The Evolution of the Digital Divide: How Gaps in Internet Access May Impact Electronic Commerce

Currently, over 43 million hosts are connected to the Internet worldwide (Network Wizards 1999). In terms of individual users, somewhere between 40 to 80 million adults (eStats, 1999) in the United States alone have access to around 800 million unique pages of content (Lawrence & Giles, 1999), globally distributed on arguably one of the most important communication innovations in history.

Enthusiasm for the anticipated social dividends of this "revolution in democratic communication" (Hoffman, 1996) that will "harness the powerful forces of science and technology"(Clinton, 1997a) for all members of our society appears boundless. The Internet is expected to do no less than virtually transform society. Nowhere is this confidence expressed more clearly than in President Clinton’s aggressive objective to wire every classroom and library in the country by the year 2000 (NetDay, 1998), followed by every home by the year 2007, so that "every 12-year-old can log onto the Internet" (Clinton, 1997b).

Yet even as the Internet races ambitiously toward critical mass, some social scientists are beginning to examine carefully the policy implications of current demographic patterns of Internet access and usage (Hoffman, Kalsbeek, & Novak, 1996; Hoffman & Novak, 1998; Hoffman, Novak, & Venkatesh, 1997; Katz & Aspden, 1996). For while Clinton’s "Call to Action for American Education" (Clinton, 1997a) may be likely to guarantee universal access for our nation’s next generation, are the approximately 200 million Americans presently over the age of 16 equally likely to have access to the Internet? The findings thus far are both obvious and surprising, with important implications for social science research and public policy.

Key demographic variables like income and education drive the policy questions surrounding the Internet. These variables are important because they are the most likely to have a differential impact on the consequences of interactive electronic media for different segments in our society. Looming large is the concern that the Internet may not scale economically (Keller, 1996), leading to what Lloyd Morrisett, the former president of the Markle Foundation, has called a "digital divide" between the information "haves" and "have-nots."

For example, although almost 70 percent of the schools in this country have at least one computer connected to the Internet, less than 15 percent of classrooms have Internet access (Harmon, 1997). Not surprisingly, access is not distributed randomly, but correlated strongly with income and education (Coley, Cradler, & Engel, 1997).A recent study of Internet use among college freshman (Sax, Astin, Korn, & Mahoney, 1998) found that nearly 83 percent of all new college students report using the Internet for school work, and almost two-thirds use email to communicate. Yet, closer examination suggests a disturbing disparity in access. While 90.2 percent of private college freshman use the Internet for research, only 77.6 percent of students entering public black colleges report doing so. Similarly, although 80.1 percent of private college freshman use email regularly, only 41.4 percent of students attending black public colleges do.

Further, although numerous studies (CyberAtlas, 1999; Maraganore & Morrisette, 1998) suggest that the gender gap in Internet use appears to be closing over time and that Internet users are increasingly coming from the ranks of those with lower education and income (Pew Research Center, 1998), the perception persists that the gap for race is not decreasing (Abrams, 1997).

Hoffman and Novak (1998) examined racial differences in Internet access and use at a single time point and found in 1997 that, overall, whites were significantly more likely than African Americans to have a home computer in their household and also slightly more likely to have PC access at work. Whites were also significantly more likely to have ever used the Web at home, whereas African Americans were slightly more likely to have ever used the Web at school. As one might expect, increasing levels of income corresponded to an increased likelihood of owning a home computer, regardless of race. Although income explained race differences in computer ownership and Web use, education did not. That is, they found that whites were still more likely to own a home computer than were African Americans and to have used the Web recently, despite controlling for differences in education.

Their most striking findings, however, were for students. Hoffman and Novak (1998) found no differences among white and African American students when students had a home computer.However, among students without a computer in the home, white students were much more likely than African American students to have used the Web, and also more likely to have used the Web at locations other than home, work or school. They concluded that access translates into usage, and that whites are more likely than African Americans to use the Web because they are more likely to have access.

In 1998, the Commerce Department’s National Telecommunications and Information Administration (McConnaughey & Lader, 1998) analyzed data on computer penetration rates from the October 1997 Census Current Population Survey (CPS) as part of an ongoing examination of the digital divide.This analysis represented an update from their 1995 study of similar data from the November 1994 CPS.The authors concluded that the gap between the technology haves and have-nots had increased between 1994 and 1997, with African Americans and Hispanics actually farther behind whites in terms of home computer ownership and Internet access and with an even wider gap between individuals at upper and lower income levels.

More recently, Babb (1998) investigated home computer ownership and Internet use among low-income individuals and minorities. She found that African Americans and Hispanics were less likely to own computers, even after adjusting for income and education and termed this finding, consistent across seven different data sets under examination, the "single most important finding" of her study.

Interestingly, some have suggested that United States policy itself may be a contributing factor in the growing digital divide. Cooper and Kimmelman (1999) argue that the Telecommunications Act of 1996 has had the unintended and unfortunate consequence of increasing the division between the telecommunications haves and have-nots” As evidence, they point to 1) increased concentration and less competition in the telecommunications and cable industries, 2) significant increases or flat prices, instead of declines, in cable, long distance, and local phone rates, and 3) a growing disparity among those market segments employing heavy use of telecommunications networks like the Internet and those whose use is more modest.

The consequences to American society of this race gap in Internet use are expected to be severe (Beaupre & Brand-Williams, 1997). Just as A. J. Liebling observed for the freedom of the press (Liebling, 1960), the Internet may provide for equal economic opportunity and democratic communication, but only for those with access. The United States economy may also be at risk if a significant segment of our society, lacking equal access to the Internet, wants for the technological skills to keep American firms competitive.

Given these concerns, we set out to conduct a systematic investigation of the differences between whites and African Americans in the United States with respect to computer access, which is the primary current prerequisite for Internet access, and Web use. We wished to examine whether observed race differences in access and use can be accounted for by differences in income and education, how access impacts use, and when race matters in the calculus of equal access. The particular emphasis of this research is on how such differences may be changing over time. We believe our results may be used as a window through which policymakers might view the job of ensuring access to the Internet for the next generation.

To address these issues, we undertook a comparative analysis of Web usage and access across racial/ethnic groups in the United States, with a major focus on differences between whites and African-Americans at three different points in time. Our analysis is based on primary data from three population projectable, nationally representative surveys of Internet use among Americans, including the first survey on Internet use to collect data on race and ethnicity.

Because we have three waves of data collection spanning a period of eighteen months, this research permits reliable survey-based answers to the question of the magnitude of race differences over time. Previously, Hoffman and Novak (1998) established clear baseline measures that document racial differences in Internet access and use, and the extent to which racial differences may themselves depend upon specific demographic variables.Our objective in the current research is to examine these differences over time and determine to what extent the observed racial gaps in access and usage have evolved.We believe that demographic disparities in Internet access and usage have important implications for the growth of electronic commerce.Therefore, documenting the evolution of the digital divide is an important first step in understanding how differential access to and usage of the Internet impacts the course of its commercial growth.

This work is intended to stimulate discussion among scholars and policy makers interested in how differences in Internet access and use among different segments in our society affects their ability to participate and reap the rewards of that participation in the emerging digital economy. For that reason, we have attempted to present the results in a manner that allows the data to speak, as it were, for itself.

This paper is organized according to specific analysis objectives. In section two we begin by comparing the demographic composition of our three samples with U.S. Census data for a comparable time period. Overall, we find that the three CommerceNet/Nielsen IDS cross-sections are representative of the U.S. population, although results for some minority groups (Asian-Americans and Native Americans) are based upon sample sizes too small to permit projectability to the U.S. population. Next, in section three, we consider differences over time in Internet access and use among African-Americans, and whites in the U.S. over age 16, with respect to key demographic variables. This is followed, in section four, by a detailed analysis of racial differences adjusting for demographic variables, including student status, home computer ownership, education, income, gender, and the presence of children in the home. In section five we explore racial differences in online consumer behavior and business uses of the Internet. This set of comprehensive analyses is followed in section six by an evaluation of the implications of technology access for learning and knowledge. Finally, in section seven we summarize the major issues surrounding the social and economic uses of the Internet. This summary is presented as a series of discussion and policy points relevant to the development of an open research agenda concerning the socioeconomic impact of the Internet and electronic commerce in the United States and globally.

Sample Composition

The three surveys used in this research are the: 1) Spring 1997 CommerceNet/Nielsen Internet Demographic Study (IDS) conducted in December 1996 and January 1997; 2) Fall 1997 IDS, conducted in August and September 1997; and 3) Spring 1998 IDS, conducted in May and June 1998 (Nielsen Media Research 1997a; 1997b; 1998). Hereafter, these studies will be referred to as the IDS 2, IDS 3, and IDS 4, respectively.

The IDS 2, conducted in December 1996/January 1997, was the first nationally projectable survey of Internet usage to collect data on race and ethnicity. That permitted us, for the first time, to obtain baseline estimates of differences in Internet and Web use among racial and ethnic groups in the United States (see Hoffman & Novak, 1998). The IDS 3 and IDS 4 also included race and ethnicity, permitting changes to be tracked over time. Additionally, the survey instruments used in IDS 3 and 4 were identical, though slightly modified from IDS 2, thereby ensuring complete comparability across two of the three surveys, and reasonable comparability from IDS 2 to IDS 3 and IDS 4.

Each CommerceNet/ Nielsen IDS is based upon an unrestricted random digit dial sampling frame, and used a computer-assisted telephone interviewing system to obtain respondents. Eligible respondents were persons 16 years and older in the U.S. and Canada. In this paper, only the data from the United States respondents were used. When weighted, the respondents represent and allow projection to the total population of individuals in the United States aged 16 and over. Respondent weights in the IDS were adjusted by Nielsen so that marginal weighted distributions of Education, Gender, Race (African-American / non-African-American), Hispanic Origin, and Age were equivalent to Census estimates for the US and Canada. All results in this paper are based on weighted analyses in which the Nielsen IDS respondent weights were applied to the raw counts, producing population projectable estimates.

Table 1 compares the three CommerceNet/Nielsen Internet Demographic Studies, IDS 2, IDS 3, and IDS 4, conducted in December 1996/January 1997, August/September 1997, and May/June 1998, respectively, to March 1995 Census Current Population Survey (CPS) estimates for key demographic variables.The Nielsen IDS samples are representative of the US population. By comparing the weighted percentages from each Nielsen IDS with corresponding weighted percentages from the Census CPS, we find that the distributions reported in Table 1 differ only slightly. These slight differences are because 1) Nielsen used more recent CPS data, and 2) the Nielsen adjustment combined US and Canada, while our analysis deals only with the U.S.

Table 1. Demographic comparisons of the three CommerceNet/Nielsen Internet demographic studies with the 1995 Census CPS (US population aged 16 and older)

Table 2 presents the distributions of key demographic variables for each IDS by race and Hispanic origin, compared to the Census CPS. Note that because each Nielsen IDS is adjusted so that the marginal percentages of African-Americans and non-African-Americans corresponds to Census results, it does not necessarily follow that race-specific demographic distributions for other racial/ethnic groups (i.e., Asian-Americans, Hispanics, and Native Americans) from Nielsen IDS and Census CPS will be necessarily comparable. We see that:

The comparisons in Table 2 are important because they point out the challenges inherent in obtaining truly representative weighted samples. Additionally, these comparisons illustrate that readers must be careful when generalizing survey sample results to determine how the weighted sample compares to the population.

Schement (1997) presents a detailed discussion of the overlap between Census race categories of white, African-American, Native American, and Asian-American and the ethnic categories of Hispanic/non-Hispanic. We attempted to form mutually exclusive race/ethnic categories (e.g. Hispanic whites, non-Hispanic whites, Hispanic African-Americans, non-Hispanic African-Americans, etc.). But, due to discrepancies between the Nielsen IDS and Census CPS demographic distributions for Hispanics, we instead opted to treat race and Hispanic/non-Hispanic ethnicity as two separate and overlapping categorizations. This means that where the Hispanic category appears in the tables to follow, Hispanics may include individuals who also characterize themselves as white or African American or another race. Note that where sample sizes for a particular segment are too small for reliable estimation of percentages, those percentages have not been reported.

Table 2. Demographic distributions over time conditional on race and ethnicity of the three CommerceNet/Nielsen Internet Demographic Studies compared to the 1995 Census CPS (US Population Aged 16 and Older)

Demographic Differences in Web Access and Use Among African-Americans and Whites

Three usage segments are compared in this section: respondents who have 1) no Internet access, 2) Internet access only, but have never used the Web, and 3) ever used the Web. Table 3a shows results for whites, and Table 3b for African Americans. The sample sizes for Hispanics with were too low for reliable reporting. The major demographic differences between whites and African-Americans occur for age and income.

·The youngest (16-24) age segment differentiates usage segments more for African-Americans than for whites. African American Web users are more likely to be under 25, across all three surveys. Overall, Web users are more likely to be under 46 years of age.

·Overall, Web users are more likely to have some college or a college degree, although African American Web users are more likely to not have a high school education than white Web users. However, our African American samples overrepresent students and some of the individuals in this group are students.

Tables 3a and 3b may provide evidence of an education-driven digital divide. However, it is larger for African Americans than for whites, though it appears to narrow in IDS 4. Of African Americans with no Internet access, 70.27% were high school graduates or had less than a high school education in IDS 2, compared with only 62.11% of whites in the same educational categories with no Internet access. Of course, African Americans are more likely, in general, to have less education, so this result requires further study.

Similarly, of African American Web users, 56.34% had some college or college degrees in IDS 2, compared to 70.89% of white Web users. This gap also does not appear to be diminishing over time. However, to the extent that African Americans are less likely to have some college or a college degree, this result is tentative.

·Occupation categories were defined as follows.The professional, homemaker, full-time student, and retired/not currently working categories were self-identified single-response. The white-collar category included those respondents who indicated their primary occupation was either technical, administrative/managerial, clerical, sales, or service worker. The blue collar category included those respondents who indicated their primary occupation was laborer or craftsman/craftswoman. Due to its extremely low representation in the sample, respondents indicating they were in the military were excluded from analyses including the occupation variable. For this reason, the Occupation column totals may not necessarily sum to 100 percent. Web users are most likely to be professional, white collar, and students, regardless of race. The percentage of African American Web users who are blue-collar workers has more than tripled over time.

Table 3b. Demographic differences over time in Web access and use for African Americans

·The lowest income group (<$40K) differentiates usage segments much more for African-Americans than for whites. Of African-Americans with no Internet access, 80.72% had household incomes less than $40,000 in IDS 2, compared with only 60.9% of whites with no Internet access. Thus, within African-Americans, the income-driven digital divide appears larger than for whites. Further, it is not diminishing over time. Again, however, African Americans are more likely to have lower incomes than whites, so these results are tentative until further study. Similarly, of African American Web users, 59.73% had household incomes above $40,000 in IDS 2, compared with 71.01% of white Web users. This gap has also not diminished over time. This concern for an ever-widening gap within African-American income segments has been identified by sociologists as a serious one, which "will continue to grow as the black middle class moves forward and poor black Americans stagnate" (Beaupre& Brand-Williams, 1997).

Katz & Aspden (1997) reported evidence of what Lloyd Morrisett of the Markle Foundation has termed a digital divide, with Internet users being generally wealthier and more highly educated. Sparrow and Vedantham (1995) summarize the broader information technology situation as follows:

In summary, Tables 3a and 3b show evidence of a digital divide for both whites and African-Americans. Within both racial groups, Web users were most likely to be among the wealthiest individuals (those with incomes above the median of $40,000), while the segment with no Internet access was the most likely to be composed of individuals with the lowest incomes (less than $40,000). The same holds true for Education. The Web user segment was most likely to consist of individuals with some college or who had completed college, while the segment with no access was most likely to be composed of those with a high school education or less. All these effects were more pronounced for African Americans than whites and these effects appear to persist over time.

Racial Differences in Web Access and Use Over Time

In this section, we analyze differences among whites and African Americans in Web use adjusting for key demographic variables. Each demographic analysis will first examine differences in Web use among all respondents. Then, we examine detailed Web use behavior for recent Web users only. Below we provide statistics on recent Web use for each IDS to facilitate these comparisons.

Table 4 presents a series of comparisons among whites and African-Americans on key indicators of Web access and use.

In IDS 2, overall, whites were more likely than African-Americans to have access to the Internet, and to have ever used the Web. Whites were also more likely to own a computer, have PC access at work, and have fax, cable, and a satellite dish at home.

As Table 4 shows, the percentages of access and use for both whites and African Americans have increased over time, but the gaps persist. In fact, the overall gap between whites and African Americans in Internet access and having ever used the Internet have actually increasedover time. In IDS 2, 35.8% of whites had Internet access, compared to 31.68% of African Americans. By IDS 4, although 49.33% of whites had access, the percentage of African Americans with Internet access had risen only a few percentage points, to 35.54%. Similarly, in IDS 2, 24.34% of whites had ever used the Internet, compared to 18.76% of African Americans. In IDS 4, eighteen months later 40.37% of whites had ever used the Internet, compared to 27.98% of African Americans.

Although whites are still more likely to own a PC and to have PC access at work, these gaps have not increased over time. Further, the gaps in cable and satellite ownership have disappeared. In fact, satellite penetration has doubled among African Americans and the penetration rate in IDS 4 equals that for whites.

Our estimate of 29% of African-American households with access to a personal computer in IDS 2, 3, and 4 compares with estimates provided by Simmons Market Research Bureau (Interactive Marketing News, 1997), which reported that 23% of African-Americans owned a personal computer. The gap between whites and African-Americans in computer ownership has been cited as the key explanation for corresponding gaps in Web usage.A Yankelovich Monitor study (Interactive Daily, 1997) "suggests that what bars entry to cyberspace among African Americans is owning a home PC, not lack of interest in the Internet"” However, a Forrester Research study (Walsh, 1999) cites "technology optimism" as an important predictor of technology adoption. Further research is required to understand these increasing gaps in access and usage.

A number of reasons have been provided in the popular press for the gap between whites and African-Americans in computer ownership. Price and value are often cited as explanations. For example, Malcolm CasSelle, co-founder of NetNoir, stated, "African-Americans just don’t perceive the value of the Internet. Many blacks would pay $500 for a TV, and you could get a computer, though maybe not a top-of the line one, for not much more than that" (Holmes, 1997). Similarly, Larry Irving, assistant secretary of Commerce, noted that WebTV is in the under-$500 price range, and "laptop and PC prices are coming down. As that continues to happen, the Internet will become more prevalent in the African-American community"” (Holmes, 1997).

However, our analysis suggests that PC penetration rates are not increasing overall among African Americans. Although the percentage of whites owning a home computer has increased slightly over time, approaching fifty percent, the overall percentage of African Americans who own a home computer has remained at 29%. Later we will investigate home computer ownership among different usage segments.

Overall, then, the gap in access and use is persistent and appears to be increasing. However, a different picture emerges when we examine recent Web users. In IDS 2, whites were more frequent and more recent Web users than African Americans. White recent Web users in IDS 2 were also more likely to have ever used the Web from home and work, while recent African American Web users were more likely to have ever used the Web from school. African Americans in IDS 2 were much more likely to be newer users, and whites were much more likely to have been using the Web for two years or more.

But over time, is apparent that African American Web users have made significant gains in Web use. Indeed, the gaps are diminishing rapidly. The gap in Web use at home has decreased dramatically (even as the overall PC penetration rate among African Americans stagnates) and African Americans now appear to be more likely than whites to have ever used the Web from work, school or other locations. Additionally, African American Web users are becoming more recent and more frequent Web users. In fact, differences between white and African American Web users in their recency and frequency of Web use have disappeared. African Americans are still among the newest users, but are now also joining the ranks of the long-term users. By IDS 4, 43.98 percent of African American recent Web users, compared to 49.47 percent of white recent Web users, had been using the Web for two years or more. This suggests that one cause of the digital divide arises from the differential lack of access.

The differences on most indicators of Internet access and use, and also home computer ownership between whites and African-Americans is greater for students (Table 5b) than non-students (Table 5a).We examine non-students first.

Non-Students.As Table 5a reveals, white non-students in IDS 2 were more likely to have access to the Web, to have ever used the Web, to own a PC, and have PC access at work. Additionally, white non-students were also more likely than African Americans to have a home fax, cable, and a satellite dish.

Over time, and similar to the previous analysis, overall differences between white and African American non-students actually appear to be increasing. Note that the percentage of white non-students who own a home computer is increasing slightly, but that the percentage of African American non-students who own a home computer is actually decreasing over time. At the same time, the percentages of African American non-students who have cable or a satellite dish have increased and are now similar to the penetration rates for whites.

Among non-student recent Web users, we see that over time African Americans are making steady gains in Web use.Since IDS 2, African Americans are increasingly likely to have been using the Web for two years or more, more likely to have ever used the Web at home, work, school, or other locations, and more likely to be more recent and frequent users of the Web.

However, white non-student recent Web users are making similar gains, and in some cases, their use still outpaces that of African Americans. For example, whites are still more likely to be among the most recent Web users. In IDS 4, 39.96 percent of whites reported using the Web in the past 24 hours, compared to 30.55 percent of African Americans.

On the other hand, differences in Web use frequency (a few times a week or more) between white and African American non-student recent Web users have disappeared. Additionally, African American non-students are more likely to have used the Web from work, school or other locations. Because African American non-students are also more likely than whites to be the newest users, we may expect increasing gains in Web use, compared to white non-students, over time.

These results point out the importance of home computer ownership and multiple access points for African Americans and other minority groups. African American non-students are significantly less likely to own a home computer than white non-students (and in fact, even less likely over time), and consequently we believe, exhibit lower overall rates of access and usage.

Yet, among recent Web users, who by definition have access somewhere, recency and frequency usage rates for African American respondents has over time come to mirror that of whites.

Students. Table 5b shows clearly that, overall, Web access, usage and PC penetration rates are considerably higher for students than non-students, regardless of race. Additionally, Web access, usage and PC penetration rates are all rising over time, for both white and African American students.

In fact, 53.83 percent of African American students owned a home computer in IDS 4, compared with only 25.67 percent of African American non-students.Eighteen months earlier, in IDS 2, the PC penetration rate was considerably lower (and similar to non-students): only 31.88 percent of African American students, compared with 28.66 percent of African American non-students, owned a home PC.

At the same time, Table 5b reveals clear disparities in Internet access and use between African American and white students. For example, though both groups have higher rates over time, white students are more likely to have Web access, and to have ever used the Web, and these differences persist over time.

What might explain the persistent gap in Internet and Web access and use between African-American and white students? White students are more likely to own home computers than African American students and this difference remains over time, despite the fact that home computer ownership over time among white students is flat and home computer ownership among African American students has been steadily and impressively rising.

However, the gaps in access and usage are clearly decreasing over time and this may in part be due to increasing rates of PC ownership among African American students. Additionally, African American students were more likely to have access to a PC at work in IDS 4. Thus, one could hypothesize that as PC ownership and PC access rates continue to rise, so will overall Web access and usage rates. Eventually, we would expect the access and usage gaps between white and African American students to disappear.

We now turn to an analysis of recent student Web users, shown in the lower panel of Table 5b. Sample sizes are very small for African American students who are recent Web users, so results for this sub-segment must be interpreted with extreme caution. Nevertheless, we include them for comparability with non-student recent Web users in Table 5a.

African American recent Web user students, like their white counterparts, are increasingly likely to be more recent and more frequent Web users. In fact, the differences between the two groups, as for non-students, have effectively disappeared. However, white students are still more likely to have ever used the Web from home, while African American students are more likely to have ever used the Web from work and school.

In this section, we examine the impact of home computer ownership on Web access and usage. Because there are such dramatic differences between students and non-students, we treat student status as a separate variable in this analysis. We first analyze non-students and then turn to students.

Non-Students.Table 6a compares Web access and usage patterns for white and African American non-students who do not own a home computer. Table 6b shows the same comparison for white and African American non-students who do own a home computer. First, it is quite clear that the presence of a computer in the home has a dramatic impact overall on access and usage. Non-student respondents, regardless of race, are much more likely to have access to the Web and to have ever used the Web if they own a home computer. Additionally, among non-students with a home computer, whites are more likely to have access and to have ever used the Web, but the differences are small.

Yet, there are striking gaps in access and usage among non-students without a home computer.In IDS 4, whites were more likely than African Americans to have access to and have ever used the Web. Further, the gap in use among non-students without a home computer is increasing over time.

Turning to an examination of recent Web users, we first examine those non-students who do not own a home computer. Sample sizes for African Americans in this sub-segment are very small, so results must be interpreted with extreme caution. Here we notice that whites and African Americans have similar rates of usage frequency and recency, suggesting once again, that given access, usage follows for both groups. African Americans without a home computer are more likely than whites to have ever used the Web at school or other locations.

Among non-students recent Web users with a home computer, both recency and frequency of Web use increase over time for both whites and African Americans, and differences are diminishing rapidly. By IDS 4, 73.85 percent of white non-students with a home computer had last used the Web within the past week, compared to 67.58% of African Americans and 64.62% of whites had used the Web a few times a week or more, compared to 59.05% of African Americans.

Both groups enjoy similarly high percentages of having ever used the Web at home; 80.55% of whites and 82.12% of African Americans in IDS 4 had ever used the Web at home. In IDS 4, African American non-students were more likely than their white counterparts to have ever used the Web at work, at school and at other locations.

Table 6a. Race differences in Web access and use by student status and home computer ownership

Table 6b. Race differences in Web access and use by student status and home computer ownership

Students. Tables 6c and 6d below show the analysis for students with and without a computer in the home.

Sample sizes are much smaller for African-American students when segmented by whether they have a computer at home, so the results must be interpreted with caution. Compared to Tables 6a and 6b, Tables 6c and 6d show that students are more likely to have access to the Web and to use the Web, compared to non-students.

Regardless of race, non-students with a home computer have access and usage rates that are similar to students without a home computer. Students with a home computer enjoy the highest levels of access and use, while non-students without a home computer have the lowest levels of access and use.

Additionally, having a computer in the home leads to much higher levels of Web access and use for students of both races, but the difference is clearly more dramatic for non-students than students. That is, the differences in access and usage for non-students with and without a home computer are larger than the differences for students with and without a home computer. Presumably this is because students have more opportunities for access at school, even if they do not own a computer at home.

However, without a computer in the home, there is a much larger gap between African-American and white students in terms of Web access, and the corresponding percentages of having ever used the Web are also smaller. In fact, the gap in access for students without a computer in the home appears to be increasing.

It also appears that the presence of a computer in the home is bringing African American and White students to parity in Internet access and Web use.

Increasing the Internet access opportunities for students, especially African-American students without home computers, may help to reduce the gaps in access and usage.

Table 6c. Race differences in Web access and use by student status and home computer ownership

Table 6d. Race differences in Web access and use by student status and home computer ownership

Tables 7a-7d show the influence of education on Web access and usage between whites and African Americans. In general, increasing levels of education lead to higher levels of Web access, usage, PC ownership and PC access at work. However, these levels are higher for whites than for African Americans and these race differences persist even after adjusting for education. In fact, the gaps in access and usage are largest for those with a college degree.

Less Than High School and High School Graduates. At the lowest levels of education, whites are more likely than African Americans to have access to the Web, to have ever used the Web, and to own a computer. What is more, these differences persist over time.

Among those with less than a high school education, African Americans are gaining ground faster than whites in Web access, use and PC ownership over time. Overall, the gaps are smaller here. Whites are more likely to have access to a PC at work, in IDS 2 and IDS 3. By IDS 4, there is no difference between whites and African Americans with less than a high school education on this variable. However, students are included in this group and that could account for some of these gains.

Among high school graduates, levels of computer ownership among both groups are stagnant, and whites are more likely than African Americans to own a PC. African American high school graduates are more likely to have access to a PC at work than are white high school graduates. Additionally, the gap in access and use between whites and African Americans is larger for high school graduates than for those without a high school degree.

Table 7a. Race differences over time on key Web usage variables for respondents with less than high school education

Some College. Table 7c shows the results for respondents with some college, but no degree. In IDS 4, whites were more likely to have Web access, to have ever used the Web, and to own a home computer. Except for access, the gaps are diminishing over time. African Americans are more likely to have PC access at work.

Larger sample sizes permitted us to examine Web usage patterns among recent Web users with some college. In IDS 2, twice as many African Americans as whites with some college education reported using the Web for the first time (50.62 percent of African Americans compared to 25.25 percent of whites). More than twice as many whites as African Americans reported using the Web for two years or more (14.23 percent whites compared to 5.97 percent African Americans). Thus, in IDS 2, whites were more likely to be long-term users and African Americans were more likely to be newer users.

By IDS 4, similar percentages of African Americans and whites have been using the Web for two years or more; 43.81 percent of whites and 46.59 percent of African Americans had been using the Web for two years or more.

In IDS 2, whites were more both recent and frequent users of the Web, compared to African Americans. Over time, however, these differences have effectively vanished, with 65.82 percent of whites and 66.87 percent of African Americans in IDS 4 reporting using the Web within the past week. Similarly, 29.13 percent of whites and 26.30 percent of African Americans with some college used the Web once a day or more in IDS 4, and African Americans with some college are actually more likely than whites to have used the Web a few times a week.

In IDS 2, whites with some college were more likely to have ever used the Web at home and work, and African Americans were more likely to have ever used the Web at school. By IDS 4, African Americans were more likely than whites to have ever used the Web at work, school or other locations. The percentage of African Americans who have ever used the Web at home has risen dramatically, going from 47.33 percent in IDS 2 to 76.52 percent by IDS 4

Table 7c. Race differences over time on key Web usage variables for respondents with some college

College Graduates.Table 7d examines differences in access and use for college graduates.In IDS 2, whites had higher levels of access, usage, and home computer ownership than African Americans. In IDS 3, it appeared that these differences had disappeared. But then in IDS 4, African American rates dropped, leading to dramatic differences and larger gaps. Thus, it appears that education does not account for the digital divide in access and usage and home PC ownership.

Interestingly, both whites and African American college graduates have high levels of PC access at work and there is no gap on this variable. Outside of this finding, it is an interesting question to ask why there is such a large gap in access and usage between educated African Americans and whites. Further examination of college educated African Americans in IDS 4 is warranted.

Among recent Web users who are college graduates, we find other interesting results. Whites and African Americans report similar levels of recency and frequency of use, as with those with some college. Whites and African American college graduates are also equally likely to have ever used the Web from home, work, or school. However, there is a small tendency for African Americans college graduates to be more likely to have ever used the Web from other locations, and this difference has persisted over time.

Table 7d. Race differences over time on key Web usage variables for college graduates

Tables 8a and 8b show the relationship between income and Web access and usage for whites and African Americans. Not surprisingly, respondents whose household income is above the median income of $40,000 report higher levels of access, use, home computer ownership and PC access at work. Below we examine race differences within income levels.

Household Income Less Than $40,000. At household incomes below $40,000, whites are more likely than African Americans to have access to the Web, to have ever used the Web, and to own a home computer. The gaps in access and use may actually be increasing, although the gap in home PC ownership appears to be improving slightly. There is no difference between whites and African Americans in PC access at work at this income level.

Table 8a. Race differences over time on key Web usage variables for respondents with household income below $40,000

Household Income of $40,000 and Above. Above the median household income of $40,000, differences between whites and African Americans in access, usage, PC ownership and PC access at work are greatly diminished compared to respondents with less household income.

We also examined recent Web users at this income category. In IDS 2, there were significant differences in the length of time that whites and African Americans had been online. We found that 170.8 percent of whites had used the Internet for two years or more, while only 9.08 percent of African Americans had. In contrast, African Americans were almost twice as likely as whites to be new users in IDS 2. By IDS 4, these gaps had disappeared with over fifty percent of both African Americans and whites to have been Web users for two years or more.

In IDS 2, 59.48 percent of whites, compared to 44.21 percent of African Americans were the most recent users. By IDS 4, this gap had largely disappeared, with 73.78 percent of whites and 75.22 percent of African Americans with incomes of $40,000 and over reporting they last used the Web within the past week.

A similar result was found for frequency of use. In IDS 2, 49.69 percent of whites, compared to 30.94 percent of African Americans, used the Internet a few times a week or more. By IDS 4, 62.86 percent of whites and 59.08 percent of African Americans had.

Table 8b. Race differences over time on key Web usage variables for respondents with household income $40,000 and over

Tables 9a and 9b report the relationship between gender and Web access and use for whites and African Americans. It is clear that overall levels of Web access and use are lower for women than men. Among recent Web users, men are more likely to have been using the Web longer, and to have used the Web more recently. Women are more likely to be newer users. While white men are more frequent users than white women, African American men are not more frequent Web users than African American women. Below we analyze differences by race for each gender.

Men. As Table 9a shows, white men are more likely to have access to the Web, to have ever used the Web, and to own a PC at home than African American men and these differences have persisted over time. Interestingly, the percentage of respondents who report owning a PC at home has grown for white men, but not African American men. There is a small tendency for white men to be more likely to have PC access at work.

Among recent Web users, African American men are much more likely than white men to be newer Web users and this finding persists over time. In IDS 2, white men were more than three times as likely as African American men to have been online for two years or more (20.31 percent versus 6.20 percent). By IDS 4, over fifty percent of white and African American men had been online that long.

In IDS 2, white men were more likely than African American men to be the most recent Web users (65.46 percent compared to 37.04 percent). By IDS 4, both white and African American men were equally and highly likely to have used the Internet recently (75.03 percent compared to 71.37 percent, respectively).

White men were also more frequent Web users than African American men in IDS 2 (55.94 percent versus 38.9 percent). White men were still more likely to be more frequent Web users than African American men by IDS 4, though the difference had shrunk considerably (67.5 percent versus 54.65 percent).

Over time, the percentage of African American men ever using the Web at home, work, school or other locations has increased considerably. In contrast, the percentage of white men who ever used the Web at school is flat, has grown only modestly for home and work, and has fallen for ever used at other locations. By IDS 4, African American men were more likely to have ever used the Web at school or at other locations, compared to white men.

Women. White women are more likely than African American women to have access to the Web, to have ever used the Web, to own a PC and to have PC access at work and these differences have persisted over time. As for men, the percentage of women owning PCs at home has increased over time for white women, but not African American women.

In IDS 2, African American women were much more likely than white women to be new users, but over time this difference has vanished. White women were more likely in IDS 2 than African American women to have used the Internet most recently (45.32 percent versus 31.75 percent). By IDS 4, 60.74 percent of white women, compared to 56.79 percent of African American women, had used the Internet within the past week. In contrast to previous results, African American women were more likely than white women to be the most frequent Internet users (44.38 percent versus 33.36 percent in IDS 2). By IDS 4, this gap had diminished with 49.02 percent of African American women, compared to 46.53 percent of white women having used the Web a few times a week or more. African American women were more likely to have ever used the Web

Tables 10a and 10b compare Web use in the presence of children in the home for whites and African Americans. Not surprisingly, overall levels of access and usage are higher for respondents who report having children under seventeen in the household. Below we examine Web access and usage by race.

No Children Under Seventeen. In IDS 4, whites without children at home were more likely than African Americans without children at home to have access to the Web, to have ever used the Web, to own a PC, and to have PC access at work. Some of these gaps were persistent.

Among recent Web users, over fifty percent of both groups had been using the Web for two years or more by IDS 4, although African Americans without children at home were much more likely than whites to be newer users, especially in IDS 2 and IDS 4.

By IDS 4, there were few differences in recency and frequency of Web use between whites and African Americans. African Americans, were more likely to have ever used the Web at school and other locations.

Whites were more likely to have access to the Web, to have ever used the Web, to own a PC, and slightly more likely to have PC access at work. The percentage of respondents owning a home PC, though higher for whites than African Americans, has remained constant over time.

Among recent Web users, whites with children at home were more likely to be long-term Web users and to be more recent users, but not more frequent users, compared to African Americans with children at home.

Whites were also more likely to have ever used the Web at home, though this has diminished over time. African Americans were more likely to have ever used the Web at school and other locations.

Table 10a. Race differences over time on key Web usage variables for respondents with no children under 17 in household

Table 10b. Race differences over time on key Web usage variables for respondents with children under 17 in household

Differences in Commercial Web Usage for African-American and White Web Users

In Table 11, we briefly consider two key Web shopping items. Note that the items used to measure these shopping behaviors are not comparable across the three surveys, because their definition changed after IDS 2. This means we can compare percentages within IDS 2, and then from IDS 3 to IDS 4. But it is not possible to draw a comparison from IDS 2 to IDS 3 or from IDS 2 to IDS 4.

Note that overall, the rate of searching for information on the Web and purchasing products on the Web is increasing over time.

In IDS 2, whites were more likely to use the Web to purchase a product or service online. In IDS 3 and IDS 4, whites were also more likely to purchase online, but the gap is closing.

However, in IDS 2, both whites and African Americans were equally likely to search for product information online. Yet in IDS 3 and IDS 4, whites are more likely than African Americans to have searched the Web for product information. This result is interesting because Novak, Hoffman and Yung (1999) find that there is more skill involved in buying products online than searching for information online. Thus, these differences may not be due to different skill sets of whites versus African Americans. Potential explanations of the differences in search behavior are that African-Americans are less interested in comparing products and services online, are obtaining this information from other sources, or are simply finding less of interest to search for online.

Katz and Aspden (1997) found that Internet users ranked business opportunities as a relatively unimportant reason for current Internet users, but a very important reason for non-users. This was phrased, however, in terms of"making money"” In Table 12, we examine business uses of the Web among recent Web users with access to a PC at work. As with Web shopping variables, items in IDS 2 are not comparable to items in IDS 3 and IDS 4, so no direct comparisons may be made from IDS 2 to IDS 3 or IDS 2 to IDS 4.

In IDS 2, whites were more likely than African Americans to use the Web to research competitors, publish product information, sell products and services, for customer service and support and for vendor support and communication.

In IDS 3, a number of these race differences in business uses of the Web have gone away and some new differences emerged. For example, whites were more likely than African Americans to use the Web for internal communication, and still more likely to use the Web for customer service support and vendor support and communication.

By IDS 4, African Americans were, for the first time, more likely than whites to have access to an Intranet at work, to use the Web for internal communication, and for customer service and support. However, whites were more likely to use the Web to buy products or services. It is likely that these differences may reflect different job functions.

Implications of the Digital Divide for Knowledge

Does the digital divide affect media consumption patterns for different segments in our society? The question is relevant because it has implications for individual knowledge and learning. One way of controlling the flow of information to different segments of society is through access (Tichenor et al., 1980; Donohue et al., 1995). However, even when information is accessible to everyone, the relative amount of knowledge gained between the haves and have-nots can widen (Tichenor et al., 1970, 1980). A variety of factors contribute to the widening of this knowledge gap. Differences across segments in their education (Tichenor et al., 1970), interest in the topics covered (Chew & Palmer, 1994; Gaziano, 1983; Robinson, 1972; Tichenor et al., 1980), and interpersonal contact with others familiar with topics covered in the media (Chaffee, 1972; Dervin & Greenberg, 1972; Tichenor et al., 1980) all affect an individual’s motivation to seek out and consume information. For instance, if the media covers topics of personal relevance to the "have-nots," then they are motivated to actively consume this information (Tichenor et al., 1980). In this instance, access narrows the knowledge gap. However, if the media covers topics that are remote to the have-nots" (which is often the case with mass media, see Donohue et al., 1995), then they only passively consume the information and the gap widens (Reagan, 1996; Tichenor et al., 1980).

A related motivating factor for consuming content regarding a topic gaining mass media exposure is whether this topic will be discussed in the individual's social circles (Tichenor et al., 1980). If those in one's social network are unlikely to mention the topic, the individual, even if she or he processes the information, has a lower probability of being able to subsequently digest, debate, ponder and ultimately remember this information than someone who is already an expert on the topic. In contrast, those who are already experts on a topic are likely to be consulted and questioned by others. The expert thus has motivation to both process the information as well as an opportunity to subsequently analyze this information in greater depth than the nonexpert.

Such reinforcing factors contributing to the knowledge gap may also be contributing to the racial divide on the Internet, in terms of both adoption and usage.For instance, mass media coverage of the Internet is likely of greater interest and relevancy to those who are already on the Internet and/or have family and friends who are online than those who have not adopted. If we start with a base rate of fewer minorities than whites using the Internet, then an information gap may exist (and potentially widen) between the races in terms of how to access, use and benefit from the Internet as mass media coverage of the Internet increases.

There may also be fewer cross-channel references across media for those in the minority, especially for those in lower socioeconomic segments. For instance, schools, churches, local retailers or other community services in urban poor neighborhoods may not have Web sites or use the Internet for communication. Consequently, Internet content may seem remote or irrelevant to the personal lives of the majority of urban poor. In fact, in a 1970s study on the urban poor's exposure to mass media, it was found that poor African-Americans were more likely to talk interpersonally or use "nonestablished" channels for local news than were poor whites (Dervin & Greenberg, 1972). As with "established" mass media, minorities may not perceive the Internet to be a useful substitute or supplement to their current sources.

Indeed, from a uses and gratification perspective (cf., Finn, 1997; Robinson, 1972), the racial divide may in part be due to differences in the perceived benefits of the Internet relative to other media currently adopted or activities currently engaged in. Research conducted in 1994 suggests that, at least at that time, home connectivity to the Internet was driven by entertainment needs as well as to pass time (Perse & Dunn, 1998). Those who have not adopted the Internet may feel that other media sources such as television already sufficiently fulfill these needs. Interestingly, using media as a pastime is said to occur when there is no functional relevancy of the media for the user (Reagan, 1996). Perhaps adopters in 1994 were using the Internet not because they had any particular need to fulfill but rather in order to determine which future needs the Internet could fulfill. Such a luxury (in terms of time and money) may only be affordable to those in the upper socioeconomic strata.

Usage of different media also depends upon people's interests as well as the topic area of interest (Reagan, 1996). The greater interest an individual has in a topic, the more media sources she or he is likely to consult. For instance, an avid NBA fan is likely to consult cable as well as local television programs, radio, print and perhaps even the Internet to find additional information about the NBA. Furthermore, certain topics are more likely to increase exposure to specific media. For instance, interest in financial news corresponds with increased exposure to print media (Reagan, 1996). It is unclear which topics would coincide with increased usage of the Internet, and whether these relations would vary across demographic differences.

Researchers examining the increased specificity of cable programming and print materials to target special interest groups forecast that such specialization of media content may only widen knowledge gaps (Tichenor et al., 1980). The Internet is largely organized around special interest groups and topics.Consequently, even if the Internet is accessible to all segments of society, the knowledge gap may continue to exist in specialized topic areas.At the same time, Internet communities break down status, time and geographical barriers, such that discussion of special interest topics are no longer limited to one's immediate geographical surroundings.Thus, knowledge gaps in the future may occur between those with different interests rather than across demographic segments.

Developing a Research Agenda

1) Computers in the home. While previous research has shown that inequalities in Internet access in schools persist (Educational Testing Service 1997, Sax, et. al. 1998), our results suggest that inequalities in Internet access at home may be even more problematic. The role of access to the Internet at home needs to be much more clearly understood (Abrams 1997). Whites are more likely to have access to the Internet and to have ever used the Web than African Americans and these gaps appear to be increasing over time. Our results are consistent with other recent research (Babb, 1998; Cooper & Kimmelman, 1999; McConnaughey & Lader, 1998) that has explored the digital divide. However, we have probed more deeply and discovered that among recent Web users, who by definition have access, the gaps in Web use have been decreasing over time. By IDS 4, in most cases there were no or only slight differences between whites and African Americans in how recently they had used the Web, how frequently, or in their length of time online.

Gaps in general Web access and use between African-Americans and whites appear to be driven by whether or not there is a computer present in the home. Access to a personal computer, whether at home, work, school or somewhere else, is important because it is currently the dominant mechanism by which individuals can access the Internet. We have shown that access translates into usage. Overall, individuals who own a home computer are much more likely than others to use the Web. This suggests that programs that encourage home computer ownership (see, for example, Roberts 1997) and the adoption of inexpensive devices that enable Internet access over the television should be aggressively pursued, especially for African Americans.

Morrisette (1999) forecasts that by the year 2003, over half of all households in the United States will have access to the Internet, but that PC penetration could stall at 60 percent of households. Research is necessary to understand what motivates individual-level adoption of home computers and related technologies, as well Internet adoption, both within and outside the home. Additionally, research is required to understand the long-term impact of home computer ownership on Internet access and use.

Footnotes

References

Atkin, C. K., Greenberg, B. S., & McDermott, S. (1983). Television and race role socialization. Journalism Quarterly, 60, 407-414.

Anderson, T. E., & Melchior, A. (1995). Assessing telecommunications technology as a tool for urban community building. Journal of Urban Technology, 3 (1), 29-44.

Babb, S. F. (1998). The Internet as a tool for creating economic opportunity for individuals and families. Unpublished doctoral dissertation, University of California, Los Angeles.

Beaupre, B., & Oralandar, B. (1997, February 8). Sociologists predict chasm between black middle-class, poor will grow. The Detroit News, pp.

Castaneda, L. (1997, April7). African American Financial Index Available on Web. The San Francisco Chronicle, pp.

Chaffee, S. H. (1972). The interpersonal context of mass communication. In F.G. Kline & P.J. Tichenor (Eds.), Current Perspectives in Mass Communication Research (pp. 95-120). Bevery Hills: Sage Publications.

Chew, F., & Sushma, P. (1994). Interest, the knowledge gap, and television programming. Journal of Broadcasting & Electronic Media, 38, 271-287.

Clinton, W. J. (1997b, April 2). Remarks by the President at Education Announcement/Roundtable. The East Room, The White House, Office of the Press Secretary. Available: http://www.iitf.nist.gov/documents/press/040297.htm.

Coley, R. J., Cradler, J., & Engel, P. K. (1997). Computers and classrooms: The status of technology in U.S. schools (ETS Policy Information Report). Princeton, NJ: Educational Testing Service. Available: http://www.ets.org/research/pic/compclass.html

Cooper, M., & G. Kimmelman (1999, February). The digital divide confronts the Telecommunications Act of 1996: Economic reality versus public policy. Yonkers, NY: Consumers Union. Available: http://www.consunion.org/other/telecom4-0299.htm

Donohue, G. A., Tichenor, P. J., & Olien, C.N. (1995). A guard dog perspective on the role of media. Journal of Communication, 45, 115-132.

Dervin, B., &. Greenberg, B. S. (1972). The communication environment of the urban poor. In F.G.Kline & P.J. Tichenor (Eds.), Current perspectives in mass communication research (pp. 195-233). Bevery Hills: Sage Publications.

Educational Testing Service (1997). Computers and classrooms: The status of technology in U.S. schools, Policy Information Center. Available: http://www.ets.org/research/pic/compclass.html

Finn, S. (1997). Origins of media exposure: Linking personality traits to TV, radio, print and film use. Communication Research, 24, 507-529.

Gaziano, C. (1983). The knowledge gap: An analytical review of media effects. Communication Research, 10, 447-486.

Harmon, A. (1997, October 25). Net day volunteers back to wire schools for Internet. New York Times

Hoffman, D. L., Novak, T. P., & Chatterjee, P. (1995). Commercial scenariosfor the Web: Opportunities and challenges. Journal of Computer-MediatedCommunication [On-line], 1 (3). Available: http://www.ascusc.org/jcmc/vol1/issue3/hoffman.html

Holmes, T. E. (1997, February 20). Seeing a future with more blacks exploring the Internet. USA Today.

Interactive Daily (1997, February 18). More african-americans plan to go online.

Internet Marketing News (1997, February 28). Web marketers beginning to focus on minority audience, 4 (9), .

Katz, J., & Aspden, P. (1997, October). Motivations for and barriersto Internet usage: results of a national public opinion survey. Paper presented at the 24^th Annual Telecommunications Policy Research Conference, Solomons, Maryland.

Keller, J. (1996). Public access issues: An introduction. In B. Kahin & J. Keller (Eds.), Public Access to the Internet (pp.). Boston, MA: The MIT Press.

Lawrence, S., & Giles, C. L. (1999). Accessibility of information on the Web. Nature, 400, 107-109.

Maraganore, N., & Morrisette, S. (1998, December 2). The on-line gender gap is closing. Data Insights (Issue No. 1, 18). Forrester Research Reports.

McConnaughey, J. W., & Lader, W. (1998, July 28). Falling through the Net II: New data on the digital divide. National Telecommunications and Information Administration: United States Department of Commerce. Available: http://www.ntia.doc.gov/ntiahome/net2/falling.html

Morrisette, S. (1999, January). Consumer’s digital decade. Forrester Report, Forrester Research, Inc. Available: http://www.forrester.com/

New Media Week (1997, March 3). BET, Microsoft Sees Potential in African-American Audience.

Nielsen Media Research (1997a). The Spring ’97 CommerceNet/Nielsen Media Internet Demographic Survey. City, State: Author.

Nielsen Media Research (1997b). The Fall ’97 CommerceNet/Nielsen Media Internet Demographic Survey. City, State: Author.

Nielsen Media Research (1998). The Spring ’98 CommerceNet/Nielsen Media Internet Demographic Survey. City, State: Author.

Novak, T. P., Hoffman, D. L., & Yung, Y. F. (2000). Modeling the Flow Construct in Online Environments: A Structural Modeling Approach. Manuscript submitted for publication.

Perse, E. M. & Dunn, D. G. (1998). The utility of home computers and media use: Implications of multimedia and connectivity. Journal of Broadcasting & Electronic Media, 42, 435-456.

Pew Research Center (1998).Online newcomers more middle-brow, less work-oriented: The Internet news audience goes ordinary. The Pew Research Center for the People and the Press. Available: http://www.people-press.og/tech98sum.htm

Reagan, J. (1996). The 'repertoire' of information sources. Journal of Broadcasting & Electronic Media, 40, 112-122.

Roberts, R. M. (1997, June 19). Program lowers costs of going online; families can get break on equipment. The Atlanta Journal and Constitution.

Robinson, J. P. (1972). Mass communication and information diffusion. In F.G. Kline & P.J. Tichenor (Eds.), Current Perspectives in Mass Communication Research (pp. 71-93). Bevery Hills: Sage Publications.

Rutkowski, A. M. (1998). Internet trends, February. Washington, D.C.: Center for Next Generation Internet. Available: http://www.ngi.org/trends.htm

Schement, J. R. (1997, October). Thorough Americans: Minorities and the new media. Paper presented at the Aspen Institute Forum, October 1996.

Schools and Libraries Corporation (1998, November 23). First wave of e-rate funding commitment letters sent. News release.

Schwartz, E. (1996). NetActivism: How citizens use the Internet. Sebastopol, CA: O’Reilly & Associates, Inc.

Sheppard, N. (1997, April 30). Free-nets reach out to communities’ needs. The Ethnic NewsWatch, .

Skriloff, L. (1997, February 17). Out of the box: A diverse netizenry. Brandweek .

Tichenor, P. J., Donohue, G. A., & Olien, C. N (1980). Community conflict and the press. Bevery Hills: Sage Publications.

Tichenor, P. J., Olien, C. N. & Donohue, G. A. (1970). Mass media flow and differential growth in knowledge. Public Opinion Quarterly, 34, 197-209.

Walsh, E. O. (1999, March 3). The digital melting pot. Forrester Research, Inc.

About the Authors

Thomas P. Novak is Associate Professor of Management at the Owen Graduate School of Management, Vanderbilt University, and co-founder and co-director of eLab. He received his Ph.D. in 1984 from the Psychometric Laboratory, University of North Carolina at Chapel Hill. His research interests include Internet and Web-based commerce, modeling consumer online navigation behavior, electronic commerce policy, and consumer behavior in online environments. His research has appeared in Communications of the ACM, The Information Society, Journal of Consumer Research, Journal of Marketing, Journal of Marketing Research, Marketing Science, and SCIENCE.
Address: Owen Graduate School of Management, Vanderbilt University, 401 21st Avenue South, Nashville, TN 37203

Ann Schlosser is Assistant Professor of Marketing at Vanderbilt University. Before coming to Vanderbilt, Dr. Schlosser was a postdoctoral research associate in interactive marketing at the National Center for Supercomputing Applications (NCSA). In 1997, she received the @d:tech Scholarship for individual contribution to understanding the influence of technology on advertising, communication, and marketing. Her research has appeared in the Journal of Consumer Psychology, Journal of Interactive Marketing, Computer Supported Cooperative Work, Handbook on Electronic Commerce and Advertising and The World Wide Web. Dr. Schlosser specializes in consumer behavior in virtual environments. Her current research projects include studying consumer interaction in computer-mediated environments, Web site design and its impact on attitude strength and brand loyalty, and consumer control in computer environments.
Address: Owen Graduate School of Management, Vanderbilt University,401 21st Avenue South, Nashville, TN 37203

	Spring 1997 IDS	Fall 1997 IDS	Spring 1998 IDS
Survey	IDS 2	IDS 3	IDS 4
Time Frame	December 1996 and January 1997	August 1997 and September 1997	May 1998 and June 1998
Number of United States respondents	5,813	7,157	4,042
Total weighted US population aged 16 and over	199.9 million	202.3 million	202.4 million

JCMC 5 (3) March 2000

Message Board