar are you with the meaning of the term Information Highway' Q53B "My knowledge of new technologies gives me the skills to move more easily in today's job market' ACCE2 'How important do you think it is to access newer server' PERSO 'How important do you think the Information Highway is to you' Barriers PBARR 'How much of a barrier is it (access to new equipment) for you personally' LBARR 'How much of a barrier is it (lack of knowledge how to use technologies) for you personally' Indirect Societal Applications Q20H 'How useful is IH for the following application - Providing information /services to help the growth of small business' Q20G 'How useful is IH for the following application - Improving access to education and training for Canadians' Q20C 'How useful is IH for the following application - Linking similar community based and cultural groups across Canada' Q20D 'How useful is IH for the following application - For individuals to have access to government information' Q20F 'How useful is IH for the following application - Delivery of medical services to rural and remote areas' Q20A 'How useful is IH for the following application - For individuals to be consulted on local and community affairs' Q21B 'Community networks should be used to help low-income Canadians' Q20E 'How useful is IH for the following application - For individuals to communicate with one another' Q20B 'How useful is IH for the following application - For individuals to vote electronically in elections or referendums' Television Q2E 'How important to have access to basic cable service - Canadian channels' Q2F 'How important to have access to basic cable service - U.S. / foreign channels' Q2D 'How important to have access to television' Cultural Activities ZQ25D2 'Over the past 23 months, how many times have you visited - a bookstore' ZQ25C2 'Over the past 23 months, how many times have you visited - a public library' ZQ25B2 'Over the past 23 months, how many times have you visited - an art gallery or art museum' ZQ25A2 'Over the past 23 months, how many times have you visited - a museum or archive' ZQ26I 'How many hours did you spend last week reading books' ZQ25E2 'Over the past 23 months, how many times have you visited - a movie shown by a commercial theatre, club' ZQ25F2 'Over the past 23 months, how many times have you visited - a folk, rock, jazz, or country music performance' ZQ25I2 'Over the past 23 months, how many times have you visited - a national park / historic site' Virtual Banking Q1E 'I prefer using banking machines over going into a bank and dealing with a teller' Q1F 'I prefer using debit cards over using cash, cheques or credit cards Q2G 'How important to have access to Internet' Telephone Use Q27C 'In the past 3 months how often have you used your phone for contacting governments' Q27B 'In the past 3 months how often have you used your phone for work / business related' Q27C 'In the past 3 months how often have you used your phone for contacting your doctor or other health-related services' Pop-Culture Activity & Homemaking Q26E 'How many hours have you spent last week watching news or education programs on TV' Q26F 'How many hours have you spent last week watching all other types of programming on TV' Q26D 'How many hours have you spent last week watching videos' Q26G 'How many hours have you spent last week reading newspapers (excluding on-line electronic newspapers)' Q26H 'How many hours have you spent last week reading magazines (excluding on-line electronic magazines)' Q26B 'How many hours have you spent last week on housework / caring for children' Concern with Internet Cost & Access REL 'Would you be more or less likely to use the Internet if Internet services were more reliable (for example, were always available) COST 'Would you be more or less likely to use the Internet if the cost of an Internet account was cheaper' RAP 'Would you be more or less likely to use the Internet if it took less time to find, download information on the Internet' EASE 'Would you be more or less likely to use the Internet if the Internet was easier to use' Informed Consent BNKGT 'How likely would you be to do transactions electronically if a bank guaranteed the transaction' SEC 'How likely would you be to do transactions electronically if there were adequate security measures to protect personal information such as credit card numbers' GOVT 'How likely would you be to do transactions electronically if there was a government framework in place that establishes laws about electronic commerce' PINF 'How likely would you be to do transactions electronically if the business clearly indicated how it will use any personal information collected' PRIV 'How likely would you be to use the Internet if you knew how your personal information would be collected and used' SECUR 'Would you be more or less likely to use the Internet if the Internet was as safe and secure as a bank machine' Safe e-commerce ICRD 'I would be willing to give my credit card number over the Internet, in order to purchase a product or service' REGI 'I do and would not mind registering personal information on Internet sites I visit' EXP 'I expect that I will buy some products or services over the Internet within the next 2 years' TCRD 'I would be willing to give my credit card number over the phone in order to purchase a product or service' Economic Security FUTUR 'Thinking about your personal economic situation in the next years, do you feel pessimistic or optimistic' RLTO10 'I think there is little chance I could lose my job in the future' SWC 'In general, I am very satisfied with my working conditions' RLOS10 "I feel I have lost control over my economic future' 1. Tele-work encompasses attitudes and actions relating to issues that range from the impact of working at home on family life, working hours and, finances, control over economic destiny, time pressures, and networks/work contacts. Tele-work also measures the appeal of working from home, the frequency with which respondents work from home, as well as the likelihood that a respondent will conduct work from home in the future. In a related measure, we also include the respondents' attitudes towards new technology and the possibility of working from home. 2. The second scaled variable, Positive Imagery, measures the extent to which the Information Highway summons constructive descriptions among respondents. This scale includes overall positive imagery, as well as feelings relating to its usefulness, helpfulness, and security. Positive imagery is also revealed by the liberating and educational potential of the Information Highway. 3. The concern with the cultural impact of the Information Highway is another theme that is broached in 5 survey questions. This third scaled variable examines the impact of the IH on Canadian identity and the mechanisms (government or otherwise) that should be introduced to ensure Canadian content on the IH. We also identify the potential positive impacts of the IH on Canadian culture. 4. The fourth item outlines the variables that are included in the technophobia (fear of technology) dimension. Complementary to the positive imagery variable, this variable measures the attitudes that are linked to the rejection of the use of new technologies, notably in the delivery of government services. The fear of new technology is also linked to the sense of loss of privacy and negative impact the Internet can have on family life. 5. The fifth grouping of variables relates to the ease and comfort respondents demonstrate with regard to the Information Highway. The competency/technophilia (at ease with technology) variable takes into account individuals' self-reported ability to work with computers as well as the value they place on the Information Highway in their lives. The demonstrated ease with computers, new technology and the Information Highway also has an effect on the skills they have to adapt in today's job market. 6. The succinctly named barriers, our sixth dimension, is a composite of two variables that measures the perceived barriers, including access and lack of knowledge, in relation to the Information Highway. 7. The next scaled variable, indirect societal applications, is a complex measure that encompasses nine distinct variables. The range of issues is broad, addressing the usefulness of the Information Highway in a variety of social sectors. These include small business, access to education and training, and the delivery of medical services to rural and remote areas. It also looks at the Information Highway in linking community based and cultural groups, enabling Canadians to communicate with one another, helping low-income Canadians, and providing government information to Canadians. We also examine the applications of the Information Highway in our political system, notably in electronic voting and for consultation between government and its citizenry on local and community affairs. 8. Turning to respondents' attitudes towards television, we reduced three variables related to television and cable services. These address the importance of access to television as well as basic cable service with Canadian and U.S. / foreign programming. 9. The Cultural activities dimension gives us a sense of how frequently these types of activities figure in the lives of Canadians. The survey asks respondents to enumerate the number of visits in the past 23 months to a bookstore, public library, art gallery, museum, archive, and national park/historic site. Canadians are also asked to recall the number of times they attended a commercial movie theatre or musical performance. In this measure of cultural activities, we also include the number of hours spent reading books over the course of a week. 10. The tenth dimension, virtual banking, combines the preference for using new technologies for simple banking and financial transactions with the importance of access to the Internet. This measure is linked to the latent (unrevealed) likelihood of adopting Internet banking as the preferred banking method. 11. Although the use of the telephone is widespread and access is hardly an issue for most Canadians, we include the following dimension to get a sense of it whether the telephone is the preferred method of communication. This scaled variable measures the frequency of telephone use for contacting government, business use and accessing medical services. 12. The following dimension, pop-culture activity and homemaking, is an effort to measure time spent on activities inside the home that fall outside of the realm of work, community or volunteer activities. This measure of passive consumerism and time management includes the number of hours spent watching television (including videos), reading newspapers and magazines, as well as the amount of time spent caring for children or doing housework. This dimension conspicuously avoids any mention of time spent on the Internet while at home. 13. We also outline the variables that are used to form an overall scale of concern with Internet cost and access. This dimension gauges whether respondents would be more or less likely to use the Internet if barriers to use were softened or conditions improved. These barriers and conditions include, cost, ease of use, reliability and time. 14. The next dimension deals with issues surrounding privacy and security of electronic transactions, including electronic banking and shopping. Informed consent also takes into account respondents' attitudes towards the security and the use of personal information collected on line by businesses. Associated with these issues of security and privacy of personal information sent or collected online is the role of government in establishing laws about electronic commerce. 15. Along the same vein as informed consent, we have grouped four variables that touch upon the perception of safe e-commerce. This measures the likelihood of giving personal information or making purchases on line or over the telephone, independent of any guarantees of safety or security. 16. And finally, there are four variables that help give a sense of respondents' feeling of economic security. We asked Canadians to think about their personal economic situation in the next few years and whether or not they could lose their job in the future. The measure of relative optimism (or pessimism) is enhanced by the sense of control they have over their economic future and their overall satisfaction with their present working conditions. These 16 dimensions help define the broad outline of attitudes and values among respondents to the Information Highway survey. They are very helpful in reducing the number of variables that need to be manipulated and making the data set more accessible. Nonetheless, these dimensions are highly correlated with one another. We cannot use these collapsed dimensions to effectively define the underlying structure of users and nonusers of the Information Highway18. The next step is therefore to re-interpret these 16 dimensions into a set of independent factors. We apply a data reduction technique called factor analysis to regroup the dimensions and extract 6 principal components19. The summary results are presented in the following table (Exhibit 5.1.2). Exhibit 5.1.2 Principal Component Analysis Principal Component Dimension 1 2 3 4 5 6 Tele-work + Positive Impact on Jobs + Positive Imagery - Concern with Cultural Impact + Technophobia + Competency - + + Barriers + Indirect Societal Applications + + + Television + Cultural Activities - + + Virtual Banking - + + Telephone Use + Pop-culture Activity and Homemaking + Concern with Internet Cost & Access + Informed Consent + Safe e-commerce - + Economic Security - In an effort to clarify the results, Exhibit 5.1.2 presents the most important factor loadings (both positive and negative). A positive loading indicates that the strength of the component varies in the same direction as the dimension. A negative loading (minus sign) indicates that a strong positive result in the dimension has an opposite effect on the factor. We see that dimensions can be in more than one component and likewise, the principal components are comprised of a combination of different dimensions. The next step is to label each of the factors so we have a basic lexicon that we can use to identify the different types of users and non-users of the information highway. FACTOR ONE Lack of confidence This factor is described by the lack of confidence in using the Internet and accessing the information highway. It is a combination of technophobia and challenges to the use of the Internet. There is a high level of concern with the cultural impact of the Information highway and is associated with lower levels of economic security. FACTOR TWO Experience and interest Factor two relies heavily on issues of informed consent, e-commerce, as well as cost and access to the Internet. Competency loads positively, as do virtual banking and societal applications of the Information Highway. These are all issues that relate to experience and interest in using the Internet. FACTOR THREE Practical tool Factor 3 isolates the effect of the tele-work dimension. We included in the factor analysis a complementary dimension to tele-work, measuring the perceived impact of the information highway on jobs. Factor 3 reveals that these two variables vary in the same direction. The overall imagery of the Information Highway is likely to be negative, although competency and societal applications rank high. This is the Internet as a practical tool. FACTOR FOUR Passive convenience Factor 4 compares the importance of television to other cultural activities. Watching television is likely to have a negative impact on the number of other cultural activities enjoyed by respondents, but increase the enthusiasm for virtual banking and societal applications of the Information Highway. This factor describes the passive convenience of the Internet. FACTOR FIVE Leisure, culture and entertainment time Factor 5 takes into account the number of hours spent consuming pop-culture at home and cultural activities outside of the home. Leisure, culture and entertainment time encompasses the effect of these variables. FACTOR SIX Telephone use Factor 6 picks up the partial effect cultural activities and combines it with the number of hours spent on the telephone for work or to contact government /medical services. Because the telephone dimension loads so heavily in this factor (and for lack of a better term) we will refer to factor 6 as telephone use. These six factors (lack of confidence, experience and interest, practical tool, passive convenience, leisure, culture and entertainment time, and telephone use) explain at least 50% of the total variance. They are by definition and construction independent from one another and can therefore be readily applied in a multivariate analysis. The next phase of the analysis is to use these factors in a predictive or explicative model of Internet users and non-users. Who is lagging behind in the Digital Divide and Why? Exhibit 5.2.1 profiles Internet users based on a number of demographic characteristics. We see that a greater proportion of men than women have used the Internet in the past three months. From June 1998 to July 1999, the rate of growth of Internet use among women is almost twice as large as among men. Younger age cohorts, increased income and education levels also indicate higher rates of Internet use. Canadians over the age of 65 have the greatest rise in Internet use from 1998 to 1999, with a staggering 143% increase in one year. Despite this impressive increase, their numbers remain far below the average Internet use among younger Canadians. Rural Canadians also increased the proportion of Internet use from 1998 to 1999. Compared to Canadians residing in urban locations, however, fewer rural Canadians have used the Internet in the past three months. We applied the demographic variables in a multivariate analysis of Internet use and non-use. A linear model would be inappropriate in this case because we are limited to values no smaller than 0 (non user) and no greater than 1 (user). On a dichotomous dependent variable, we use a logistic regression on the independent (explicative) factors and variables to predict the likelihood of being an Internet user (1) or non-user (0)20. Using age, income, education and gender in the logistic regression, we find that they are all significant parameters in determining the likelihood of Internet use. Exhibit 5.2.2 displays the coefficients of the variables in the model. The model itself still has quite a bit of unexplained variance (pseudo R-squared of 0.21). Exhibit 5.2.2: coefficients of logistic regression model Dependent variable: Internet use in the past three months Demographic Variables B s.e. Sig. (Constant) -0.76 0.14 0.00 *** Age -0.41 0.03 0.00 *** Gender (men) 0.51 0.08 0.00 *** High Income 0.65 0.08 0.00 *** Education 0.65 0.05 0.00 *** *** significant < 0.01; ** significant < 0.05; * significant < 0.10 There are some definite cleavages in the demographic breakdown of Internet users and non-users. But beyond the demographic profile of who is on-line and who is still lagging behind, there may be some useful information that can be applied from the factor analysis in the previous section. The broad attitudes, defined and extracted in the factor analysis, may give us a better sense of the characteristics of Internet users and non-users. The results from the logistic regression are found in Exhibit 5.2.3. As we see in Exhibit 5.2.3, factors 1, 2, 3, and 5 are significant parameters in determining the likelihood of Internet use in a household. A higher degree of confidence in Internet abilities (negative coefficient on lack of confidence), experience and interest, seeing the Internet as a practical tool, and increased time for leisure, culture and entertainment are likely to increase the propensity towards Internet use. Compared to respondents who have not used the Internet in the past three months, Internet users have a unique attitude and value set towards the Information Highway. Exhibit 5.2.3: coefficients of logistic regression model Dependent variable: Internet use in the past three months Variable B s.e. Sig. Constant 0.13 0.44 0.78 1. Lack of confidence -1.23 0.15 0.00 *** 2. Experience and interest 0.33 0.13 0.01 ** 3. Practical tool 0.51 0.12 0.00 *** 4. Passive convenience 0.14 0.11 0.21 5. Leisure, culture, and entertainment time 0.36 0.12 0.00 *** 6. Telephone use 0.20 0.14 0.16 Age -0.18 0.10 0.07 * Gender (men) -0.15 0.23 0.50 High Income 0.58 0.25 0.02 ** Education 0.10 0.15 0.52 *** significant < 0.01; ** significant < 0.05; * significant < 0.10 Combining attitudinal factors and demographic characteristics, we can provide a more complete picture into the differences between Internet Users and Non-users. By including age, gender, income and educational level in the regression model, we find that higher incomes have a positive effect on the likelihood of being an Internet user. To a lesser (but still significant) extent, respondents in older age cohorts are less likely to be Internet users than their younger counterparts. The effect of the age variable is likely influenced by the sharp decrease in Internet use among respondents over the age of 55. When all other factors are taken into account, gender and education have little or no impact on the propensity towards Internet use. Not only do income levels affect the likelihood of Internet use, further analysis shows that socio-economic status may also play in important part in determining some of the attitudes expressed towards the information highway. Exhibit 5.2.4 summarizes our findings of logistic regression on Internet use (dimensions and demographics) based on income. This is evidence to support the hypothesis that SES is generating a different set of attitudes towards the information highway. Respondents with higher incomes have a different set of parameters that influence the likelihood of Internet use than do respondents with lower incomes. For both groups, the variables point in the same direction however the intensity and magnitude of the coefficients vary somewhat between groups. Significant variables for higher income earners include factors 1, 2, and 3. Among low-income respondents, we find that factors 1, 3, 4, and 5, as well as the age demographic, are significant determinants in estimating the likelihood of Internet use. Exhibit 5.2.4: coefficients of logistic regression model based on income Dependent Variable: Internet use in the past three months Income greater than $60K Income less than $30K Variables B s.e. Sig. B s.e. Sig. 1. Lack of confidence -1.07 .018 0.00 *** -1.73 0.33 0.00 *** 2. Experience and interest 0.31 0.15 0.04 ** 0.26 0.26 0.32 3. Practical tool 0.36 0.15 0.01 ** 0.96 0.26 0.00 *** 4. Passive convenience 0.13 0.14 0.36 0.38 0.22 0.08 * 5. Leisure, culture, and entertainment time 0.20 0.14 0.16 0.73 0.26 0.01 ** 6. Telephone use 0.20 0.18 0.27 0.32 0.25 0.21 Age -0.10 0.13 0.44 -0.37 0.19 0.04 ** Gender (men) -0.04 0.28 0.88 -0.51 0.46 0.27 Education -0.03 0.17 0.86 0.36 0.30 0.24 *** significant < 0.01; ** significant < 0.05; * significant < 0.10 Having explored some of the differences between Internet users and non-users, particularly the effect of the respondents' socio-economic status, it would be appropriate at this time to explore the different attitudes and beliefs present among non-users of the Internet. Population Array of Non-Users In the first wave of the questionnaire, non-users were asked to identify the main reason why they do not have an Internet account. Responses were grouped into three broad categories: cost issues, lack of interest, and no apparent need. There is obviously room in these categories for some overlap. It is easy to imagine that the distinction between the level of interest and the perceived need for an Internet account may be somewhat blurred for a number of respondents. But fundamentally, they are different types of reactions. If we were to change the stimuli (i.e., create a need for Internet use much the same way the telephone has become an indispensable household or personal technology), we might still find a segment of the population that displays no interest in getting an Internet account. Of all identified non-users, approximately one in five indicate cost as the main reason why they do not have an Internet account. An additional 20 per cent display no interest in acquiring Internet access. Nearly three in ten indicate that they do not need Internet access from home. TYPE ONE: COST TYPE TWO: INTEREST TYPE THREE: NEED We know that users and non-users score quite differently on a number of scaled factors we identified in the previous section of this report. For instance, Internet users are more likely to be keen on tele-work and have a greater sense of confidence in the safety of e-commerce. Across the groups (or typologies) of non-users, we examine whether these factors apply with the same intensity or whether there are significant differences in the characteristics latent to each type of non-user. Exhibit 5.3.1 provides a summary of the results based on the attitudes and behaviours extracted in the factor analysis. A plus (+) symbol can be interpreted in such as to infer that the variable has a positive loading on the factor. In terms of demographics, there are some interesting differences between the identified types of non-users. Exhibits 5.3.2 through 5.3.5 show the distribution within demographic characteristics (column percentages) and across the typology of non-users (row percentages). Exhibit 5.3.1: Typology of Non-Users Reasons why no Internet Access Cost / Affordability No Interest Don't Need Factors 1. Lack of confidence +++ ++ + 2. Experience and interest + -- - 3. Practical tool + -- - 4. Passive convenience - -- --- 5. Leisure, culture, and entertainment time -- + - 6. Telephone use + -- - Respondents who identify cost as the main reason why they don't have an Internet account are likely to have a greater lack of confidence in their technical and computer skills. Compared to other types of non-users, they have some interest and experience with the information highway, and are more likely to see the societal applications of the Internet. They are less likely to have a lot of time for leisure and cultural activities. Telephone use is highest among this group of non-users. Respondents with no interest in acquiring an Internet account also have a lack of confidence in their high tech skills. Particular to this group is their lack of interest and experience; they are also less likely to see the Internet as a practical tool or derive any benefit from its societal applications. They have time to take in entertainment and cultural activities in and out of the home and among all non-users, are the least likely to have a high telephone use. Our third major type of non-users, respondents who indicate they don't need an Internet account, are not likely to view the Information Highway as passive convenience or having any direct or indirect societal applications. All other indicators are similar to the second type of non-users (no interest), with the exception of the time variable for leisure and entertainment. Like their counterparts who indicate cost as the major barrier, they are less likely to have a positive loading on that factor. Exhibit 5.3.2 displays the results of typology of non-user by SES category. Overall, a majority of non-users (57%) are of low SES. Socio-economic status does not appear to be a factor in the distribution of non-users across two of the three typologies. Only on the issue of cost/affordability of Internet access are low-SES respondents over-represented among non-users. Although a plurality of non-users indicate they don't need the Internet, the proportion that indicate cost as a major factor is heavily weighted by a margin of two to one among low SES respondents. Exhibit 5.3.2 Cost / Affordability No Interest Don't Need Overall Socio Economic Status Low SES 63% 31% 53% 28% 55% 42% 57% 100% High SES 37% 23% 47% 33% 45% 44% 43% 100% Overall 100% 27% 100% 30% 100% 43% The main reasons for not having an Internet account vary somewhat depending on the age of the respondent. We see in the far right column of Exhibit 5.3.3 the overall distribution of ages among non-users. Exhibit 5.3.3 Cost / Affordability No Interest Don't Need Overall Age < 25 7% 39% 5% 31% 3% 30% 5% 100% 25-34 22% 35% 13% 24% 15% 40% 16% 100% 35-44 34% 33% 24% 27% 25% 40% 27% 100% 45-54 20% 27% 22% 34% 18% 39% 20% 100% 55-64 11% 19% 16% 32% 18% 49% 16% 100% 65+ 4% 7% 19% 41% 18% 53% 15% 100% Overall 100% 26% 100% 31% 100% 43% Aside from the low number of non-users below the age of 25, the distribution of non-users is fairly well spread out across the identified age categories. The proportion in each 10-year age cohort (over the age of 25) ranges from 15% (over the age of 65) to 27% (35-44 year olds). We know that from overall age sampling and results from the previous section (factors of users and non-users), the oldest age cohort is over-represented among non-users. The older the respondent, the less likely they are to mention cost as the main factor why they do not have an Internet account. Conversely, younger non-users are less inclined to indicate they have no interest in accessing the Internet. Close to four out of ten respondents under the age of 25 mention cost/affordability as the main reason. This proportion drops to two out of ten respondents among respondents aged 55-64, and less than ten percent among respondents over the age of 65. Older respondents are far more likely to indicate that they don't need access to the Internet. In Exhibit 5.3.4, we see that women are over-represented among non-users. In our multivariate analysis combining demographics and attitudinal factors, the gender of the respondent was not a significant parameter in determining the likelihood of using the Internet. Most of the variance, therefore, was explained by other factors in the model. It is interesting to note that the distribution of men and women across the typologies of non-users is very balanced. Men mention cost, interest and need with the same pattern of frequencies as women. Exhibit 5.3.4 Cost / Affordability No Interest Don't Need Overall Gender Men 37% 24% 39% 31% 41% 45% 39% 100% Women 63% 27% 61% 31% 59% 42% 61% 100% Total 100% 26% 100% 31% 100% 43% There is little evidence to show that the level of education has a significant effect on the typology of non-users (Exhibit 5.3.5). A majority of non-users have an education level of High School or less, and approximately three in ten have some post secondary education. Exhibit 5.3.5 Cost / Affordability No Interest Don't Need Overall Education High School or less 55% 27% 56% 32% 53% 42% 55% 100% Some Post Secondary 29% 27% 30% 32% 27% 41% 29% 100% University 15% 24% 15% 27% 19% 49% 17% 100% Overall 100% 26% 100% 31% 100% 43% To summarize, there are some definite attitudinal factors at play in the different typologies of non-users. There is also some evidence that age and SES may also be factors in determining the main reason why respondents do not have access to the Internet. Fundamentally, there are different types of non-users, each with its distinct structure. Any effort made to address the needs of non-users would have to recognize this, either in enabling them to get on-line through alternative means or simply recognizing that a segment of the population is not interested in acquiring Internet access. The multidimensional analysis provides us with meaningful insights into the different sets of attitudes and demographic characteristics of Internet non-users. This is an especially practical tool for developing policy responses to suit the unique character of different types of Canadians. There are a number of general basic access, technical training, content development and alternative service delivery implications that should be part of a public policy response. Type 1 non-users recognize some value of the Internet in meeting some of their needs, but face the primary obstacles of cost/affordability and technical skill development. This group is considered to be 'near-users'. Social literacy is an important factor for these individuals, as well as those comprising the other two Types. Public access sites (CAP) and community learning services (CLN) have important roles to play in meeting the needs of Type 1 non-users. Public education and awareness initiatives by organizations (community, NGO's, government) providing these services would likely facilitate and improve the use of the Internet by this group of non-users. Broadly speaking, Type 1 non-users tend to cluster around younger Canadians (44 and younger), but this is not exclusive. There is an important gender gap in the non-user typology (overall 61% women, 39% men). Women are over-represented in Type 1 by a ratio of 2 to 1. Type 2 non-users similarly face technical and cost barriers, but as great a challenge is the perceived or very real lack of personal or social benefit and value of Internet service. Improved content (economic, social and cultural), designed and developed to better meet the existing needs of these non-users, may facilitate greater online activities. Technical skills training, content relevant to social needs and the availability of public access sites are also important components of an access strategy for this group. There is a tendency (again not exclusive) for the Type 2 non-user to be in the pre-retirement cohort (45-55 yrs). At the federal government level, major roles exist in access and content activities for HRDC (CLN), Industry Canada (CAP), Heritage (social, cultural content), and Treasury Board Secretariat (TBS) (government services). Type 3 non-users are also relatively far removed from online activities, particularly in the near future. In particular they are not likely to see or derive any personal benefit from access, and are not likely to have the resources or social skills and interests to benefit from current Internet access. Broadly speaking, seniors and retired Canadians (55+ yrs) tend to make up much of Type 3 non-users. For both Type 2 and Type 3 non-users, strategic needs assessments (social needs) combined with content development initiatives may increase the propensity for Net use in the longer term. The proportion of women is slightly higher in Type 2 and Type 3 non-user than men. It is important to note that all the non-user types are clearly dominated by women. From a public policy perspective, the reality is that these non-users will need to be provided with communication and information services in a variety of traditional and standard formats or means, and accessible for those with different skill and income levels, for some time to come. The federal departments with mandates in these primarily social policy issues include HRDC, TBS and Heritage. Social Context The Immutability of the Dual Digital Divide? During the transition to an 'information society' over the next decade or so, an ongoing tension will exist due to those 'have-nots' who have no interest in the Internet or other aspects of the Information Highway and the others who desire to be connected to meet very real educational, literacy, skills, economic and social needs but face serious obstacles. This tension, evident in much of the discussion about the research findings above, gives rise to several questions. How serious is the digital divide? How can we ameliorate the outstanding cost and skill disadvantages of those who remain unconnected? What are the content dimensions of the divide? Is this divide immutable or subject to resolution through developmental initiatives by government and/or industry? It is important that the digital divide and 'have-not' problem is not simply underrated because at this relatively early stage of development a large number of non-users express little interest in the technology. Nor should this problem be down played based on arguments, such as there is little content of value in terms of mass appeal and benefit, and that this will change, thereby attracting more users. Advanced interactive communications will be pervasive throughout our social and economic activities over the long term, whether this is experienced implicitly or explicitly by individuals. This is much the case now. Central to the core strategies of governance and the market, technological access and proficiency will be necessary for individuals to maintain a competent level of participation in society and to derive the benefits thereof. Access to the Internet will not, on its own, overcome the social and economic inequalities and cleavages in society or communities, and to the contrary may even aggravate these. But the ability to use the new technologies as tools and resources, and to have a chance to realize potential opportunities, means that some amelioration of some disadvantages and inequalities is possible. As a matter of policy, both economic and social, it will be necessary for government to continue to address the challenges arising from both sides of the dual divide. Access, technological literacy and content development initiatives will need to continue over the near term to address the near-user segment of the 'have-nots', and these initiatives will need to evolve over the longer term to address the needs of the distant-users as connectivity becomes more important to their lives. Multiple modes for access to information and services will also need to continue to meet the communication needs of all Canadians, whether connected to the Internet or not. A useful way of understanding some of these trends is by considering how people consume communications. The context of the complexity of our lives and society should not be lost in this exercise. People concentrate their lives and activities in life in different ways. As part of this, people use a variety of communication technologies in different ways and different purposes. These options are not just based on choice, but also on the means or resources people have available for communications (e.g., affordability, literacy, etc.). Literacy The concept and practice of literacy extends beyond a consideration of whether people can use computer hardware, software or Web pages. As discussed above, about half of non-users have indicated that such factors as disinterest or lack of need, and the perceived difficulty in using technology were important reasons for not being online (see Figures 8 and 12). The difficulties identified in use, indicate that statements about disinterest or lack of need mask other factors such as literacy. The concept of literacy means more than the basic ability to read and write. Literacy also entails individuals' abilities to understand and use information in a way that is beneficial and meaningful to their every day lives. Literacy levels also affect the choices people make, or limits their options, in the means they use to interact to gain information and participate in society. The highly variable levels of literacy in Canada are one of the explanations that a large number of non-users have no interest or need in the Net. In Canada, about one in five (22%) of Canadians adults fall into the lowest level of literacy. This means they have serious difficulty dealing with, or reading, basic printed materials. Another one in four (24-26%) have a somewhat improved level of literacy, but are only able to deal with "material that is simple and clearly laid out, and material in which the tasks involved are not too complex" . 21 Literacy, then, means much more than the ability to use a technology, but the capability of the individual to have and use social and cultural skills in ways that allows them to benefit from the use of technology, whether this be Net access, a book, newspaper, television, and so on. 22 Along with cost, lack of perceived need and value, literacy is one of the important variables which help explain the low level of Internet use by seniors. About six out of ten seniors have never completed high school. In 1994, over half of seniors in Canada performed at the lowest level of literacy as measured using the International Adult Literacy Survey (IALS) with just over half (53%) able to perform simple reading tasks. A minority of seniors, none-the-less, have gone online. In 1998, less than one in ten seniors (9%) reported having Internet at home, and of the seven per cent of seniors stating that they had used the Internet somewhere in the previous three months, most (84%) had done so from home, with the remainder gaining access through schools (8%) and public access sites (5%). A similar minority of the senior population is involved in some type of education program (5% or 175,000). 23 In this context, depending upon the level of literacy for each individual, they may use any one or several of a mix of different technologies to meet their communication and information needs. For example, where someone has difficulty with print or computers/Internet, they may rely instead on television or telephone for their information needs. The preferences expressed by Canadians to use a number of different technologies to access information and services (see Figure 16) will have less to do with unfamiliarity with the Internet, and more to do with the capability of using different technologies to meet needs. To accommodate the range of literacy skills in Canada, and the preferences of the public to use different means of access for different types of information and services, requires that these different means of access continue to be made available. As well, it is important that the users of these alternative forms of access and distribution are able to receive the same level or quality of service and information that is being made available on the Internet. There is another disturbing trend involving literacy and the Internet. There tends to be a rather uncritical optimism and faith on the part of Internet proponents about the reliance on technology and the Internet as a primary tool to overcome literacy, skills and job training challenges. To some extent, this theme is apparent with both the CAP and CLN programs. Such assumptions risk diminishing the need to provide learning, training, skills and literacy materials in other formats and by other means of access (e.g., in-person, print). These options are necessary to facilitate individuals' existing capacities and skills, and address obstacles for Canadians who use new services in a basic way, or who could benefit from these services if a combined approach to learning was better employed. This is not to imply that technology does not have an important role to play as one resource among many. For example, a Statistics Canada study of computer literacy observed that inequalities of access meant that those who could benefit the most from new technology by improving their skills (low income, unemployed, lower education) risk becoming further marginalized as other Canadians benefit from the development of these new skills.24 But caution in the introduction and reliance on new technology is important. Historically, for example, when other ostensibly revolutionary technologies, such as film, radio and television, were introduced to society, it was thought that these would not just add new skills, but would also be able to replace traditional forms of learning, and even replace schools. This did not turn out to be the case, because the conditions and practices of successful literacy, learning, etc. are not technology centric. 25 Research also suggests that another literacy problem may be developing due to an over reliance or sole reliance on computers and the Internet for access to information, education and training. The over reliance on computers and network-based learning and training has meant that other forms of learning and social capacity development have been greatly reduced or displaced altogether. Research in education over the past decade has found that a major cost of this technological determinism has been a deskilling or increase in illiteracy in the traditional basic areas of reading and writing, as well as in the capability or capacity of individuals' social, community and general life skills and abilities. In other words, youth, who as a social segment are the major users of the Internet and new technology, while proficient in this, have much downgraded skills and abilities in the fundamental practical use of the arts, sciences, and the ability to socially interact and participate in a meaningful way. These involve those many activities in daily life away from the computer and the Internet. While these high technology users have been exposed to much information, they have not actually learned or formed the same level of knowledge and developed the ability to apply it as a social capacity in their daily lives as have preceding generations. So, while it will be important that while policy attempts to create technological based literacy and get more Canadians online, at the same time, these initiatives don't, as a consequence, erode the standard and socially necessary forms of literacy. 26 Incomes, Cost and Affordability The greater divide in Canada than that of the Information Highway is that of income. Cost barriers to accessing the Internet are multi-dimensional. Cost as a barrier needs to be considered in a wider context than just that of the cost of online service. The affordability of the Net and other goods and services in people's lives very much depends on their overall circumstances. The trend in recent years in Canada has been towards greater financial demands and stress for Canadians. This has implications for which products and services they are able to spend money on, and which level of service (first or second class) they are able to afford. In addition to the obvious costs of computer hardware or set-top boxes, there are also the costs of software, and monthly ISP charges. These costs all present immediate barriers or obstacles for many non-users. For others, the cost of the underlying communication networks necessary to use these Net access technologies and services can also present affordability challenges. For example, many Canadians who have telephone and cable service do not necessarily find these services affordable. However, they take great pains to keep them because they are considered essential and necessary. 27 In 1994, Ekos Research Associates Inc. provided a useful typology of the changing class structure in Canadian society. This typology differentiated segments of class in Canada using characteristics and perceptions of Canadians on a range of social, economic and cultural issues. This analysis identified a major schism, most often around core values, between the elites in society and government, and those in the middle and lower income groups in society. This research also echoed the findings of Statistics Canada and others showing an emerging class polarization in Canada. In the Ekos typology, the Elite Insiders (19%) forming the upper class represented the information economy knowledge workers. These individuals are generally well educated, highly skilled, and well paid. The middle class was represented by two groups: the Secure Middle (24%) and the Insecure Middle (16%). The secure middle were economically secure with higher than average education levels, and fairly high incomes. The Insecure Middle reflected the erosion of the middle class in Canada down to the expanding lower class. This group had middle incomes and better than average education levels. However, this group was also facing an erosion of economic security, jobs and income. The lower class was subdivided between the Disengaged Dependents (22%) and the Outsiders (19%). The former group are largely detached from the world of work and experience poverty, low literacy and skill levels, unemployment, etc.. These Canadians are also the least well educated, but value highly skill development. The Outsiders, are largely removed from the mainstream of society and are at the margins of the economy. This group is poorly educated and experiences high levels of unemployment.28 The Information Highway study has shown that with affordability/cost issues and the Internet, Canadians from all classes face some degree of economic stress. This is more aggravated for those in the lower class. Evidence since the Rethinking Government (1995) study shows that the trend of the erosion of the middle class and the growing income disparities in Canada became even more dramatic in the 1990's. Middle class shrinkage and the swelling of the lower class, and the commensurate stagnation or shrinkage of income at both the middle and lower class levels suggests that optimism about all Canadians being connected to the Internet is exaggerated. As discussed above, differential levels of access are likely to become a structural feature of the Internet and other new technologies and services in Canada. While this type of structural barrier may be partially overcome by falling prices (temporal barrier) of access fees and access technologies, these trends will not put everyone online. Other structural and class based barriers, such as literacy and education, or obstacles, such as lack of interest or lack of perceived value, will persist. Middle class shrinkage has meant that where 60 per cent of families with children under the age of 18 earned between $24,500 and $65,000 in 1973, by 1996 this had dropped to 44 per cent. At the same time, the core middle class earning between $37,600 and $56,000 per annum, had dropped from 40 per cent of the population in 1973 to 27 per cent by 1996. Tied to this was a fundamental change in employment, both in terms of types of jobs and hours worked. From the 1970's to the 1990's, the percentage of the labour force working on a full time basis dropped from two thirds to fifty per cent. Moreover, one in five jobs were part time in the 1990's, roughly double that of the previous generation. The fastest growing segment in the labour force in the 1990's was casual jobs (15% of all jobs) with self employment accounting for about half of all the new jobs created during that period. Employment patterns also featured a shift away from a manufacturing to a service economy. Research indicates that many of these new types of jobs do not require, or make available, Internet access as part of the work activities. As such, work as an alternative means of access for those without access at home will not be a broadly based mitigating factor of the connectivity problem. 29 In terms of people, this meant that by 1997, there were over 5.1 million poor people in Canada, with a poverty rate of 17.2%. Unattached individuals (36.3%) were three times more likely than families (14.3%) to be in poverty. Most of the poor families (82%) were accounted for by what is considered fairly mainstream definitions of a family. This included: married with household head over 65 (8.6%); married, under 65 with children (11.9%); married, under 65 no children (10.3%); and single parent mothers (61.4%). Added to this number were the working poor. In 1996, over 200,000 people were poor even though they worked 49-52 hours per week. 30 Canada tends to be very democratic with the distribution of our poor. There are identifiable neighborhoods, communities and regions in which residents clearly suffer economic disadvantage. But poverty exists across all neighborhoods and communities, and in both rural and urban areas. In 1997, 681,000 poor families or slightly more than half (57%) of all poor families, and 891,000 singles (60%) of all poor singles, lived in cities of 500,000 population or more. In rural Canada, almost one in four (22.4%) of unattached individuals were poor. 31 As with literacy, these demographics have implications for how information and services are made available in a diversity of formats, often more affordable formats, such as print or in-person, and how this is done on a geographical basis, in cities, communities, neighborhoods, and rural areas, to facilitate those who choose not, or are not able, to gain access through the Internet. Diversity of Content While there is an increasing abundance of commercial content on the Net, a major weakness in Canada is the lack of a diversity of quality, indigenous, socially and culturally relevant content. The experience of many using the Internet is often frustration in being able to find information, and as part of this, finding quality and useful information. In addition, many non-users indicate that one of the reasons they are not online is that they perceive little content of any value to meet their needs. This is not to say that that there is a shortage of information; if anything the Internet is an information glut! But much of the information is not catalogued in a user friendly searchable manner, and, as well, increasingly a considerable amount of information being fore grounded on portals and mega sites, whether by industry or government, tends to reflect the interests of those information providers as opposed to the public user or potential user. With plans by governments to ramp up the provision of information and services online in the next few years, the public's need for both quality and a diversity of information should be incorporated into information development and dissemination strategies. Without this, the resulting 'fire-hose' information provision strategy may get quantity on the Net, but this may not meet the quality 'drinking water' needs of the public. 32 In particular, there is a need for more public support in the development, cataloguing in a Canadian context, and dissemination of citizenship and other socially relevant information content. This support should not be restricted just to the Internet. The Net is a means for access. Content development occurs in many forms and at different levels in society. It is this diversity that also requires support in order to contribute the content made available on the Net. A number of programs already exist as part of industrial policy to support the Canadian big "C" Cultural content - commercial infotainment products and services produced for domestic and international markets. However, there is much less support for small 'c' content or indigenous content. Small "c" content can generally be described as social and cultural information (local, regional, national) that is necessary for individual Canadians to meet their daily social, educational, economic and cultural needs. This information already exists in other formats, but has only partially been reproduced for online access. As well, this information needs to be designed and offered with a number of accessible interfaces to accommodate different levels of literacy, and those with disabilities, in Canada. At the same time, some information will not lend itself to easy duplication or use on the Internet. Information strategies and policies will need to carefully assess the needs and capacities of different interests and segments of society in decision making about the provision of information in different formats. 33 The concept of diversity of content also raises the issue of the importance of certain types of information (its availability, how it is packaged, etc.) for different segments of the population, for example, seniors, unemployed, part time employed, youth, etc. If the presence of needed content is one of the key criteria for attracting people to the Internet, consequently, the quality of this content will be important in keeping them online. Two of the problems in getting more people to use the Internet are: the development and packaging of information to meet the specific needs of different segments of the population, and to do so in a way that is of comparative value to other means, and; where such information does exist, the lack of awareness by non-users of the range of information that is available and how these resources may help to meet their needs. The development of relevant information resources is clearly one of the important roles for public policy. The governments and public service organizations involved in promoting and facilitating public access also need to develop better communication strategy's to create higher levels of awareness in different user segments about the availability of public access, the types of information available, and the availability of literacy training and related public Internet services. At the same time, it should not be simply assumed that Web-based content or delivery of content will be a sufficient replacement or displacement of this information in other formats. As Figures 18 and 19 below demonstrate, how different demographic segments use the Internet tells us less about technology and more about needs and the social context or daily lives of individuals. In other words, how relevant is the information or service on the Web for application and benefit in the lives of the users? Figure 18 Source: Ekos Research, 1998. Figure 19 Source: Ekos Research Associates, 1998. Understanding social context is important to understand how readily new technologies are accepted and used. This can often create a tension with the emancipatory or revolutionary claims that are sometimes made about new technologies. People generally judge a new technology in relation to the benefits enjoyed from the technologies or services they already know and use, for example, television, cable-TV, automobile, telephone, catalogues, books, etc. The value of a new technology is often measured by people in relation to its performance, the satisfaction derived, the costs and the degree to which a new technology meets existing needs or improves daily life. Research in the U.K. on the acceptance and use of the Internet and other technologies has demonstrated the importance to individuals of the degree to which a technology benefited existing values or activities (the social value). For example, earlier technological improvements, such as washing machines and the microwave oven improved some household activities, often by saving time. However, contrary to the promises of the capabilities of the microwave when it was first introduced, cooking by other methods (e.g., stove top, conventional oven) was not displaced. The point being that the take-up of new technology by individuals is not based on the speculative benefits of a technology offered up by those promoting these, but on whether or not the technology satisfies existing needs and can be easily incorporated into an individuals' life activities. Technological possibility does not necessarily mean social or individual desirability. While the Internet may offer specific communicative benefits in some activities or areas, such as the work place or distance interaction/transaction, its socially transformative capability has yet to be demonstrated as possible, or desirable, for all activities or types of information/services. 34 Cost is also closely tied to use value assessments by individuals. In the competition at home for consumption of entertainment and communication services, the Internet is at some disadvantage because unlike the one time purchasing of radios, televisions, etc., the costs involve purchasing (fairly often!) a relatively expensive computer or digital decoder, as well as ongoing monthly fees. The cost/affordability and social value dimensions of the Internet (let alone quality of content/service) as compared to the existing empirically measurable household/ individual activities raise questions about the rather extensive speculation on the revolutionary potential of the Web. Social value considerations also offer some insights into better understanding why many choose not to be online, or why many of those online add the Internet to their existing communication practices as opposed to displacing these. While Web-based content/services will be important, and increasingly so for some, the ability to access and use communication and information resources in a diversity of formats and through a diversity of means of access will be more important for the majority of people for years to come. 35 Changing Patterns in Consumption? Recent trends in the development of the Internet, such as how it is being made available, accessed and used, suggest ways to think about how the digital divide may need to change and how actions to address the issues of access and content may need to evolve. To generalize, the use of the Internet is tending to form around three models of consumption: passive, active and work. Passive consumption is a well established pattern, with the most obvious example television viewing. With the Internet now a central line of business for media companies, the trend is to produce, package and distribute content in a similar fashion to that of other standard media commodities. Thus, we have programming and channels, selected from electronic shelf space on mega-sites or portals, e.g., AOL. This marketing strategy (information push) combined with the likelihood of the Internet being integrated as a commodity with telephone service, cable television and other video services, creates conditions whereby Internet content can be passively consumed by individuals as viewers, instead of as surfers, e.g., click and play, click and watch, click and print, or click and write an email, and click back to TV. The ease of access to, and use of, the Internet in a passive information and entertainment format will create conditions to make this more accessible to a part of the current non-user population. However, interest, literacy, skills and affordability will remain as issues in varying degrees for many non-users. Affordability of service will be a continuing concern to lower income users. Active consumption, where you really have to make some effort to acquire something, is a second dominant theme of current Internet evolution. This largely involves conducting transactions online, whether this be e-commerce, filing or retrieving forms, documents or other types of information. This will overlap with passive consumption to some degree, e.g., e-commerce browsing, but requires not only the ability to afford access, but higher levels of technological proficiency and literacy than passive consumption. The third theme is consumption as work. This will likely remain the purview of information elites and geeks. Consuming the Internet as work involves the often frustrating and time consuming surf and search for a particular bit of information necessary for a specific task or need. In addition to dedication, technological proficiency and possibly more expensive technology than Internet through the TV will likely be required to undertake these tasks. Issues of the Dual Digital Divide will need to be continually rethought in the context of how different types of information and services, particularly those considered essential and necessary, are made available and accessed given the different ways people consumer communication products, and the resources they have available to do this. The actual and potential demand and use of the Internet and its services will not be satisfied for the 'have-nots' if important variables such as content value, available time, social context, technological literacy, affordable access and related resources are not addressed. U.S. Digital Divide As countries at the forefront of the development of information economies and society's, it is useful to consider the similarities or differences in Internet use between Canada and the United States. In general, a similar pattern of Internet connectivity has emerged in the United States to that of Canada. At the end of 1998, just under half (40%) of U.S. households had computers and one quarter (25%) had Internet access. As with Canada, there has been remarkable growth in Internet penetration, but again the greatest growth exhibits a social class pattern whereby an information rich group (higher income, better education, dual parent households) is clearly differentiated from an information poor group (younger, lower income and education, minorities, rural and central city locations). As with Canada, while there is growth in all income segments, the digital divide is widening between upper and, the lower middle and the lowest income segments. The expanding digital divide based on income group indicated that the divide was widening between the upper income groups and all groups below $50,000 income. The growth of the spread in access between the highest income ($75k +) and the lowest income (<$10k) segments was twenty nine percent (from a 42% difference to 52%). The spread between the upper income and other lower income segments also increased during this period. As in Canada, this lag is expected to persist for some years to come. 36 Source: NTIA, 1999. In numbers similar to Canada (see above) the main reasons cited for not having Internet access from home were lack of perceived need (25.7%) and cost (16.8%). The main cost factors involved the monthly Internet service charge (9.7%) and toll fees to reach an ISP. About one in ten (9.6%) of respondents indicated that they didn't have home Internet because they could use this service somewhere else, whether at work, school or a public access centre. 37 In the United States, public access is made available through Community Access Centres. Community Access Centres (CAC) are comparable to Canada's Community Access sites (Community Access Program - CAP). In the U.S., these are located in schools, libraries and other community access points. They are seen as continuing to play an important role as part of public policy to ensure all have some form of access, until every household can afford access. CAC's tend to be used by the unemployed, low income and those with lower education to search for jobs and take courses. 38 The types of uses of the Internet also varies by social class in the U.S., for example, in the lower income strata (under $20k) about half (45% - 50%) take courses on the Internet as compared to about a third (35%) of those in higher income groups (over $25k). Similarly, while about a quarter (20% - 25%) of the lower income segments use the Internet for job searching, a somewhat smaller number (15% - 20%) in the over $25k segments use it for this purpose. 39 United States' policy relies on the promotion of a competitive market and universal service goals as the primary measures to rework the country into an information society. The underlying justification for this goal is similar to that of Canada. While the Internet is not currently an essential service, in other words necessary for survival, nonetheless it is considered to be approaching this in the U.S. where it, with the telephone and computers, will become "necessary for success". 40 The Internet's value is contextualized in how it is being developed and used to change work, communications and consumption. While competition in communications is seen as lessening the problem of a digital divide to some degree, such as through price reductions, it is not considered the only solution. Cost of equipment and service, affordability, and ability to use will continue to be obstacles and barriers. In the near term, initiatives including the development and support for CAC's, universal service support expenditures, literacy initiatives, language initiatives, product design for those with disabilities, and communication/awareness strategies (awareness of access sites/services for target groups) are identified as necessary ongoing public initiatives. 41 Conclusion The coasts of history are strewn with the wrecks of predictions. James Bryce, U.K. Ambassador to the U.S., 1893. As the Internet joins the mix of mainstream communication technologies, for the majority of individuals, some level of access will be necessary to maintain a competent level of participation in society, particularly where individual economic and social needs are involved (e.g., employment, access to important information and services). Over time, the Internet can be expected to take its place as a complementary means of access with other forms of communication and social interaction, as opposed to displacing these, at least for the foreseeable future. The social context of people's lives will be the major determinant of how the Internet will be incorporated into the broader mix of communication technologies and practices, and this will be based on such variables as cost, existing needs, the use value of the information and services, etc. People will still go to stores, watch television, read books, meet friends in public, go to concerts, and attend schools. With changes in technology, pricing and related issues, who will be accessing and using this technology on a mainstream basis and by what different means remains subject to some speculation. Currently, the Internet is approaching close to what is likely to be the maximum penetration levels for upper middle and upper income households, but with many in the lower social classes and a minority in the upper classes not connected. The optimism about all Canadians being connected to the Internet is exaggerated. Over time, unless there is something about the Internet that will make it as essential and valuable as local basic telephone service or, affordable, easy to operate and ubiquitous as radio and over-the-air television, a Dual Digital Divide featuring differential levels of access will persist as is the norm with most other products and services in Canada, including communications. Differential levels of access are informed by such variables as: needs, income, affordability, perceived value, education, gender, location and literacy. Income (cost) has the most significant impact on Internet use. Age has a lesser, but still significant effect, with some older age cohorts less likely to be Internet users than their younger counterparts. Gender and education tend not to have major impacts on the propensity towards Internet use, but still have some importance. Females are over-represented in the non-user population. In addition to income, socio-economic status plays an important part in determining and understanding attitudes expressed by individuals towards the Information Highway and how this relates to use. Higher income individuals have a different set of parameters (attitudes) than lower income Canadians that affect the likelihood of Internet use. Non-users generally fall into three distinct groups or Types. Type 1 non-users are interested in online service but face obstacles and barriers, such as cost and technical literacy. This group can be considered near users. Improved content, the availability of public access sites (e.g., CAP, CLN) and, increased awareness through promotional initiatives of the existence of these public services and the content available may facilitate and improve the use of the Internet by this group. Type 2 and Type 3 non-users similarly face cost and technical literacy obstacles, but other important factors are the lack of relevant content and the perceived lack of personal benefit and social value of Internet service. Over the longer term, Type 2 non-users may use the Internet more, but this will require some public role through access sites (CAP, CLN), relevant content development, training assistance, etc. Type 3 non-users are far removed from likely being Internet users. In the mid and long term, government and other organizations will need to continue to serve and communicate with this group, as well as many in the Type 1 and Type 2 groups, using a variety of existing services, such as in-person, mail, telephone, etc. As with current Internet users, when non-users do start using the Internet, they will most likely be adding this to a mix of communication access methods, as opposed to displacing all of these. The speed of the introduction of the Internet, and the central role it is playing in the strategies of industry and government, necessitates a public policy role to facilitate access, technological literacy and the development of relevant and diverse content resources. To this point in time, the Information Highway and the Internet have largely been subject of 'push' strategies' by government and industry - the public has not been concerned with home Internet access in any comparable way to that of quality heath care, good jobs, fairer taxes, or cheaper gas prices. More recently demand has become as an important factor for some social segments, and with some professions, and this may extend further if perceived relevance and value increases. A central role for the Internet in the provision of government information and services to the public, and as part of national industrial, social and cultural development policy, means that, unlike other home oriented communication technologies which are discretionary in nature, access to the Internet will gain importance such that individuals are able to satisfy particular needs, as discrete as these may be at times, in an effective and timely fashion. There are many different policy options available to resolve the issue of technical access to the Internet. For example, in the United States, a number of companies are offering combinations of free computer/Internet access that are tied to use and advertising commitments. In another approach, La Grange, a city of 27,000 in Georgia, has addressed the digital divide problem by paying for all of its citizens to be connected to the Internet. The city has realized a cost efficiency through a group purchasing approach whereby it costs less than $10 per year per citizen for broadband connectivity. A similar approach has been employed in Canada by community networks. 42 Like La Grange, Sweden has similarly adopted a policy of providing broad band access to everyone in the country. In Canada, the government of the province of Quebec recently announced a $121 million program to subsidize the cost of getting more families on the Internet. The program subsidizes either the rental cost of a computer and Internet service, or Internet services through a PC or the television. The program covers up to 75 per cent of the costs to citizens to a maximum of $450 per year. 43 However, in the absence of a broadly-based policy approach in Canada whereby different levels of government directly subsidize technical access from the home, there will be an ongoing role for the federal government to provide a large number of public access facilities. A federal role will also exist for literacy and content support. In this context, for non-users, occasional and regular Internet users, core policy objectives across government departments will need to be concerned with such issues as: access; technological literacy; social competency, capacity and application; content development initiatives at different levels of society; and the continued availability of multiple means of access and information distribution. These efforts will also need to evolve and change over time as the needs and practices of individuals change, and as technologies are changed. For those who cannot afford, or choose not to access the Internet from home, public access sites will continue to play in important public policy role. The primary purpose of these sites will be that of fulfilling a public service role within the overall federal social and economic policy frameworks as these relate to the Information Highway. Variable public Internet access through public sites conforms with existing social practices and the use of public services in society, for example, using a library, a post office, or a government office, etc. On a forward going basis, strategic analysis and planning will need to be devoted to such questions as: the appropriate number and locations of public access sites; community networking models; local management responsibility; quality of service (both infrastructure, content, and staffing); services available to the public; the terms of access for different services/uses; and government funding obligations and responsibilities. These should be primary concerns of the federal departments of Industry Canada (CAP program) and HRDC (CLN program) and are also relevant to their community, municipal and provincial government partners. In the area of content, there is a need for greater cooperation and coordination of activities between federal governmental departments, and with provincial, regional and local partners. For federal departments, such as Industry, HRDC, TBS, Heritage, Health, among others, in the context of Canadians' social needs and capacities, policy and program initiatives will be required to ensure that traditional forms of diverse content (community, regional, national) and a diversity of content in electronic format be developed in complementary ways to meet existing and evolving communication needs. Where appropriate for communities, governments, as part of their statutory obligations and responsibilities, should support the integration of community access and networking activities into formal or informal collaborations and partnerships. Training and local content development support should be included in this. These integration activities should be led by community groups and driven by the particular needs of individuals and groups in each community. Community groups and organizations have developed extensive expertise over time to meet the needs of their community. As well, they have also established successful working relationships with many other groups in their community. These existing strengths and relationships permit the incorporation of computer technology and online communications as resources to improve these services. More simplistic options, such as imposing models from outside that may be antagonistic to these experiences and this expertise, or introducing new technology with a view to displacing existing services and expertise, risks ill-serving communities and individuals. As opposed to technology-centric approaches, initiatives by community based organizations and different levels of government in the development and provision of content and services need to be undertaken within a context of how information is currently accessed and used by different social segments to meet different needs. Different content development initiatives will need to evolve, be replaced by others and perhaps even disappear altogether, as online activity moves from the margins to become one of the mainstream communication tools in individuals lives. As part of the design and development of electronic content resources, efforts are required to better organize and catalogue socially, culturally and economically relevant Canadian resources. This will facilitate access and create value for users, and increase the attractiveness of service for non-users. Other problems that have had some effect on the relatively low levels of use of public sites by non-users are a lack of awareness of the availability of public sites in the community, and the availability of electronic content that may meet the specific needs of individuals. Improved communication strategies, both at the local and national levels, with goals of increasing the awareness by the public in general, and specific underrepresented social segments, of the availability of sites and relevant content resources would, to some degree, facilitate increased connectivity. For example, communication strategies may be useful in linking the existing, known needs of population segments, such as seniors, under employed and unemployed, lower education/skills, etc., to awareness by these groups of the availability of useful information resources on the Internet. The ability for individuals to understand why and how they can integrate the Internet into their daily life practices to meet their existing needs is likely to prove more successful in encouraging use than rather abstract predictions and promises about the information revolution. Looking forward, a recognition that a Dual Digital Divide may be a permanent feature of communications in Canada over the long term, creates opportunities for governance through policy, programs and other initiatives to ensure that diverse means are available such that all Canadians are able to fully participate in society in meaningful and beneficial ways. References Canada (1999) Speech from the Throne, September 23, 1997. ----- (1999) Community Learning Networks Initiative: Guidelines for Applications, Human Resources Development Canada. ----- (1999) Frequently Asked Questions, Community Learning Network Initiative, Human Resources Development Canada. ----- (1999) A Portrait of Seniors in Canada. Statistics Canada, Cat. No. 89-519-XPE. -----(1998) Making a Difference in Human Development: A Vision for HRDC, Practical Vision and Action Plan, working draft, HRDC, June 22. ----- (1997) Preparing Canada for a Digital World. Report of the Information Highway Advisory Council. ----- (1997) Backgrounder: Canada's Information Highway Strategy. Industry Canada, Connecting Canadians Web page www.connect.gc.ca. ----- (1997) Spending Patterns in Canada. Ottawa, Statistics Canada, Cat. No. 62-202-XPB. ----- (1996) Building the Information Society: Moving Canada into the 21st Century. ----- (1996) Reading the Future: A Portrait of Literacy in Canada. Statistics Canada. ----- (1996) "Computer Literacy - a growing requirement", Education Quarterly Review. Centre for Social Justice (1998) The Growing Gap: A report on growing Inequality between the rich and poor in Canada. Toronto, October. Crane, D. (2000) "Sweden's broadband for all shames Canada", Toronto Star, March 23. Dickenson, P. & Sciadas, G. (1999) Canadians Connected. Statistics Canada. Ekos Research Associates Inc., (1998) The Information Highway and the Canadian Communication Household. Ottawa. Ekos Research Associates Inc., (2000) Rethinking the Information Highway: Privacy, Access and the Shifting Marketplace. Ottawa. Ekos Research Associates Inc. (1995) Rethinking Government. Ottawa. Financial Post (2000) "A Georgia City closes the digital divide by giving its Citizens free Internet access", March 23, p. C 12. Manley, J. (1998) Speaking Notes: Connecting Canadians. Notes delivered by the Minister of Industry to the Empire Club, Ottawa, February 26. Morrison, D., Svennevig, M. & Firmstone, J. (1999) "The Social Consequences Of Communication Technologies in the United Kingdom", in (Ed. R. Boyce) The Communications Revolution at Work: the Social, Economic and Political Impacts of Technological Change. Montreal, McGill-Queen's Press. Mosco, V. (2000) Public Policy and the Information Highway: Access, Equity and Universality. National Library of Canada. Network Letter (2000) "Quebec seeks to bolster Internet penetration with tax credits and subsidies", March 27, p. 5. NTIA (1999) Falling Through the Net. Washington, July. Ottawa Citizen (1999) "Wired Canada to cost $5B: Federal Departments' services, information to go on Web in 2004", December 12, p. A1. National Council of Welfare (1999) Poverty Profile, 1997. Ottawa, Autumn. ----- (1998) Poverty Profile, 1996. Ottawa, Spring. Reddick, A. (1999) "The Dual Digital Divide", presentation at the 1999 Atlantic Association of Sociologists and Anthropologists Conference, St. Thomas University, Fredericton, New Brunswick, October 23, 1999. ----- (1999) "Access and the Information Highway", in (Ed. R. Boyce) The Communications Revolution at Work: the Social, Economic and Political Impacts of Technological Change. Montreal, McGill-Queen's Press. ----- (1998) Community Networking and Access Initiatives in Canada, Ottawa, Public Interest Advocacy Centre. ----- (1995) The Information Superhighway: Will Some Canadians be Left on the Side of the Road?, Ottawa, Public Interest Advocacy Centre. ----- (1995) Sharing the Road: Convergence and the Canadian Information Highway. Ottawa, Public Interest Advocacy Centre. Sen, A. (1999) Development as Freedom. New York, Knopf. Upitis, R. (1999) "Impact of the Communications Revolution on Education", in (Ed. R. Boyce) The Communications Revolution at Work: the Social, Economic and Political Impacts of Technological Change. Montreal, McGill- Queen's Press. 1 The Canadian Dual Digital Divide, presentation by Andrew Reddick, October 23, 1999, 1999 Atlantic Association of Sociologists and Anthropologists Conference, Fredericton, Canada. 2 Canadians Connected, P. Dickenson and G. Sciadas, Statistics Canada, February, 1999, p. 3.4. 3 Canada, Building the Information Society: Moving Canada into the 21st Century, 1996; Speech from the Throne, September 23, 1997, p. 16.. 4 Canada, Building the Information Society: Moving Canada into the 21st Century, 1996, pp.3,4; Preparing Canada for a Digital World, IHAC, 1997, pp. 46-51. 5 The action areas include: Canada On-Line; Smart Communities; Canadian Content On-Line; Electronic Commerce; Canadian Governments On-Line; and Promoting a connected Canada to the World, see Speaking Notes for the Honourable John Manley, Minister of Industry to the Empire Club, Connecting Canadians, February 26, 1998. 6 Backgrounder, Canada's Information Highway Strategy, Industry Canada, 1997; Connecting Canadians, web page www.connect.gc.ca. 7 Community Learning Networks Initiative: Guidelines for Application, HRDC, 1999; Frequently Asked Questions, Community Learning Networks Initiative, 1999. 8 Making a Difference in Human Development: A Vision for HRDC, 1998; Practical Vision and Action Plan, working draft, HRDC, June 22, 1998. 9 Reddick, A. (1995) The Information Superhighway: Will Some Canadians be Left on the Side of the Road?, Ottawa, Public Interest Advocacy Centre. 10 The Information Highway and the Canadian Communication Household, Ekos Research Associates Inc., 1998, Final Report, p.9. 11 Spending Patterns in Canada, Statistics Canada, 1997, Cat. No. 62-202-XPB, pp. 34, 36-37. 12 The Information Highway and Canadian Communication Household, Ekos Research Associates Inc., frequency tables, 1998. 13 The Information Highway and Canadian Communication Household, Ekos Research Associates, Final Report, 1998, p. 12. 14 The Information Highway and Canadian Communication Household, Ekos Research Associates Inc., frequency tables, 1998. 15 Ekos Research Associates Inc., forthcoming. 16 Product substitution is where the market for a product is recreated by supplanting an existing technology with a new and perhaps improved one, but not necessarily for reasons of obsolescence. See "Access and the Information Highway", A. Reddick, in the Communications Revolution at Work: the Social, Economic and Political Impacts of Technological Change, (Ed. R. Boyce), McGill-Queen's University Press, 1999. 17 Ekos Research Associates Inc., 1998. 18 Applying these dimensions in a multiple regression reveals very high multicolinearity. Most of the 16 explicative variables (dimensions) do not have independent variance: they are basically explaining the same thing. This increases the standard error of the model, thus greatly reducing the explicatory power. 19 The number of factors was determined by evaluating the relative contribution of including an additional factor into the model. The sixth factor improved the model while the relative contribution of the seventh factor was insignificant, as revealed by Fisher's statistical test. 20 Likelihood estimation expresses the probability of Internet use as a function of model parameters. Logistic regression refers to the log-odds of the probability that (Y = 1). Although the probability is non-linear, by taking the natural logarithms of the odds, we can construct a linear function. The probability (P) that (Y=1) = 1/(1+e-L); where L is the logistic regression, a linear function of K variables. L= B0 + B1X1 + B2X2 + ... + BkXk. 21 Reading the Future: A Portrait of Literacy in Canada, Statistics Canada, 1996. 22 See, for example, Mosco, V., (2000) Public Policy and the Information Highway: Access, Equity and Universality, for the National Library of Canada; or, Sen, A., (1999) Development as Freedom. New York, Knopf. 23 Statistics Canada, A Portrait of Seniors in Canada, 1999, Cat. No. 89-519-XPE, pp. 83-84; Ekos Research Associates Inc., The Information Highway and the Canadian Communication Household, 1998. 24 Statistics Canada, "Computer Literacy - a growing requirement", Education Quarterly Review, 1996, p. 9. 25 Upitis, R. (1999) Impact of the Communications Revolution on Education in (Ed. Boyce, R.) The Communications Revolution at Work. Montreal, McGill-Queen's University Press. 26 Upitis, R. (1999) Impact of the Communications Revolution on Education in (Ed. Boyce, R.) The Communications Revolutions at Work. Montreal, McGill-Queen's University Press. 27 Sharing the Road: Convergence and the Canadian Information Highway, Reddick, A., PIAC, 1995, pp.43-44. 28 Rethinking Government, Ekos Research Associates Inc., 1995, pp. 5-6. 29 The Growing Gap: A report on growing inequality between the rich and poor in Canada, Centre for Social Justice, October, 1998, p.p. x, 24. 30 Poverty Profile 1996, National Council of Welfare, Spring, 1998; Poverty Profile 1997, National Council of Welfare, Autumn, 1999, pp. 10, 16, 17, 19, 52.. 31 Poverty Profile 1996, National Council of Welfare, Spring, 1998; Poverty Profile 1997, National Council of Welfare, Autumn, 1999, p. 52. 32 "Wired Canada to cost $5B: Federal departments' services, information go on Web in 2004", The Ottawa Citizen, Sunday, December 12, 1999, p. A1. 33 Sharing the Road: Convergence and the Information Highway, A. Reddick (1995), Public Interest Advocacy Centre. 34 "The Social Consequences of Communication Technologies in the United Kingdom", D. Morrison, M. Svennevig, J. Firmstone in (Ed. R. Boyce), The Communications Revolution at Work, Montreal, McGill-Queen's University Press, 1999, pp. 63-67. 35 Ibid, pp. 67, 72, 74. 36 Falling Through the Net: Defining the Digital Divide, NTIA, July 1999, pp. xiii, 1, 9. 37 Falling Through the Net: Defining the Digital Divide, NTIA, July 1999, p.38. 38 Falling Through the Net: Defining the Digital Divide, NTIA, July 1999, p. xiv. 39 Falling Through the Net: Defining the Digital Divide, NTIA, July 1999, p. 60. 40 Falling Through the Net: Defining the Digital Divide, NTIA, July 1999, p. 77. 41 Falling Through the Net: Defining the Digital Divide, NTIA, July 1999, p. 77-79. 42 "A Georgia City closes the digital divide by giving its citizens free Internet access", Financial Post, March 23, 2000, p. C 12; Community Networking and Access Initiatives in Canada, A. Reddick, 1998. 43 "Sweden's broad band for all shames Canada", D. Crane, Toronto Star, March 23, 2000; "Quebec seeks to bolster Internet penetration with tax credits and subsidies", Network Letter, March 27, 2000, p. 5. 74 The Dual Digital Divide: The Information Highway in Canada, PIAC, 2000.