See discussions, stats, and author profiles for this publication at: [Link]

net/publication/304215685

The context of ICT4D and development in rural areas: a case study of South
Africa

Conference Paper · September 2015

CITATION READS

1 4,643

4 authors, including:

Tapfuma Pashapa Ulrike Rivett

University of Cape Town University of Cape Town
2 PUBLICATIONS 14 CITATIONS 71 PUBLICATIONS 770 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Tapfuma Pashapa on 21 June 2016.

The user has requested enhancement of the downloaded file.

 The context of ICT4D and development
in rural areas: a case study of South
Africa
Tapfuma Pashapa
University of Cape Town, Department of Information Systems, iCOMMS
South Africa

A/Prof Ulrike Rivett

University of Cape Town, Department of Information Systems, iCOMMS
South Africa

Abstract
Rural places are predominantly described as remote and sparsely populated areas outside
urban centers where agriculture is the main economic activity. This paper reflects on the
rurality classifications used by Statistics South Africa by exploring the link between
improved public service delivery and ICT proliferation. The services referred to in the study
are water, sanitation, education and health care. The study is based on a literature review
and a statistical analysis on the central theme of ICTs and access to improved service
delivery in South African rural places. Logistic regression models were fitted to the 2014
South African General Household Survey (GHS) data to investigate the significance of
variables which determine access to Internet in rural households. The study showed that
the education level of the household head and household income are the most influential
factors in the determination of access to household Internet particularly in rural farm areas,
while female headed households are more likely to have access to the Internet in both
rural farm areas and rural tribal areas. The study also showed the significance of access to
household electricity as a determinant of access to the Internet in tribal areas.

Keywords
ICTs, service delivery, rurality

Introduction
South Africa’s Department of Rural Development and Land Reform (2009) describe rural
South African areas as sparsely populated places where agriculture is a prominent activity
and economic activities are focused on the production of food and raw materials.
As is the case in other developing countries, South African rural places are associated with
poor service delivery (water, sanitation, electricity, education and health) because of
economic disadvantages to providing services in vast areas with low population density.
On the other hand, studies have shown that ICTs can be used to bridge distances and
support development by introducing services in areas with sparse populations.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
96
 Established information and communications technologies (ICTs) include radio
and television, while new ICTs relate specifically to mobile phones and the
Internet...ICT represents the mobile network, and mobile tools used for data
collection and analysis, and the technology (hardware, software and services)
that expedites the data flow
(Schouten 2013).
Heeks (2009) states that ICTs are the basis for development because they provide
information without which development is not possible.
ICTs are quickly changing relationships, facilitating the measurement and monitoring of
interventions, and enabling practitioners at a local level to use evidence to guide decision
making for the equitable and sustainable extension of water, sanitation and hygiene
services (Schouten 2013). This paper presents the findings of a study investigating the
context of ICT proliferation and access to improved service delivery in South African rural
places.

Study aim
Rurality is the extent to which an area bares the characteristics of rural places defined in a
country. The concept of rurality changes as the defining features of rural places evolve with
time. Emerging and improving ICTs characterize the current South African rural context.
There have been various debates about how ICTs stimulate development in Africa. For
example, some authors argue that there can be no development without information, ICTs
improve the planning and management in development projects, and that information
acquired through ICTs is important for development (Heeks (2009), Chapman & Slaymaker
(2002)). On the other hand, Langmia (2005) argues that in a country like South Africa, rural
communities cannot benefit from ICTs before the infrastructure which supports ICTs are
set up.
This study aimed at investigating the context of rural development in South Africa by
analyzing literature and statistical data in order to identify the context of rural development
and ICTs. The variable used to define development was the provision of public services.

Methods
The study used the methods of a systematic literature review and a statistical analysis.
The literature review focused on publications about information and communication
technologies for development (ICT4D) in South Africa. The literature searches were done
using the following academic databases: Google Scholar, JSTOR, Mendeley and
Worldcat. The searches were limited to the themes, “ICT4D”, “service delivery” and “South
African rurality”. Only literature published after the year 2000 was considered. Literature
reviews were also carried out using the South African government publications, which
focus on rural development. For example, the National Development Plan 2030, the
comprehensive rural development programme (CRDP) and its predecessor the integrated
sustainable rural development plan (ISRDP).
The literature review was supported by analyzing descriptive statistics on the 2011 South
Africa Census, 2014 General Household Survey (GHS) data and the 2001 to 2013 South
Africa Afrobarometer data sets. Logistics regression models were used to investigate the
determinants of access to the Internet in rural households. Other methods such as the

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
97
discriminant analysis were not used because the data did not pass diagnosis tests. As
such, logistic regression modelling is best suited to the study because it has less strict
data assumptions. Logistic Regression has been used in a similar study by Askar (2015) to
predict the use of ICTs for task-related teaching. Equation 1 represents the model which
was fitted to the 2014 GHS data.
logit (pi) = α + β1x1 + β2x2 + β3 + … + βkxk + ei,Equation 1
Where x1, x2…xk represent the k number of variables determining access to the Internet
α is a constant,
β1, β2, β3,… βk are the coefficients
and ei is an error term for each estimation.
pi is the probability of access to the internet for ith scenario of the variables. Hence,
logit (pi) represents the logarithm of the odds of event pi.
The odds of the event occurring is given by the exponent of [α + β1x1 + β2x2 + β3 +
…+ βkxk]. Therefore, the larger the value of any given βk value, the larger the odds
of event pi occurring when the effects of other variables are held constant (the
converse is true).
Access to the Internet at home was chosen as the dependent variable with the response
variables being: age of household head, education level of household head, annual
household income and gender of head. The statistical analysis was done by using STATA
statistical package and Excel Spreadsheets.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
98
 Figure 1. Structure of study

Household income and the gender of the household head are taken to represent
measures of the household wellbeing. Access to electricity is important because it
supports access and use of the Internet. Education also determines the ability of
individuals to use and accept ICTs. Budlender (2003) outlines that household head
variables are sometimes used as household indicators because there are no alternatives.
For example, the education level of the household.
Figure 1 shows a diagrammatic representation of the structure of the study. First, an
analysis of ICTs and service delivery in rural places was done in order to understand the
differences of the characteristics of South African rural places. The study investigated the
various types of rural places as identified by sector classification frameworks and their
representativeness of the current rural context. In stage 2, the theme of ICT as a current
agent of rural development was explored. In stage 3, statistical analysis of the primary
data sources was carried out to complement the literature review. Finally, the findings of
the study were discussed in stage 4.

South African rurality

Rurality is sometimes used as a basis for allocating development resources. The nature
and character of rural places changes over time. As such, it is important to investigate the

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
99
nature of rural places consistently (Prieto-Lara & Ocaña-Riola 2010). ICTs can stimulate
development because they improve the connectivity between rural and urban areas. ICTs
also provide platforms on which services can be provided.

Urban-rural classifications in South Africa

Laldaparsad (2011) defines three types of areas used by Statistics South Africa, namely:
rural farm areas, rural tribal areas and urban areas.
Laldaparsad states that rural farm areas were described by the following characteristics in
the 2011 South African census:
 coloured as the main race group of the household head
 persons with some primary schooling
 households using water from borehole
 low income persons
 households using gas, paraffin, wood and coal as the main source of energy for
cooking
Tribal areas were defined by
 african as the main race group of household head
 households using animal dung or solar as the main source of energy for cooking
 households using river spring water as the main water supply
 persons living in a hut or traditional dwelling
Urban areas where defined by
 persons with complete primary and secondary education
 households accessing water from water schemes operated by municipality or other
private providers
 relatively smaller number of children ever born per household
 high population density
 households with flush toilets connected to sewer, septic tank, pit with ventilation or
bucket toilets
 higher income persons
 persons living in a formal house or shack in squatter camp/backyard
The two types of rural places are characterized by poor service provision. That is, farmland
areas are characterized by households which use borehole water and the use of gas,
paraffin or coal as the main source of energy for cooking. The inhabitants of tribal lands
typically use spring water as the main water supply and they dwell in traditional structures.
In the initial urban-rural classifications of places by the Department of Water and Sanitation
(DWS), former homeland areas where classified as rural. DWS classifications were based
on population size and density, historical background (homeland and township) and
economic activities.
Other departments use the integrated sustainable rural development program (ISRDP)
development nodes as a basis for determining ruralness. ISRDP nodes consist of areas
ranging from large towns, farming districts and ‘deep’ or ‘very rural’ local municipalities
(Jacobs & Hart 2012). The ISRDP was developed to contribute towards poverty reduction
by ensuring that all programmes implemented in a district operate in an integrated manner
(Public Service Commission 2010). However, Schmidt (2012) points out that the selection

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
100
of the nodes was influenced by political considerations such as the equitable
implementation of rural allowances amongst provinces.
ISRDP nodes are for example used by the Department of Health to identify rural places
where rural allowances are used as a strategy to attract health [Link] is despite
the fact that the nodes were designed on the basis of poverty indicators and not
specifically the shortage of health personnel. This highlights the need to understand the
current South African rural context and to update existing area classifications in relation to
the current context of ICT proliferation.

ICTs and service delivery in rural South Africa

Limited telephone line infrastructure is a major barrier to connecting remote places to
Internet services. Langmia (2005) highlights this by asking a critical question, “How can
technology progress be possible when the infrastructure that go along with the Internet like
basic telephone lines are still a distance dream to a vast majority of the African
population?” Besides the challenge of telephone lines, aspects such as the availability of
electricity impact on the success of ICTs. Electricity is another important precondition for
the effectiveness of ICTs. Langmia refers to fitting indigenous South African languages into
ICT systems in order to create relevance of ICTs for indigenous population groups. The
most popular mechanism for communication in rural South Africa is the cell phone. The
2011 South Africa Census data shows that 84 percent of the households in rural areas
(both tribal and farming) have access to cell phones. However, the widespread use of cell
phones does not necessarily translate to Internet connectivity or development. The ability
of people to use a cell phone for more than communication is seen as defining factor for
development. For instance, Brown and Molla (2005) explain that mobile banking in rural
South Africa was less successful due to the lack of skills and negative perceptions by cell
phone users.
ICTs are seen to be a driving force of the modern globalization by connecting people and
businesses across vast geographical areas (Borghoff 2011). Langmia (2005) notes that
globalization pushed African countries to digitalize radio and television broadcast systems
as a means of connecting to the whole world. South Africa responded by digitalizing
television and radio transmission. As per requirement by the international
telecommunications union (ITU), all countries where required to make the transition from
analogue to radio and television digital broadcasting systems. Digitalization improved the
quality of television and radio reception as well as the number of channels offered by
South Africa Broadcasting Commission (Langmia 2005).
Locally, the South African development plan by National Planning Commission (2012)
explicitly outlines the importance of building ICT infrastructure to support economic
development and connecting schools, hospital and individuals to affordable information
services. The National Planning Commission’s strategic plan is aided by programs such as
the Digital Doorway, which was designed to provide rural and disadvantaged communities
with computer equipment and allow them to experience computers and other ICTs without
formal learning (Mulder et al. 2008).
Other ICT projects include setting up of multi-purpose community centers (MPCCs) or
telecenters in rural and disadvantaged areas. MPCCs provide facilities which offer
computers, the Internet, telephones and other forms of ICTs to rural communities.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
101
 Telecenters were initiated in 1999 as one of the primary vehicles for the
implementation of development communication and information and to integrate
government services into primary rural communities. This was done to address
historical, social and economic factors, which limited access to information,
services and participation by citizens, as they had to travel long distances to
access these services
(Government Communication and Information Systems [GIS] 2014).

Among other uses, the centers provided ICTs for rural students enrolled to distance
learning institutions such as the University of South Africa (UNISA). MPCCs have
successfully provided ICTs to some rural communities. For example, Dlodlo (2009) cites
Moutse village in Mpumalanga, where villagers receive training on using ICTs at Ndlovu
medical clinic’s IT center. The area has a cellular network coverage of 90 percent and
most households in the village have access to electricity. Hence most households have a
television or a radio. The area also has a community radio station.

How are ICTs used in the provision of basic services?

ICTs are inherently designed to create connections through communication. In the context
of basic services, communities and service providers rely on each other and service
delivery can be improved by supporting aspects such as monitoring of infrastructure,
supporting feedback and integrating data collection. For instance, the Telehealth Project
for rural KwaZulu-Nata connects rural hospitals to specialists at University of KwaZulu
Natal’s Medical School (Mars (2008)). Wood et al. (2008) also cites the Western Cape’s
provincial health recording system, eKapa, which is used by health practitioners to keep
track of patients on TB and AIDS treatment.
In South Africa, cell phones are used to offer e-learning services such as Dr Maths and
M4Lit (Butgereit & Botha (2011), Vosloo, Deumert, & Town (2008)). Dr Maths was
developed by the council for scientific and industrial research (CSIR). It is an application
that is used on a social network platform called MXit. Dr Maths provides guidelines to
solve mathematics problems by providing an interface through which students interact with
tutors. M4Lit is used to encourage a reading and writing culture amongst leaners by
providing a platform to publish content in indigenous languages.
Cellular service providers in collaboration with financial institutions offer money transfer
and bill payment services by providing mobile banking and Internet banking facilities
through which customers can send and receive money or pay utility bills. The majority of
South Africa’s banks agree that online banking and cell phone transactions have been
growing at an impressive rate. ABSA’s cell phone banking customers [had] increased by
113 percent to 1.5 million, while Internet banking grew 11 percent to 1.1 million customers
(Mawson 2010).

Quantitative data analysis

South Africa’s rural population is shaped by a history of labor migration to cities and towns.
Since 1994, there has been a continuing flow of migrants from rural places to large urban
centers such as Johannesburg, Durban and Cape Town. Due to this, the population in
rural South Africa is characterized by a larger percentage of the aging population and a
relatively small percentage of the working population. In addition, age distributions differ

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
102
between people living in farm areas compared to those in tribal rural areas. The analysis is
based on data from the 2011 Census, 2014 GHS data and the 2001-2013 Afrobarometer
data. Statistics South Africa define two types of rural areas in the 2011 Census and the
2014 GHS. These are, tribal or farm areas in the Census data and tribal or formal rural
areas in the GHS data. Besides the difference in terminology, both surveys identify tribal
rural areas. Hence, formal rural areas in the GHS survey are taken to be the same as farm
areas in the 2011 Census data.

Case study of the Eastern Cape: service delivery in tribal and farm rural
areas
The Eastern Cape Province was selected as a case study because of its history of service
delivery backlogs and because it’s diversity in terms of geographic characteristics.
Amathole, Chris Hani, OR Tambo, Cacadu, Joe Ggabi and Alfred Nzo are districts in the
Eastern Cape Province. Cacadu district will not be used for district level area comparisons.
55 percent of the population in Eastern Cape lives in rural places. Of this percentage, 52
percent live in tribal areas, while the remaining three percent live in farm areas. The modal
household size in tribal areas is 3 people and 2 people for farm areas. 56 percent of the
households in tribal areas are female headed and 21 percent of the households in farm
areas are female headed. 15 percent of the population in tribal lands is employed
compared to the corresponding value of 75 percent on farm areas. As such, the median
income range is higher in farm lands compared to tribal lands.

Figure 2. Age distributions of farm and tribal land dwellers in the Eastern Cape (2011
Census)

Figure 2 shows the age distributions of people in farm and tribal areas of the Eastern Cape
Province according to the 2011 South African census data. The age distribution of the farm
land dwellers is characterized by a large percentage of people in the economically active
ages and such a distribution is expected to favor higher access to ICTs because of higher
income obtained from employment.

Comparison of access to ICTs and service delivery in the farm and tribal
areas of the Eastern Cape Province
Table 1 presents the percentages of Eastern Cape households with access to water,
sanitation, electricity, household goods and the Internet.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
103
 Province Amathole Chris Hani OR Tambo Cacadu Joe Gqabi Alfred Nzo

Tribal Farm Tribal Farm Tribal Farm Tribal Farm Tribal Farm Tribal Farm Tribal Farm
WATER
Pipe water 77.3 43.8 78.4 31.3 83.3
38.2 74.1 42.6 57.3 54.8 77.5 24.8 81.6
within 200m -
Operated by 17.1 36.5 4.8 23.1 4.0
30.2 20.4 36.5 9.4 41.1 9.8 17.0 33.4
WSP -
TOILETS
Basic Toilets 33.1 44.4 21.3 30.4 32.1 35.3 37.9 35.0 50.0 37.0 33.7 37.1 34.0
-
ELECTRICITY
Electricity for 63.1 53.1 49.1 24.6 51.1
40.9 54.0 48.0 43.5 48.0 55.1 37.6 24.8
cooking -
HOUSEHOLD
GOODS
Computer 2.2 16.9 2.3 12.7 1.9 12.9 2.3 7.5 16.5 2.4 13.7 38.9 50.1
-
Satellite TV 5.7 20.3 4.0 15.4 5.5 17.3 6.0 8.3 22.2 13.7 19.7 4.7 13.3
-
Television 47.2 57.1 51.2 53.0 49.1 54.4 46.6 19.9 63.0 41.9 47.4 38.9 50.1
-
Radio 55.0 66.1 59.7 70.8 55.9 69.1 49.4 47.3 63.1 57.6 63.8 54.6 56.4
-
Landline/phone 1.7 15.2 1.9 12.3 1.7 13.0 1.8 2.2 15.5 1.1 14.2 1.5 10.1
-
Cell phone 78.5 77.7 76.2 81.8 79.6 78.8 79.2 89.4 75.3 78.7 75.6 79.2 74.9
-
INTERNET
From home 0.8 7.9 0.7 6.0 0.7 6.0 1.0 2.5 7.9 0.8 8.0 0.8 4.4
-
From cell 8.9 10.3 6.1 10.7 8.6
10.5 9.2 9.7 11.4 10.3 6.9 11.6 9.7
phone -
From work 0.8 3.0 0.7 2.3 0.7 1.0 0.9 2.9 3.3 0.7 1.0 0.8 3.9
-
From 1.8 1.9 1.6 2.4 0.8
2.2 1.8 1.8 1.2 2.2 1.5 2.4 1.1
elsewhere -
No access to 21.9 13.7 16.6 14.8 17.7
14.4 21.8 12.8 20.9 13.8 15.5 15.9 16.1
internet -
Table 1. Percentage distribution of households with access to services and household
goods in Eastern Cape Province (Source: 2011 South Africa Census)

The table shows that farm communities have a higher percentage of households with
access to water within 200m of a homestead throughout the province although the data
show that tribal land dwellers are more likely to be connected to water sources operated
by water service providers. However, this disparity is largely explained by the fact that farm
dwellers draw their water from borehole (35.4 percent relative to 5.4 percent for tribal
dwellers). 37.3 percent of tribal dwellers draw their water from rivers or streams compared
to 5.5 percent for farm dwellers. Drawing water from a stream means that tribal land
dwellers are more likely to access unsafe drinking water. This pattern is consistent in the
five districts of the province.
Larger percentages of farm area dwellers have access to computers, satellite TV, radio
and telephone in the province except in OR Tambo district. The percentage of households
with access to the Internet is higher for farm households although Internet connectivity is
still low even among farm area dwellers since only 7.9 percent of the households have
access to the Internet from their homes. In sum, farm area dwellers are generally more

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
104
likely to have access to ICTs compared to tribal area dwellers.

Logistic regression for household access to ICTs in South African rural

households (2014 GHS data)
Table 2 and Table 3 show the coefficients of logistic regression models which predict the
likelihood of household access to the Internet in the tribal and formal rural areas of South
Africa.

Logistic regression No of obs 8122

Log likelihood = -4420.8679 LR chi2(5) 855.27

Prob>chi2 0.0000

Pseudo R2 0.0882

InternetAccess Odds Ratio Std. Err. Z P>|z| [95% Conf. Interval]

hhIcome 1.8403180 0.0524945 21.38 0.00 1.7402540 1.9461360

head_age 0.9781143 0.0016873 -12.83 0.00 0.9748129 0.9814269

Q527Access 0.4126833 0.0450309 -8.11 0.00 0.3332238 0.5110904

head_sex 1.2387660 0.6624460 4.00 0.00 1.1155010 1.3756510

educlevel 1.3693930 0.1064539 4.04 0.00 1.1758650 1.5947730

_cons 0.0149956 0.0043783 -14.38 0.00 0.0084613 0.0265762

Table 2. Logistic regression model coefficients for Internet household access in South
African tribal areas (2014 GHS)
Key:
InternetAccess is a dichotonous variable reflecting household access to Internet. i.e. 0 represents no access and 1
represents access. Internet access refers to connection within the household, library or community hall or telecenter,
school or university and Internet café within 2km of household
educlevel is a dichotonous variable such that 0 means no education and 1 means at least primary education
hhIcome is the logararithm of the monthly income
head_age is the age of the household head
head_sex is the gender of the household head such that 1 represents males and 2 represents females
_cons is a constant
Significant tests for the model and the coefficients: Tests are done at the 5 percent level of significance. The p-value
for the model is less than 0.025 and all the p-values for the four variables are less than 0.025. Therefore, all the variables
have significant influence on access to household Internet.

Logistic regression No of obs 1020

Log likelihood = -441.4252 LR chi2(5) 323.74

Prob>chi2 0.0000

Pseudo R2 0.2683

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
105
 InternetAccess Odds Ratio Std. Err. Z P>|z| [95% Conf. Interval]

hhIcome 3.5481850 0.3758620 11.96 0.00 2.8829540 4.3669140

educlevel 3.7370530 1.3389460 3.68 0.00 1.8516240 7.542332

Q527Access 0.2813089 0.0836002 -4.27 0.00 0.1571162 0.5036700

head_age 0.9776439 0.0062579 -3.53 0.00 0.9654552 0.9899864

head_sex 1.8283610 0.3700830 2.98 0.00 1.2296120 2.7186650

_cons 0.0000156 0.0000168 -10.30 0.00 0.0000002 0.0001284

Table 3. Logistic regression model coefficients for access to household Internet in South
African formal rural areas (2014 GHS)
Key:
InternetAccess is a dichotonous reflecting household access to Internet. i.e. 0 represents no access and 1 represents
access. Internet access refers to connection within the household, library or community hall or telecenter, school or
university, Internet café within 2km of household.
educlevel is a dichotonous variable such that 0 means no education and 1 means at least primary education.
hhIcome is the logararithm of the monthly income.
head_age is the age of the household head.
head_sex is the sex of the household head such that 1 represents males and 2 represents females
_cons is a constant
Significant tests for the model and the coefficients: Test are done at the 5 percent level of significance. The p-value
for the model is less than 0.025 and all the p-values for the four variables are less than 0.025. Therefore, all the variables
have significant influence on access to household Internet.

Household income, education status of household head and gender of household head all
increase the likelihood of access to household Internet in both types of areas although the
factors have a more pronounced effect in formal rural areas. Access to electricity has a
higher effect in tribal areas (coefficient of 0.4126833) compared to rural formal areas
(coefficient of 0.2813089). Household income and the education level of the household
head both have the highest effect on access to household Internet in both areas. However,
the effect of the two variables is higher in farm lands i.e. 1.840318 and 1.369393 for tribal
areas compared to the corresponding coefficient values of 3.548185 and 3.737053 in farm
areas. Age also has a positive significant effect on access to the Internet. The analysis
also shows that female headed households are more likely to have access to the Internet
although such a finding would require verification with other data.

Equipment or
No Access subscription Exposure TOTAL
No interest knowledge Elsewhere cost Concern
Tribal 32.5 47.7 0.6 19.0 0.2 100
Rural formal 37.3 44.6 1.0 16.9 0.3 100
Table 4. Percentage distribution of reasons for no access to the Internet in the tribal and
formal rural households of South Africa (2014 GHS)

Table 4 shows the distribution of the reasons for no access to household Internet in tribal
and rural areas as reflected by the GHS survey data of 2014. The table shows knowledge

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
106
and income as major reasons for not having access to household Internet as indicated in
the variables “no knowledge”, “no interest” in the Internet and lack of “equipment or
subscription cost”.

Distribution of rural people who received news over radio and TV, and
participation in political discussions

Figure 3. Percentage of rural people who received news through radio and TV, and
participation in political discussions (1999 to 2013 Afrobarameter data)

Figure 3 shows line graphs derived from Afrobarometer data. It shows the percentage of
people who accessed news by radio, television and those who reported participating in
political discussions over the years 2001 to 2013. Afrobarameter data are used to
investigate political perception and highlight how communities receive information about
governance. The graph shows an increase in the percentage of rural people who use
television. On the other had, the use of radio decreased probably as a result of emerging
technologies such as the Internet or people prefering television. Therefore, increasing
participation in political discussion can be explained by improved access to information
through radios, TVs and other forms of ICTs.

Discussion
In South Africa, farm areas are generally inhabited by higher income commercial farm
dwellers, while tribal areas are generally characterized by diverse challenges such as
poverty, poor service delivery and low literacy levels. As seen in the data analysis, in both
rural farm areas and formal rural areas, the likelihood of household Internet access is
largely determined by household income and the education level of the household head.
The Afrobarameter data showed increasing percentages of people receiving news through
radio and television in rural areas. This can be attributed to the improvement of radio and
television signaling by the digitalization of broadcast systems as revealed in the literature
review.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
107
ICT innovations are expected to impact rural development in different ways. For example,
the study revealed factors which determine ICT proliferation and use in rural areas. That is,
the pre-existence of service infrastructure such as phone lines for Internet connection and
electricity to operate ICT devices. The logistic regression analysis showed that electricity
has a significant effect on household Internet access in both tribal and formal rural areas.
The GHS data show telecenters and libraries as alternative modes of access to ICTs. For
example, telecenters are used by rural students who enroll for distance learning with
learning institutions such as UNISA. Applications such as MXit and M4Lit are also revealed
as potential modes of education. However, they need to be developed so that they
facilitate the learning of science subjects in indigenous languages (Langmia (2005),
Butgereit & Botha (2011)). Telehealth facilities such as eKapa and UKZN telehealth
facilities have been developed to improve monitoring and communication within health
facilities. However, such networks need to be extended to the most remote areas.
An analysis of the 2011 Census data of the Eastern Cape Province showed a high level of
ICT use particularly television and mobile communication devices. Mobile communication
devices in particular are a popular mode of communication in all rural areas. Therefore, a
possible substitute is assessing smartphone distribution. Ideally, the mobile technologies
available to the rural population should provide easy access to the Internet. The
smartphone distribution was used by Mukandatsama (2015) as a more reliable indicator of
the distribution of mobile communication devices in African countries for the purpose of
assessing the contribution of mobile phones to development.
The logistic models fitted to the GHS data for the two types of areas showed that female
headed households are more likely to have access to household Internet. The Swedish
International Development Cooperation Agency (2001), state that female headed
households should be viewed as heterogeneous units which should not necessarily be
associated with poverty. Hence the significance of female headed households can be
explained by the history of labor migration which meant that female households became
common and thus were forced to become self-reliant. However, such a finding needs to be
clarified by further studies.

Conclusion
This study was based on the notion that ICTs stimulate development because of their
ability to improve information flow between service providers and the public. Essential
services are increasingly being provided over ICTs. In addition, those excluded from the
virtual world are not only deprived of wider access to information, public services, and
other economic benefits, but also of an opportunity to pursue their rights as citizens,
including political participation (Polat 2012). At the same time, it is argued that developing
countries have more urgent development challenges such as eradicating poverty. While
other studies have argued that governments need to develop infrastructure and services
such as electricity and telephone lines so that ICTs can be used more efficiently. In sum,
the impact of ICTs on development various based on the ability of people to use ICTs, their
acceptance of the technologies and the accessibility of ICTs.
In the South African context, the provision of various services such as online learning,
administration and management of patience in the health sector, online payment of service
bills and reporting of service disruptions are increasingly being done using ICTs. In

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
108
addition, rural dwellers are able to access services via ICTs at MPCCs and thus avoid
travelling to urban centers to acquire services.
The findings of the statistical analysis show that access to household Internet is
predominantly affected by household income and the education of the household head.
The effect of income and education is more pronounce in farm areas. In addition, the study
notes the importance of electricity particularly in tribal areas. Hence, reflecting the
importance of improving access to electricity before ICTs are used effectively. The
reviewed literature also showed that another obstacle to the effectiveness of ICTs is the
language barrier. For example, Butgereit & Botha (2011) explained that the Dr Maths
learning platform encountered challenges because it employed foreign tutors who could
not communicate to leaners in indigenous languages. Ngwenyama et al. (2006) has
previously argued that education is a prerequisite for the effectiveness ICTs as a stimulant
of development.
The statistical analysis showed that cell phones are widely used in both farm and tribal
areas. However, the impact of cell phone technologies needs to the studied closely, for
example, Mukandatsama (2015) proposed the use of the smartphone distribution as a
more effective measure of mobile phone penetration.
The development of telehealth facilities has improved the health sector. However,
telehealth facilities are still under development throughout the country and in particular
their success depends on the ability of rural dwellers to access and use ICTs. In addition,
the UKZN telehealth facility can connect hospitals and health facilities with developed ICT
infrastructure. Several challenges exist with regards to extending such a facility to remote
places. For example, the capacity to build and maintain a large network.
There is a need to establish a direct link between access to ICTs and improved service
delivery by conducting an independent study on household ICT use. This paper proposes
a further study to understand rurality with respect to the current context of ICT proliferation
by using rurality index measures which combine connectivity variables such as ICTs and
access to services variables which would be combined with qualitative data.

Bibliography
Askar, P., 2015. Logistic regression modeling for predicting task-related ICT use in
teaching. Journal of Educational Technology and Society, 9(2), pp.141–151. Available
at: [Link]

Borghoff, T., 2011. The role of ICT in the globalization of firms. Journal Of Modern
Accounting & Auditing, 7(10), pp.1128–1149.

Brown, I. & Molla, A., 2005. Determinants of Internet and cell phone banking adoption in
South Africa. JIBC, 20, pp.20–29. Available at:
[Link]

Budlender, D., 2003. The debate about household headship. Social Dynamics, 29(2),
pp.48–72.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
109
Butgereit, L. & Botha, R.A., 2011. A model to identify mathematics topics in MXit lingo to
provide tutors quick access to supporting documentation. Pythagoras, 32(2), pp.1–11.
Available at: [Link]

Chapman, R. & Slaymaker, T., 2002. ICTs and Rural Development: Review of the
Literature, Current Interventions and Opportunities for Action, London. Available at:
[Link]

Department of Rural Development and Land Reform, 2009. Comprehensive Rural

Development Programme (CRDP). In Strategic Plan 2009-2012. Pretoria, pp. 36–40.

Dlodlo, N., 2009. Access to ICT education for girls and women in rural South Africa: A case
study. Technology in Society, 31(2), pp.168–175. Available at:
[Link]

Government Communication and Information Systems [GIS], 2014. About Government

Communication and Information Systems. Available at:
[Link]

Heeks, R., 2009. Development Informatics Where Next for ICTs and International
Development?, Available at:
[Link]

Jacobs, P. & Hart, T., 2012. Rural Innovation Assessment Tool (RIAT): skills development
in rural areas-a brief review of evidence.

Laldaparsad, S., 2011. Urban and rural trends in South Africa. South Africa, Pretoria.

Langmia, K., 2005. The Role of ICT in The Economic Development of Africa : The case of
South Africa. International Journal of Education and Development using Information
and Communication Technology, 2(4), pp.144–156. Available at:
[Link]

Mars, M. & Seebregts, C., 2008. Country Case Study for e-Health: South Africa, Available
at: [Link] Case Study for eHealth South
[Link].

Mawson, N., 2010. Cellphone banking surges. TELECOMS. Available at:

[Link]

Mukandatsama, V., 2015. Smartphone distribution in Zimbabwe - whose line is it anyway?

TECHZIM. Available at: [Link]
zimbabwe-whose-line-is-it-anyway/ [Accessed July 20, 2015].

Mulder, I. Bohle, W., Boshomane, S., Morris, C., Tempelman, H. and Velthausz., 2008.
Real-world innovation in rural South africa. The Electronic Journal for Virtual
Organizations and Networks, 10(Special Issue on Living Labs), pp.7–20.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
110
 National Planning Commission, 2012. National Development Plan 2030: Our future - make
it work, Available at: [Link]

Ngwenyama, O., Andoh-Baidoo, F.K., Bollou, F. and Morawczynski, O., 2006. Is There A
Relationship Between ICT, Health, Education And Development? An Empirical
Analysis of five West African Countries from 1997-2003. The Electronic Journal on
Information Systems in Developing Countries, 23(5), pp.1–11.

Polat, R.K., 2012. Digital exclusion in Turkey: A policy perspective. Government

Information Quarterly, 29(4), pp.589–596. Available at:
[Link]

Prieto-Lara, E. & Ocaña-Riola, R., 2010. Updating rurality index for small areas in Spain.
Social Indicators Research, 95(2), pp.267–280.

Public Service Commission, 2010. An evaluation of integration and coordination in the

integrated sustainable rural development programme, Pretoria. Available at:
[Link].

Schmidt, M.I., 2012. A critical evaluation of the analytical operational application of various
settlement typologies in South Africa. University of Stellenbosch.

Schouten, T., 2013. Changing relationships: ICT to improve water governance. In

International Water and Sanitation Centre. Stockholm, pp. 1–16.

Swedish International Development Cooperation Agency, 2001. Briefing paper of

ferminisation of poverty.

Vosloo, S., Deumert, A. & Town, C. 2008. m4Lit : a teen m-novel project in South Africa. ,
pp.25–26. Available at: [Link]
content/uploads/2012/04/[Link].

Wood, R.,Bekker., L-G., Brown, S., and Rivett, U. 2008. The utility of pharmacy dispensing
data for ART programme evaluation and early identification of patient loss to follow-up.
Southern African Journal of HIV Medicine, 9(2), p.44.

Steyn, J., Van Belle, J.P. (Eds.). 2015. Beyond development. Time for a new ICT4D paradigm?
Proceedings of the 9th IDIA conference, IDIA2015, Nungwi, Zanzibar: 96-111
ISBN: 978-0-620-68395-1.
111

View publication stats