797082
NMS0010.1177/1461444818797082new media & societyvan Deursen and van Dijk
research-article2018
Article
The first-level digital divide
shifts from inequalities in
physical access to inequalities
in material access
new media & society
1–22
© The Author(s) 2018
Article reuse guidelines:
sagepub.com/journals-permissions
https://doi.org/10.1177/1461444818797082
DOI: 10.1177/1461444818797082
journals.sagepub.com/home/nms
Alexander JAM van Deursen
and Jan AGM van Dijk
University of Twente, The Netherlands
Abstract
For a long time, a common opinion among policy-makers was that the digital divide
problem would be solved when a country’s Internet connection rate reaches saturation.
However, scholars of the second-level digital divide have concluded that the divides in
Internet skills and type of use continue to expand even after physical access is universal.
This study—based on an online survey among a representative sample of the Dutch
population—indicates that the first-level digital divide remains a problem in one of
the richest and most technologically advanced countries in the world. By extending
basic physical access combined with material access, the study finds that a diversity
in access to devices and peripherals, device-related opportunities, and the ongoing
expenses required to maintain the hardware, software, and subscriptions affect existing
inequalities related to Internet skills, uses, and outcomes.
Keywords
Device diversity, devices, digital divide, digital inequality, first-level digital divide,
Internet access, Internet outcomes, Internet skills, material access, peripherals
Introduction
As of 2018, the diffusion of the Internet has reached as high as 95% in several countries (worldbank.org). In countries with such high connection rates, the Internet has
Corresponding author:
Alexander JAM van Deursen, Department of Communication Science, University of Twente, PO Box 217,
7500 AE Enschede, The Netherlands.
Email: a.j.a.m.vandeursen@utwente.nl
2
new media & society 00(0)
become a basic utility for social inclusion. Arguments regarding the Internet and its
effects on social inclusion are reflected in the “digital divide” discourse, with the
premise being that Internet access provides benefits and that not having access to the
Internet has negative consequences. With Internet connection rates reaching high levels, a great deal of attention in digital divide research has moved away from indicators
related to having an Internet connection (referred to as the first-level digital divide) to
indicators of the second-level digital divide (Hargittai, 2002), namely, Internet skills
and usage. More recently, attention on the outcomes of Internet use or tangible benefits
has increased, referred to as the third-level digital divide (Van Deursen and Helsper,
2015; Wei et al., 2011).
Because of the shift from the first-level digital divide in countries where fixed and
mobile broadband are nearly ubiquitous, there is the potential that new technology characteristics are being ignored (Sylvester et al., 2017). The first-level digital divide still
requires attention because, besides having or not having an Internet connection, differences in material access must be considered (Gonzales, 2016; Van Dijk, 2005). Material
access includes the means required to maintain the use of the Internet over time, such as
computer devices (e.g. desktops, tablets, Smart TVs), software (subscriptions), and
peripheral equipment (e.g. printers, additional hard drives). New material divides appear
as a result of rapidly changing technology, the large variety of devices available to the
general public, and the reality that not all of the materials provide the same online
opportunities.
The current contribution empirically investigates various factors related to material
access in a large-scale representative sample of the Dutch population. First, we consider
the devices used to access the Internet and the ownership of peripherals that make the use
of the Internet more convenient, such as additional screens and docking stations. To
obtain a general idea of who is using the various devices and peripherals, we link these
to important personal and positional categories and resources.
Second, we focus on specific aspects of material access, namely, device opportunity,
device and peripheral diversity, and maintenance expenses. Device opportunity relates to
the replacement of devices. For example, we consider those who only use smartphones
or tablets. Diversity relates to the total number of devices and total number of peripherals
used. Maintenance expenses are important to sustain the level of subscriptions and
devices (Gonzales et al., 2016). Accordingly, we examine whether these three aspects
differ among various segments of the populations.
Finally, we focus on the role of the three material access aspects in relation to the
second- and third-level digital divides. We are interested in whether they relate to
Internet skills, to the things people do online, and to the outcomes they achieve from
Internet use.
The following research questions are addressed:
1.
2.
3.
What factors explain the differences in the specific devices and peripherals used?
What factors explain device opportunity, device diversity, and maintenance
expense?
To what extent do device opportunity, device diversity, and maintenance expense
affect the diversity of Internet use and its outcomes?
van Deursen and van Dijk
3
Background
Material access in resources and appropriation theory
Resources and appropriation theory (Van Dijk, 2005) argues that categorical inequalities
in society produce an unequal distribution of resources and that an unequal distribution
of resources causes unequal access to the Internet. Access to the Internet is defined as a
process of appropriation that starts with general attitudes toward the Internet and
advances to having physical and material access. The process of appropriation is influenced by the social and technological context of Internet use. The social context consists
of personal and positional inequalities among users. Personal categorical inequalities
such as age and gender are frequently observed in digital divide research. This is also
true for the positional categorical inequalities such as job position, level of educational
attainment, and household size. Personal and positional inequalities, according to the
theory, produce different resources. For example, resources that are frequently considered in digital divide research, albeit under other labels such as economic, social, and
cultural capital, include possession, income, and access to a social network (for a review
of inequalities between persons, positions, and resources, see Scheerder et al., 2017).
According to the theory, the differences in Internet access that resources produce, reinforce inequalities of participation in society and this effect therefore reflects on greater
inequalities between persons, positions, and resources.
At the core of the resources and appropriation theory are the types of access to digital
technology. Accordingly, material access is part of the second phase, which extends the
basic physical access to computers and the Internet. It is this phase that is the focus of the
current study, and thus, it will be further elaborated upon, and the resources and personal
and positional categories will be linked to this phase. In the final section, the material
access phases are linked to the subsequent phases of Internet skills and Internet use and
to the tangible benefits that are derived from this enhanced access, such as the improvement of one’s position in society.
Material access
To study the material access divide in the Netherlands, we must first understand who is
using some specific materials. For example, the typical devices used for accessing the
Internet and going online are desktop and laptop computers, tablets, and smartphones.
Other devices that are increasingly used to access the Internet are game consoles and
Smart TVs. Furthermore, there is a large variety of peripheral equipment available,
among which the most common are printers, scanners, additional screens and hard
drives, and docking stations. Each of these additions support the user by offering a more
gratifying online experience.
Three aspects are important when focusing on material access inequalities, namely,
(1) differences in device opportunities, or the use replacement of a device by other
devices with different technical capacities; (2) differences in the diversity of devices and
peripherals; and (3) differences in the maintenance costs of devices and peripherals. All
three aspects are dependent on technical characteristics. The most studied differences in
technical capacities are the use of desktops and laptop computers versus smartphones
4
new media & society 00(0)
and tablets. Although the latter have advantages such as their mobility, convenience,
price, more continuous Internet use (Mossberger et al., 2012), and increasingly provide
additional possibilities besides location services, game playing, and video streaming,
they are not a substitute for desktop or laptop computers. They offer lower memory, storage capacity, and speed; less advanced applications (Akiyoshi and Ono, 2008; Mossberger
et al., 2012); and limited content availability (Napoli and Obar, 2014). Smartphones and
tablets also offer less control over Internet use as specific platforms embedded in closed
systems are used in which manufacturers have high levels of control (Napoli and Obar,
2014). They also provide less support as the platforms are often less compatible with
protocols and standards (Murphy et al., 2016). Their smaller screen sizes, greater scrolling requirements, and reduced functionality when typing result in an increased cognitive
burden (Bao et al., 2011; Murphy et al., 2016; Napoli and Obar, 2014). These differences
impact behavioral patterns and tendencies, and a broad array of Internet outcomes
(Napoli and Obar, 2014). Overall, Internet access on smartphones and tablets contributes
to diminished levels of user engagement, content creation, and information seeking
(Napoli and Obar, 2014). Desktop and laptop computers allow users to participate in a
greater breadth of activities (Davison and Cotton, 2003; Mossberger et al., 2012; Zickuhr
and Smith, 2012). Smartphones and tablets are used relatively often for leisure purposes
(gaming and social networking) and personal safety, and desktop and laptop computers
are more task- or work-based (Murphy et al., 2016; Pearce and Rice, 2013; Zillien and
Hargittai, 2009). Information seeking, for example, is more superficial when using a
smartphone or tablet, whereas searches on a desktop are much more immersive
(Humphreys et al., 2013; Isomursu et al., 2007).
Following the above reasoning, when considering differences in device opportunity,
some combinations of devices are less likely to be beneficial than others in providing a
wider variety of Internet uses and outcomes. Using only desktop or laptop computers
results in relatively few opportunities for continuous communication, entertainment, and
location-based activities, while smartphone and tablet users will only miss out on more
immersive Internet work in a relative sense, even as mobile devices further increase in
functionality. However, a common trend among younger generations and minority
groups is the replacement of desktops and laptops with tablets and smartphones. In terms
of digital inequality, these replacements may have fewer potential uses (Napoli and Obar,
2017), while the combination of desktops or laptops with smartphones or tablets offers a
broad range of opportunities. For this reason, those who only use smartphones or tablets
are referred to as a “mobile underclass” (Napoli and Obar, 2014).
The technical characteristics are also important when considering differences in
device and peripheral diversity. The use of more devices offers opportunities for a better
and more diverse Internet experience (Donner et al., 2011). The same is true with respect
to peripheral equipment in that the use of a greater variety of peripherals will lead to a
more satisfying Internet experience. Being able to print materials found online at home,
scan files to be uploaded, use an additional screen for multitasking and have access to an
additional hard drive for storing downloaded or created multimedia all provide specific
advantages.
The final aspect to consider is the maintenance expenses of devices and peripherals (Gonzales, 2016; Gonzales et al., 2016; Graham and Thrift, 2007). Devices can
van Deursen and van Dijk
5
become unstable, breakdown, or require software updates. Accordingly, the actual
purchase price of a device is only a small part of the total cost of ownership, as one
must consider the additional expenses associated with the hardware such as devices
and peripherals, repair costs, and software licensing costs. Even though the vast
majority of people in western countries are now able to access the Internet, these
numbers do not accurately reflect the ability to reliably maintain that access
(Gonzales, 2016).
Resources as a cause for differences in material access
For material access, economic resources are likely to play an important role, specifically,
the income required to acquire and sustain Internet use. We expect that people with high
incomes own a multitude of (the best) devices and peripherals. They own more desktop
and laptop computers, as well as more game consoles compared to those with lower
incomes (e.g. Jansen, 2010). People with low incomes are more likely to own secondhand devices and to experience malfunctioning hardware and software. Although most
of the lower economic groups now use Internet technology, access is unstable and characterized by frequent periods of disconnection (Gonzales, 2016). Similar situations may
occur in relation to Internet subscriptions, which are likely to be better and more expensive among those with higher incomes. Those with lower income are more likely to only
have Internet access on their smartphone, whereas those with higher incomes have
Internet access on smartphones and other devices such as laptops and desktops (Tsetsi
and Rains, 2017). We hypothesize the following:
H1a. Income contributes positively to device opportunity, device diversity, peripheral
diversity, and device maintenance.
In addition to income, we expect social resources to play a role in material access.
Social support is a general indication of social capital—the accumulated sum of
mutual acquaintances—that due to its durability, is a structural resource embodied
by one’s social network (Bourdieu and Wacquant, 1992). When devices or peripherals are purchased, social resources enable the receipt of support and maintenance in
the event of a malfunction. People with access to other Internet users in their own
household are also in a better position to share devices, peripherals, and subscriptions. In this respect, a lack of social support can have a negative impact on the adoption and purchase of devices and peripherals. Hence, the availability of support
(quantity) and the level of satisfaction with the support received (quality) are dependent on the available resources (Helsper and Van Deursen, 2017). We hypothesize the
following:
H1b. Support quantity contributes positively to device opportunity, device diversity,
peripheral diversity, and device maintenance.
H1c. Support quality contributes positively to device opportunity, device diversity,
peripheral diversity, and device maintenance.
6
new media & society 00(0)
Personal and positional inequalities as causes for differences in material
access
Digital divide research suggests that the most-observed personal categories affecting
Internet access are gender, age, and ethnicity (Scheerder et al., 2017). In the developed
world, the use of the Internet by both men and women continually increases as they are
faced with the same digital technology at work, at school, and at home (Van Dijk, 2005).
Nonetheless, gender remains a significant factor impacting Internet use over time in
countries with high Internet penetration (Helsper and Reisdorf, 2017). Gender-based stereotypes suggest that women are less technologically competent (Dholakia and Chiang,
2003). Furthermore, while it has been argued that women are at a natural disadvantage
because they purportedly lack technological skills and are more technophobic, there is
also the argument that the technology itself does not meet women’s needs (Hilbert,
2011). Research demonstrates that while both males and females access the Internet and
mobile devices, substantive differences exist in the particular devices they use (Odaci
and Kalkan, 2010). We hypothesize the following:
H2a. Men are more likely than women to have more access to device opportunities,
device diversity, peripheral diversity, and device maintenance.
With respect to age, young people have more and earlier access to all types of devices
that are used to connect to the Internet. For example, mobile communication is now a
condition in the everyday lives of young people (Ling, 2012), and as a result, they such
as new inventions and innovations and are far more likely to experiment with new
devices and peripherals. Accordingly, seniors remain the section of the population that is
the least likely to have access to the Internet, despite the benefits that such access affords
them (Baker, 2013). Think of transcending social and spatial barriers, maintaining and
enhancing social networks, and providing a greater sense of connection to the world
(Winstead et al., 2013). Unfortunately, a lack of interest, high costs, ergonomic impediments, complexity (Carpenter and Buday, 2007), knowledge (Gitlow, 2014), and anxiety
about new technologies and low technology self-efficacy (Lam and Lee, 2006; Vroman
et al., 2015) are all important barriers that interfere with the willingness of seniors to
adopt new technology. We hypothesize the following:
H2b. Age contributes negatively to device opportunity, device diversity, peripheral
diversity, and device maintenance.
In countries populated by minorities born in other countries (parents or children),
there are often significant gaps in Internet access between the majority population and
these minority groups. Related research on ethnicity, though not equivalent to immigration status, suggests that they are very active mobile device users, in contrast with the
relatively low levels of desktop and laptop ownership among minority groups (Cotton,
Anderson and Tufekci, 2009; Watkins, 2009).
H2c. Majorities are more likely than minorities to have access to device opportunity,
device diversity, peripheral diversity, and device maintenance.
van Deursen and van Dijk
7
Factors driving the distribution of resources and Internet access are positional categories,
the first of which fall under work and education. Physical access hinges on several resources
that include education and employment (Gonzales, 2016). Those who are more highly educated and who are employed are more digitally engaged, and benefit more from technology
in their everyday lives (Clayton and Macdonald, 2013). We hypothesize the following:
H2d. Those who are employed are more likely to have access to device opportunity,
device diversity, peripheral diversity, and device maintenance expenses than those
who are not employed.
H2e. Educational level of attainment contributes positively to device opportunity,
device diversity, peripheral diversity, and device maintenance.
Lower levels of social isolation (e.g. not living alone or being in a relationship)
improve one’s chances of engaging with the Internet (e.g. Van Deursen and Helsper,
2015). Marital status and household composition are often considered in digital divide
research. We expect that the presence of children affects material access, as they often
have their own devices for tasks such as school work. We hypothesize the following:
H2f. Married people or households with two or more adults are more likely than those
who are single, divorced, widowed, or living alone to have access to device opportunity, device diversity, peripheral diversity, and device maintenance.
H2g. Having children in the household contributes positively to device opportunity,
device diversity, peripheral diversity, and device maintenance.
Internet attitudes as a cause for inequalities in material access
The core of resource and appropriation theory describes the process of Internet appropriation in which having material access is a second phase. The first phase is having
motivational access. Van Dijk’s (2005) notion of motivational access is primarily shaped
by attitudes toward technology, which are crucial to using it (Venkatesh et al., 2003).
Negative attitudes such as computer anxiety, decrease the likelihood that an individual
will access the Internet (Dutton and Reisdorf, 2017; Reisdorf and Groselj, 2017; Van
Dijk, 2005). Internet anxiety is characterized by avoidance, by the expression of negative
comments about the Internet and its effects on society, and by attempts to minimize the
time spent using the Internet (Durndell and Haag, 2002; Rockwell and Singleton, 2002).
Thus, we hypothesize the following:
H3. Internet attitude is positively correlated with device opportunity, device diversity,
peripheral diversity, and device maintenance.
Material access as a cause for differences in skills, uses, and outcomes
The final objective is to test the role of the three aspects of material access in the total
process of appropriating digital technology. Prior research revealed that the quality of
8
new media & society 00(0)
computer equipment, the age of the computer used, and Internet pricing affect the
quality of Internet use at home (Zillien and Hargittai, 2009). Technology appropriation is a process with four successive phases or access types. After adopting a favorable attitude toward the Internet and acquiring physical and material access, one must
possess Internet skills (Van Dijk, 2005). Internet skills are known to vary considerable
among segments of the population (e.g. Hargittai et al., 2018; Van Deursen and Van
Dijk, 2011). In addition to Internet skills, Internet usage is part of the second-level
digital divide and can be defined in terms of frequency and the type of activities performed, both of which have their own grounds of determination (e.g. Blank and
Groselj, 2014; Van Deursen and Van Dijk, 2014). The process of Internet access eventually affects the outcomes of Internet use (Helsper, 2012; Van Dijk, 2005). We add
three types of material access to the discussed resources and positional and personal
determinants to predict the subsequent phases of Internet access—skills, uses, and
outcomes. The goal is to establish the independent effect of material access aspects on
the second- and third-level digital divide. We hypothesize the following:
H4. (a) Device opportunity, (b) device diversity, (c) peripheral diversity, and (d)
device maintenance contribute positively to Internet skills, Internet use diversity,
Internet use frequency, and Internet outcomes.
Method
Sample
The present study uses an online survey and draws on a sample collected in the
Netherlands. To obtain a representative sample of the Dutch population, we used
Panelclix, a professional organization for market research to provide a panel of approximately 110,000 people. Members of the panel receive a small incentive for every
survey they complete. In the Netherlands, 98% of the population uses the Internet;
therefore, the user population is very representative of the general population in terms
of its sociodemographic makeup. The panel includes novice and advanced Internet
users. In total, we aimed to obtain a dataset with approximately 1600 respondents over
the age of 18. Eventually, this resulted in the collection of 1698 responses over a
2-week period in February 2018. During the data collection period, amendments to the
sampling frame were made to ensure the representativity of the Dutch population.
Accordingly, the analyses revealed that the gender, age, and formal education of our
respondents largely matched official census data. As a result, only a small post hoc
correction was needed. The sample had the following composition: gender: 49.1%
female; age: M = 49.7, SD = 18.1; education: low (e.g. primary school) 32%, middle
(e.g. high school) 39%, high (e.g. college and university) 29%; occupation: employed
53%, unemployed 4%, disabled 7%, retired 23%, housewife/husband 6%, student 6%;
ethnicity: born in the Netherlands 95%, born outside the Netherlands 5%.
The online survey used software that checked for missing responses and then
prompted users to respond. The survey was pilot tested with 10 Internet users over two
rounds. Amendments were made based on the feedback provided. No major comments
van Deursen and van Dijk
9
were provided in the second round. The average time required to complete the survey
was 20 minutes.
Measures
To answer the first research question, the first set of dependent variables are the specific
devices and peripherals used. We included desktop, laptop, tablet, smartphone, Smart
TV, and game console. All are coded as dichotomous variables, indicating “no use” or
“use.” The same is done for the peripheral equipment used. We considered printer, scanner, additional screen, additional hard drive, and docking station.
To study device opportunity, we considered two conditions that both provide a limited
set of opportunities. The first condition consists of only using a desktop and/or laptop
(13%), and the second condition of only using a tablet and/or smartphone (6%).
To account for the diversity of devices and peripherals, we considered the total number of devices used to connect to the Internet (M = 2.94, SD = 1.35) and the total number
of peripherals used (M = 1.96, SD = 1.01).
The final dependent variable considered maintenance expenses and was measured by
the annual amount spent in euros on hardware and software (categorized as < 50, 51 to
100, 101 to 150, 151 to 200, and >200).
Material resources were incorporated as income and were measured by using the total
family income over the last 12 months, assessed on an 8-point scale. Based on the census
data, three categories were created—low, middle, and high income.
Social resources were represented by support quantity and was measured by asking
respondents how many people in the home use the Internet (M = 2.31, SD = 1.16). The
home or household is the most proximate context for support. Other contexts include
work, social networks, neighbors, and so on. The survey offered no data regarding these
contexts. Support quality was measured by asking the respondent to think about the last
time they asked for help when using the Internet, followed by the question “How much
confidence do you have that you can now solve the same problem yourself?” A 5-point
scale was used that ranged from “no confidence at all” to “high confidence” (M = 2.31,
SD = 1.16).
The first personal indicator considered was age. This was computed by subtracting
the reported year of birth from the survey year. Age was subsequently divided into four
categories 18–35, 36–50, 51–65, > 65). Gender was included as a dichotomous variable
(M/F), as was majority (born in the Netherlands) or minority (born outside the
Netherlands) status.
Several positional categories were considered. First, employment statuses were coded
as dummy variables, specifically, employed, retired, disabled, househusbands/wives,
unemployed, and students. Education was collected by degree as 1 of 10 categories and
followed the Dutch educational system classification. These data were subsequently
divided into groups of low, middle, and high educational level.
Internet attitude was measured by four items adapted from the digital motivation
scale of Helsper, Smirnova and Robinson (2017). All items are balanced for the direction
of response (5-point scale, M = 3.88, SD = 0.76). A sample statement included the following: ‘Technologies such as the Internet and mobile phones make life easier’.
10
new media & society 00(0)
Internet skills were measured using the Internet Skills Scale (Van Deursen et al., 2016).
This 20-item instrument covers operational, information navigation, social, and creative
components and is used in this study as a single skill set. All items were scored on a
5-point scale that ranged from “Not at all true of me” to “Very true of me” and exhibited
high internal consistency (α = .89; M = 3.66; SD = 0.73). Examples of items are “I know
how to open downloaded files,” ‘I find it hard to decide what the best keywords are to use
for online searches’, and ‘I know which information I should and shouldn’t share online’.
Internet use diversity was adapted from Van Deursen et al. (2017), and covers a range
of 36 different uses including economic (savings, earnings, employment, finances, and
degrees), cultural (identity and belonging), social (political and civic participation, and
bridging and bonding ties), and personal (health, leisure, and self-actualization) domains.
Respondents were asked to indicate how often they conducted each specific activity. A
“No use” answer was coded as a 0, and all other frequencies as 1. A composite measure
of Internet use diversity was created by summing the items (M = 18.82; SD = 5.94).
Internet use frequency was measured by asking respondents how often they use the
Internet, using a 5-point scale ranging from “less than monthly” to “multiple times a
day” (M = 2.31, SD = 1.16).
Internet outcomes was adapted from Van Deursen et al. (2017) and covers a range of
26 different outcomes related to the same domains as considered for Internet usage.
Respondents were asked, for each outcome, to what extent they agreed that they had
achieved or received that outcome (5-point scale). Agreeing and fully agreeing were
coded as 1, and all other options as 0. A composite measure was created by summing the
items (M = 5.91; SD = 4.42). Examples of items are “I save money by buying products
online’, ‘I have a better relationship with my friends and family because I use the
Internet’, and “Online entertainment made me feel happier.”
Analysis
Regarding the first research question, we used logistic regression analyses to identify the
significant predictors of the separate devices and peripherals considered. For the second
research question, we used linear regression analysis to analyze the contributors to device
diversity, peripheral diversity, and device maintenance expenses. Logistic regression was
used to analyze the contributors to the two conditions of device opportunity. In all analyses, we added the independent personal and positional variables, the resources, and
Internet attitude. For the final research question, we conducted linear regression analyses
where device opportunity, device diversity, and peripheral diversity were added to the
list as independent variables. Dependent variables included Internet skills, Internet use,
and Internet outcomes.
Results
Differences in devices and peripherals used
Among the Dutch population, the following devices are used to go online: desktop
(50%), laptop (70%), tablet (53%), smartphone (76%), Smart TV (26%), and game
11
van Deursen and van Dijk
Table 1. Logistic regression analysis to predict the use of devices (odds ratios).
Desktop
Gender (M/F)
0.63***
Age (ref. 18–35)
36–50
1.41*
51–65
2.14***
66 +
2.45**
Nationality
1.06
(NL/other)
Employment (ref. Empl.)
Unemployed
1.17
Retired
0.66
Disabled
0.62*
Househusbands/
0.59*
wives
Student
0.91
Education (ref. low)
Middle
0.96
High
1.12
Children (N/Y)
1.12
Marital status (ref. Mar’d.)
Single
1.00
Widow
0.72
Divorced
0.98
Income
1.02
Support quantity
1.07
Support quality
0.93
Internet attitude
1.08
Constant
1.17
.06
Nagelkerke R2
Chi-square
73.42***
Laptop
Tablet
Smartphone
Smart TV
Game
console
0.95
1.34**
1.27
0.75*
0.31***
0.87
0.67*
0.79
0.87
1.71**
1.59**
1.62
0.52**
0.62
0.30***
0.16***
0.56*
0.60**
0.31***
0.23***
1.04
0.69
0.20***
0.04**
0.97
0.50*
0.78
0.69
0.63*
0.53*
0.82
0.40***
0.84
0.68
0.74
0.91
0.60
0.60
0.96
0.89
0.48*
0.96
0.73
1.51
1.08
1.71
0.94
0.48
0.90
2.07*
1.00
1.52**
0.58*
1.09
1.23
1.15
1.14
0.97
0.67
0.96
0.85
0.93
1.79*
1.24
1.49
0.98
0.62
1.09
1.36***
0.95
1.14
1.24**
0.21*
.13
161.71***
1.44
0.00
1.14
0.96
1.07
1.50**
1.21
0.06*
.29
273.32***
1.13
1.44
1.14
1.06
1.06
1.17**
1.24**
0.64
.08
95.49***
0.94
1.87**
0.68
1.09
0.66*
2.07**
1.35***
1.71***
1.12
1.43***
1.08
1.24***
1.39***
1.62***
0.11***
0.15**
.11
.32
149.30*** 406.48***
*p < .05, **p < .01, ***p < .001.
console (12%). Table 1 indicates that those between 18 and 36 years are less likely to use
desktops compared to the older age groups and are also less likely to use tablets compared to those between 36 and 65 years old. Furthermore, those between 18 and 36 years
are more likely to use smartphones, Smart TVs, and game consoles. In addition, Table 1
reveals that men are more likely to use desktop computers, Smart TVs and game consoles to go online, while women are more inclined to use tablets. Majorities are more
likely to use tablets and smartphones than minorities.
Table 1 indicates that compared to those who are employed, those who are unemployed are less likely to use laptops and tablets; those who are disabled are less likely to
use a desktop or tablet computer, and househusbands/wives are less likely to use
12
new media & society 00(0)
desktops, laptops, and Smart TVs. Students are more likely to use game consoles. The
level of educational attainment affects the use of laptops, which is higher for highly educated individuals than it is for those with lower levels of education, whereas the use of
game consoles to access the Internet is more common among those who are somewhat
(mid-level) educated compared to those who have a low level of education. Those with
children in the household are less likely to use a laptop than those without children, and
widowers and divorced people are more likely to use a smartphone than those who are
married. The latter are less likely to use a tablet.
Income positively affects the use of tablets, smartphones, and Smart TVs. Support
quantity positively contributes to the use of smartphones, and support quality is positively related to the use of laptops, smartphones, and game consoles. Finally, Internet
attitude positively affects the use of laptops, tablets, smartphones, and Smart TVs for
online access.
The following peripherals are used: printer (89%), scanner (50%), additional screen
(10%), hard drive (37%), and docking station (6%). Table 2 indicates that personal inequalities emerge for gender, age, and majority status. Men are more likely to use additional screens. Compared to those aged 16–35, older age groups are more likely to use
printers and scanners. Those aged over 65 are less likely to use additional screens, and
those aged over 50 are less likely to use additional hard drives. Minorities are less likely
to use additional screens and hard drives.
Compared to those who are employed, those who report a disability are less likely to
use additional screens, and students are more likely to use scanners. Those with higher
levels of education are more likely to use scanners, additional screens, or docking stations compared to those who have lower educational qualifications, while those who are
somewhat well-educated are more inclined to use additional screens.
Finally, income positively contributes to the use of additional hard drives and docking
stations, while support quantity affects the use of printers and additional hard drives.
Support quality positively impacts the use of scanners, additional screens, and additional
hard drives. Similarly, Internet attitude positively affects the use of scanners, additional
screens, and additional hard drives.
Device opportunity. With respect to device opportunity, on one hand, we considered people
who only use desktop and/or laptop computers, and on the other hand, people who only
use a tablet and/or smartphone. Both situations result in a limited variety of uses, and thus
provide fewer opportunities compared to situations in which desktop and/or laptop computers are combined with smartphones and/or tablets. Table 3 reveals that people aged
over 50 are more likely to only use desktop and/or laptop computers, while the youngest
age group is more likely to only use a smartphone and/or tablet. People in the Dutch
majority, lower income, less access to support, and higher levels of Internet attitude groups
are less likely to only use desktop and/or laptops. Internet attitude also contributes negatively to only using smartphone and/or tablets. Finally, compared to employed people,
those who report a disability are more likely to suffer from lower device opportunity.
Device and peripheral diversity. With respect to the diversity of devices used to access
the Internet, users on average connect to the Internet with 2.94 devices (SD = 1.35),
13
van Deursen and van Dijk
Table 2. Logistic regression analysis to predict the use of peripherals (odds ratios).
Printer
Gender (M/F)
0.90
Age (ref. 18–35)
36–50
2.15***
51–65
2.91***
66 +
5.17*
Nationality
0.56
(NL/other)
Employment (ref. Empl.)
Unemployed
1.23
Retired
1.97
Disabled
0.87
Househusbands/
1.92
wives
Student
1.61
Education (ref. low)
Middle
1.02
High
1.22
Children (N/Y)
0.64
Marital status (ref. Mar’d.)
Single
0.91
Widow
0.71
Divorced
0.70
Income
0.80
Support quantity
1.47***
Support quality
0.99
Internet attitude
0.98
Constant
5.29*
.11
Nagelkerke R2
Chi-square
97.95***
Scanner
Additional
screen
Additional
hard drive
Docking
station
0.83
0.46***
1.27
0.76
1.55**
2.39***
2.78**
1.12
1.18
0.69
0.19**
0.18*
0.62
0.30***
0.16***
0.56*
0.69
0.55
0.33
0.44
1.65
0.92
1.23
0.90
1.55
1.70
0.28*
1.01
0.68
0.74
0.91
0.60
0.71
0.36
0.58
0.57
2.34**
1.68
0.48
0.90
0.98
1.43*
0.75
1.89*
2.63***
0.60
1.14
0.97
0.67
1.38
2.31*
0.78
1.87**
1.09
2.07**
1.71***
1.43***
1.24***
1.62***
0.15**
.10
124.63***
1.14
0.80
0.92
1.43*
0.87
1.29
1.20
0.03**
.13
76.25***
0.78
0.58*
0.73
1.09
1.00
1.20**
1.26***
0.12***
.06
100.12***
0.98
0.09
0.74
0.92
0.95
1.33*
1.62***
0.08*
.17
145.23***
*p < .05, **p < .01, ***p < .001.
and they use 1.93 peripherals (SD = 0.98). Columns two and three in Table 4 indicate
the predictors of device and peripheral diversity. Whereas both device diversity and
peripheral diversity are higher among men, age is a negative contributor in device
diversity but a positive factor in peripherals diversity. Furthermore, both device and
peripheral diversity are greater among citizens of the Dutch majority. Compared to
the employed, househusbands/wives use fewer devices to connect to the Internet, and
students use more peripherals. Those with higher levels of education have higher
device and peripheral diversity than those with low levels of education, and married
people exhibit greater peripheral diversity than do widowers and people with no children in the household. Finally, income, support quality, support quantity, and Internet
attitude contribute positively to both device and peripheral diversity.
14
new media & society 00(0)
Table 3. Logistic regression analysis to predict device opportunity (odds ratios).
Desktop and/or
laptop only users
Gender (M/F)
Age (ref. 18–35)
36–50
51–65
66 +
Majority/minority
Employment (ref. employed)
Unemployed
Retired
Disabled
Househusbands/wives
Student
Education (ref. low)
Middle
High
Children (N/Y)
Marital status (ref. married)
Single
Widow
Divorced
Income
Support quantity
Support quality
Internet attitude
Constant
Nagelkerke R2
Chi-square
Smartphone and/or
Tablet only users
0.78
1.56
1.69
3.42***
6.15***
2.28**
0.34**
0.34**
0.33*
1.37
1.92
1.32
1.85*
1.82
2.71
1.16
0.83
2.34*
2.30*
0.20*
0.85
0.85
1.79
0.98
0.51
2.74**
0.88
1.19
0.47*
0.59**
0.63***
0.92
0.55***
1.96
.27
261.98***
0.69
0.76
0.48
0.73
0.97
0.92
0.76*
0.37
.11
69.13***
p < .05, **p < .01, ***p < .00
Expenses for maintenance. The last column in Table 4 shows the contributors to annual
maintenance expenses. For example, the table indicates that men have higher maintenance expenses than women (which might relate to the fact they are more likely to own
desktops, game consoles, and Smart TVs) and that employed people have more maintenance expenses than do unemployed and disabled individuals. Similarly, widowers have
higher expenses than married individuals and those in relationships. Finally, income,
support quantity and quality, and Internet attitude have positive effects on annual maintenance expenses.
Overview of the hypotheses test results (Table 5).
Effect of material access on skills, usage, and outcome diversity
Table 6 indicates that for Internet skills, in addition to the significant contributions of age,
employment, education, support quantity, support quality, and Internet attitude, there are
15
van Deursen and van Dijk
Table 4. Linear regression analysis to predict the diversity of devices and peripherals used, and
maintenance (annual expenses for hardware and software) (β).
Device
diversity
Gender (M/F)
Age (ref. 18–35)
36–50
51–65
66 +
Majority/minority
Employment (ref. employed)
Unemployed
Retired
Disabled
Househusbands/wives
Student
Education (ref. low)
Middle
High
Children (N/Y)
Marital status (ref. married)
Single
Widow
Divorced
Income
Support quantity
Support quality
Internet attitude
R2
F
Peripheral
diversity
Maintenance
expenses
−.08***
−.11***
−.10***
−.01
−.12***
−.14*
−.05*
.10**
.13***
.13*
−.05*
−.04
−.03
−.04
.03
−.07**
−.10*
−.09***
−.08***
.01
.04
.00
.00
.00
.10***
.02
.05
−.02
.05
−.02
.01
.12***
.07*
.09***
.16***
.22
24.29***
−.04
−.08
−.08**
−.04
.06*
.03
.15***
−.07**
−.01
.02
−.01
−.05
−.06*
−.04
.06*
.02
.13***
.11***
.11
9.97***
−.01
.05
−.02
.12***
.06*
.04
.13***
.10
9.37***
*p < .05, **p < .01, ***p < .001.
Table 5. Overview of hypotheses H1 to H3.
H1a income
H1b support quantity
H1c support quality
H2a Age
H2b Gender (m/f)
H2c Majority/minority
Device
Device
opportunity diversity
Peripheral
diversity
Maintenance
expenses
+
+
Ns
−/+
Ns
−
+
+
+
−
−
−
+
+
+
−
−
Ns
+
+
+
−
−
−
Supported
Supported
Partially supported
Partially supported
Partially supported
Partially supported
(Continued)
16
new media & society 00(0)
Table 5. (Continued)
H2d Employment
(employed vs rest)
H2e Education
H2f Marital status
(married vs rest)
H2 g Children
at home (n/y)
H3. Internet attitude
Device
Device
opportunity diversity
Peripheral
diversity
Maintenance
expenses
−
−
Ns
−
Partially supported
Ns
−
+
Ns
+
−
Ns
−
Partially supported
Partially supported
−
Ns
−
Ns
Rejected
+
+
+
+
Supported
+: significant positive contribution; –: significant negative contribution; Ns: no significant contribution.
also contributions from device opportunity (smartphone/laptop only), peripheral diversity,
and maintenance expenses. The next step is Internet usage, which entails both diversity and
frequency of use. With respect to diversity, besides the added Internet skills, it has also been
determined that device, device opportunity (smartphone/laptop only), device diversity, and
maintenance expenses contribute as expected. Furthermore, device opportunity (smartphone/laptop only) and maintenance expenses are positively related to use frequency. While
Internet skills and Internet use diversity are important contributors to gaining Internet outcomes, device diversity was found to make a positive, independent contribution.
Table 7 provides an overview of the hypotheses test results.
Discussion
Main findings
A common opinion among policy-makers is that the digital divide problem is solved when
a country’s Internet connection rate reaches saturation. However, scholars of the secondlevel digital divide have concluded that the divides in Internet skills and type of use continue to expand even after physical access is universal. Furthermore, several studies have
highlighted the continued importance of attitude and material dimensions (e.g. Gonzales,
2016; Hargittai and Kim, 2010; Napoli and Obar, 2014; Reisdorf and Groselj, 2017). This
study confirms the importance of both dimensions in the Netherlands, a rich and technologically advanced country where 98% of the population now has home Internet access.
More specifically, in the current study, material access was conceptualized as device
opportunity, device and peripheral diversity, and maintenance expenses, which are all
important dimensions that differ among segments of the population.
In light of the first two research questions, age, gender, majority status, employment,
education level, and household composition, as well as material resources (income),
social resources (quantity and quality of support), and Internet attitudes are all related
to unequal ownership of devices and peripherals. Furthermore, there are systematic differences in the material access dimensions, as these are strongly associated with Internet
17
van Deursen and van Dijk
Table 6. Linear regression analysis to predict internet skills, use diversity, use frequency, and
outcomes (β).
Internet
skills
Gender (M/F)
Age (ref. 18–35)
36–50
51–65
66 +
Majority/minority
Employment (ref. employed.)
Unemployed
Retired
Disabled
Househusbands/wives
Student
Education (ref. low)
Middle
High
Children (N/Y)
Marital status (ref. married)
Single
Widow
Divorced
Income
Support quantity
Support quality
Internet attitude
Desktop/laptop only
Smartphone/tablet only
Device diversity
Peripheral diversity
Maintenance expenses
Internet skills
Internet use diversity
Internet use frequency
R2
F
Internet use
diversity
Internet use
frequency
.05
Internet
outcomes
−.01
−.03
.03
−.06*
−.10**
−.21***
−.02
−.12***
−.24***
−.40***
.03
−.01
−.03
−.15*
−.04
−.16***
−.16***
−.17***
−.02
.01
−.04
.02
−.02
.06*
.00
.12**
.01
.02
.02
.00
.03
.06*
−.03
.03
.01
−.01
.01
−.02
−.05*
.07**
.13***
−.04*
.02
.06*
.02
.01
.03
−.03
.02
.03
.03
.02
−.03
−.00
.02
−.05*
.25***
.20***
−.07**
−.03
.04
.15**
.06*
−.06*
−.01
.03
.05*
.03
−.08**
.05**
−.05*
−.01
.07*
.01
.17***
.18***
.04
−.01
.06*
.11**
.01
.01
.07**
−.09**
−.03
.05
.04
.06*
.09**
.37
39.04***
.27
23.15***
.15
11.10***
.04
.05*
.01
.01
.03
.02
.30***
.04
−.01
.09***
.01
.00
.08**
.38***
.00
.44
47.00***
*p < .05, **p < .01, ***p < .001.
attitude, income, support quantity, and support quality. Furthermore, males, majorities,
and the employed, well-educated, and married people benefit most in terms of material
access. In most cases, this is also true for younger people. However, we also found that
younger people are more likely to only use a tablet or smartphone (while the overall
percentage is low), confirming the existence of an age-related “mobile underclass”
18
new media & society 00(0)
Table 7. Overview of hypotheses H4a to H4d.
H4a Device
opportunity
H4b Device
diversity
H4c Peripheral
diversity
H4d Maintenance
expenses
Internet
skills
Internet use
diversity
Internet use
frequency
Internet
outcomes
+
+
+
Ns
Ns
+
Ns
+
+
Ns
Ns
Ns
+
+
+
Ns
Partially
supported
Partially
supported
Partially
supported
Partially
supported
+: significant positive contribution; –: significant negative contribution; Ns: no significant contribution.
(Napoli and Obar, 2014) in the Netherlands. While mobile devices rapidly increase in
functionality, they should not be considered equal to desktop and laptop computers.
Both offer their own specific advantages, suggesting that using only mobile or nonmobile devices limits the potential range of Internet activities and outcomes. Overall, the
findings suggest that common digital divide indicators appear in our conceptualization
of material access, suggesting that attention to the first-level digital divide remains
important.
This becomes even more apparent in light of the third research question, which was
posed to better understand what these differences in material access imply for Internet
skills, use, and outcomes. The findings show that device opportunity is related to Internet
skills and Internet-use frequency and diversity. Using a higher diversity of devices is
related to a higher diversity of Internet use and more Internet outcomes. Furthermore,
having more maintenance expenses contributes to Internet skills, and Internet-use frequency and diversity. Thus, the first-level divides accounted for in the current study
contribute to second- and third-level divides. By extending basic physical access combined with material access, the study finds that the opportunities these devices offer,
diversity in the access to devices and peripherals, and ongoing expenses required to
maintain the hardware, software, and subscriptions all relate to inequalities in Internet
skills, uses, and outcomes.
Noteworthy is the role of Internet attitude, which is not only important for all material access variables but also for Internet skills and uses. This confirms that Internet
attitudes are stronger than many socioeconomic variables and that they are very important at the beginning of the process of Internet appropriation (Van Dijk, 2005). Internet
attitude should also be considered as part of the first-level digital divide—the psychological part. Apparently, attitudes toward the Internet are still as important as they were
two decades ago (see also Helsper and Reisdorf, 2017; Reisdorf and Groselj, 2017).
This study confirms that the digital divide is a reality that is here to stay, not only in
terms of attitude, skills, type of use, and Internet outcomes but also in terms of material
access. Accordingly, this also applies to rich and connected countries in the world.
Policy-makers who aim to reduce the digital divide must not neglect or forget about the
first level.
van Deursen and van Dijk
19
Limitations and future research
We provided a large-scale, empirical investigation of material access inequalities in the
Netherlands. Although the considered aspects provide a comprehensive view of material
access, we did not account for the quality of the devices under investigation. For example, processor speeds differ between laptops. Future studies may account for more technical details regarding the devices under investigation, as it is likely that differences
between tablet use, for example, are greater among population categories when the technical capacities of tablets are considered. Furthermore, given the rapid changes in smartphone capabilities, it is important to revisit questions of the digital divide with respect to
mobile technology (Hargittai and Kim, 2010). In this case, differences in device opportunity might decrease between devices. However, it could also mean that inequality in
the technical potential of devices is growing.
In this study, we have revealed that inequalities in material access are present in the
Netherlands. Structural differences appear when particular segments of the population
systematically and over longer periods of time take advantage of better devices and
peripherals, while others use devices that enable fewer opportunities. Future research
should determine whether there is growth or reduction in the multiple material access
gaps identified in this article, from a longitudinal perspective.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this
article.
References
Akiyoshi M and Ono H (2008) The diffusion of mobile Internet in Japan. The Information Society
24(5): 292–303.
Baker C (2013) A connection for all ages: enabling the benefits of high-speed Internet access for
older adults. Report, AARP Public Policy Institute, Washington, DC, vol. 79, issue 18.
Bao P, Pierce J, Whittaker S, et al. (2011) Smart phone use by non-mobile business users. In:
Proceedings of the 13th international conference on human computer interaction with mobile
devices and services, Stockholm, 30 August–2 September, pp. 445–454. New York: ACM.
Blank G and Groselj D (2014) Dimensions of Internet use: amount, variety, and types. Information,
Communication & Society 17(4): 417–435.
Bourdieu P and Wacquant LJ (1992) An Invitation to Reflexive Sociology. Chicago, IL: University
of Chicago Press.
Carpenter BD and Buday S (2007) Computer use among older adults in a naturally occurring
retirement community. Computers in Human Behavior 23(6): 3012–3024.
Clayton J and Macdonald SJ (2013) The limits of technology: social class, occupation and digital
inclusion in the city of Sunderland, England. Information, Communication & Society 16(6):
945–966.
Cotton SR, Anderson WA and Tufekco Z (2009) Old wine in a new technology, or a different type
of digital divide? New Media & Society 11(7): 1163–1186.
Davison E and Cotton S (2003) Connection discrepancies: unmasking further layers of the digital
divide. First Monday 8(3).
Dholakia RR and Chiang KP (2003) Shoppers in cyberspace: are they from Venus or Mars and
does it matter? Journal of Consumer Psychology 13(1–2): 171–176.
20
new media & society 00(0)
Donner J, Gitau S and Marsden G (2011) Exploring mobile-only Internet use: results of a training
study in urban South Africa. International Journal of Communication 5: 574–597.
Durndell A and Haag Z (2002) Computer self efficacy, computer anxiety, attitudes towards the
Internet and reported experience with the Internet, by gender, in an east European sample.
Computers in Human Behavior 18(5): 521–535.
Dutton WH and Reisdorf BC (2017) Cultural divides and digital inequalities: attitudes shaping
Internet and social media divides. Information, Communication & Society. Epub ahead of
print 19 July. DOI: 10.1080/1369118X.2017.1353640.
Gitlow L (2014) Technology use by older adults and barriers to using technology. Physical &
Occupational Therapy in Geriatrics 32(3): 271–280.
Gonzales AL (2016) The contemporary US digital divide: from initial access to technology maintenance. Information, Communication & Society 19(2): 234–248.
Gonzales AL, Ems L and Suri VR (2016) Cell phone disconnection disrupts access to healthcare
and health resources: a technology maintenance perspective. New Media & Society 18(8):
1422–1438.
Graham S and Thrift N (2007) Out of order: understanding repair and maintenance. Theory,
Culture, & Society 24(3): 1–25.
Hargittai E (2002) Second-level digital divide: differences in people’s online skills. First
Monday 7.
Hargittai E and Kim SJ (2010) The prevalence of smartphone use among a wired group of young
adults. Chicago: Institute for Policy Research Northwestern University.
Hargittai E, Piper AM and Morris MR (2018) From internet access to internet skills: digital inequality among older adults. Universal Access in the Information Society. Epub ahead of print
3 May. DOI: 10.1007/s10209-018-0617-5.
Helsper EJ (2012) A corresponding fields model for the links between social and digital exclusion.
Communication Theory 22(4): 403–426.
Helsper EJ and Reisdorf BC (2017) The emergence of a “digital underclass” in Great Britain and
Sweden: changing reasons for digital exclusion. New Media & Society 19(8): 1253–1270.
Helsper EJ and Van Deursen AJAM (2017) Do the rich get digitally richer? Quantity and quality
of support for digital engagement. Information, Communication & Society 20(5): 700–714.
Helsper EJ, Smirnova S and Robinson D (2017) Digital Reach Survey. Available at: http://www.
lse.ac.uk/media-and-communications/research/research-projects/disto/disto-youth.
Hilbert M (2011) Digital gender divide or technologically empowered women in developing countries? A typical case of lies, damned lies and statistics. Women’s Studies International Forum
34: 479–489.
Humphreys L, Von Pape T and Karnowski V (2013) Evolving mobile media: uses and conceptualizations of the mobile Internet. Journal of Computer-Mediated Communication 18(4): 491–507.
Isomursu P, Hinman R, Isomursu M, et al. (2007) Metaphors for the mobile Internet. Knowledge,
Technology & Policy 20(4): 259–268.
Jansen J (2010) Use of the Internet in higher-income households. Report, 24 November.
Washington, DC: Pew Research Center’s Internet & American Life Project.
Lam JC and Lee MK (2006) Digital inclusiveness—longitudinal study of Internet adoption by
older adults. Journal of Management Information Systems 22(4): 177–206.
Ling R (2012) Taken for Grantedness. The Embedding of Mobile Communication into Society.
Cambridge, MA: MIT Press.
Mossberger K, Tolbert CJ and Hamilton A (2012) Broadband adoption| measuring digital citizenship: mobile access and broadband. International Journal of Communication 6: 2492–2528.
Murphy HC, Chen M-M and Cossutta M (2016) An investigation of multiple devices and information sources used in the hotel booking process. Tourism Management 52: 44–51.
van Deursen and van Dijk
21
Napoli PM and Obar JA (2014) The emerging mobile Internet underclass: a critique of mobile
internet access. The Information Society 30(5): 323–334.
Napoli PM and Obar JA (2017) Second class netizens. In: Lind RA (ed.) Race and Gender in
Electronic Media: Content, Context, Culture. New York: Routledge, pp. 293–311.
Odaci H and Kalkan M (2010) Problematic Internet use, loneliness and dating anxiety among
young adult university students. Computers & Education 55(3): 1091–1097.
Pearce KE and Rice RE (2013) Digital divides from access to activities: comparing mobile and
personal computer Internet users. Journal of Communication 63(4): 721–744.
Reisdorf BC and Groselj D (2017) Internet (non-) use types and motivational access: implications
for digital inequalities research. New Media & Society 19(8): 1157–1176.
Rockwell SC and Singleton L (2002) The effects of computer anxiety and communication apprehension on the adoption and utilization of the internet. Electronic Journal of Communication
12(1–2).
Scheerder A, van Deursen AJAM and van Dijk JAGM (2017) Determinants of Internet skills, uses
and outcomes. A systematic review of the second- and third-level digital divide. Telematics
and Informatics 34(8): 1607–1624.
Sylvester A, Toland J and Parore P (2017) Is the digital divide still relevant in 2017? Two cases
from marginalised communities in Aotearoa-New Zealand. In: PACIS 2017 Proceedings,
123. Available at: http://aisel.aisnet.org/pacis2017/123
Tsetsi E and Rains SA (2017) Smartphone Internet access and use: extending the digital divide and
usage gap. Mobile Media & Communication 5(3): 239–255.
Van Deursen AJAM and Helsper EJ (2015) A nuanced understanding of Internet use and non-use
among the elderly. European Journal of Communication 30(2): 171–187.
Van Deursen AJAM and Van Dijk JAGM (2011) Internet skills and the digital divide. New Media
& Society 13(6): 893–911.
Van Deursen AJAM and Van Dijk JAGM (2014) The digital divide shifts to differences in usage.
New Media & Society 16(3): 507–526.
Van Deursen AJAM, Helsper EJ and Eynon R (2016) Development and validation of the Internet
skills scale (ISS). Information, Communication & Society 19(6): 804–823.
Van Deursen AJAM, Helsper EJ, Eynon R, et al. (2017) The compoundness and sequentiality of
digital inequality. International Journal of Communication 11(2017): 452–473.
Van Dijk JAGM (2005) The Deepening Divide. Thousand Oaks, CA: SAGE.
Venkatesh V, Morris MG, Davis GB, et al. (2003) User acceptance of information technology:
toward a unified view. MIS Quarterly 27(3): 425–478.
Vroman KG, Arthanat S and Lysack C (2015) “Who over 65 is online?” Older adults’ dispositions toward information communication technology. Computers in Human Behavior 43:
156–166.
Watkins SC (2009) The Young and the Digital: What the Migration to Social Network Sites,
Games, and Anytime, Anywhere Media Means for our Future. Boston, MA: Beacon Press.
Wei KK, Teo HH, Chan HC, et al. (2011) Conceptualizing and testing a social cognitive model of
the digital divide. Information Systems Research 22(1): 170–187.
Winstead V, Anderson WA, Yost EA, et al. (2013) You can teach an old dog new tricks: a qualitative analysis of how residents of senior living communities may use the web to overcome
spatial and social barriers. Journal of Applied Gerontology 32(5): 540–560.
Zickuhr K and Smith A (2012) Digital differences. Report, Pew, Washington, DC. Available at:
http://www.pewinternet.org/Reports/2012/Digital-differences.aspx
Zillien N and Hargittai E (2009) Digital distinction: status-specific types of Internet usage. Social
Science Quarterly 90(2): 274–291.
22
new media & society 00(0)
Author biography
Alexander JAM van Deursen is an associate professor at the Department of Communication
Science at the University of Twente in the Netherlands. Most of his research focuses on three lines
of research with the overarching theme of digital inequality. In a scientific manner he map barriers
of online engagement and explain differences in the outcomes that people get from engaging with
the Internet (of Things). Research projects Alexander leads are Digital Inequality in the Netherlands,
21st century digital skills in the creative industry, inequality in Internet of Things skills and studying digital inequality in the social context of the home. He is involved in the From Digital Skills to
Tangible Outcomes Project, holds Visiting Scholar positions at the London School of Economic
and Political Science and Arizona State University.
Jan AGM van Dijk is emeritus professor of Communication Science and the Sociology of the
Information Society at the University of Twente, the Netherlands. He is Director of the Center for
eGovernment Studies at the same university.