Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

Digital gender divide or technologically empowered

women in developing countries?
A typical case of lies, damned lies, and statistics
Martin Hilbert*
*University of Southern California (USC); United Nations Economic
Commission for Latin America and the Caribbean (ECLAC/CEPAL)

Published as:
Hilbert, M. (November). Digital gender divide or technologically
empowered women in developing countries? A typical case of
lies, damned lies, and statistics. Womens Studies International
Forum, 34(6), 479-489. http://dx.doi.org/10.1016/j.wsif.2011.07.001

ABSTRACT

The discussion about womens access to and use of digital Information and
Communication Technologies (ICT) in developing countries has been inconclusive so
far. Some claim that women are rather technophobic and that men are much better
users of digital tools, while others argue that women enthusiastically embrace digital
communication. This article puts this question to an empirical test. We analyze data sets
from 12 Latin American and 13 African countries from 2005-08. This is believed to be the
most extensive empirical study in this field so far. The results are surprisingly consistent
and revealing: the reason why fewer women access and use ICT is a direct result of
their unfavorable conditions with respect to employment, education and income. When
controlling for these variables, women turn out to be more active users of digital tools
than men. This turns the alleged digital gender divide into an opportunity: given womens
affinity for ICT, and given that digital technologies are tools that can improve living
conditions, ICT represent a concrete and tangible opportunity to tackle longstanding
challenges of gender inequalities in developing countries, including access to
employment, income, education and health services.

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

What is the digital divide? .............................................................................................................. 4

What do we know about the digital gender divide? ........................................................ 5
What about women in developing countries?.................................................................... 6
How can misleading statistics mask the reality about the gender divide? ............. 8
What does the data say? .................................................................................................................. 9
Controlling correlations in Latin America ........................................................................ 10
Women and ICT in Africa ........................................................................................................ 14
How do men and women use the Internet ....................................................................... 16
A word of caution on the presented statistics ................................................................. 18
A small change in mindset can sometimes make a large difference ........................... 19
References .............................................................................................................................................. 22

***The author would like to thank the support of Canadas International Development
Research Centre (IDRC), which has been the driving force behind the creation of
important statistics throughout the developing world for decades. Without the long-term
vision, dedication and trust of its professionals, like Ben Petrazzini, this, and many other
studies of this kind, would not exist.

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

During the second half of the last century, human kind has turned to the massive task of
making our bewildering store of knowledge more accessible (Bush, 1945). The result
has brought on irrevocable social, productive, political and cultural transformations,
which are based on a global communication infrastructure that includes innovations like
the Internet, mobile telephony and social networking applications in all shapes and sizes.
During the beginning of this new century, society at large is starting to embrace these
new tools, changing forever the way we communicate, coordinate our activities and
organize social interactions (Bell, 1973; Perez, 1983; Webster, 1995; Negroponte, 1995;
Castells, 1996; Freeman and Loua, 2001). At the core is the question of access to digital
networks, and, in particular, who gets empowered and who is informationally
marginalized by use of these new tools.
As a contribution to this ongoing discussion, this article analyzes the differences between
mens and womens access to and use of Information and Communication Technology
(ICT) in developing countries. We start with a literature review that shows that some see
digital technologies as practical and tangible tools for women to overcome longstanding
inequalities. ICT can help women to gain employment (for example through telework or
newly created information jobs), obtain cost-effective health services and education (such
as through online courses or software-based literacy programs) and to increase their
income (such as through e-business channels and online transactions). In contrast to this
glass-half-full outlook stands the pervasive and persistent counterargument that women
are at a natural disadvantage to benefit from the digital revolution because they are less
tech savvy, and more technophobic, and because the technology is not built for their
needs and intuition. If this were the case, the increasing socio-economic importance of
ICT would add a new dimension to the already existing vicious circle between
discrimination and womens backwardness, which can be expected to be particularly
severe in developing countries, where four out of five women live worldwide.
Unfortunately very few of the related studies control for potentially confounding
variables.1 We know that the lack of employment, income and education affect ICT usage
negatively (e.g. NTIA, 1999; Cullen, 2001; Warschauer, 2003; Mossberger, et.at., 2003;
OSILAC, 2007). We also know that women are discriminated against in many aspects of
social life, including employment, income and education. Given these potential
confounders, it is not clear if being a woman per se has a negative, neutral or positive
A notable exception for the case of the United States is Rice and Katz (2003), which, after control, do
not detect any significant digital divide between men and women. The author is not aware of any
controlled studies for the digital gender divide on the international level.
1

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

effect on ICT usage. In the first case, the digital revolution would pose a severe threat to
women. In the latter case, the increasing socio-economic importance of ICT would pose a
unique opportunity: the new tools would be a perfect tool to fight existing inequalities
between men and women.
To this point, lack of adequate statistical data had prevented us from testing this question
empirically. Arguments were often based on anecdotal evidence from case studies or
uncontrolled correlations, which sometimes lead to contradictory results. In recent years,
statistical institutes and academic research centers in the developing world have made a
significant effort to collect adequate statistical information. For this study we employ 25
datasets from 12 Latin American and 13 African countries (total of 1,176,816
observations), which allows us to execute a series of uncontrolled and controlled
empirical tests that will provide further insight into this unresolved question.

What is the digital divide?

The unfolding of the digital revolution is happening at unprecedented speed (for ICT
penetration rates during the past 15 years, see ITU, 2010). However fast, it is not
immediate and the related diffusion process follows the form of a well-known S-shaped
curve, which distinguishes between early adopters and latecomers (Rogers, 2003). While
this process unfolds, a new form of inequality is added to all the existing forms of
discrimination: an inequality in the power to communicate and to process information
digitally. The term digital divide has been coined to refer to this concept (e.g. NTIA,
1995, 1999; OECD, 2001).
Studies on the digital divide differ in their focus and methodological approach. Despite
their differences, all of them answer (part of) the following questions: who (individuals
vs. organizations/communities, vs. societies/countries/ world regions, etc.), with which
attributes (income, education, geography, age, gender, or type of ownership, size,
profitability, sector, etc.), connect how (pain access vs. usage vs. real impact), to what
kind of technology (phone, Internet, computer, digital TV, etc.) (see Hilbert, 2011). In
this article we test for one specific attribute of ICT users: their gender.
The main focus of this article is set on analyzing access to ICT in Latin America and
Africa, while we also sneak an exploratory peak into how men and women use the
Internet in Latin America. This is important because literature has shown that access and
usage foster the well-being in multiple aspects of life (e.g. Castells, 1996; Webster, 1995;

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

Waverman, et.al. 2005; Cimoli, et.al, 2010; Hilbert and Peres, 2010). Previous research
has shown that ICT adoption patterns are characterized by the same long established
determinants of inequality as other aspects of social life, such as those related to income,
education, skills, employment, geography, age and ethnicity, and gender, among others
(e.g. Cullen, 2001; Compaine, 2001; OECD, 2002; Warschauer, 2003; Mossberger, et.at.,
2003; van Dijk, 2005; OSILAC, 2007; Hilbert, 2010).

What do we know about the digital gender divide?

Let us begin with clarifying that most literature in this field refer to ICT access and usage
patterns among biologically identifiable men and women (sex), not the self-identified
gender identity of an individual, such as understood in the field of gender studies. While
it would be very interesting to explore the relationship between the digital divide and
gender identity, the paucity of data on the last variable forces us to follow most existing
research and equate gender with sex in this article.
During the 1990s, researchers were quick to observe that women tend to be latecomers to
the digital age (e.g. Dholakia, 1994; NTIA, 1999). As a consequence, the new technology
was popularly portrayed as a male domain (Badagliacco, 1990). Bimber (2000: 2)
concluded that the gap in ICT usage between women and men is the product of both
socioeconomic differences and some combination of underlying, gender-specific effects.
Researchers claimed that those gender-specific differences had their origins in the fact
that women underestimated their actual usage skills, which lead to lower self-efficacy to
use ICT (Busch, 1995; Joiner, et.al, 1996; Hargittai and Shafer, 2006), as well as in their
general attitudes toward computers (Shashaani, 1994). It was concluded that men are
more interested in technology than women, and they are also more tech savvy (Fallows,
2005: 5). In short, women were seen as being more likely to be technophobic and were
ascribed a certain computer anxiety. This type of reasoning is in line with a longstanding
argument that technology is gendered (Lohan and Faulkner, 2004; Puente, 2008). ICT are
seen as yet another toy for the boys (Faulkner, 2001).
As more statistics became available and Internet and mobile telephony penetration rates
began to rise, women started to catch up in many developed countries (Rice and
Katz, 2003). In the United States, most new users were women around the year 2000
(Cummings and Krout, 2002). Gender differences remained, but were smaller (Leggon,
2006) and mainly concentrated on marginalized groups, such as ethnic minorities
(Tolbert, et.al., 2007). However, once online, women remained less frequent and less
5

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

intense users of the Internet (Ono and Zavodny, 2003; Wasserman and RichmondAbbott, 2005). The focus of attention started to shift towards differences in how men and
women use ICT (Bonfadelli, 2002). For example, it was found that girls use the Internet
for instant messaging and chat-rooms, whereas boys downloaded games and music,
engaged in online trading, and created Web pages (Lenhart, Rainie, & Lewis, 2001;
Roberts and Foehr, 2004). Fallows (2005: 1) summarized a survey in the United States
with the conclusion: men like the internet for the experiences it offers, while women like
it for the human connections it promotes. As already mentioned, we will look at both
aspects in our subsequent analysis: access and usage, with a focus on the first one.
Statistical data from the USC led World Internet Project (2009) reconfirm these findings.
In Canada, 79% of men and 75% of women were online in 2007. This difference grows
to 56% to 46% for citizens of 60 years and older. The study also confirms differences in
usage. In 2004, Canadian men spent on average more time online than women (14.3 to
12.0 hours per week). This difference increased from 2.3 to 3.5 hours in 2007 (18.8 hours
to 15.3 hours). As the main reasons for non-usage, Australian women state lack of
interest (35%), not having a computer or Internet connection (26%) or lack of skills
(16%). The percentage of men to women who use the Internet is reported for the
following developed countries: Australia: 74% to 71%. Czech Republic: 55% to 46%;
Hungary: 45% to 39%; Israel: 71% to 64%; New Zealand: 78% to 77%; Singapore: 69%
to 54%; United Kingdom: 68% to 65%. The two exceptions to this trend seem to be
Sweden (with 75% of men online and 78% of women) and the United States (71% to
73%). However, even in these countries, men are more frequent and more intense users.
In 2008, men from the U.S. are more likely than women to surf the web at least daily
(54 to 41 percent) and men spend 1.5 hours more than women at their monitors reading.
In short, differences have become smaller in developed countries, but still remain,
especially in usage.

What about women and ICT in developing countries?

Due to the paucity of adequate statistics about the worlds poor, technology-related
research and respective policy-advice is often exclusively focusing on the roughly 20 %
of the world population living in the most industrialized countries2, while the remaining
In 2006, the countries member countries of the Organisation for Economic Co-operation and
Development (OECD), which represents the worlds industrialized countries, was home to 1,184
million inhabitants, within a world population of 6,555 million (18 %).
2

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

80 % of the global population is frequently ignored or inappropriately subsumed under

these findings. This is delicate, because living conditions, opportunities and threats differ
decisively in developed and developing countries. The vast majority of women live in
developing countries and they often suffer even more gender related discrimination than
their counterparts in developed countries. At the same time, if ICT were to hold a
promise to empower women, than this promise is much larger in the developing world,
given that the lower starting point provides for greater potential gains.

ICT: a threat for women

Similar to the above-cited data from developed countries (World Internet Project, 2009),
existing data from developing countries show that women are less likely than men to use
ICT. This leads related research to the conclusion that a digital gender divide clearly
exists and is a severe threat to women: In many countries such gaps become dramatic,
putting women at a significant disadvantage (Hafkin and Huyer, 2007: 33). Similar to
findings in developed countries, this divide applies to access and to the frequency and
intensity of usage (Park, 2009). Looking for reasons, researchers normally fall back on
anecdotal case studies and local evidence, which found that women face barriers that
include lack of access and training, and that they were confronted with software and
hardware applications that did not reflect their female interests and needs (Arun and
Arun, 2002; Ng and Mitter, 2005; Best and Maier, 2007). In this sense, the same
technophobic arguments that had been raised in the developed world during the 1990s,
have been transferred to women in the developing world in recent years. It is argued that
women have a negative attitude toward ICT (Varank, 2007) and that the introduction of
technologies has often implicitly been designed to meet the needs of men, not of women
(Basu, 2000; Hafkin, 2000).

ICT: an opportunity for women

In contrary to these findings, some case studies and anecdotal evidence show that ICT
can and are empowering women in developing countries. For example, ICT provide
women entrepreneurs with access to worldwide e-business channels, which and can be
operated 24 hours a day from home in real-time (Heeks, et.al. 2004; Schaefer Davis,
2007; Brodman and Berazneva, 2007). Ng and Mitter (2005) look beyond ICTs
contributions to economic well-being, and show how ICT are used by women for the
purposes of community building and political organization. ICT enable meaningful
7

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

participation and make female voices heard, as proven by the role of digital networks in
feminist movements (Harcourt, 1999). Others have argued that ICT have the potential to
completely redefine traditional gender roles, especially for women who have limited
skills or who lack the resources to invest in higher education (Kelkar and Nathan, 2002).
In short, ICT can be powerful tools for women to overcome discrimination, achieve full
equality, well-being and participation in the decisions that determine their lives and the
future of their communities. [] ICT [...] opens up a direct window for women to the
outside world. Information flows to them without distortion or any form of censoring,
and they have access to the same information as their counterparts (Sharma 2003: 1).
However, this potential to empower women in the developing world depends on access to
and actual usage of these technologies, which is a necessary first step (see e.g. Scott,
2001).

How can misleading statistics mask the reality about the gender
divide?

We have seen that the literature is inconclusive. We do not know if ICT are a severe
threat or an opportunity for women. What could be the reason for this apparent
contradiction?
There is a subtle message that can often be read between the lines of research related to
the digital gender divide. For example, Sharma (2003) points out that women have less
online access than men, for all the usual gender-related reasonstime, money, control,
learning opportunities, other commitments, prioritising others needs. Arguing that
longstanding gender-related inequalities are the reason for less usage is very different
from arguing that women are naturally technophobic. It has widely been measured that
women around the world are discriminated in fields like employment, income and
education (see e.g. Anand and Sen, 1995). It is therefore not clear if these existing
inequalities lead to the fact that women make less usage of ICT or, if being a woman per
se has a negative effect on ICT usage. This problem is well known in statistics and is
treated under the topic of so-called confounding variables (e.g. Freedman, et.al., 2007).
Often the confounder is easy to spot. For example, if somebody would realize that
childrens ICT usage is positively correlated to the size of their shoes, most people would
become suspicious and reason that age, and therefore literacy skills, might confound this

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

relation. There is no reason to believe that the shoe size of children with the same level of
schooling would make any significant difference. Often it is not as easy. But the cure
remains the same: as soon as there is a suspicion of confounding variables, it is wise to
control for them and to compare subjects on the same level of such variables. If the result
still makes a difference, it is more probable that the original variable has explanatory
power. If not, the confounder made the difference.3

What does the data say?

The most frequently analyzed statistics so far have been collected by telecommunication
administrative authorities and have been harmonized by the United Nations
Telecommunications Union (e.g. ITU, 2010). Traditionally, these administrative registers
collect the national aggregates of the numbers of subscriptions, connections and devices
and therefore do not allow detailed cross-tabulations with user attributes (like gender,
income, employment, education, etc). Those are provided by the household surveys that
we will use.
Our databases are both products of the initial seed funding of Canadas International
Development Research Centre (IDRC). In Latin America, IDRC has cooperated since
2002 with the United Nations Economic Commission for Latin America and the
Caribbean (ECLAC) to operate OSILAC (Observatory for the Information Society in
Latin America and the Caribbean)4. During the last decade, OSILAC has successfully
worked with National Statistics Offices all over the region to include ICT indicators in
existing household surveys. Given the large samples of official household surveys, this
data is very robust (all used sample sizes are between 21,000 and 408,000)5. Parts of these
databases are publicly available (OSILAC, 2009). In Africa, IDRC is cooperating with
the Research ICT Africa Network6, which has conducted their own household and
individual user surveys of ICT access and usage between 2007 and 2008 (sample sizes

This does not change the fact that children with larger shoes will have better ICT usage scores.
Same accounts for the case of ICT and women: might be that women use ICT less than men, but the
question is why: because they are women, or because of some other reason that come with being a
woman?
4 OSILAC: http://www.eclac.org/SocInfo/OSILAC/
5 Sample sizes: Brazil 2005: 408,148; Chile 2006: 268,873; Costa Rica 2005: 43,682; Ecuador 2006:
55,666; El Salvador 2006: 68,312; Honduras 2007: 100,028; Mexico 2007: 21,292; Nicaragua 2006:
40,190; Panama 2007: 48,295; Paraguay 2007: 21,053; Dominican Republic 2005: 20,610; Uruguay
2006: 64,164.
6 Research ICT Africa: http://www.researchICTafrica.net/
3

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

between 819 and 2,355).7 Despite the smaller sample size, this is nonetheless an
important effort, as Africa is normally considered a black hole for technology related
statistics (for overview of these surveys see Gillwald and Stork, 2008).

Controlling correlations in Latin America

Let us start with a series of simple correlations between gender and both, Internet usage
and mobile phone usage. It is important to underline that we use the question of active
usage by a specific person as an indicator for access, not the plain existence of equipment
in a household. We use the Pearson correlation coefficient to measure the degree of
association between two correlation coefficients (e.g. Williams and Monge, 2001;
Freedman, et.al., 2007), in our case between being a woman and using the Internet. We
code in a way that a negative correlations (r < 0) means that women use less Internet than
men, while r < 0 implies that Internet usage is positively correlated with being a woman.8
A first look at the upper two rows of Table 1 reveals that most of the correlations
between ICT usage and being a woman turn out to be negative. In agreement with
previous findings, the overall data show that women are less likely than men to use the
Internet or a mobile phone. In Brazil, for example, the regions largest country with over
a third of the regions GDP and population, being a women is negatively correlated with
using the Internet (with a correlation coefficient of r = -0.022), and with using a mobile
phone (with r = -0.029).9
In the following rows of Table 1 show two kinds of the inequalities between men and
women in percentage points. It is shown that there are real inequalities between men and
women regarding their working status (being employed or self-employed) and their
current attendance at an educational institution. Continuing with the example of Brazil,
92.2% of all men are actively working, compared to only 83.7% of women, and 31.6% of
Sample sizes: Benin: 1101; Botswana: 818; Cote dIvoire: 1112; Ethiopia: 2355; Ghana: 1092;
Kenya: 1461; Mozambique: 1131; Namibia: 885; Rwanda: 1078; Senegal: 1081; South Africa: 1771;
Tanzania: 1490; Uganda: 1127.
8 Pearsons r is normalized between +1 and -1, where +1 is a perfect positive association and -1 is a
perfect negative association. A correlation near zero indicates that there is no relationship between
the two variables.
9 A second look at the data reveals that the identified correlations are not very strong (even though
they are all statistically significant, weighted and stratified samples with p < .001). Squaring
Pearsons correlation coefficient tells us how much of the variation in ICT usage can be explained by
variation in being male or female. In all cases, less than 0.5% of the variation in ICT usage can be
explained by gender (for example, in Chile r2 = -0.0472 = 0.2%). There must be much more powerful
explanatory variables that determine ICT usage than gender.
7

10

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

all men currently attend an educational establishments, compared to 30.8% of all women.
These differences do not seem to be very large, but let us see what happens when we
control for them. This can be done with a partial correlation, which measures the degree
of association between two variables when the effects of a third variable are removed (see
e.g. Williams and Monge, 2001). To be more precise: what is the relationship between
being a women and ICT usage when the effect of work or current schooling is removed?
Table 1 shows that in the controlled environment, being a woman is positively correlated
with using the Internet (for Brazil r = +0.056) and with using a mobile phone (for Brazil r
= +0.033). While this correlation is still very low, it is striking that this turn-around effect
is consistent throughout almost all analyzed countries, which represent a very
heterogeneous group of socio-demographic and cultural societies. There are some
countries in which women are more active Internet users to begin with (Panama,
Honduras, Nicaragua) or more active mobile phone users (Dominican Republic, Panama,
Nicaragua), which of course naturally argues in favor of women being more active ICT
users to begin with. There are also cases in which women are not discriminated in the
fields of employment status (Nicaragua) or current attendance at an educational
establishment (Panama and Ecuador). This does not affect the logic of our result. The
overwhelming majority of the cases show that, when controlling for working and
educational enrollment conditions, women make more use of digital ICT than men. The
only exception in the 20 changes in tendency that can be observed in Table 1 is mobile
phone usage in Ecuador: the correlation coefficient becomes weaker in the controlled
test, but continues to stay negative (r = -0.037). This reminds us of the fact that social
science is not an exact science.

11

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

-.033

.020

.008

.004

-.026

Mobile use with being a

woman

-.004

-.029

n.a.

n.a.

n.a.

n.a.

-.044

.009

.011

-.029

.013

-.070

98.0

92.2

98.7

95.1

88.0

89.3

98.3

93.7

95.8

87.3

88.3

86.5

98.0
% of
Women

83.7

96.3

90.4

74.0

79.3

96.5

78.0

85.0

81.3

91.7

64.8

% of
Men

31.1

31.6

29.6

33.0

44.2

34.5

35.0

34.2

28.6

31.4

34.7

10.8

% of
28.2
Women

30.8

27.6

29.3

44.1

30.3

34.2

34.3

29.9

31.2

33.1

11.3

.050

.056

.047

.048

.088

.030

.047

.082

.148

.093

.066

.007

.039

.033

n.a.

n.a.

n.a.

n.a.

.019

.094

.139

.069

.085

-.037

Ecuador 2006

-.032

Nicaragua
2006

-.002/ -.023

Honduras
2007

-.045

Panama 2007

-.024

Dominican
Rep. 2005

Mexico 2007

-.022

Costa Rica
2005

Uruguay 2006

-.047

El Salvador
2006

Brazil 2005

Internet use with being a

woman

Paraguay
2007

Chile 2006

Table 1: Correlations and controlled correlations of gender with ICT usage in

Latin America; working and studying populations by men and women.

Correlation
coefficient, r

% of
Men
Actively
working
Real world
inequalities
(in %)
Attending
educational
establishment

Internet use
Correlation
coefficient, r:
being a woman,
controlled for
Mobile use
working and
assisting educ.
establishment

Source: own elaboration, based on OSILAC, 2009.

Let us dig deeper into this question and open up this statistical black box to see what
actually accounts for these results. The first two rows of Table 2 show the actual
percentages that lead to the previous finding that in most countries more men than
women use the Internet. Continuing with our example of Brazil, 22.0% of men use the
Internet compared to 20.2% of women. In mobile phone usage the divide is at 38.5% to
35.4%. In agreement with the results from Table 1, the notable exceptions for Internet
usage are Panama, Honduras and Nicaragua, and for cell phone users, Dominican
Republic, Panama and Nicaragua. The following rows show what happens if we put men
and women on equal footing regarding their working condition. We only consider men
and women who are either employed or self-employed, neglecting those who are
unemployed, retired or stay at home without salary. Based on this condition, it turns out
that in all countries more women than men use ICT actively, again with the sole
exception of mobile phone usage in Ecuador. In Brazil, only 22.8% of all working men
use the Internet, while 28.5% of all working women are online. Only 47.0% of all
Brazilian working men use a mobile phone, while 50.6% of all working women
telecommunicate on the go. The same general change in direction accounts for ICT when
12

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

controlled for current attendance at an educational establishment. Once in school, women

turn out to be more active users of digital opportunities (35.6% to 36.2% for Internet use
in Brazil; 32.5% to 39.9% for mobile usage). The exceptions to the general rule for
Internet usage are again Ecuador, as well as Costa Rica and Dominican Republic.
Generally speaking, the differences are much more pronounced for mobile phone usage
than for Internet use. Once set on equal footing in terms of employment and education,
women seem to embrace mobile voice communication quite a bit more enthusiastically
than men.

Nicaragua
2006

Honduras
2007

Dominican
Rep. 2005

39.6

22.0

30.5

24.3

11.3

5.6

23.4

17.1

22.8

9.7

11.6

7.9

Women

35.1

20.2

28.3

20.6

11.1

4.6

20.8

14.7

24.5

10.2

11.9

6.6

Men

54.2

38.2

n.a.

n.a.

n.a.

n.a

35.0

56.6

45.0

26.1

40.9

41.6

Women

53.8

35.4

n.a.

n.a.

n.a.

n.a

30.9

57.4

46.1

23.6

42.1

34.8

Men

31.0

22.8

30.9

21.0

10.6

4.6

24.0

17.0

21.1

8.9

10.7

7.3-

Women 36.6

28.5

37.3

25.8

15.1

5.7

30.4

24.7

37.2

15.4

15.1

8.2-

Men

68.9

47.0

n.a.

n.a

n.a.

n.a.

44.2

62.5

53.7

41.3

49.2

49.0

Women 73.1

50.6

n.a.

n.a

n.a.

n.a.

46.8

71.9

68.3

47.6

56.4

45.2

Men

70.2

35.6

49.8

39.2

19.9

11.5

35.8

32.6

42.4

16.9

19.1

26.8

Women 70.3

36.2

53.1

41.2

23.8

12.6

35.5

29.6

48.5

18.8

22.0

26.1

Men

39.1

32.5

n.a.

n.a.

n.a.

n.a.

29.4

60.7

37.4

13.6

37.7

58.9

Women 44.2

39.9

n.a.

n.a.

n.a.

n.a.

33.6

66.8

49.7

17.0

41.8

60.5

Men

14.3

11.7

13.6

7.9

3.3

1.4

7.9

3.8

6.7

1.9

0.7

n.a.

Women 16.4

14.2

16.4

10.2

4.4

1.5

9.4

5.5

11.5

2.9

0.8

n.a.

Men

14.5

14.8

15.4

9.8

4.6

2.0

12.3

7.8

9.8

3.8

4.3

n.a.

Women 18.1

17.8

18.3

12.9

5.1

2.8

15.5

13.2

19.9

6.3

6.6

n.a.

Men

0.3

1.6

0.5

0.8

n.a.

0.1

0.1

1.7

0.7

0.01

0.0

n.a.

Women 0.3

2.3

0.8

0.6

n.a.

0.01

0.1

1.2

1.3

0.01

0.1

n.a.

Men

6.9

4.2

12.3

7.0

3.5

0.9

9.3

7.1

6.9

5.7

5.8

n.a.

Women 7.6

4.3

14.9

6.9

5.8

1.0

12.1

7.5

10.3

10.1

7.7

n.a.

Men

n.a.

6.2

2.6

0.2

0.2

0.01

1.1

4.9

1.1

n.a.

0.2

n.a.

Women n.a.

7.4

2.8

0.4

0.5

0.01

1.2

5.7

1.5

n.a.

0.1

n.a.

Ecuador 2006

Men

Panama 2007

Mexico 2007

Costa Rica
2005

Uruguay 2006

El Salvador
2006

Brazil 2005

Mobile use

Chile 2006
Men and women
Men and women
attending educational actively working
establishment

Overall ICT
inequalities

Internet use

Paraguay
2007

Table 2: Percentage of man/women that use the Internet and own a mobile
phone in Latin America; place of Internet usage, Internet use frequency.

Internet use

Mobile phone

Internet use

Mobile use

Place of Internet usage, given that

the person is actively working

At home

At work

Communal
public access

Commercial
public access
Other
persons
home

Source: own elaboration, based on OSILAC, 2009.

13

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

These results seem to indicate that women are the more enthusiastic ICT users. However,
one could argue that this tendency originates in the fact that women are more likely to be
forced to use computers at work for unsophisticated and repetitive secretarial tasks (e.g.
Kaplan, 1994). In this case, force, not enthusiasm would be the reason for our results.
Saying it very bluntly, the argument would be that men at construction sites do not need
Internet access, while female secretaries are forced by their employers to execute trivial
typing jobs and routine office activities, such as banal word processing and spreadsheet
work. This general tendency could also affect ICT-enthusiasm in school, since girls and
boys often already anticipate their future job. While this sounds like a possible
hypothesis, this argument cannot explain the detected differences in mobile phone usage.
Besides, as shown by the lower rows in Table 2, working women do not only access the
Internet when forced to do so by their employers in their working environment, but
women are also more active online users at home, at public access centers or commercial
cyber cafes, and even at other peoples homes. Continuing with our example of Brazil,
Table 2 confirms that more working women use the Internet at their job than men (14.8%
to 17.8%), but at the same time more women also go online at home (11.7% to 14.2%) at
a communal access center (1.6% to 2.3%), a commercial public access center (4.2% to
4.3%) or at the home of family and friends (6.2% to 7.4%). While ICT access at work
might still have a catalyzing role, it can be seen that working women also make use of
their digital skill outside the working environment. Rather than being forced to ICT usage
against their will by an external force, it seems that women naturally enjoy the use of
digital communication wherever they get the opportunity to do so.

Women and ICT in Africa

Let us now compare uncontrolled and controlled usage rates in Africa (see Table 3).
Sample sizes are much smaller in these surveys and ICT usage rates in Africa are lower,
making it more difficult to detect differences. While all results of the weighted samples
turn out to be significant (weighted stratified samples with p < .01), they are less robust
than the ones from Latin America. Having said this, the general tendencies are the same.
In agreement with the traditional findings of literature, the overall correlation between
gender and ICT usage shows that in 11 of the 13 countries, a larger percentage of men
use the Internet than women (with the exception of Rwanda and Tanzania, in which
women already represent the larger share). In Kenya, for example, one of the larger and

14

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

technologically most advanced African countries of our sample, 21.1% of all men have
been online in 2007/8, while only 11.5% of all women use the Internet. 56.0% of all men
use a mobile phone, versus only 46.9% of all women.
Notwithstanding, the following rows of Table 3 show that, in general, African women are
also less literate10 (in Kenya 77.2% of men to 68.0% of women), and that fewer women
are actively working or studying (employed, self-employed or full-time student) (81.4%
of Kenyan men to 49.9% of women). Women also have less income (29.8% of all
Kenyan men belong to the top 25% income group of the country, while only 16.6% of all
women do).
On the basis of these characteristics, a new group was created. We will refer to it as
women on equal footing, simply for the sake of giving it a name. In this group we only
consider men and women who are literate, are actively working or studying and who
belong to the top 25% income group11. Controlling for these three inequalities, we can
see that the gender divide disappears in most African countries for women on equal
footing. In the case of Kenya, the divide in Internet usage is erased at 29.7% for both
men and women, while women on equal footing turn out to be more active mobile phone
users (90.0% to 92.7%). When placed on equal footing, the ratio of women versus men
turns around for Internet usage in four of the 13 analyzed countries (Namibia, Ethiopia,
Mozambique, Senegal). For another six countries, men continue to use the Internet more,
but the relative difference diminishes in all cases (South Africa, Benin, Botswana, Ghana,
Uganda, Cote dIvoire). For example, in South Africa, in the uncontrolled environment,
the share of men online is almost twice as large (20.2% of men to 11.3% of women),
while it shrinks to a difference of merely five percent for men and women on equal
footing (39.9/37.7 = 1.05).
This observed change in tendency is again much more pronounced for mobile phone
usage. In nine of the 13 countries, these controls turn the inequality around. With the
exception of Senegal and Tanzania, women on equal footing tend to embrace mobile
telephony more than men.

Literacy was defined by including all respondents that claimed to be able to read the newspaper
easily and to write a letter easily.
11 In the case of Africa it is necessary to focus on this high-income group of the top-25%, since income
levels in general are relatively low (in absolute terms) and ICT are tradable goods with prices levels
that are only accessible to segments that reach a certain absolute level of income (see Hilbert, 2010).
10

15

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

Kenya

Namibia

Ethiopia

Rwanda

Mozambique

Senegal

Tanzania

South Africa

Benin

Botswana

Ghana

Uganda

Cote dIvoire

Overall ICT
inequalities

Table 3: Percentage of man/women that use the Internet and own a mobile
phone; literacy, working and income inequalities; in Africa 2007/08.

Men

21.1

11.2

0.9

1.8

1.1

14.4

1.9

20.2

11.9

8.1

7.9

10.1

3.7

Women

11.5

7.2

0.4

2.1

0.9

6.7

2.3

11.3

5.3

4.0

3.2

4.0

1.1

Men

56.0

53.3

3.7

11.8

21.9

55.1

26.2

56.3

37.9

42.5

60.7

59.4

26.3

Women

46.9

45.4

2.5

7.5

32.4

26.2

17.6

64.9

20.5

37.9

57.2

58.9

12.2

Men

77.2

56.2

33.5

47.1

38.9

33.4

72.5

75.9

42.0

45.1

49.1

77.5

49.9

Women

68.0

58.5

26.5

40.0

23.5

19.9

67.7

74.4

21.2

38.0

39.8

72.5

25.2

Men

81.4

58.3

93.0

79.4

86.6

84.4

77.5

66.4

94.0

86.7

89.0

64.8

87.5

Women

49.8

42.5

32.3

62.8

35.8

53.6

53.9

38.5

49.7

51.6

80.7

43.4

47.0

Top 25% Men

income
Women

29.8

34.2

50.1

30.5

27.2

44.4

39.2

37.0

38.3

32.6

31.3

34.5

32.2

16.6

17.9

10.7

21.0

15.2

9.3

20.9

17.4

14.5

9.0

19.0

21.1

10.6

Men

25.3

21.4

13.9

16.5

13.8

18.1

25.1

28.5

18.1

14.8

15.0

31.1

21.7

Women

13.6

12.2

4.4

8.0

2.7

3.0

10.5

12.8

3.8

3.9

8.1

17.8

6.3

E
Men

29.7

26.9

3.8

6.2

2.8

31.0

4.0

39.9

27.6

23.9

26.7

23.9

13.5

Women

29.7

37.8

6.3

7.6

14.2

37.4

12.6

37.9

26.9

17.5

11.7

17.5

7.4

Men

90.0

90.3

18.7

39.9

57.7

91.1

56.4

89.9

86.9

84.4

82.3

84.4

62.6

Women

92.7

93.1

34.3

43.4

92.5

87.9

47.8

94.8

95.9

94.9

94.7

94.9

71.4

Internet
use

Mobile use

Man and women

on equal footing

Real world inequalities

Literate
Actively
working/
student

Equal
footing

Internet
use

Mobile
phone

Source: own elaboration, based on Research ICT Africa, 2008.

How do men and women use the Internet

As seen during the literature review, studies from developed countries reported that men
and women use the Internet for different ends, which can lead to diverse definition of the
digital divide. Let us now take a look at the kind of online services used in Latin
America. This is of particular interest because we have already seen that women welcome
the use of digital tools; therefore, the kinds of services they use might give us hints about
possible digital opportunities for women (Table 4). The first row confirms the previously
mentioned finding from developed countries that men seem to be more frequent online
users than women. This accounts for Chile, Uruguay, Costa Rica, Dominican Republic
and Nicaragua, while in Mexico and Honduras more women tend to be online every day.
When evaluating these statistics, we have to remember that usage frequency does not tell
us anything about the length of each session. Longer sessions could by far offset lower

16

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

frequency. Unfortunately the available statistics do not give us insight into the overall
intensity of usage.
When asked about the kinds of services used online, men reveal that they are much more
enthusiastic about using the Internet for entertainment reasons than women. When it
comes to using digital channels for education and training, the data is clear that women
tend to make much better use of the existing opportunities than men. This is especially
encouraging when considering the previously presented results of female disadvantages
in terms of literacy and educational attendance throughout the developing world (see
Tables 1 and 3). It shows that women already started to make use of the digital
opportunities to fight those existing inequalities.
Table 4 also shows that women still do not yet fully exploit many of the other
opportunities the digital world provides for them. Women are less enthusiastic about
applications of e-business and e-government. The use of e-business and online banking
channels could provide women with important steps to improve their financial
independence, while e-government services facilitate necessary, but often burdensome
interactions with public authorities. The use of the Internet for plain communication
purposes provides a mixed picture, as do the statistics on health services. Women from
Mexico and Dominican Republic are already using online networks to improve the health
conditions for themselves and those close to them. Overall, there still seems to be a large
potential to take advantage (or maybe create) adequate online content to improve living
conditions for women in Latin America.

17

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

Nicaragua
2006

Honduras
2007

Dominican
Rep. 2005

Costa Rica
2005

n.a.

25.2

34.3

n.a.

n.a.

35.8

36.7

n.a.

33.8

35.6

Women

34.3

n.a.

22.6

34.5

n.a.

n.a.

33.1

30.1

n.a.

34.2

29.0

Men

54.7

74.1

49.5

19.9

11.0

5.7

51.5

60.7

4.9

41.7

61.8

Women

50.5

67.2

34.5

14.2

4.9

1.9

43.2

51.1

1.9

32.8

56.1

Men

12.1

68.4

41.4

41.5

39.7

53.7

58.5

67.6

1.3

60.9

58.7

Women

12.4

75.0

46.5

44.9

49.3

65.0

66.8

72.5

1.8

63.0

62.3

Men

7.3

16.5

5.8

7.2

2.3

3.7

9.9

10.7

1.4

5.0

3.5

Women

5.5

10.8

2.6

3.3

1.2

2.5

5.7

5.7

1.2

3.3

2.2

Men

7.1

21.7

4.6

2.4

n.a.

2.2

21.8

14.8

0.8

n.a.

5.3

Women

5.6

16.4

3.0

1.2

n.a.

2.2

17.3

11.4

0.9

n.a.

4.3

Men

9.9

29.4

n.a.

3.4

n.a.

0.7

n.a.

13.2

0.5

n.a.

n.a.

Women

8.9

25.5

n.a.

2.3

n.a.

0.01

n.a.

9.2

0.2

n.a.

n.a.

Men

58.8

68.8

79.0

48.4

51.8

18.1

73.5

63.1

17.9

69.6

77.8

Women

60.2

68.5

81.1

49.5

55.7

13.9

74.3

55.0

18.4

71.5

77.6

Men

n.a.

n.a.

n.a.

6.3

1.8

1.8

n.a.

18.7

n.a.

n.a.

n.a.

Women

n.a.

n.a.

n.a.

8.9

0.9

1.5

n.a.

25.1

n.a.

n.a.

n.a.

Panama 2007

Mexico 2007

El Salvador
2006

Uruguay 2006

39.5

Brazil 2005

Men

Chile 2006
Internet users that use daily

Paraguay
2007

Table 4: Frequency of Internet usage; online service used by men and women in
Latin America.

Given that the person uses the Internet

Entertainment

Education and
training

Buying and
contracting

Online banking

Government
interaction

Communication

Health

Source: own elaboration, based on OSILAC, 2009.

A word of caution on the presented statistics

The humorist Mark Twain (1835-1910) has popularized the wisdom that there are three
kinds of lies: lies, damned lies, and statistics. This does not only apply to the statistical
practices of not controlling for confounding variables, such as criticized in this article,
but a word of caution is also in order when interpreting the statistics in the presented
Tables.
Given the large sample sizes of the weighted household samples, all results turn out to be
statistically significant. However, all results are based on stratified samples, meaning that
the survey organizers took the sample according to their knowledge about how the
population is distributed in a particular country. Once collected, stratified samples are
18

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

weighted according to the proportions of the actual society. The answers of one survey
correspondent from a more common socio-demographic group might be multiplied with a
factor much larger than a person from a minority. This weighting turns a small sample
into the representative of a large population. Nevertheless, it also affects significance
tests. The theory of significance test is based on random sampling, not on stratified
samples that are subsequently expanded. If the weighted number of cases exceeds the
sample size, tests of significance tend to be inflated, which is our case. Therefore, even
though all our results are statistically significant, meaning that it is very probable that the
observed differences between men and women are real and not just due to chance, these
tests are inflated.
As a consequence, results that are very close (such as 49.5% to 50.5% or a correlation of
0.008), have to be taken with a large grain of salt, as pure luck of sample drawing might
play us a trick here. Given the much smaller sample size in Africa than in Latin America,
the Latin American results are more reliable and stable than the African surveys (in the
Latin American samples each observation was weighted with hundreds of people on
average, while in Africa, factors of thousands were applied). Having said this,
differences for one country on the decimal level surely would not make a strong case by
itself. However, the consistency of our results across a large number of very
heterogeneous societies makes a relatively strong case: even though some results are
close and might be influenced by chance, they tend to show the same direction in general.
In other words, the presented results should be interpreted as a mutually confirming
whole, while specific results for particular countries might be subject to small variations.
Particularly close results from one specific country should not be used as a standalone
argument and might require more detailed sampling and further analysis.

A small change in mindset can sometimes make a large

difference
We have analyzed a very heterogeneous group of 25 countries, representing different
levels of development, geography, culture and social structure. According to the United
Nations Human Development Index (UNDP, 2008), Chile is the most developed country
of our sample, reaching a rank 40 of the 179 countries included in the 2008 Index.
Mozambique is the least developed with a rank of 175. Independent from these
differences, our results have been surprisingly consistent: ICT per se does not have
anything on them that might keep women and girls from using it in developing countries.
19

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

In fact, when controlled for existing inequalities, it shows that women embrace digital
technology more enthusiastically than men. One might be tempted to speculate that
women are simply better communicators and that therefore the use of these technologies
seems more intuitive for women than for men. Unfortunately, the presented data do not
tell us why women use ICT more than men; they just tell us that this is the case.
Notwithstanding, women continue to be discriminated in many other aspects of social
life, including employment, literacy and income. These inequalities also throw their
shadows on ICT usage. More specifically, being a woman is positively correlated with
ICT usage, and negatively correlated with employment, income and education (see
Tables 2 and 3). Uncontrolled correlations mix both effects, resulting in the fact that
underemployed, underpaid and undereducated women use ICT less than men. Traditional
discrimination in the fields of employment, income and education turn the positive
correlation between women and ICT into a negative one. At the same time, as shown
during the literature review, ICT have the potential to provide access to employment,
education and income. Therefore, ICT provide women with a bootstrapping opportunity
to pull themselves out of these unfavorable starting conditions. In other words, if woman
are provided with ICT, digital tools represent an opportunity for women to fight
longstanding inequalities.
The resulting logic is schematized in Figure 1. Traditionally, longstanding inequalities
prevent women from accessing ICT, leading to a vicious circle between digital exclusion,
unemployment, low income and lacking education. However, once having access to ICT,
this vicious circle can be turned into a virtuous circle, whereas the identified positive
attitudes of women toward ICT enable them to circumvent and fight existing inequalities.

20

WOMEN
less(2011)
Hilbert
employment

who have
ICT
less
less
Permitted Scholarly Posting of the Accepted Author Manuscript
income
education
FIGURE 1: Fighting longstanding discrimination with digital means

WOMEN
negative

employment

negative

income

ICT

negative

education
positive
positive

positive
Source: authors own elaboration.

This finding is by no means the end, but leads to the question of how to provide more
women with access to digital opportunities. For example, Table 2 indicates that
communal or commercial public access centers might be a viable option (see also Maeso
and Hilbert, 2006). Others have pointed to the need of regulations and incentives to
facilitate the actual usage of applications that would favor women, such as legislation to
promote telework (see e.g. Boiarov, 2008). Besides, the development of adequate content
becomes a major concern, especially in key areas such as education (see e.g. RELPE,
2008).
Summing up, the empirical evidence in this article argues for a re-thinking about women
and ICT usage. This rethinking should also affect policy making, which is unfortunately
still influenced by the superficial and unsustainable argument that women are
technophobic. For example, in the final declarations of the United Nations World Summit
on the Information Society (2003-2005), heads of States and governments have
recognized that a gender divide exists as part of the digital divide in society (WSIS,
2005) and declared a need for enhancing communication and media literacy for women
with a view to building the capacity of girls and women to understand and to develop ICT
content (WSIS, 2003). These statements seem to be based on the idea that women are
less digitally capable. Based on the results here presented, this is not at all the case and a
change in mindset seems appropriate. These policy statements should rather be
reformulated to something along the following lines: a digital gender divide exists only

21

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

as a direct reflection of existing gender-related inequalities and policy actions should

make use of the natural communication skills and media capacities of women and their
proven embrace of the new digital opportunities to overcome longstanding gender
inequalities. Such re-thinking is necessary to create policies and projects that truly allow
girls and women to become equal members of an information society, digital society,
network society, knowledge society, or simply equal members of society, independent
from the forename it may be given.

References
Anand, S., & Sen , A. (1995). Gender Inequality in Human Development. In Occasional Paper 19 (Vol.
1995). Presented at the Gender inequality in human development: theories and measurement,
United Nations.
Arun, S., & Arun, T. (2002). ICTs, gender and development: women in software production in Kerala.
Journal of International Development, 14(1), 39-50.
Badagliacco, J. M. (1990). Gender and Race Differences in Computing Attitudes and Experience. Social
Science Computer Review, 8(1), 42-63.
Basu, M. (2000). Feminist Perspectives and Gender Link in Development: The Critical Role of Women's
Organisations. The Ralph Bunche Institute on the United Nations. Retrieved August 17, 2009.
Bell, D. (1976). The Coming Of Post-industrial Society. Basic Books.
Best, M. L., & Maier, S. G. (2007). Gender, Culture and ICT Use in Rural South India. Gender Technology
and Development, 11(2), 137-155.
Bimber, B. (2000). Measuring the gender gap on the internet. Social Science Quarterly, 81(3), 86876.
Boiarov, S. (2008). Informe sobre legislacin y normativa vinculada al teletrabajo en Amrica Latina y el
Caribe. eLAC2007, goal 5. United Nations ECLAC, Information Society Programme. Retrieved
August 23, 2009, from
http://www.cepal.org/socinfo/noticias/noticias/2/32222/GdT_eLAC_meta_5.pdf.
Bonfadelli, H. (2002). The Internet and Knowledge Gaps: A Theoretical and Empirical Investigation.
European Journal of Communication, 17(1), 65-84..
Brodman, J., & Berazneva, J. (2007). Transformina Opportunities for Women Entrepreneurs. Information
Technologies and International Development, Special issue on Women's Empowerment and the
Information Society, 4(2), 3-10.
Busch, T. (1995). Gender differences in self-efficacy and attitudes toward computers. Journal of
Educational Computing Research, 12(2), 147-158.
Bush, V. (1945, July). As We May Think. The Atlantic Monthly, , 176(1). Retrieved August 17, 2009, from
http://www.theatlantic.com/doc/194507/bush.
Castells, M. (1996). The Rise of the Network Society. Blackwell Pub.
Cimoli, M., Hofman, A., & Mulder, N. (Eds.). (2010). The Impact of Information and Communication
Technologies in Latin America (In Association with the United Nations Economic Commission
for Latin America and the Caribbean.). Edward Elgar. Retrieved from http://www.eelgar.co.uk/Bookentry_Main.lasso?id=13996
Compaine, B. M. (2001). The Digital Divide: Facing a Crisis or Creating a Myth? (illustrated edition.).
The MIT Press.
Cullen, R. (2001). Addressing the digital divide. Online Information Review, 25(5), 311 - 320.

22

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

Cummings, J. N., & Kraut, R. (2002). Domesticating Computers and the Internet. The Information Society:
An International Journal, 18(3), 221.
Dholakia, R. R., Dholakia, N., & Pedersen, B. (1994). Putting a byte in the gender gap. American
Demographics, 16(12), 20.
Fallows, D. (2005). How Women and Men Use the Internet | Pew Internet & American Life Project. Pew
Internet and American Life Project. Retrieved August 16, 2009, from
http://www.pewinternet.org/Reports/2005/How-Women-and-Men-Use-the-Internet.aspx.
Faulkner, W. (2001). The technology question in feminism : : A view from feminist technology studies.
Womens Studies International Forum, 24(1), 79-95.
Freedman, D., Purves, R., & Pisani, R. (2007). Statistics, 4th Edition (4th ed.). W.W. Norton & Co.
Freeman, C., & Lou, F. (2001). As time goes by: From the Industrial Revolutions to the Information
Revolution (p. 407). Oxford University Press.
Gillwald, A., & Stork, C. (2008). Towards Evidence Based ICT Policy and Regulation: ICT access and
usage in Africa (Policy Paper 2 No. Vol. 1). ResearchICTAfrica (RIA). Retrieved from
http://www.researchictafrica.net.
Hafkin, N. (2000). Convergence of Concepts: Gender and ICTs in Africa. In E. Rathgeber & E. Ofwana
Adera (Eds.), Gender and the information revolution in Africa (pp. Chapter 1, document 4).
Ottawa: IDRC. Retrieved August 17, 2009, from http://www.idrc.ca/en/ev-32939-201-1DO_TOPIC.html.
Hafkin, N., & Huyer, S. (2007). Women and Gender in ICT Statistics and Indicators for Development.
Information Technologies and International Development, Special issue on Women's
Empowerment and the Information Society, 4(2), 25-41.
Harcourt, W. (Ed.). (1999). Women@Internet (p. 240). Palgrave Macmillan.
Hargittai, E., & Shafer, S. (2006). Differences in Actual and Perceived Online Skills: The Role of Gender.
Social Science Quarterly, 87(2), 432-448.
Heeks, R., Arun, S., & Morgan, S. (2004). Researching ICT-based enterprise for women in developing
countries: a gender perspective. In Women's ICT-Based Enterprise for Development project.
Institute for Development Policy and Management (IDPM), Manchester. Retrieved August 17,
2009, from http://www.womenictenterprise.org/LivelihoodsResearch.doc.
Hilbert, M. (2010). When is Cheap, Cheap Enough to Bridge the Digital Divide? Modeling Income Related
Structural Challenges of Technology Diffusion in Latin America. World Development, 38(5),
756-770.
Hilbert, M., & Peres, W. (Eds.). (2010). Information Societies in Latin America and the Caribbean:
Development of Technologies and Technologies for Development. ECLAC institutional books.
Santiago: United Nations Economic Commission for Latin America and the Caribbean. Retrieved
from http://www.cepal.org/SocInfo/
Hilbert, M. (2011). The manifold definitions of the digital divide and their implications for policy
responsibility. inedito.
ITU (International Telecommunication Union). (2010). Measuring the Information Society 2010 - The ICT
Development Index 2010. Geneva. Retrieved August 4, 2010, from http://www.itu.int/publ/DIND-ICTOI-2010/en.
Joiner, R., Messer, D., Littleton, K., & Light, P. (1996). Gender, computer experience and computer-based
problem solving. Computers & Education, 26(1-3), 179-187.
Kaplan, R. (1994). The gender gap at the PC keyboard. American Demographics, 16(1), 18.
Kelkar, G., & Nathan, D. (2002). Gender Relations and Technological Change in Asia. Current Sociology,
50(3), 427-441.

23

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

Leggon, C. (2006). Gender, race/ethnicity, and the digital divide. In M. F. Fox, D. G. Johnson, & S. V.
Rosser (Eds.), Women, gender, and technology (p. 204). University of Illinois Press.
Lenhart, A., Lewis, O., & Rainie, L. (2001). Teenage Life Online . Pew Internet & American Life Project.
Retrieved August 16, 2009, from http://www.pewinternet.org/Reports/2001/Teenage-LifeOnline.aspx.
Lohan, M., & Faulknet, W. (2004). Masculinities and Technologies: Some Introductory Remarks. Men and
Masculinities, 6(4), 319-329.
Maeso, O., & Hilbert, M. (2006). Centros de acceso pblico a las tecnologas de informacin y
comunicacin en Amrica Latina: caractersticas y desafos. United Nations, Santiago Chile:
United Nations ECLAC, Information Society Programme. Retrieved August 7, 2009, from
http://www.cepal.org/SocInfo/publicaciones
Maier, S., & Nair-Reichert , U. (2007). Empowering Women Through ICT-Based Business Initiatives: An
Overview of Best Practices in E-Commerce/E-Retailing Projects. Information Technologies and
International Development, Special issue on Women's Empowerment and the Information Society,
4(2), 43-60.
Mossberger, K., Tolbert, C. J., & Stansbury, M. (2003). Virtual Inequality: Beyond the Digital Divide (1st
ed.). Georgetown University Press.
Negroponte, N. (1995). Being digital (p. 255). Vintage Books.
Ng, C., & Mitter, M. S. (2005). Gender and the Digital Economy: Perspectives from the Developing World.
SAGE Publications.
NTIA (National Telecommunications and Information Administration). (1995). Falling through the Net: A
Survey of the "Have Nots" in Rural and Urban America. Retrieved August 2, 2009, from
http://www.ntia.doc.gov/ntiahome/fallingthru.html.
NTIA (National Telecommunications and Information Administration). (1999). Falling through the Net:
Defining the Digital Divide. A Report on the Telecommunications and Information Technology
Gap in America. Revised. U.S. Government Printing Office, Superintendent of Documents.
http://www.ntia.doc.gov/ntiahome/fttn99/contents.html.
OECD (Organization for Economic Co-operation and Development). (2001). Understanding the Digital
Divide. Paris, France: OECD (Organization for Economic Co-operation and Development).
OECD (Organization for Economic Co-operation and Development). (2002). Measuring the Information
Economy 2002 (p. 94). Paris, France: OECD (Organization for Economic Co-operation and
Development).
Ono, H., & Zavodny, M. (2003). Gender and the Internet. Social Science Quarterly, 84(1), 111-121.
OSILAC (Observatory for the Information Society in Latin America and the Caribbean). (2007).
Monitoring eLAC2007:Progress and current state of development of Latin American and
Caribbean information societies. United Nations, Santiago Chile: United Nations ECLAC,
Information Society Programme, OSILAC. Retrieved August 6, 2009, from
http://www.cepal.org/SocInfo/publicaciones
OSILAC (Observatory for the Information Society in Latin America and the Caribbean). (2009). ICT
Statistical Information System - OSILAC. Santiago Chile: United Nations ECLAC. Retrieved
August 5, 2009, from http://www.cepal.org/tic/flash/default.asp?idioma=IN.
Park, S. (2009). Concentration of internet usage and its relation to exposure to negative content: Does the
gender gap differ among adults and adolescents? Women's Studies International Forum, 32(2), 98107.
Perez, C. (1983). Structural change and assimilation of new Technologies in the economic and social
systems. online www.carlotaperez.org. Retrieved April 26, 2009, from
http://www.carlotaperez.org/papers/basic-structuralchange.htm#abstractcontents.

24

Hilbert (2011)

Permitted Scholarly Posting of the Accepted Author Manuscript

Puente, S. N. (2008). From cyberfeminism to technofeminism: From an essentialist perspective to social

cyberfeminism in certain feminist practices in Spain. Women's Studies International Forum, 31(6),
434-440.
RELPE, (Red Latinoamericana Portales Educativos) (2008). Analisis comparativo Portales Educativos
Latinoamericanos | RELPE (p. 35). RELPE. Retrieved January 20, 2010, from
http://www.relpe.org/relpe/documentos-tecnicos/analisis_comparativo_portales.php.
Rice, R. E., & Katz, J. E. (2003). Comparing internet and mobile phone usage: digital divides of usage,
adoption, and dropouts. Telecommunications Policy, 27(8-9), 597-623.
Roberts, D. F., Foehr, U. G., & Foundation, H. J. K. F. (2004). Kids and media in America (p. 380).
Cambridge University Press.
Rogers, E. M. (2003). Diffusion of Innovations, 5th Edition (5th ed.). Free Press.
Schaefer Davis, S. (2007). Empowering Women Weavers? The Internet in Rural Morocco. Information
Technologies and International Development, Special issue on Women's Empowerment and the
Information Society, 4(2), 17-23.
Scott, A. (2001). (In)Forming politics: Processes of feminist activism in the information age. Women's
Studies International Forum, 24(3-4), 409-421.
Sharma, U. (2003). Women Empowerment Through Information Technology. Authors Press.
Shashaani, L. (1994). Gender-differences in computer experience and its influence on computer attitudes.
Journal of Educational Computing Research, 11(4), 347-367.
Tolbert, C., Mossberger, K., King, B., & Miller, G. (2007). Are All American Women Making Progress
Online? AfricanAmericans and Latinas. Information Technologies and International
Development, Special issue on Women's Empowerment and the Information Society, 4(2), 61-88.
UNDP (United Nations Development Programme). (2008). Human Development Report 2007/8. United
Nations Development Programme (UNDP). Retrieved August 23, 2009, from
http://hdr.undp.org/en/reports/global/hdr2007-2008/.
van Dijk, J. A. (2005). The Deepening Divide: Inequality in the Information Society (1st ed.). Sage
Publications, Inc.
Varank, I. (2007). Effectiveness of Quantitative Skills, Qualitative Skills, and Gender in Determining
Computer Skills and Attitudes: A Causal Analysis. The Clearing House, 81(2), 71-80.
Waverman, L., Meschi, M., & Fuss, M. (2005). The Impact of Telecoms on Economic Growth in
Developing Countries | ICT Regulation Toolkit (ICT Regulation Toolkit). InfoDev, ITU.
Retrieved from http://www.ictregulationtoolkit.org/en/Publication.3532.html
Warschauer, M. (2003). Technology and Social Inclusion: Rethinking the Digital Divide. The MIT Press.
Wasserman, I. M., & Richmond-Abbott, M. (2005). Gender and the Internet: Causes of Variation in
Access, Level, and Scope of Use. Social Science Quarterly, 86(1), 252-270.
Webster, F. (1995). Theories of the information society (p. 304). Routledge.
Williams, F., & Monge, P. R. (2001). Reasoning With Statistics: How To Read Quantitative Research (5th
ed.). Wadsworth Publishing.
World Internet Project. (2009). World Internet international Report 2009. USC Annenberg School Center
for the Digital Future. Retrieved August 18, 2009, from http://store.digitalcenter.org/worldinternet-report.html.
WSIS (World Summit on the Information Society). (2003, December). WSIS: Plan of Action. United
Nations. Retrieved August 17, 2009, from http://www.itu.int/wsis/docs/geneva/official/poa.html.
WSIS (World Summit on the Information Society). (2005, November). Tunis Commitment. United
Nations. Retrieved August 17, 2009, from http://www.itu.int/wsis/docs2/tunis/off/7.html.

25