Excerpts from Digital Citizenship: The Internet, Society, and Participation
First Monday

Excerpts from Digital Citizenship: The Internet, Society, and Participation
(Cambridge, Mass.: MIT Press, 2007)

by Karen Mossberger, Caroline J. Tolbert and Ramona S. McNeal

 

Contents

Chapter 1: Defining Digital Citizenship
Chapter 2: The Benefits of Society Online: Economic Opportunity
with Kimberly Johns

 


 

Chapter 1: Defining Digital Citizenship

Citizenship is a status that is bestowed on those who are full members of a community.
 — T.H. Marshall, “The Problem Stated with the Assistance of Alfred Marshall,” 1949.

“Digital citizenship” is the ability to participate in society online. What, however, does it mean to invoke the notion of citizenship in relation to the use of a technology? More than half a century ago, British sociologist T.H. Marshall defined citizenship as endowing all members of a political community with certain civil, political, and social rights of membership, including “the right to share to the full in the social heritage and to live the life of a civilized being according to the standards prevailing in the society” (1992, 8). Information technology, we argue, has assumed a secure place today in the civilized life and prevailing standards of U.S. society. In much the same way that education has promoted democracy and economic growth, the Internet has the potential to benefit society as a whole, and facilitate the membership and participation of individuals within society. We contend that digital citizenship encourages what has elsewhere been called social inclusion (Warschauer, 2003).

We define “digital citizens” as those who use the Internet regularly and effectively — that is, on a daily basis. Previous research has defined a “digital divide” in terms of access to technology (Norris, 2001; Bimber, 2003) or the skills to use technology as well as access (Mossberger, Tolbert, and Stansbury, 2003; Warschauer, 2003; Van Dijk, 2005). Daily Internet use implies sufficient technical competence and information literacy skills for effective use along with some regular means of access. In 2006, digital citizens accounted for a little under half of the U.S. population. Twenty–seven percent of Americans still do not go online at all, and are completely excluded from participation in society online (Pew Internet and American Life Project, 2006).

This book examines three aspects of participation in society online: the inclusion in prevailing forms of communication through regular and effective use; the impact of Internet use on the ability to participate as democratic citizens; and the effects of the Internet on the equality of opportunity in the marketplace. Digital citizens are those who use technology frequently, who use technology for political information to fulfill their civic duty, and who use technology at work for economic gain. To understand the potential and challenges for digital citizenship, we turn to Rogers Smith’s three traditions of citizenship in U.S. history: Lockean liberalism (equality of opportunity), civic republicanism (politics), and ascriptive hierarchy (inequality). These traditions demonstrate how Internet use is integral to citizenship in an information age, and why political and economic uses of the Internet differ from other activities online. The ability to participate in the civic sphere and compete in the economic realm are both central to U.S. conceptions of citizenship as embracing political community and equality of opportunity.

The following pages present new evidence that Internet use does indeed have significant benefits for democratic participation and economic welfare. We find that Internet use increases the likelihood of voting and civic engagement; it also promotes higher incomes for African Americans and Latinos in particular. Our findings establish that patterns of exclusion endure even as Internet use has grown, and that they are linked to other inequities. Economist Amartya Sen (1993) has argued that poverty and inequality should be viewed not in terms of material possessions but in light of the capacities and functioning of the members of a society. The capacity to use the Internet includes access to technology at home and in other settings, and educational and technical skills. Drawing on Sen and our empirical findings, we view digital citizenship as representing capacity, belonging, and the potential for political and economic engagement in society in the information age.

 

++++++++++

Chapter 2: The Benefits of Society Online: Economic Opportunity
with Kimberly Johns

There is strong evidence that the Internet has played a major role in the productivity revival experienced by the U.S. since the early 1990s ... . [ Yet] there is increasing concern that unequal access to Internet services is contributing to widening inequalities in income, wealth, and power.
  — Charles Ferguson, Brookings Institution, 2002.

The development of Web browsers and commercial applications for the Internet contributed to what has been characterized by some economists as the “Roaring Nineties,” which produced robust economic growth (Krueger and Solow, 2001). When viewed over the longer term, however, technology has also reshaped the terrain of the U.S. job market and contributed to widening inequality. In order to understand the role of information technology for economic opportunity, we ask whether computer and Internet use on the job benefits U.S. workers by raising wages beyond what they would otherwise receive, given their other qualifications and characteristics. There has been some prior evidence indicating that this is so, along with some debates over this research. Our findings update earlier studies on the benefits that individuals receive from technology use at work, as most research predates the Internet and the visible productivity gains from technology in the 1990s.

A more pointed question, though, is whether or not public policy should encourage the acquisition of digital skills among those who are least likely to have them — low–income workers, the least educated, and the unemployed. Information technology use on the job is most prevalent in occupations requiring higher education levels, such as professional and managerial jobs (U.S. Department of Commerce, 2002). Do technology skills matter for less–educated Americans in terms of their economic well–being? If so, those who are already disadvantaged in the job market because of discrimination and lesser skills may have this disadvantage compounded. Part of our analysis in this chapter highlights the benefits of information technology use for less–educated workers. Prior research on the benefits of technology use among less–educated employees — those who have a high school diploma or less — is based on employer surveys in several cities. Using national data on actual earnings, we find that technology is in many ways even more important for income among these individuals. We offer new and comprehensive evidence that technology skills are critical for wages, among disadvantaged workers as well as all workers. This provides a strong case for digital citizenship as a societal concern.

While all of the advanced industrialized countries experienced rapid increases in income inequality during the 1980s and 1990s, this inequality has grown most extensively in the United States (Morgan and Kim, 2006; Jacobs and Skocpol, 2005; Friedman, 2003). This is not merely the result of an expansion of wealth at the top of the income distribution but of the declining fortunes of many Americans as well. Compared to other industrialized nations, the United States has substantially more poverty, whether calculated in relative terms in distance from the median income or in absolute terms based on cross–national standard–of–living measures (Smeeding and Rainwater, 2001). For some Americans, especially those at the bottom of the wage distribution, there has been a decline in real wages during this same period (Krueger, 2003).

In the liberal tradition, the equality of opportunity is a more important value than the equality of outcomes. If workers are rewarded differentially for the choices they make based on their interests, natural talents, or the amount of effort they are willing to devote to economic success, then income inequality is fair in the view of most Americans. But such inequality is sanctioned by public opinion only if there are equal opportunities to succeed. As the following section will show, widening income inequality in the United States is at least partially caused by changes in skills. The consequences of technology disparities are unequal chances to participate in the economy and prosper.

We review prior research on the importance of technology for economic opportunity and then present new findings based on multiple sources. First, we examine the comprehensive, large–sample CPS. The most recent CPS that includes questions about information technology use was conducted in 2003, and its large size yields representative samples even for smaller subgroups in the population, such as minorities or less–educated employees. We supplement these analyses with data drawn from the Pew Internet and American Life Project in 2002 and 2005. In contrast to the CPS, the Pew opinion surveys contain information about the frequency of technology use at work. We have argued that the frequency of use is a proxy for skill, and so this should provide a more nuanced test of the role of technology use in the labor market.

Beyond Amazon and Silicon Valley

The Internet has had highly visible impacts on business and the economy as well as more subtle but far–reaching effects. There is evidence that information technology has increased productivity and growth, while checking inflation rates (Welfens and Jungmittag, 2003). Productivity gains during the 1990s were attributed to industries that produce technology or those outside the information technology industry that use it intensively (Stiroh, 2001). Information technology has continued as a “major driver” of growth well beyond the 1990s, according to the U.S. Department of Commerce (Gallagher, 2005). Computer use and higher productivity reduced the inflation rate by 0.3 to 0.4 percentage points in the period between 1987 and 1998 — that is, goods and services in the economy were cheaper because of the increased efficiency with which they were produced (Crepon and Heckel, 2002). Some scholars argue that information technology will lead to long–term deflationary trends, reducing the costs of doing business in every industry (Tapscott, Ticoll, and Lowy, 2000). Lower costs combined with greater productivity also produce increases in real wages for individuals (Tanaka, 2004).

Throughout the 1980s and the early 1990s, economists puzzled over the absence of any discernible productivity gains from the widespread application of computer technology — a mystery that Robert Solow termed the “productivity paradox” (Blinder, 2000). This paradox disappeared by the late 1990s, as it became clear that information technology was powering growth and productivity (Mishel, Bernstein, and Schmitt, 2001, 19–20; Barrington, 2000). The appearance of Web browsers in the early 1990s and the growing popularity of the Internet were not the only factors enhancing the use of information technology by businesses and consumers during the 1990s. Better and cheaper hardware and software increased the utility of computers in many other ways (Atkinson, 2004). But the Internet provided a newfound ability to network operations, communicate with suppliers and consumers instantly, and market products on the Web.

Perhaps the most obvious development was the emergence of e–commerce, conducted on Web sites from Amazon.com to zworld.com (which sells computer keyboards). But e–commerce is still dwarfed by old–fashioned commerce, as it accounts for only three percent of the total adjusted sales (U.S. Department of Commerce 2007). Technology can also have an impact because of investment and use throughout the economy, in manufacturing, retail, banking, transportation, health care, and other sectors. The need for high–tech applications varies across industries, but the demands for technology innovation and proficiency reach well beyond Silicon Valley. Almost two–thirds of the growth in investment can be attributed to information technology (Welfens and Jungmittag 2003, 15), and this investment continued to expand in 2004 and 2005 (Gallagher, 2005). The emergence of broadband is predicted to accelerate the adoption of Internet strategies by firms (OECD, 2003).

The integration of Internet applications throughout the economy is fueling economic growth and has the broadest implications for understanding the role of technology skills in the workforce. The gains in productivity that first appeared in the 1990s are likely to continue, albeit at a slower pace (Litan and Rivlin, 2002). Experts predict that it is the “old economy” sectors that will account for future growth, and that this will be “not from new activities, but from faster, more efficient conduct of existing ones — faster, cheaper handling of information needed in ordinary business transactions such as ordering, billing, and getting information to employees, suppliers, and customers” (Litan and Rivlin 2002, 6). Delta Airlines, for example, has cut the time for loading planes in half and reduced the number of workers by half with the introduction of Internet terminals at gates to direct baggage handling, cleaning, and fueling. Federated Department Stores uses the Internet to disseminate information from the floor throughout the company, and has decreased inventory costs and improved pricing decisions (Sommers and Carlson, 2003).

Whether increases in productivity and growth will persist over the long term depends on how that technology continues to be used, with sustained innovation, investment in capital, the reorganization of work processes, and worker training (Bresnahan, 1999; Autor, Katz, and Krueger, 1998; Allen, 2001; Arnal, Ok, and Torres, 2003). The increased use of the Internet has changed work processes and required more training as part of the reorganization of work practices (Arnal, Ok, and Torres, 2003; OECD, 2003). Technology creates a need for the continuous updating of employee skills through human resource development (Wentling, Waight, and King, 2002, 11). This contributes to the stock of human capital in the economy.

Information Technology and Economic Change

At the same time that the economy as a whole has clearly benefited from the introduction of new technologies, less–skilled workers have borne the costs of economic change. Information technology has been a contributing factor in rising income inequality over past decades (Autor, Katz, and Krueger, 1998; Acemoglu, 2002). Other explanations for greater wage inequality and the simultaneous decline in real wages for less–educated workers include trade, globalization, the weakening of unions, and changes in the supply of skilled workers; but a consistent theme is the role of technological change and rising skill demands (for reviews, see Katz, 2000; Acemoglu, 2002).

Declining wages for less–educated workers are in part due to shifts toward more knowledge–intensive industries and away from manufacturing. This creates a demand for higher levels of education as well as technology use. Knowledge–intensive (and information technology–intensive) industries such as telecommunications, finance, business, and insurance comprise more of the economic activity of the United States than ever before. In 2003, these knowledge–intensive industries accounted for 25 percent of the value added in the United States (OECD 2005b). Information is increasingly important in the economy and leads to competitive advantage (Wentling, Waight, and King, 2002, 15).

There have been changes within industries as well. Information technology has had contradictory effects, raising the skills needed for some jobs and spurring the development of new occupations, while lowering the skills and compensation for other jobs, or eliminating them entirely (Autor, Levy, and Murnane, 2003; Capelli, 1996). Frank Levy and Richard Murnane (1996) explain that work can be categorized as consisting of routine tasks that computers can perform at practical costs or as exceptional tasks that entail a higher cost when performed through computers rather than human labor. Innovations such as online banking have eliminated some routine tasks performed by bank clerks, for example. But such Internet applications have also created new occupations or increased the demand for some existing job categories. More highly skilled technicians, systems analysts, security specialists, Web designers, and others are needed to implement online banking.

The overall effect of technological change has been to raise the level of skill in the workforce. The demand for college graduates has increased within industries and is not just a reflection of a shift away from manufacturing. Occupations with higher average pay and higher educational requirements expanded more rapidly between 1984 and 1993 in those sectors that adopted computer technology at a faster rate (Autor, Katz, and Krueger, 1998; see also Dunne, Haltiwanger, and Troske, 1997). Economists view the spread of computers as not only an increase in the demand for computer users and technicians but more broadly as part of a technological change that has altered the organization of work and thereby affected the need for workers with various skills (Autor, Katz, and Krueger, 1998). In a review of research from member countries of the Organization for Economic Cooperation and Development, Young–Hwa Kim (2002) concluded that there has been a general effect of “up–skilling” since the 1980s, and that there is a positive relationship between this upskilling of the workforce and the use of information technology in the economy.

In their study of wages and skills, David Brauer and Susan Hickok (1995) found that technological change is the most important factor driving the rising wage inequality between low–skilled and high–skilled workers. Because highly educated workers are more likely to employ computers, the growth in computer use alone accounts for as much as 40 percent of the increase in the return to education, or the “wage premium” enjoyed by more educated workers (Brauer and Hickok, 1995). Using CPS data from 1979 and 1989, the authors found that technological change had reduced the total of wages paid by industry for all skill levels, except for college graduates. This suggests a significant displacement of low–skilled or less–educated workers (Brauer and Hickok, 1995).

Subsequent to the time period discussed by the Brauer and Hickok study, the Internet has decentralized the distribution of work within organizations and across space, and this leads to pressures eliminating jobs for lower–skilled workers as well. Outsourcing, facilitated by the Internet, is most detrimental for less–educated workers. Networking allows job functions to more easily move to distant locations to capture the lowest prices for labor. Less–educated workers are less likely to relocate than college–educated workers (Bound and Holzer, 2000). From the perspective of the economy as a whole, relocation and/or outsourcing may lower costs. Yet the brunt of the impact may be shouldered by individual workers who have lost their jobs and find that there are relatively fewer opportunities available for low–skilled workers (Autor, 2001).

This description depicts a complex set of changes in which higher levels of skill include educational attainment as well as computer competencies. Technological change has also emphasized the importance of building “human capital” through education, training, and skills development. The remainder of this chapter will explore the effects of computer and Internet use on wages, reviewing existing research and presenting new analysis.

Impacts for Individual Workers

A growing percentage of workers at differing educational levels use computers and the Internet at the workplace. Frequencies from the most recent (2003) CPS data used in the multivariate analysis for this chapter indicate that 72 percent of Americans who are employed and have more than a high school education use computers at work, and 58 percent of employed Americans with more than a high school education use the Internet on the job. This compares with 35 percent of workers with a high school education or less who use computers at the workplace, and 21 percent of less educated employees who use the Internet at work. There is about a 37 percent point gap between high– and low–skilled workers for both computer and Internet use. Still, more than a third of less educated workers use computers at work, and more than a fifth go online at their jobs.

What influence do information technology use and skills have on individual economic opportunity? Some existing studies indicate that technology use at work increases wages, but this is subject to some debate, and there are real gaps in the research in this area. Yet it is one of the most important questions to ask if we want to justify expanding technology access.

Wage growth in occupations in the 1980s and early 1990s was associated with computer use (Card, Kramarz, and Lemieux, 1996; Autor, Katz, and Krueger, 1998). Prior research (predating the Internet) indicates that individual workers enjoy higher wages in return for computer use, beyond what their education and occupation would predict. In a widely cited study of 1980s’ CPS data, Alan Krueger (1993) estimated the premium for computer use to be wages that were 14 percent higher in 1984 and 16.5 percent higher in 1989 than for similarly situated workers who did not use computers. He explained these findings as the result of greater productivity for workers with technology skills. David Autor, Lawrence Katz, and Alan Krueger (1998) found a similar premium for the early 1990s. Other studies showing increased wages for technology use at work include research on Canada (Reilly, 1995; Morissette and Drolet, 1998), Australia (Miller and Mulvey, 1997), the Netherlands (Oosterbeek, 1997), and the United Kingdom (Arabsheibani, Emami, and Marin, 2004) as well as for older workers in the United States (Friedberg, 2001). Most of these studies indicated that the wage premium attributable to technology use ranged between 10 and 15 percent (Arabsheibani, Emami, and Marin, 2004), but there are some exceptions. The studies cited above examined data from earlier time periods when there was little Internet use outside some scientific and academic circles.

The pioneering study by Krueger (1993) was criticized by John DiNardo and Jörn–Steffen Pischke (1997), who used their analysis of German data to argue that workers using pencils or sitting down on the job enjoy higher wages as well. Although Krueger controlled for observable differences such as educational attainment and occupation, there may be unobservable factors other than computer use that contribute to higher wages in certain occupations (for example, more talented workers being assigned to jobs using computers). In fact, one study using panel data indicated that French workers who were among the first to employ computers and other new technologies on the job tended to be the most qualified workers, and that controlling for this, the wage premium for computer use was approximately two percent rather than 15 percent (Entorf, Gollac, and Kramarz, 1999). The French research contained some unique data that would be difficult to replicate. Yet the French study also focused on employees who were in the vanguard of the early diffusion of a technology. During a period of widespread use, unobserved individual differences among workers may be less of a threat to validity. One group of scholars responded to the “unobserved variables” critique of Krueger’s work by using two–stage models (Heckman’s regression) to control for endogeneity in their study of British computer use, and they found a wage premium comparable to Krueger’s study over the same period in Britain (Arabsheibani, Emami, and Marin, 2004).

Another criticism is that technology use represents only one part of the rising skill requirements in the workforce. Timothy Bresnahan (1999) concludes that cognitive abilities and people skills account for more of the return to increases in education and skills than information technology use, although he does find some positive effects for technology use as well. While we are most concerned here with the impact of technology skills, we acknowledge that they may be just one part of the changing skill set demanded in the new economy.

This highlights the need to better understand whether technology use significantly increases the wages of less–educated workers. Phil Moss and Chris Tilly (2001) conclude from a review of the literature that skill needs are indeed rising for jobs at all levels, not just managerial or professional jobs. According to a telephone survey of employers in four cities, the needed competencies include computer skills as well as other “hard” skills such as reading, writing, and math, and “soft” or social skills (Holzer, 1996; Moss and Tilly, 2001). Forty percent of the employers who were surveyed mentioned some increase in the level of skills needed for jobs requiring a high school diploma or less, and computer use was cited by about 70 percent of these employers as the reason for the rising requirements. Across occupations, computer use was mentioned as a reason for increasing skills in 92 percent of the clerical occupations that had experienced an increase, 63 percent of the customer service jobs with rising skills, and 48 percent of the blue–collar jobs with changing skills (Moss and Tilly, 2001, 55). In a separate set of face–to–face interviews with employers, “the most common skill change reported” was the requirement for computer skills, but the need for other hard and soft skills was also commonly cited as accompanying these changes (Moss and Tilly, 2001, 63–64).

The multicity survey cited by Moss and Tilly in the preceding paragraph was also analyzed by Harry Holzer (1996). Between 1992 and 1994, employers were randomly sampled in four cities: Boston, Detroit, Atlanta, and Los Angeles. Holzer (1996, 116–117) found that computer skill requirements were a significant determinant of wages for non–college jobs across all racial, ethnic, and gender groups, but that white females were the most heavily rewarded for computer use at work. Those who were the least likely to experience higher wages for computer use were African American and Latino males (Holzer, 1996, 125, 127–128). Other factors that were significant across the models for all workers were requirements for reading and writing, a high school diploma, vocational training, and experience (Holzer, 1996, 116–117). The dependent variable in this study was the log of the weekly starting wage of the last person hired in each of the firms responding to the survey. While suggestive, these data are neither as comprehensive nor as precise as the national CPS, which is based on the current wages of individual respondents. Additionally, there may have been considerable change since the early 1990s. With the emergence of the Internet and the more widespread use of information technology in the workforce, a more recent assessment of the impact of technology is needed.

There is some initial research on Internet use in the United States. Ernest Goss and Joseph Phillips (2002) found that in the manufacturing sector, Internet users were paid more highly — a wage premium of 13.5 percent. Controlling for other factors influencing pay, Internet use was still a significant predictor of higher wages. This study was based on the 1998 CPS and was limited to one economic sector. The work by Goss and Phillips (2002), and earlier research on computers by Krueger and his colleagues, all indicate that Internet use at work might have similar effects across the economy.

Using more recent and complete data, we test whether the income gap due to Internet use is significant beyond the manufacturing sector. Does the frequency of use matter, given that we have defined digital citizenship as regular and effective use? We are also interested in how information technology affects workers in different occupations. Are the benefits of technology concentrated primarily in high–level managerial and professional jobs, or are they spread to a greater extent throughout the workforce? To better understand the consequences of digital citizenship, it is also important to investigate differences in the benefits of technology use for less–educated workers and minorities. Women, African Americans, and Latinos are even more likely than others to view information technology as an avenue for increasing economic opportunities in the United States (Mossberger, Tolbert, and Stansbury, 2003). Can Internet skills confer some advantages in the job market that might offset, to some extent, other inequities? There is currently a lack of recent national research that directly evaluates the effects of the Internet on the wages of U.S. workers.

In the next section, we use the 2003 CPS to test the impact of information technology use at work for employee earnings, first using a general sample, and then examining a subsample of only less–educated workers (with a high school diploma or less). The sample of less–educated workers has two advantages. First, it allows us to test whether technology use at work is consequential for this group, which is also most likely to experience digital inequality. Second, analyzing the subsample permits us to better isolate the effects of Internet use from education — the endogeneity problem, when education leads to both better–paying jobs and a greater likelihood of Internet use. Next, we supplement this analysis with survey data collected by the Pew Internet and American Life Project in 2002 and 2005 to examine the significance of the frequency of Internet use at work for income. This is important, given our emphasis on the frequency of use for digital citizenship.

Approach

We explore the impact of Internet use at work using the 2003 CPS, which is the most recent survey conducted by the U.S. Census Bureau that includes a supplement on information technology use. The large–sample survey not only provides accurate estimates of the population as a whole but also information on weekly earnings that is not found in most other sources. The CPS does not, however, include data on the frequency of computer or Internet use at work. Additionally, we examine the effects of frequent use and online training using national opinion data collected in 2002 and 2005 by the Pew Internet and American Life Project: The May 2002 Workplace Email Survey and The Internet and American Life Major Moments Survey, February–March 2005. These surveys feature questions about Internet use at work, job training activities, and income. The Princeton Survey Research Associates conducted the random national telephone surveys for Pew, and the U.S. Census Bureau collected the data for the CPS. Our primary hypothesis is that Internet use at work leads to higher incomes for employees, controlling for other factors, including education, occupation, and age. The following section provides a detailed explanation of our methods and variable coding for all three surveys. For those who are less interested in methods, you may skip to the findings in the section (“Results”) that follows.

Discussion of Methods and Variable Coding

Data and Methods: 2003 CPS

In order to explore the impact of technology access at work on wages, we turn to the 2003 CPS March Supplement on information technology conducted by the U.S. Census Bureau. The national random sample survey includes over 103,000 respondents. This sample (a hundred times larger than a typical national opinion survey) provides accurate estimates of the population as a whole, with detailed questions about occupations and employment as well as technology use. This unique data set allows a rigorous empirical test of whether computer and Internet use at work leads to increased income, especially among subpopulations, such as those with a limited education. We estimate multivariate regression models to predict weekly earnings for the population as a whole and less–educated Americans.

We begin by filtering our sample population for only employed workers in the labor force. Of the 103,000 respondents in the sample, 62 percent (or 64,259) are employed at work and 2 percent (or 2,193) are employed/absent from the job. These individuals are included in the analysis. The remaining respondents in the sample are unemployed due to a layoff (.34 percent), unemployed but looking for a position (3 percent), not in the labor force due to retirement (17 percent), not in the labor force because of a disability (4.5 percent), or not in the labor force for some other reason (11 percent). These respondents were excluded from the analysis. Additionally, we include a binary variable in our models coded 1 if the respondent is employed full–time, and 0 if the respondent is employed part–time. We expect full-time workers will earn more than those in the labor force part–time.

The primary dependent (or outcome) variable measures weekly earnings of the respondent in dollars. A limitation of these data is missing values for the variable measuring income. Of the 103,000 respondents, 90 percent had missing values on the weekly earnings question, because the CPS rotates the percentage of panel respondents who are asked about earnings. Because of missing data on the dependent variable, our models included 14,851 cases/individuals. This sample is still almost 15 times larger than a typical thousand–person survey, and is still randomly selected. As a follow–up analysis, we measure the annual household income of the respondent as the dependent variable. Unlike weekly earnings, almost all respondents in the survey answered questions about annual household income yielding a full sample of a hundred thousand cases.

Three questions are used as the primary explanatory (independent) variables, each measuring technology use at work. The respondents were asked if they used a computer at work, engaged in “computer use at work for Internet or e–mail,” and had used the Internet this year to take courses. The latter question was included to find out whether Internet use for increasing skills had any effect on wages. Affirmative responses to each question were coded 1 (yes) and 0 (no). These three binary variables serve as our explanatory variables, and separate our sample among those who use technology on the job and those who do not. Separate regression models are estimated for the three types of technology use on the job.

Beyond technology use at work, many other factors are known predictors of income and earnings, especially occupation. An advantage of the CPS data beyond standard surveys is detailed employment information. We use the 11 industry and occupation job categories measuring a respondent’s primary occupation. A series of binary (1/0) variables was created for each occupation, with production as the reference (left–out category). We expect that management and professional occupations will have the highest earnings. As an additional control, we include a binary variable measuring whether the respondent is employed in the job sector that the U.S. Census defines as the “information industry,” which includes technology/computing jobs as well as publishing. We would expect those employed in the information industry to have a higher probability of using computers and the Internet at work.

Our models also include standard demographic controls given known earnings gaps based on gender, race, age, and education. We expect that white males who are older with a higher education will earn more than minority females who are younger with a lower education. By including these demographic variables in the models, we control or hold constant the effect of demographic factors on earnings. A binary variable measures gender, with females coded 1 and males 0. Compared to standard surveys, our national data include large and representative samples of African Americans and Latinos. Of the 103,000 total sample, 10 percent (or 10,113) reported being of Hispanic origin, and almost 10 percent (or 9,920) reported being black. Additionally, almost 5 percent (or 5,037) were Asian American. Three binary variables measure whether the respondent is an African American (coded 1), Latino (coded 1), or Asian American (coded 1), with white non–Hispanic as the reference group. Age is measured in years. It serves as a proxy for experience; we presume that older employees have greater job experience and will earn more. The educational attainment of the respondent is measured on a 5–point ordinal scale ranging from 1 (less than a high school degree) to 5 (a bachelor’s degree or higher). Geography/location is measured with binary variables for urban and suburban residents, with rural residents and those who did not identify their location as the reference group (coded as 0). Private sector and federal government jobs tend to pay more than local governments and non–profits. We use a series of binary variables to measure the job sector (federal government, private, or local government), with state government and non–profit sectors as the reference category coded 0. Including a different grouping of binary variables for the job sector does not change the substantive findings reported here.

Data and Methods: 2002 and 2005 Pew Surveys

As a robustness check, we also examine the 2002 and 2005 Pew surveys with multivariate regression used to model the effects of the frequency of Internet use and other factors on the respondent’s personal income. The dependent variable is an 8–point ordinal scale, where 1 indicates that the family income in the previous year ranged from $0 to $10,000, and 8 signifies a family income of $100,000 or higher. We use two alternative measures of Internet use at work. Internet use at work is measured with a binary variable, where yes responses are coded 1, and no responses are coded 0. This coding is comparable to the coding used in the above CPS analysis.

Because we have emphasized the importance of the frequency of Internet use as a preferable way to measure skills and digital citizenship, we measure the use of technology at work on an ordinal scale. In the 2002 survey question, the wording was: “Counting all of your online sessions, how much time did you spend using the Internet yesterday [at work]?” The responses were coded on an 8–point scale from 1 (less than 15 minutes) to 6 (two to three hours) to 8 (four or more hours). For the 2005 survey, the question wording was: “In general, how often do you use the Internet from work — several times a day, about once a day, three to five days a week, one to two days a week, once every few weeks, or less often?” The responses were coded on a 6–point scale, with 6 equal to several times a day, and 1 equal to “less often.” These detailed questions on the frequency of Internet use at work create a measure of skills associated with access and their impact on employee income.

The models also include a number of demographic and socioeconomic factors that are known to influence income, which are coded to be similar across the two years of the national opinion data and comparable with the analysis of the CPS data. These variables include education measured on a 7–point scale, ranging from an eighth–grade education or less coded 1 to postgraduate work coded 7 as well as age measured in years. Gender is measured using a binary variable coded 1 for males and 0 for females. We expect males to earn higher incomes than females. To control for race and ethnicity, dummy variables were included for African Americans, Asian Americans, and Latinos, each coded 1, with non–Hispanic whites as the reference group (coded 0). Because of differences in metropolitan and rural labor markets, two dummy variables were included to measure the respondent’s geographic location, with residents of suburban and urban areas coded 1, and residents of rural areas coded 0.

Like the CPS data, the 2002 Pew survey included a question on the occupation of the respondent. A series of dummy variables measure the job category of the respondent, including professional, manager or executive, clerical or office worker, business owner (with two or more employees), and sales (either store clerk or manufacturer’s representative). Variables for the respondents who named each one of these job categories are coded 1, and 0 for the respondents who did not name this as their job type. The reference group is composed of service workers, skilled trades, semiskilled labor and laborers, all coded 0. Unique in this survey is a series of binary variables that measure employer type and size. These variables include large corporations, medium–size companies, small businesses, schools or other educational institutions, and other (including non–profits). The employees of these organizations are coded 1, and if the individual did not work for this type of organization they are coded 0. The reference group is government workers, including federal, state, and local government employees.

 

Results: Appendix Tables
 
Table 2.A.1 (CPS, general population, technology use at work, earnings)
Table 2.A.2 (CPS, less–educated population, technology use at work, earnings)
Table 2.A.3 (Pew, general population, frequency of technology use at work, income)
Table 2.A.4 (CPS, general population, technology use at work, household income)
Table 2.A.5 (CPS, general population, technology use at work, household income)
 

 

Finally, the models control for economic conditions in the respondent’s state that affect employment opportunities. State unemployment rates in 2002 and 2005 are from the Economic Census. The models also include a measure of the number of information technology jobs in the respondent’s state from the State New Economy Index conducted in 2002 (Progressive Policy Institute, 2002). States with a larger share of workers trained and skilled in the use of information technology are expected to foster higher incomes than states with a smaller share. The Progressive Policy Institute explains that this measure includes workers in a variety of industries. The variable used in this analysis measures the number of information technology jobs in the information technology sectors and then subtracts this number from the total number of workers in information technology occupations in a state. This creates a more accurate measure of the extent to which traditional industries employ information technology professionals. Nevertheless, states with high scores are high–tech locations such as Colorado, Washington, and Massachusetts. Low–scoring states tend to have economies based on natural resources or traditional manufacturing.

Results, CPS: Information Technology Use for All Workers

Since the dependent variables in Tables 2.A.1 and 2.A.2 are weekly earnings in dollars, ordinary least squares regression is reported, with robust standard errors to control for heteroskedasticity. Column 1 (Table 2.A.1) tests whether computer use at work is associated with increased weekly earnings for the general population, holding other factors constant, while column 2 includes an identical set of control variables, but swaps computer use at work for Internet/e–mail use at work. Finally, column 3 includes a variable measuring whether the respondent took courses online. Across the three models in Table 2.A.1 (total population sample) we see strong and consistent evidence that technology use at work is related to higher wages, even after controlling for a battery of factors known to increase earnings, including education, age, and occupation.

The substantive magnitude of the effects of technology use at work on economic opportunity is substantial. Average weekly earnings are $692.35 (standard deviation $519.32), which equals roughly $2,768 a month or $33,000 a year before taxes. Holding other demographic, occupational, economic, and job sector factors constant, an individual who uses the computer at work is predicted to earn $101 more per week than the same individual who does not use the computer at work (column 1). This is a 14.5 percent boost in earnings based on technology use at work, and is consistent with Krueger’s earlier (1993) findings of a 14 to 16.5 percent wage premium for computer use.

Internet/e–mail use at work creates a larger boost in wages, all else being equal (see column 2). Weekly earnings are $118.27 higher for those individuals using the Internet at work than those employed individuals who do not use the Internet on the job — a 17 percent boost in weekly earnings, all else being equal. Even taking courses online appears to increase weekly earnings by a predicted $39 a week over those who have not taken online courses (see column 3). This is strong and consistent evidence that technology use at work may increase wages for the U.S. population.

Many of the control variables are in the expected direction, lending validity to our findings. Females earn on average approximately $200 less per week than their male counterparts, while older individuals earn more than the young. Racial and ethnic minorities (African Americans, Latinos, and Asians Americans) earn less than whites. Employees with bachelor’s degrees gain a bonus of approximately $350 per week, in comparison with those who only have high school diplomas. The effect of Internet use is therefore around one–third of the impact of having a four–year college degree rather than high school only — a considerable amount, given the literature on the increasing returns to education. Geographic location matters as well, with suburban residents (who have more employment opportunities) earning roughly $100 more per week than their rural counterparts, all else being equal, while urban residents earn roughly $50 more per week than rural residents. Occupation also matters significantly for wages, with those in management and professional occupations earning considerably more than the reference category (production). Sales and construction occupations also earn more than our baseline occupation (production). As predicted, federal government and private sector employees earn between $200 and $100 more per week, respectively, than those working in non–profits or state government. As expected, full–time workers earn almost $400 more per week than those who work only part–time.

So far the analysis provides fairly robust evidence that technology use at work is associated with increased economic opportunity among the employed segment of the U.S. population, and that the substantive size of the effect rivals that of increased education, place (suburban/rural/urban), occupation, or job sector (working in the private sector or for the federal government). The models in Table 2.A.1 are robust, accounting for 41 percent of the variation in weekly earnings among the sample of 15,000 respondents.

Results, CPS: Information Technology Use for Less-Educated Workers

The more important question, for our purposes, is whether technology use at work can increase the wages of the less–educated employees. Table 2.A.2 replicates the models in Table 2.A.1, but includes only those respondents in the CPS survey with a high school degree or less. Box 2.1 below is drawn from the analyses reported in Tables 2.A.1 and 2.A.2, and it compares the average dollar amounts attributable to computer and Internet use for the general population and less–educated workers with the effects of demographic and other variables. The models control for a wide range of occupations, as shown in Tables 2.A.1 and 2.A.2, but for simplicity of comparison, we have listed only selected occupations that involve a high degree of Internet use: management and secretarial workers.

Technology use on the job is associated with even greater proportionate wage increases for less–educated employees. Within this segment of the population, technology use at work was less common than for those with education beyond a high school degree. Yet we see that technology use continues to have a positive and statistically significant effect in increasing weekly earnings. Less–educated workers who use the computer at work are predicted to earn $90 more per week than the same less–educated worker who does not use the computer on the job, all else being equal. Again, Internet use at work leads to even larger economic gains — a $111 increase in weekly earnings. These dollar figure increases are comparable to those for the population as a whole, but because average weekly earnings are significantly lower for this population, these increases account for a larger percentage change.

The control variables largely mirror the U.S. population as a whole, but with some notable exceptions that we would expect among less–educated workers. While women and racial minorities continue to earn less than males and whites (although the gender and racial gaps are smaller among the less educated), urban residents are now statistically no different than rural residents in earnings, while those in suburbs continue to earn roughly $50 more per week, all else being equal. Among the less educated, the trades earn higher wages (construction, repair, and transportation), while those in service, sales, and secretarial positions earn considerably less than the baseline (production) occupations. Full–time workers earn almost $300 dollars more per week than part–time employees among the less–educated population.

 

Box 2.1: What Matters for Weekly Earnings, CPS 2003
 
The variables reported are all statistically significant with a 95 percent confidence interval for predicting weekly earnings. The dollar amounts are based on regression coefficients in Tables 2.A.1 and 2.A.2, and represent the independent effect of each variable, holding other factors constant.
 
 Weekly Earnings
VariableModel 1: Computer UseModel 2: Internet Use
 
General Population (Table 2.A.1)+$101.60+$118.27
Education (difference, 4 yrs. college vs.
h.s. diploma)
+$354.72+$343.72
Age (per year)+$4.86+$4.83
Female-$208.36-$205.22
Latino-$52.30-$55.38
Asian American-$51.92-$52.99
African American-$65.70-$64.12
Urban+$49.90+$48.55
Suburban+$99.37+$98.33
*Management vs. Production+$319.29+$311.82
*Secretarial vs. Production-$40.81-$37.82
Federal Government vs. State/Non–profits+$189.68+$195.96
Private Sector vs. State/Non–profits+$88.76+197.14
Full–time+$379.59+$373.93
 
Less–Educated Workers (Table 2.A.2)+$89.76+$111.33
Age (per year)+$2.92+$2.92
Female-$133.73-$133.78
Latino-$72.15-$74.13
Asian American-$46.45-$50.98
African American-$27.07-$26.89
Suburban+$44.68+$44.86
*Management vs. Production+$223.69+$219.24
*Secretarial vs. Production-$26.02-$23.96
Federal Government vs. State/Non–profits+$76.71+$76.71
Full–time+$290.63+$289.01
 
* Selected Occupational Categories: See Tables 2.A.1 and 2.A.2 for other categories where the difference between the occupation and the reference category (production) is statistically significant.

 

 

Box 2.2: Wage Premium for Internet Use for Less–Educated Workers
 
The figures below are the expected percentage difference that Internet use at work makes for wages, controlling for other factors. Predicted values estimated from Table 2.A.1.
 
 Wage Premium/Internet Use at Work
African American Men18.36%
African American Women17.31%
Latinos16.99%
Latinas16.11%
White Men14.77%
White Women13.56%

 

In order to compare the magnitude of information technology’s impact for different groups of workers, expected wage premiums are estimated based on the regression coefficients in Table 2.A.1 varying race, ethnicity, and gender, with all other variables set at their modal or mean values. These categories are associated with Smith’s definition of ascriptive hierarchy.

Information technology use at the workplace matters even more for minorities. Among less–educated workers who use the Internet in their jobs, African Americans and Latinos enjoy a higher premium for Internet use than similarly situated whites, even though Internet use does not begin to compensate for otherwise lower wages. African American men with a high school education or less earn 18.36 percent more than similarly situated African American men who do not use the Internet at work, while African American women gain a bonus of 17.31 percent. For less educated Latino workers, the wage premium is 16.99 percent for men and 16.11 percent for women. Among less educated white workers, the Internet increases earnings 14.77 percent for men and 13.56 percent for women. This demonstrates that information technology use at work is indeed related to economic opportunity for less educated workers, and that the effects are slightly greater for minorities. In contrast to Holzer (1996), women do not enjoy the greatest advantage from technology use, at least not in terms of the percentage gain in wages. Differences based on gender, however, are slight. This shows that the Internet can be a mechanism for leveling the playing field among less educated workers, who have generally fared poorly in the new economy. For minorities, digital skills are even more important for economic gain.

Online courses increase weekly earnings, according to the 2003 CPS — particularly for less educated workers. Additionally, the 2003 CPS includes a question about taking an online course, which allows us to estimate changes in employee earnings as well as the significance of online education (see column 3 in Tables 2.A.1 and 2.A.2). The research on upskilling in the workforce as a result of technological change suggests that Internet use in the workplace encourages individuals to learn new skills as jobs and work processes are reorganized (requirements may be increased for mobility within a firm or an occupation), and this in turn increases wages. It is reasonable to expect Internet use at work to stimulate further training or education. Higher incomes for workers may also be the result of continued human capital development.

How substantive is the magnitude of the effects of distance learning or online job training on economic opportunity? For the general population, Table 2.A.1 (column 3) indicates that even taking courses online appears to increase weekly earnings by a predicted $39 a week over those who have not taken online courses. In contrast, less educated workers who have taken online courses (see Table 2.A.2, column 3) enjoy a larger gain in income than the general population, with less educated employees experiencing a $63 per week increase in earnings from online courses. Again, this suggests that Internet use can have a greater marginal effect for certain groups such as the less educated.

Results, Pew: The Frequency of Use and Income

A further test of the effects of Internet use at work is to ask whether or not the frequency of use affects income, based on 2002 and 2005 survey data from the Pew Internet and American Life Project (Table 2.A.3). The Pew survey does not report individual earnings, and so household income is the best available alternative outcome measure. The variables are described in the previous “Data and Methods” section in this chapter, and are shown in Table 2.A.3. The results for the 2002 and 2005 Pew surveys are reported using a dependent variable that is measured on an ordinal 8–point scale, where 1 indicates that the family income in the previous year ranged from $0 to $10,000, and 8 signifies a family income of $100,000 or higher. Higher scores are associated with higher personal income. Since the dependent variable is ordinal and there are enough categories that it approaches a continuous variable, the model is estimated using Ordinary Least Squares (OLS) regression.

Frequent use at work is related to higher incomes for workers. As shown in columns 2 and 4 in Table 2.A.3, income increases with the frequency of computer and Internet use at the workplace in both 2002 and 2005. Income increases as information technology use becomes more central in accomplishing job–related tasks, thereby enhancing productivity. This further supports the case that information technology use is important for economic opportunity (and perhaps for mobility into more technology–intensive and higher–paying jobs). More frequent use may suggest a higher level of skill and a greater range of activities using the Internet. Frequent use may be accompanied by other types of expertise or skill. But controlling for occupational differences, education, and other factors, the frequency of use is significantly related to higher income.

Additional Support: Other Models Using Income

To further test the impact of Internet use at work, we conducted several other analyses that measure effects on income. The CPS data only asked weekly earnings for a limited sample of respondents, but included annual household income for the full sample of 103,000 individuals. We replicate the models presented in Tables 2.A.1 and 2.A.2, where the dependent variable is annual household income, rather than weekly earnings in Tables 2.A.4 (general population) and 2.A.5 (low–educated sample). In the 2003 CPS, annual household income is measured on a 16–point scale from $2,500 to $150,000. Both computer and Internet use at work are related to higher household income in the 2003 CPS, holding constant the same factors used in the CPS analysis of weekly earnings. As expected, the magnitude of the effect on household income is somewhat smaller, as this may include multiple wage earners and other sources of income. Across the models in Tables 2.A.4 and 2.A.5, we see strong and consistent evidence that technology use at work is related to increased family household income, even after controlling for a number of factors known to increase earnings.

Second, we replicated the Pew 2002 and 2005 income analysis based on the binary variables of whether or not the respondent used computers or the Internet at work, rather than the frequency of Internet use at work (Table 2.A.3, columns 1 and 3). This provides a check on the validity of the income findings for the frequency models reported above. Income may include earnings from multiple household members, so it is a less precise outcome measure of the impact of Internet use than the individual wages used in the first CPS analysis (Tables 2.A.1 and 2.A.2).

The Pew 2005 survey lacked the occupational and employer information included in the 2002 Pew survey, but allowed us to assess whether the impact of Internet use has diminished over time, as suggested by Goss and Phillips (2002), given the recent data. The results for both years were similar to those obtained from the CPS analysis discussed above, and so this increases our confidence in the finding that technology use at work boosts wages and income. Those respondents who use the Internet at work have significantly higher incomes than those who did not in 2002, even when controlling for individual level demographics, state economic conditions, occupation, the size of a business, and geography. The 2005 data indicate that Internet use continues to lead to higher incomes. In fact, Internet use at work may have greater economic payoffs over time. This is suggested by the unstandardized regression coefficient for 2005 in Table 2.A.3, which is nearly double the one for 2002. Without similar control variables, though, it is difficult to evaluate whether the wage premium for Internet use has indeed grown between 2002 and 2005.

Summary of Results

Just as computer use at work produced a wage premium in earlier studies, Internet use at work is also clearly associated with economic gains, whether they are measured in terms of weekly earnings or annual household income. This pattern emerges in both the large–sample CPS and the smaller–sample Pew surveys, using a number of controls in both studies. It appears in 2002, 2003, and 2005. Both use at work and the frequency of use are related to higher wages or incomes. The consistent results across models increase our confidence that computer and Internet use are important contributors to economic opportunity in the digital economy.

Digital Citizenship and Economic Opportunity

Together, our findings indicate that technology use at work advances the economic prospects for individuals. We update older research (such as Krueger, 1993) and show that the development of the Internet has in fact lifted the fortunes of some. Those who use the Internet on the job are more likely to have higher weekly earnings and incomes. The frequency of Internet use at work implies differing levels of capabilities online, and we find that as the frequency of use rises, so does income. As Krueger has indicated, this suggests that higher levels of skill (and productivity) are rewarded in the marketplace. Our findings are consistent across years and surveys, in 2002, 2003, and 2005.

Perhaps most notable, however, is the magnified effect of Internet use for workers who are lower paid and often disadvantaged: less educated workers, African Americans, and Latinos. It would be reasonable to expect that the payoff for Internet skills is most concentrated in high–paying, knowledge–intensive jobs, where Internet use is most pervasive. Instead, the potential gains are relatively greater precisely for those groups in society that are also most likely to lack regular Internet access and effective skills. No prior research has demonstrated the benefits of digital skills for less educated workers, using national data and evidence based on individual earnings.

The use of the Internet for distance learning is also associated with higher incomes and weekly earnings, especially for less educated workers. The literature on skill–biased technological change suggests that Internet use may be just one dimension of the more general human capital development encouraged by technology diffusion.

Our findings provide powerful evidence that digital citizenship matters for economic participation and technology disparities are not a trivial concern for future equality of opportunity. The growth of income inequality through the development of the new economy is in part the result of fundamental technological change that has increased the need for information technology skills as well as education. A higher premium is also placed on education and cognitive skills, with the adoption of technology–intensive practices within manufacturing and other “old economy sectors,” and with the shift toward more knowledge–intensive industries such as finance. Yet even among less educated workers, technology skills garner higher wages.

The skills needed to adapt to these changes are not evenly distributed, as our analysis of digital inequality in chapter 5 will show. They are bound together with existing inequalities, such as disparities in educational opportunities in low-income communities (see chapter 5; Mossberger, Tolbert, and Gilbert, 2006). The liberal tradition of citizenship in the United States has produced a prevailing view that social justice requires equal chances, if not equal results. Firmly within this tradition, Krueger (2003) advocates education and training as a form of “redistribution” to narrow the inequalities of the new economy. In the context of the information age, equal justice requires that everyone in the United States has the ability to develop the digital and educational skills to participate fully in the economy. End of article

 

About the authors

Karen Mossberger is Associate Professor in the Graduate Program in Public Administration, College of Urban Planning and Public Affairs, University of Illinois at Chicago.

Caroline J. Tolbert is Associate Professor in the Department of Political Science at the University of Iowa.

Ramona S. McNeal is Visiting Assistant Professor in the Political Studies Department at the University of Illinois at Springfield.

 

Acknowledgements

These excerpts are reprinted with the kind permission of MIT Press and the authors of Digital Citizenship, Karen Mossberger, Caroline J. Tolbert and Ramona S. McNeal. For more details about Digital Citizenship and other MIT Press books and journals, see http://mitpress.mit.edu/.

 


 

Copyright © 2007, MIT Press.

Copyright © 2007, Karen Mossberger, Caroline J. Tolbert and Ramona S. McNeal.

Excerpts from Digital Citizenship: The Internet, Society, and Participation
(Cambridge, Mass.: MIT Press, 2007)
by Karen Mossberger, Caroline J. Tolbert and Ramona S. McNeal
First Monday, Volume 13, Number 2 - 4 February 2008
http://www.firstmonday.org/ojs/index.php/fm/article/viewArticle/2131/1942





A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2014.