Two-Speed Britain: Rural Internet Use

Abstract

By using detailed data on Internet access and use in rural and urban areas of Britain, we show the effect of low-speed broadband connection on people’s use of the Internet and the services it provides. We use a three-fold definition of deep rural, shallow rural, and urban areas to explore the nature of the digital divides between these areas, and the consequences for people’s relation with the Internet. We find that while overall access to the Internet varies little geographically, though with very different connection speeds, there are consistent differences in urban versus deep rural Internet use that provide evidence not only of an urban-rural digital divide, but also ways in which this divide is manifested. In particular, through this analysis, we have discovered that the ‘urban-rural divide’ is generally manifested between ‘deep rural’ Internet use on the one hand and ‘shallow rural and urban’ Internet use on the other hand. The existence of this divide - a two-speed Britain - means that over 1 million people in Britain are potentially excluded from, or at best find it challenging to participate in, what is generally regarded as ‘normal’ online social, commercial, creative and civic life, because they live in deep rural areas of the Britain.

Full text

Two-Speed Britain:
Rural Internet Use
John Farrington, Lorna Philip,
Caitlin Cottrill, Pamela Abbott,
Grant Blank & William Dutton

3
TWO-SPEED BRITAIN REPORT
Two-Speed Britain:
Rural and Urban Internet
John Farrington*^, Lorna Philip*^, Caitlin Cottrill *^, Pamela Abbott*, Grant Blank**,
and William Dutton***
*RCUK Digital Economy Research Hub dot.rural, University of Aberdeen
*^ Geography and Environment, University of Aberdeen
** Oxford Internet Institute, University of Oxford
*** Quello Centre, Michigan State University
Aberdeen University Press 2015
Second imprint
The research described herein was supported by the award made by the RCUK Digital Economy programme
to the dot.rural Digital Economy Hub; award reference: EP/G066051/1.
Acknowledgements
The authors are grateful to Fiona Williams for permission to use the photographs
reproduced on the front cover. They were taken in South-West Shropshire in 2014.
We are also grateful to Claire Wallace, Anne Roberts, Fiona Ashmore and Rob Craig for contributing
vignettes from their research, and to Craig Morton for his valuable assistance with data analysis.
We also wish to thank Robert Kenny for his timely and constructive input to an earlier draft of the report.

4TWO-SPEED BRITAIN REPORT

5
TWO-SPEED BRITAIN REPORT
Contents
Executive Summary
1. Introduction
1.i Background context
1.ii A rural focus
1.iii A ‘Rural Boost’ to OxIS 2013
 1.ivDening‘Deep’and‘Shallow’RuralAreasoftheUK
2. Patterns of Internet Access and Use Across Britain in 2013
2.i Access to the Internet in Rural Areas
2.ii The Demographic Characteristics of Internet Users by Location
2.iii Use of the Internet
2.iv Use of Information and Online Services
2.v Use of the Internet at Work and Home
2.vi Key Findings
3. ICT Infrastructure: Internet Use, Broadband Speeds and Mobile Internet Connectivity
3.i Broadband Speeds for Respondents to the Oxford Internet Survey 2013
3.ii Broadband Speed, Location and Internet Behaviour
 3.iiiDoesBroadbandSpeedInuenceInternetUsers’AbilitytodoWhatTheyWantOnline?
3.iv Key Findings
4.DoesBeingDigitallyConnectedMatterinRuralBritainToday?
5. Conclusion
References
Appendices
 Appendix1:UrbanandRuralClassicationsinScotland,EnglandandWales
Appendix 2: Contributors of Case Vignettes and the Research Projects Referred To
6
8
18
34
46
52
54
56

TWO-SPEED BRITAIN REPORT
6
There is a digital divide that separates urban and rural areas in Britain.
The quality of Internet connection varies depending on geography,
leaving many rural areas with a fraction of the service that is enjoyed in
urban areas. By systematically analysing data produced by the RCUK
Digital Economy Research Hub at the University of Aberdeen and
the Oxford Internet Institute, this report is the rst detailed study
highlighting the constraints that “deep rural” communities face when
trying to use the Internet, and the impact of a divide which is not only
stark, but likely to grow as broadband speeds in well-connected areas
increase at a faster rate than in rural Britain.
BACKGROUND
The regulator Ofcom says that Superfast broadband
of at least 30Mbps is available to 78% of the UK
population over the age of 14. But while it is generally
understood that quality of Internet access varies
between urban and rural areas, the true nature of this
divide may be under-estimated. Approximately 48.4
million people in the UK – or 80.7% of the population
–arebasedinurbanareas,andthedenitionofa
“rural” area can include communities close to urban
centres as well as those far from them.
By separating the 11.6 million people in rural
communities into “shallow” and “deep” rural,
this report sheds more light on a group of
1.3 million people that are “digitally excluded”
by their location. Many of these areas struggle to
receive 3G and 4G mobile reception, and some do
not have access to a broadband connection stronger
than 2Megabits (Mbits/s). In an era in which services
are moving online, this can have a marked social and
economic impact.
Two Speed Britain: Rural Internet Use
John Farrington, Lorna Philip, Caitlin Cottrill, Pamela Abbott, Grant Blank and William Dutton

7
TWO-SPEED BRITAIN REPORT
FINDINGS
• Superfast broadband (30+Mbits/s) was not
available to any of the deep rural respondents
sampled in a survey of 1090 rural residents
across Britain.
• The highest broadband speed in any of the
sampled deep rural areas, 17.4Mbits/s, was lower
than the average speed for the urban areas.
• 48% of Internet users in deep rural areas and
36% in shallow rural areas, who had speeds of
3.5Mbits/s or less, believed their connection was
too slow.
• Those with connection speeds of less than
3.5Mbits/s were less likely to try “data-heavy”
activities such as streaming, gaming, and creation
of content such as video. They were also less
likely to get involved in social networking online.
• As urban areas receive faster speeds, the “speed
gap” between urban and deep rural areas that
have not yet received an adequate connection is
likely to continue to widen.
• Poor connection has an impact on businesses, from
creative businesses unable to create and send video
and music for clients, to farmers unable to complete
online forms, cattle passports and registrations.
• It also affects the Government’s aim to make
certain services “Digital by Default” and the
“universal” broadband target of 2Mbits/s is
inadequate for this to work.
• Young people can feel excluded from peer groups
who have better connections, especially after
becoming accustomed to faster connection at
school or university. Older people are also excluded
fromtheconnectionstheycanndinsocial
networking, and the savings from shopping online.
• These issues can have a sizeable impact on
poorly-connected rural areas, including the loss of
businesses, failure to attract new businesses, and
increased household or business costs. They can
also contribute to the loss of young people from
rural areas and deter people from moving in.
CONCLUSION
The gap between the service provided to urban
and remote rural areas will become an increasingly
signicantproblemforcommunitiesthatnd
themselves with limited or non-existent connections,
as digital services take on a more prominent role
in our society. The UK’s rural areas are now at a
“tipping point”, to which poor digital connection
is contributing. Poorly-connected “deep rural”
communities will suffer increased costs, economic
disadvantage and a population drain if services are
not improved. The Governments’ commitments to
making some services “digital by default” will not be
feasible when some areas struggle for connection of
2Mbit/s or less. It is recognized that state intervention
in infrastructure provision is constrained by EU
competition law, and that state-supported rural
broadband rollout will help in the short term, but the
report recommends that the interfaces between public,
private and community sectors be improved, led by
governments, to improve collaborative working and
informationowstohelpndsuitablesolutionsforthis
issue, and to further improve speeds in deep rural areas.

1. Introduction
8TWO-SPEED BRITAIN REPORT

9
TWO-SPEED BRITAIN REPORT
It is widely understood that an urban-rural digital divide exists, and
this is supported by study of the geography of telecommunication
services in Great Britain (Ofcom 2013a, 2013b). This divide is related
to the consequences of differing speeds of connection to the Internet
in different areas. Nevertheless, studies based on survey research of
individuals and households in the UK have not been able to document
an urban-rural divide (e.g., Dutton and Blank 2011).
Table 1.1: Selected attributes of the British rural population
England Wales Scotland
Total population c. 51.8 million c. 3 million c. 5.2 million
Rural population 18.6% 18.4% 33.9%
‘Remote’ or ‘sparse’ rural population 1.2% 2.4% 6.5%
% of land area dened as rural 79.1% 94% 87%
% land area dened as ‘remote’ or ‘sparse’ rural 15.3% 68.8% 59.9%
Derived from population estimates contained in Pateman (2011) derived from mid-2009 population estimates.
A large majority of the British population – 81%
or almost 48.4 million people – live in urban areas
and any surveys that proportionally sample urban
and rural households might not incorporate a large
enough number of rural households in the total
sample to be able to look meaningfully for urban-
rural patterns in Internet use. Furthermore a simple
‘urban’ and ‘rural’ sampling frame overlooks the
fact that rural Britain is not homogenous: there are
considerable differences between the characteristics
of the more accessible, or ‘shallow rural’ (close
to urban areas) and remote, or ‘deep rural’ areas.
Remotenessinparticularintroducesspecic
challenges to the development of infrastructure, be
it transport or the communications infrastructure
required to be able to use the Internet.
Differences in Internet use between different types
ofgeographicalareasthathavebeenidentied
in earlier research have been marginal, or could
be explained by controlling for other factors (also
related, in part, to geography), such as age and
socioeconomicstatus.Suchnullndingshaveledtoa
relativeneglectofresearchwithanexplicitaimtond
systematic evidence of an urban-rural digital divide,
to substantiate abundant anecdotal evidence – such
as that presented in the illustrative case vignettes in
Section 4 of this report – of such a divide.
Opposite: Photo courtesy of Dr Lorna Philip;
not to be reused without prior permission.

TWO-SPEED BRITAIN REPORT
10
Approximately 19.3% of the British population
–c.11.6millionpeople–liveinareasdenedby
the Government as being ‘rural’1 (Pateman, 11).
Although a minority of the British population, the
rural population is a very sizeable minority whose
characteristics, we argue, should not be overlooked.
As reported in Table 1.1 there are marked differences
between England, Wales and Scotland masked by
the British average. Of particular importance in
terms of infrastructure provision, including the
ICT infrastructure necessary to facilitate Internet
connectivity, is the proportion of the total land area
denedasruralandthatdenedasremoterural.To
date, the digital connectivity experienced by the
c. 1.3 million people who live in the ‘remote’ and
‘sparse’ rural areas that comprise such a large
proportion of the British land area is inferior to that
serving the larger numbers of people who live in a
much smaller proportion of the land area.
In the research reported here we have systematically
examined the attributes of urban, ‘shallow rural’ and
‘deep rural’ areas (deriving ‘deep’ and ‘shallow’ rural from
ofciallydened‘rural’classicationsystems)and,for
thersttimeforresearchconcernedwithInternetuse,
have explored characteristics of the (sizeable) minority
of the British population who live dispersed across the
vast majority of the British land area.
A key focus of the 2013 Oxford Internet Survey
(Oxford Internet Institute, 2013) was to address this
lack of strong evidence about the urban-rural divide
inBritain.Specically,astratiedsurveysamplewas
designed, which included a disproportionately larger
number of rural residents, so that any real urban-rural
patternscouldbeidentied.Thissampleallowed
crucial distinctions to be made not only between
urban and rural areas, but also between ‘deep’ and
‘shallow’ rural areas (‘deep’ rural areas are more
remote and more sparsely populated, as described in
detail below).
This report draws upon analysis of data produced as
the outcome of a partnership between the Oxford
Internet Institute (OII) and the RCUK Digital Economy
Research Hub (dot.rural) at the University of Aberdeen.
OII’s authoritative biennial survey of Internet use,
the Oxford Internet Survey (OxIS) (part of the World
InternetProject)identiesandexploresnewtrendsin
Internet use across Britain (http://microsites.oii.ox.ac.
uk/OxIS/). Dot.rural’s research focuses on the use of
digital technologies in transformative change in rural
Britain’s society and economy (www.dotrural.ac.uk).
This collaborative research partnership enabled the
disproportionately boosted rural sample for the OxIS
2013 survey, which was critical to discovering the
differencesidentiedinthissurvey.
1.i Background Context
Much has been written and said in academic and
policy domains, as well as in day-to-day discourses
of communities in rural Britain, about the potential
for digital technology to play a key role in achieving
a more sustainable and resilient rural society and
economy (e.g., Department of Culture, Media and
Sport 2009). However, sample surveys of UK adults
havenotidentiedastrongurban-ruraldivide(e.g.,
Dutton and Blank 2011, 2013), and differences can be
explained in part by other factors that are related to
urban-rural life differences, such as age, income, and
educational experience.
If conventional wisdom is correct – and as
demonstrated in the illustrative examples from
research conducted in various projects funded
recently by the RCUK Digital Economy Research
Hub presented in this report – a sizeable minority of
rural residents cannot participate fully in the digital
economy and society that is enjoyed by the majority of
residents in Britain, where 78 percent of adults aged
14 and over have access to the Internet. If incorrect,
however, a great deal of resource and effort in public
and/or private sectors could be wasted in trying to
close a divide that does not exist, such as by investing
in unnecessary infrastructure projects. Clearly it is
importanttohavemoredenitiveinformationonthe
existence of any urban-rural digital divide, and what it
might mean for those who may be experiencing it.
A recent study of communication infrastructures and
services in the UK, based on industry data assembled
by the communications regulator, Ofcom, shows
generally less availability of high speed broadband
and mobile phone networks in rural areas compared
with urban areas across the UK. The local authority
level maps available on Ofcom’s website to visualize
the ICT infrastructure data they hold show that 3G
mobile phone coverage and fast, reliable broadband
coverage is very poor across large swathes of
northern and southern Scotland, northern England,
East Anglia, south-west England, and Wales (Ofcom
2013a, b) – reproduced as Map 1.1. While the three
maps conceal large variations in connectivity they
usefully highlight regional variations.
1SeeAppendix1fordetailsaboutthedifferentgovernmenturban-ruraldenitionsusedforEngland,WalesandScotland.

11
TWO-SPEED BRITAIN REPORT
1.ii A Rural Focus
An ‘urban-rural digital divide’ in the availability
and up-take of digital technologies and services,
particularly broadband Internet access and use,
is the product of both infrastructure and socio-
economic capabilities. (Hindman, 2000; Furuholt
and Kristainsen, 2007). The technical, infrastructure
divide has been the focus of UK government-led
initiatives (in England, Wales and Scotland) which, in
partnership with Internet Service Providers (ISPs),
such as the dominant provider, BT, are rolling out
broadband infrastructures to rural areas, but not
of an equivalent capacity to that installed in more
densely populated urban areas. One consequence is
that the available speeds, especially in the rural (and
most especially in the more remote and more sparsely
populated) areas, are often low in comparison with
those available, and being deployed, in urban areas.
It is natural that market forces will tend towards
this situation, but it means that the rural-urban
digital divide in terms of access to broadband speeds
in excess of 2Mbit/s (let alone the 10 to 30Mbit/s
regarded increasingly as a basic requirement for
effective digital participation) is increasing, and is
likely to continue to do so: the ‘faster’ areas with
better connection are getting ‘faster, faster’. The
implications for rural users of such differences in
speed are the focus of the analysis presented in
Section 2.
This geographical digital divide is being addressed
through a variety of approaches and policy initiatives
including large-scale deployments of high speed
broadband through the Broadband UK (BDUK)
programme. Current UK Government policy includes,
for example, a commitment to provide superfast
broadband to at least 90% of premises in the UK, to
ensure universal access to standard broadband of at
least 2Mbit/s, and funding to provide mobile phone
coverage to the 0.3% of premises in the UK that are
not currently served at all by a mobile phone operator
(Ofcom, 2013b). Current commitments to roll out
superfast broadband exclude as much as 10% of the
UK population, in the region of 6.5 million people.
In some rural areas, communities have organised
themselves and raised the funds to develop their own
broadband infrastructure: such bottom-up activity is
exempliedbytheactivitiesofcommunityenterprises
such as Cybermoor in Cumbria, B4RN (Broadband
For the Rural North, which is deploying connection
with 1Gbit/s capability) in north-west England and
B4GAL (Broadband for Glencaple and Lowther) in
south-central Scotland. Vignette number 5 in Section
4 illustrates how feeling ‘badly served’ by low speed
Internet led to some individuals becoming involved
with the B4RN community broadband project.

TWO-SPEED BRITAIN REPORT
12
The % of premises at which all operators have
3G coverage (outdoor reception). Data published
November 2013.
The percentage of broadband connections that
have sync speeds of less than 2.2Mbits/s. Data
published November 2013.
Map 1.1: Territorial Coverage of 3G Mobile
Services in the UK
Map 1.2: Territorial Coverage of Broadband
Connections of Less than 2.2Mbit/s in the UK
95% or more
80% - less than 90%
90% - less than 95%
60% - less than 80%
less than 60%
1
3
2
4
5
less than 5%
10% - less than 15%
5% - less than 10%
15% - less than 20%
20% or more
1
3
2
4
5
3G coverage by premises.
Each area has been ranked from 1 to 5 on the level of
mobile coverage.
Percentage receiving less than 2.2Mbit/s.
Each area has been ranked from 1 to 5 on the percentage
of broadband connections that have modem sync speeds
of less than 2.2Mbit/s.

13
TWO-SPEED BRITAIN REPORT
The % of residential and non-residential premises
where either Virgin Media cable, Openreach Fibre to
the Cabinet or Digital Regional networks (superfast
broadband) is available. Data published November 2013.
Map 1.3: Territorial Coverage of Superfast Broadband
Availability in the UK
90% availability or more
1
50% - less than 70%
3
70% - less than 90%
2
30% - less than 50%
4
less than 30%
5
Superfast broadband availability.
Each area has been ranked from 1 to 5 on the percentage of
residential and non-residential premises where either Virgin
Media cable, Openreach Fibre to the Cabinet or Digital Region
networks are available.

TWO-SPEED BRITAIN REPORT
14
Ofcom (2014b) reported that UK average broadband
speed in November 2013 was 17.8Mbit/s, up from
11.1Mbit/s in November 2008. This increase is largely
due to the take up of ‘superfast’ (30Mbit/s or higher)
services, and ISPs upgrading customers to faster
broadband packages. In May 2010, the headline speed
of 76% of UK residential broadband connections
was ‘up to and including 8Mbit/s to 10Mbit/s’ and no
residential connections were 30Mbit/s or higher. In
November 2013 the proportions were 11% and 25%
respectively (Ofcom, 2014b). However, despite the
overall picture being one of residential customers
having faster broadband over this time period, the
gap between average download speeds in urban
and rural areas is marked and was reported to have
widened from 9.5Mbit/s in May 2011 to 16.5Mbit/s
in May 2013 (Ofcom, 2013a). Average urban speed
in May 2013 was 26.4Mbit/s, and average ‘suburban’
speedwas17.9Mbit/s.Contrastthesegureswiththe
average rural speed, cited as 9.9Mbit/s (ibid.).
This widening of the urban-rural gap is claimed
to be due to: “the lower availability of superfast
broadband services in rural areas compared to urban
areas, and because ADSL broadband speeds are also
generally slower in rural areas because the average
line between the home and the nearest telephone
exchange needs to be longer2. Broadband speeds
are also generally slower in rural areas because the
average line between the home and the nearest
telephone exchange needs to be longer” (Ofcom,
2013a:3).
Ofcom expect that this gap will widen in the short
term, but that it will “begin to decline over time, as the
availability of superfast broadband increases in rural
areas” (Ofcom ibid). Overall, Ofcom note that:
“Theavailabilityandspeedofxedbroadband
Internet access is subject to much greater variation
[thanxedtelephonyandpostalservices,anddigital
terrestrial television]…partially because of variability
in the speed provided by current generation
broadband, and partially because the deployment of
superfast broadband is still underway, especially in
more rural areas…. [and] the same is true of mobile
services…” (Ofcom 2013b).
Section 3 of this report considers attributes of the
ICTinfrastructure,withaspecicfocusonbroadband
speeds as publicised by Ofcom, and explores how the
infrastructure capability affects Internet use.
Against this digital landscape, rural areas in Britain
have, in recent decades, changed, and are still
changing, in several fundamental ways. There are
well-established movements of people in and out
of rural areas. For example, younger age groups are
moving from rural to urban areas, and middle aged
and older residents from urban to rural areas. Some
rural areas have witnessed a ‘population turnaround’
whereby the population decline evident since the
latter decades of the nineteenth century has been
reversed, but others remain areas of long term
population loss. Incomers to many rural areas often
haveabove-averageeducationalqualicationsand
wealth. A demographically ageing rural population
is resulting from sustained out-migration of young
adults, ageing of pre-existing, long term residents and
the ageing of those who move into rural areas in mid
or later life (Philip et al 2012). These ageing trends
addresource,servicedeliveryandstafngchallenges
to public sector services in particular. Innovative
solutions to these challenges often rely on new ICT
applications which cannot be deployed in areas with
poor digital infrastructure capability.
A further consequence of these population
movements is the isolation that can often result from
ageing, especially as other family members are likely
to be located more distantly. Working from home,
either as a ‘removed’ person within a business or
institution located elsewhere, or as a local producer
of physical or (especially) digital products becomes
increasinglydifcultornon-viablewithoutagood
Internet connection. Participation in globalising
markets poses particular challenges for small rural
businesses. Even downloading and installing the
software that (urban) clients require the (rural)
producer to use, as a condition of business, can be
either impossible or extremely time-consuming.
Moreover, participation in many day-to-day activities,
from education to civic society, to retail shopping and
professional services such as banking, is increasingly
associated with online delivery. Thus, exclusion from
digital connections implies exclusion from a wide
range of activities regarded as normal in a networked
society. Compounding this issue is the fact that those
in rural Britain who are most disadvantaged are least
likely to be connected (Royal Society of Edinburgh
2010), presenting a further layer of exclusion within
rural society. Section 4 presents case vignettes drawn
2(A)DSL–(Asymmetric)DigitalSubscriberLine*:DSLlinesuseexisting2-wirecoppertelephonelinestocarrydigitaldataoverthe‘nalmile’to
theusers'home/businesspremises.Thelengthofthenalmilecopperconnectiondeterminesthespeedoftheconnection.Anextgenerationof
copperlines,ADSLlines,wereintroducedintheUKin2000butevenonVery-high-bit-rateDSLdownload,speedsarecompromisedifthenal
mile exceeds 1.2km.

15
TWO-SPEED BRITAIN REPORT
from research projects conducted under the auspices
of the dot.rural Digital Economy Hub at the University
of Aberdeen. These illustrative examples demonstrate
both how important those who live in rural areas
consider it is to be online, but also highlight the
challenges,frustrationanddifcultiesexperiencedby
Internet users who live in rural areas.
The implications of territorial variations in digital
infrastructure for those living, working, and running
businesses in rural areas, and/or those visiting rural
communities, are considerable in terms of rural
social and economic development, sustainability
and resilience. A move to on-line service delivery,
suchasbanking,retailandPostOfceservices
has contributed to service decline in rural areas.
However, the impact of poor digital infrastructure
and low connection speeds as a blocker to economic
developmentisarguablymoresignicant.Ifnew
and existing businesses, those with young families,
those with greater educational experience, and those
with incomes capable of adding to the economic
base of rural areas, are not able to move into, or
remain in, rural Britain, then the socio-economic and
demographic sustainability and resilience of rural
Britain will be further challenged.
For rural areas to respond effectively to the various
challenges associated with an increasingly digital
society, a better understanding of Internet use in
urban and rural communities is required. The work
reportedhereprovidesoneoftherstattemptsto
provide systematic survey evidence of Internet use
across areas of rural Britain.
In order to improve on past attempts to explore
geographical patterns in Internet use across Britain,
the Oxford Internet Survey in 2013 developed a
disproportionatestratiedsamplethatboostedthe
number of respondents from rural areas, aged 14
years or over.
By having more rural respondents than we would
have from a strictly random sample, it is possible to
have better, more statistically robust estimates of
rural patterns, and to examine differences within rural
areas.Tothatend,thesamplewasalsostratiedto
ensure adequate numbers of respondents from deep
and shallow rural areas.
1.iii A ‘Rural Boost’ to OxIS 2013

TWO-SPEED BRITAIN REPORT
16
In order to provide a large enough response set for
meaningful analysis of patterns between these three
types of geographical area to be possible, rural areas
were oversampled. The number of those sampled in
both shallow and deep rural areas was higher than
would have been drawn from a random sample that
wasdesignedtoreectthecharacteristicsofthe
British population as a whole. However, our sampling
strategy enables us to move from a main sample,
drawn to represent the general population, to the
boosted sample, depending on the type of analysis to
be conducted (Table 1.2). Weighting of respondents
enables us to use the boosted sample without
distorting results for the population as a whole. The
Oxford Internet Survey 2013 report (Oxford Internet
Institute, 2013) contains further information about
sampling.
Thereisnosingleaccepteddenitionof‘rural’
globally, across Europe or across Britain. The
academic literature, for example, suggests that
attemptstodene‘rural’draw,variously,upon
functional attributes, political economy approaches,
and social representations (Cloke and Thrift, 1994),
among other approaches. In government and across
public policy (the most relevant to this context, as
they comprise the framework within which digital
infrastructure and applications operate), functional
denitionsofruralarecommon.Quantiable
attributes such as population size, density, proximity
to urban centres, and land use, are variously combined
to classify territorial units as small as census output
areas or as large as local authority areas. (Appendix
1 provides further details on the government
classicationscurrentlyinuseacrossEngland,Wales
and Scotland that have been used in this research).
Ofcialclassications,includingthoserelieduponfor
this research, differentiate between different types of
rural areas, making an important distinction between
rural areas in close proximity to large urban centres,
and those associated with more remote areas. In
terms of digital infrastructure, it is the most remote,
least densely populated areas that studies reported
by, for example, Ofcom (2013b), identify as being most
affected by the urban-rural digital divide. For the OxIS
2013 survey sample, in order to capture these most
remote areas, we therefore move away from a simple
urban-rural binary to employ three geographical
categories: urban, shallow rural and deep rural. These
weredenedasfollows:‘remoteruralareas’and‘very
remote rural areas’.
Urban
1.iv Dening ‘Deep’ and ‘Shallow’ Rural Areas of the UK
Urban
EnglandandWales:urban/ruralclassicationcategories‘urban–lesssparse’and‘urban–sparse’;
Scotland:urban/ruraldenitioncategories‘largeurbanareas’and‘otherurbanareas’.
Shallow Rural
EnglandandWales:urban/ruralclassicationcategories‘town&fringe–lesssparse’and‘village,hamlet&isolated
dwelling – less sparse’;
Scotland:urban/ruraldenitioncategories‘accessiblesmalltowns’and‘accessibleruralareas’.
Deep Rural
EnglandandWales:urban/ruralclassicationcategories‘town&fringe–sparse’and‘village,hamlet&isolated
dwelling – sparse’;

17
TWO-SPEED BRITAIN REPORT
In numerical terms the boosted deep rural sample
represents 1.3 million residents in Scotland, Wales
and England and covers in the region of 50% of the
British land area. The shallow rural sample represents
approximately 10.3 million people.
This report presents data from the 2013 Oxford
Internet Survey for the three urban, shallow rural
and deep rural areas. A geographical weighting has
been applied throughout for the analysis reported in
Sections 2 and 3. The data were weighted to allow for
the disproportionate sampling of urban, shallow rural
and deep rural populations. All survey estimates are
calculated using the weighted data so that averages
are weighted averages and percentages are weighted
percentages. In essence, this means that we can be
sure that any differences observed between the
three geographical area types are true differences,
andwheretheyareidentiedasbeingstatistically
signicantwehaveusedthe95%condencelimit
throughout.
In order to reproduce population proportions we
usedpost-straticationweightingbasedongender,
age, ACORN type, region, number in household, and
urban/rural. We used two different weights. WALL, a
mnemonic for ‘weight all’, weights the entire survey
N = 2,657 to the population proportions for Great
Britain. This weight is used when we analyse the
dataset as a whole. WUDS, a mnemonic for ‘weight
urban-deep-shallow’, weights each stratum separately
to the population proportions for that stratum only.
This weight is used when the three strata are used
as independent variables in tables and analyses. In
Section 2, WUDS is used in all reported analyses. In
Section 3, we indicate in each Figure whether WALL
or WUDS is used.
Table 1.2: Oxford Internet Survey 2013 sample size: main and boost
Main Sample Boosted Sample Total
Deep rural 32 232 264
Shallow rural 454 372 826
Urban 1567 0 1567
Total 2053 604 2657
Derived from population estimates contained in Pateman (2011) derived from mid-2009 population estimates.

2. Patterns of
Internet Access
and Use Across
Britain in 2013
18 TWO-SPEED BRITAIN REPORT

19
TWO-SPEED BRITAIN REPORT
The Oxford Internet Survey 2013 data reveal a variety of similarities
and differences between urban, shallow rural and deep rural areas.
These are reported below under the following headings:
1. Access to the Internet in urban and rural areas.
2. Characteristics of Internet users by location.
3. What do urban and rural Internet users do online and where do they access the Internet?
4. Use of information and services online in urban and rural areas.
5. Use of the Internet in working lives and family lives.
2.i Access to the Internet in Urban and Rural Areas
Over three quarters of households in Britain
had home-based access to the Internet in 2013,
accordingtothesurveyndings.Thisgureisvery
similartothatreportedbytheNationalAuditOfce
(2013), drawing on a survey conducted in England.
Households in shallow rural areas are slightly more
likely than those in deep rural and in urban areas to
have access to the Internet at home. However, deep
rural households are the most likely to have never had
Internet access, 18% compared with 16.5% in urban
areas and only 14% in shallow rural areas.
As is apparent from Figure 2.1, the geographical
differences in household access to the Internet are
not simply urban–rural differences. The attributes of
shallow and deep rural areas differ, a variation that
isreectedthroughoutthendingspresentedinthis
report.
Source: OxIS 2013, n = 2657
Figure 2.1:
Does the Respondent’s Household have Access to the Internet by Urban, Shallow Rural and Deep Rural
Yes
Deep Rural
100
% of all respondents
80
60
40
20
0
Not Now Never
Shallow Rural
Urban

TWO-SPEED BRITAIN REPORT
20
Less than 10% of non-users of the Internet reported
that they are planning to get access to the Internet and
there is little difference by location. Ex-users (the small
‘not now’ group in Figure 2.1) are much more likely than
non-users to say that they are planning to get access in
the near future. Ex-users living in deep rural areas were
more likely to express a desire to be Internet users
again than respondents living elsewhere.
TheageproleofInternetuserscapturedinthe
Oxford Internet Survey 2013 accords with patterns
reportedelsewhere(e.g.OfceforNationalStatistics,
2013; Scottish Government, 2014) where it has
been reported that Internet use decreases with
increasing age. It is worth noting, however, that the
proportion of older Internet users has increased
notably across the UK in recent years and age related
differences in Internet use are less marked now than
a decade ago. The most recent analysis of Scottish
Household Survey data reported that there is a
marked difference between the proportions of those
aged 60- 74 and those aged 75+ who are Internet
users: 63% compared with 25% respectively (Scottish
Government, 2014). As noted in section 1, deep rural
areas have a higher proportion of their population in
the older age cohorts than do urban and shallow rural
areas and, because the Oxford Internet Survey sample
was a proportional sample this demographic pattern
isreectedinthesurveyresponses.
Figure 2.2:
Use of the Internet at Any Location by Urban, Shallow Rural and Deep Rural
Current User
Deep Rural
100
% of All Respondents
80
60
40
20
0
Ex-User Non-User
Shallow Rural
Urban
Source: OxIS 2013, n = 2657
Table 1.2: Oxford Internet Survey 2013 sample size: main and boost
Does this household have
access to the Internet?
Is this household planning to get access to the Internet in the next year?
Deep rural Shallow rural Urban Total
No Yes No Ye s No Yes No Yes
Not now, in the past 27.7% 72.3% 51.9% 48.1% 50.3% 49.7% 49.6% 50.4%
No, never 96.8% 3.2% 86.9% 13.1% 91.6% 8.4% 90.9% 9.1%
n=724

21
TWO-SPEED BRITAIN REPORT
As shown in Figure 2.2, 78% of people in Britain
currently use the Internet (at home, work, school,
college or elsewhere). Differences between deep
rural, shallow rural and urban areas are negligible.
Broadband Internet has replaced all but a handful of
dial-up connections in the UK. It is a technology that
allows the simultaneous transfer of voice and data
over a single line and its introduction has facilitated
higher speed Internet connections than were possible
over dial-up connections. However, broadband
reliability and speed varies considerably across Great
Britain, with some connections being no better than
an old dial-up connection (less than 2Mbit/s) whilst
othersare‘superfast’(denedbytheUKGovernment
as download speeds of more than 24Mbit/s and by
the EU and Ofcom as download speeds in excess
of 30Mbit/s). We asked users if they thought their
Internet connections were fast enough to do what
they wanted to do online, or too slow to do some
things they would like to do. We expected rural users
tohavemoredifcultyindoingwhattheywishedto
do online, and this was the case.
Overall, a majority of Internet users thought that
their connection was fast enough all of the time (60%
of users, see Figure 2.3). However, those Internet
userslivinginruralareasaresignicantlylesslikely
to say that the speed of the Internet is fast enough
for what they want to do all of the time. Fifty two
per cent in shallow rural areas and 48% in deep rural
areas reported that their connection was fast enough
compared with two thirds in urban areas. Notably,
nearly a third (32%) of those living in deep rural areas
say that their Internet speed is always too slow for
what they want to do compared with only 6% in urban
areas and 22% in shallow rural areas.
The association between area type and perceptions
oftheadequacyofspeedwasstatisticallysignicant
(x2=147,p=0.00).Thesendingsarequitedramatic.
Theyreinforcendingsfromstudiesoftheavailability
of infrastructures and services which show that a
large proportion of the British land mass does not
haveasufcientlyfastInternetconnectiontoallow
those who live in remote communities to do what they
expect to be able to do online; by inference, they are
signicantlydigitallydisadvantagedcomparedwith
the large majority of people, in urban and shallow
rural areas.
Figure 2.4 shows that almost all Internet users access
the Internet at home with negligible differences
between the 3 location types. While over half of
Internet users access the Internet using a mobile or
wireless dongle, those living in urban areas (59%)
are more likely to do so compared to those living
in shallow rural areas (51%) and deep rural areas
(52%). This illustrates an urban-rural divide but the
geographical difference was not large enough to be
statisticallysignicantatthe95%condencelevel.
About a third of Internet users access the Internet
via a computer at work, with negligible differences
by location observed. Only a minority of users access
the Internet at public libraries (10%) and Internet
cafes (7%). Thirty-eight per cent access the Internet
at someone else’s house, shallow rural dwellers (40%)
being more likely to do so than urban or deep rural
dwellers (both 37%). These differences were not
statisticallysignicant.
Figure 2.3:
Adequacy of Speed of Internet as Reported by Internet Users in Urban, Shallow Rural and Deep Rural
Too Slow
Deep Rural
100
% of Internet Users
80
60
40
20
0
It Depends Fast Enough
Shallow Rural
Urban
Source: OxIS 2013, n= 1839

TWO-SPEED BRITAIN REPORT
22
Therearenostatisticallysignicantdifferencesin
the gender of Internet users by urban-rural location.
However,asshowninFigure2.5,theageproles
across the three area types differed in a statistically
signicantmanner(x2 = 82, p = 0.02). Internet users
in deep rural areas are older than those living in
both shallow rural and urban areas. Thirty-six per
cent of Internet users living in deep rural areas are
over 55 years, and 58% are over 45 years, compared
with 21% over 55 years and 38% over 45 years in
urban areas. Demographically, shallow rural areas
sitbetweendeepruralandurbanareas.Thisnding,
becauseoftheweightingused,isagoodreectionof
the demographic structure of deep rural areas, whose
populations are demographically older than those
in shallow rural and urban areas (Philip et al 2012).
Vignettes number 4 and 9 in Section 4 illustrate
contrasting views of older generation non-Internet
users and highlight some barriers to becoming
Internet users as perceived by older rural residents.
2.ii The Demographic Characteristics of Internet Users by Location
Figure 2.5:
Proportions of Internet Users in Different Age Groups: Overall and by Urban, Shallow Rural and Deep Rural
14-17 18-24 25-34 35-44 45-54 55-64 65-74 75+
25
% of Internet Users
Overall Pattern of Internet User Age
20
15
10
5
0
Figure 2.4:
Where Respondents Use the Internet by Urban, Shallow Rural and Deep Rural
Deep Rural
100
% of Internet Users
80
60
40
20
0
Shallow Rural
Urban
Internet
Cafe
School/
University
Mobile/
Wireless
Dongle
Library WorkAnother
Person’s
Home
Home
Source: OxIS 2013, n= 1839
Source: OxIS 2013, n= 1839

23
TWO-SPEED BRITAIN REPORT
Figure 2.6 presents educational attainment levels for
Internet users. Almost a quarter of Internet users,
regardlessoflocation,havenoqualications.However,
those in deep rural areas are the best educated,
being the most likely to have a higher educational
qualicationcomparedtothoselivinginurbanor
shallow rural locations (43% compared to 29% and
32% respectively). The differences in educational
attainmentlevelarenotstatisticallysignicant.
While the income patterns of Internet users do not
vary noticeably by location (the incomes of Internet
users who responded to the survey broadly follow
the UK income distribution pattern), the educational
attainmentpatternsnotedabovearereectedinthe
socioeconomic characteristics of Internet users, shown
in Figure 2.7, which vary between the three area types
(x2 = 50.66, p = 0.08). Deep rural Internet users are
more likely to be in upper middle and middle-grade
socio-economic groups (38% in total) than Internet
users in urban (24%) or shallow rural areas (24%). This
could be because in deep rural areas it is the lower
social grades who are more likely not to be Internet
users, perhaps due to income status and the higher
likelihood of those on the lowest incomes not being
able or willing to pay for an Internet connection and
the hardware necessary to access the Internet.
14-17 18-24 25-34 35-44 45-54 55-64 65-74 75+
25
% of Internet Users
20
15
10
5
0
Deep Rural
Shallow Rural
Urban
Source: OxIS 2013, n= 1839
Figure 2.6:
Educational Attainment Level of Internet Users by Urban, Shallow Rural and Deep Rural
NoQualication Secondary Further Higher
45
% of Internet Users
30
15
0
Deep Rural
Shallow Rural
Urban
Source: OxIS 2013, n= 1839

TWO-SPEED BRITAIN REPORT
24
OxIS2011identiedNext Generation Internet Use as
being related to the emergence of portability and access
throughmultipledevices,andofferedthedenitionof
a Next Generation User as “someone who accesses the
Internetfrommultiplelocationsanddevices.Specically,
weoperationallydenethenextgenerationuseras
someone who uses at least two Internet applications out
of the four applications queried in the survey, namely
browsing the Internet, using email, updating a social
networkingsite,orndingdirections,orwhotstwoor
more of the following criteria: they own a tablet, own
an e-reader (such as a Kindle), or own three or more
computers” (Blank and Dutton, 2011: 4).
Almost two thirds of Internet users who responded
to the OxIS 2013 are next generation users. However,
Figure 2.8 shows that deep rural Internet users are
much less likely to be Next Generation Users, and thus
more likely to be ‘First Generation Users’ (49%) than
urban dwellers (32%) and shallow rural dwellers (38%).
Thesedifferenceswerestatisticallysignicant(x2 =
21.43, p = 0.02). After controlling for age it was found
that those living in urban and shallow rural areas are
both around 1.5 times more likely to be Next Generation
Users of the Internet than those in deep rural areas.
Inotherwords,ourndingsclearlyreportalocational
effect.Thisreectstheinfrastructurelimitationsinthe
more remote and sparsely populated parts of the UK,
where connectivity on the move is limited (even if a
deep rural resident wanted to use the Internet on the
move, they would often be unable to do so) and where
lowbroadbandspeedsmakeitdifcultformorethan
one user per connection to be online at any one time.
Vignette 7 in Section 4 illustrates how ICT infrastructure
makesitdifcult,ifnotimpossible,tobeanext
generation user household in a remote rural area.
Figure 2.8:
Next Generation Internet Users by Urban, Shallow Rural and Deep Rural
First Generation Users
Deep Rural
% of Internet Users
80
60
40
20
0
Next Generation Users
Shallow Rural
Urban
Figure 2.7:
Social Grade of Internet Users by Urban, Shallow Rural and Deep Rural
Deep Rural
% of all respondents
40
30
20
10
0
Shallow Rural
Urban
Upper Middle
(A)
Lower Middle
(C1)
Middle
(B)
Skilled
(C2)
Working
(D)
Subsistence
(E)
Source: OxIS 2013, n= 1839
Source: OxIS 2013, n= 1839

25
TWO-SPEED BRITAIN REPORT
Ninety-one per cent of all people living in Britain
(aged 14 years and over) have a mobile phone. At
87%, the proportion of those living in deep rural areas
is lower than in urban and shallow rural areas (91%
and 92% respectively). Scrutiny of responses from
only those who lived in a household with Internet
access showed that in these households mobile
phone ownership was higher, at 99%, and that there
were no geographical differences in ownership rates.
Respondents were not asked to distinguish between
different types of mobile phone (e.g. basic model,
smart phone, 4G enabled etc.).
As Figure 2.10 shows, for all the mobile phone
features recorded, there is a general pattern of use
being highest in urban areas and lowest in rural
areas. For example, rural residents are less likely than
urban residents to use their mobile phones to send or
read emails (55% and 54%, respectively, in deep and
shallow rural areas compared to 66% in urban areas,
astatisticallysignicantdifference–x2 = 24.85, p =
0.00), post pictures or video online (40% in deep rural,
41% in shallow rural and 47% in urban areas), browse
the Internet (56% and 55%, respectively, in deep
and shallow rural areas compared to 62% in urban
Individuals in households that use the Internet have a
wide range of digital devices and, overall, have higher
levels of media ownership than do individuals in
households without an Internet connection.
Eighty-three percent of households that use the
Internet have a digital camera and over three-quarters
have at least one computer. More than half of Internet
user households have satellite TV (62%). However,
Figure 2.9 shows a general pattern whereby deep
rural households are generally less likely to own digital
devices than households in other locations: the notable
exception is of a TV set with a built-in connection to
the Internet, where deep rural household ownership,
at 30% is almost the same as ownership in urban areas
but 8% higher than the rate in shallow rural areas. This
could be an attempt to overcome the more limited
Freeview service in deep rural areas (the number of
channels available on Freeview varies considerably
across Britain, with the number being lowest in the
least densely populated areas). The gap between deep
rural household digital device ownership compared
with other locations is especially noticeable in the
cases of a games machine (x2 =19.36, p = 0.01) and a
tablet computer (x2 = 12.82, p = 0.04). The former could
berelatedtotheageproleofthedeepruralsample
(fewer young adults, the demographic most likely to
useagamesmachineetc.)andthelattercouldreect
the fact that the coverage of the 3G mobile Internet
signals required to use the Internet on the move on a
tablet is much poorer in deep rural than in other types
of area. While just over a quarter of households in
urban locations have a cable TV connection, only 12%
in shallow rural locations and 5% in deep rural locations
do (x2 = 69.52, p = 0.00). This is almost certainly a
reectionofcableTVinfrastructure,whichisrelatedto
population density across Britain.
2.iii Use of the Internet
Figure 2.9:
Information Communication Technologies in Internet Users’ Households by Urban, Shallow Rural
and Deep Rural
Deep Rural
100
% of Internet Users
80
60
40
20
0
Shallow Rural Urban
Cable
TV
TabletTV with
Built-in
Internet
Connection
Games
Machine
Reader Web-Cam Portable
MP3
Player
Satellite
TV
Digital
Camera
Computer
Source: OxIS 2013 Internet Users, n=1839

TWO-SPEED BRITAIN REPORT
26
areas), listen to music (42% in deep rural areas, 44% in
shallow rural areas and 51% in urban areas) and send
text messages (90% in deep rural areas, and roughly
95% in both shallow rural and 91% in urban areas, a
statisticallysignicantdifference(x2 = 8.66, p = 0.04)).
Internet based applications on mobile phones can only
be used ‘on the move’, away from a home or public
Wi-Fi network, in areas with outdoor 3G or 4G
coverage: deep rural areas have the most limited 3G
coverage in the UK and have negligible 4G coverage.
Rural Internet users are not using Internet enabled
mobile phone features in the way urban Internet
users do because they live in areas without the
infrastructure to fully support these ‘on the move’
activities.
Taking age into consideration, there are statistically
signicantdifferencesbetweenInternetusers’use
of mobile phone features by age for all uses except
making calls or sending texts. This is unsurprising:
younger adults – those who have grown up with
technologies such as mobile phones and the Internet –
are, in general, more likely than older adults to use the
non-telephony functions of mobile phones.
The most common creative online activities
(associated with Web 2.0 and user generated content
activities) are visiting social network sites and posting
photographs taken by the user online. Deep rural
dwellers are less likely to participate in these two
popular activities than those living in other locations
(Figure 2.11) but the differences were not large
enoughtobestatisticallysignicant.Interestingly,
deep rural Internet users were the most likely to
post messages on discussion or message boards
(42% compared to 34% in urban areas and only 30%
in shallow rural areas) – this difference was only
statisticallysignicantat90%.Overall,deeprural
Internet users reported lower levels of activity in the
types of Internet functions (e.g. posting photographs)
that require higher upload and download speeds
and/or more reliable Internet connections than are
often available to households in more remote rural
areas. Vignette number 2 in Section 2 illustrates the
frustration felt by young adults in remote rural areas
who cannot use 3G functionality on their mobile
phones where they live. Vignette number 11 includes
the image of a hillwalker using their online banking
application on top of a mountain. It would be very
unusual for someone to actually be able to use an app
on their mobile phone in this way: advertising may
promote potentially misleading pictures of what online
activities can be undertaken where. Other research
has suggested that deep rural residents are less likely
than others to use online social networking (OxIS 2011,
Wilson 2012), being more accustomed to being socially
connected through local, off-line networks.
Figure 2.10:
Internet Users’ Use of Features on Mobile Phones by Urban, Shallow Rural and Deep Rural
% of Internet Users
Deep Rural
80
100
60
40
20
0
Shallow Rural Urban
Post
Photos
Online
Browse
Social
Networking
Sites
Browse
the
Internet
Listen
to
Music
Sending
Photos
Play
Games
EmailFinding
Location
Send
Text
Message
Software
Application
Make
Phone Calls
Source: OxIS 2013, n= 1839

27
TWO-SPEED BRITAIN REPORT
Just over a half of all Internet users access the
Internet at more than one location (Figure 2.12).
However, deep rural dwellers are the most likely to
access the Internet from a single location only (38%
compared with 33% and 29% respectively). The
corollary is that deep rural dwellers are the least
likely to access the Internet from multiple locations.
Itisprobablethatthesendingsreectthefactthat
deepruralresidentsaremorereliantuponxed/
home based Internet connections than those who
live elsewhere because of poor 3G mobile coverage.
In other words, they are accessing the Internet
from fewer locations than their urban and shallow
rural counterparts most probably because it is more
difcultindeepruralareastobeonline‘onthemove’
due to a lack of mobile connectivity. Figure 2.13
provides some evidence that supports this conclusion:
it shows that deep rural dwellers are the most likely
not to use their mobile phone to access the Internet
(25% in deep rural areas compared with 16.9% in
shallow rural and 15% in urban areas).
Figure 2.12:
Number of Locations Where Internet is Accessed by Urban, Shallow and Deep Rural
Deep Rural
45
% of Internet Users
30
15
0
Shallow Rural
Urban
1 3 5+2 4
Figure 2.11:
Creativity and Productivity Online by Urban, Shallow Rural and Deep Rural
Deep Rural
100
% of Internet Users
80
60
40
20
0
Shallow Rural
Urban
Visit Social
Network
Site
Read
a Blog
Write
a Blog
Post Photos
You Have
Taken
Post
Messages
Maintain a
Personal
Website
Source: OxIS 2013, n= 1839
Source: OxIS 2013, n= 1839

TWO-SPEED BRITAIN REPORT
28
Internet users mainly use search engines or a
combinationofsearchenginesandspecicwebsites
to access the content they want to view online. As
shown in Figure 2.14, most Internet users, regardless
ofwheretheylive,makeuseoftheInternettond
information about topics such as local events, news,
travel planning etc., but deep rural dwellers are less
likely than those living in other locations to use the
Internet for any of the activities reported in Figure
2.14.Thereisastatisticallysignicantdifference
between those in the different types of area who
report looking for information about jobs or work
online (x2 = 19, p = 0.04). Those living in deep rural
areas are much less likely use the Internet for job
seeking than those in urban areas. This could be
because of the higher proportion of Internet user
respondents in deep rural areas who are retired (i.e. not
lookingforemploymentopportunities)orcouldreect
the fact that more traditional methods of advertising
local jobs (e.g. word of mouth, local print newspapers)
are most common in deep rural areas and their use has
notbeensupplantedbyonlinenotications.
2.iv Use of Information and Online Services
Figure 2.14:
Information Seeking Online by Urban, Shallow Rural and Deep Rural
Deep Rural
90
% of Internet Users
60
30
0
Shallow Rural
Urban
Travel Plans News Sports
Information
Information
About Local
Events
Health
Information
Jobs/Work
Figure 2.13:
Use of Mobile Phone & Other Devices for Mobile Access to the Internet by Urban, Shallow Rural
and Deep Rural
Mostly Use
Something Else to
Access Internet
Use Mobile Phone and
Something Else Equally
to Access Internet
Mostly Use Mobile
Phone to Access
Internet
Do Not Use
Mobile Internet
60
% of Internet Users
45
15
30
0
Deep Rural
Shallow Rural
Urban
Source: OxIS 2013, n= 1839
Source: OxIS 2013, n= 1839

29
TWO-SPEED BRITAIN REPORT
Figure 2.15 shows that urban and shallow rural
residents are much more likely to use the Internet
for various forms of entertainment than those in
deep rural areas, with the exception of posting a
video where the differences are not statistically
signicant(listeningtomusiconlinex2 = 16.63,
p = 0.02; downloading music x2=13.7, p=0.08;
download videos x2 = 30, p = 0.03). Once again
thisislikelytoreectconnection/infrastructure
capabilities: deep rural areas are least likely to have
fast enough and reliable enough download speeds to
downloadorstreamTV,lms,orvideo.
Figure2.16reportsndingsrelatedtotheuse
of online services, such as e-commerce. A high
proportion of Internet users engage in e-commerce.
There are only small geographical differences in
the proportions of Internet users who make travel
reservations, compare products or buy products
online. Noticeable geographical differences are,
however, reported for selling online, for online
grocery shopping and for paying bills. At 36%, shallow
rural dwellers are the most likely to buy groceries on-
line whilst deep rural dwellers are, at 26%, the least
likely. Penetration of supermarkets’ home delivery
services is variable across Britain and, in many remote
areas, not available. Shallow rural residents are also
the most likely, at 45%, to sell goods online compared
with those living in the other locations. The lower
proportion of deep rural respondents purchasing
orsellingitemsonlinecouldreecttheoftenhigher
delivery/postal charges of non Royal Mail carriers
that are imposed on consumers living in many remote
andrurallocations.Itisalsolikelytoreecttheage
proleofdeepruralareas:theover65sarelesslikely
than those in younger age groups to make purchases
overtheInternet(OfceforNationalStatistics,2013)
and deep rural areas have the highest proportion
of residents in this age group. Interestingly, deep
rural Internet users are the most likely to pay bills
online;perhapsthisreectsthegreaterdifcultiesin
accessingpaypointssuchasbanksorpostofcesin
the deep rural areas.
Figure 2.15:
Entertainment and Leisure Online by Urban, Shallow Rural and Deep Rural
Deep Rural
100
% of Internet Users
80
60
40
20
0
Shallow Rural
Urban
Listen to
Music
Watch
TV
Download
Videos
Download
Music
Watch
Films/Videos
Post Videos
Created
Source: OxIS 2013, n= 1839

TWO-SPEED BRITAIN REPORT
30
A third of Internet users access the Internet at
work:thelackofsignicantdifferencesbylocation
probablyreectstheubiquityofbeingonlineinmany
workplaces (Figure 2.17). Deep rural residents are
the most likely to often or always work at home:
32% compared to 24% of shallow rural dwellers
and17%ofurbanones.Thisislikelytoreectself-
employment (and, by inference, working from home)
patterns: the proportion of self-employed (and thus
those most likely to work from home) is highest in
remote rural areas across the UK (for example, the
Scottish Government (2012) reported that 29% of
economically active men in ‘remote’ rural areas are
self-employed compared to 23% in ‘accessible’ rural
and 13% in urban Scotland).
Deep rural Internet users who are in employment are
alsosignicantlymorelikelytooftenusetheInternet
at home for work related activities: 42% compared to
27% of the in employment Internet users as a whole
(Figure2.18).Thiscouldbeanotherreectionof
patterns of home working (deep rural Internet users
aremorelikelytoworkfromhome),butitcouldreect
the fact that many deep rural residents cannot access
2.v Use of the Internet at Work and Home
Figure 2.16:
Buying and Using Services Online by Urban, Shallow Rural and Deep Rural
Deep Rural
100
% of Internet Users
80
60
40
20
0
Shallow Rural
Urban
Order Food/
Groceries
Pay Bills Compare
Products
Sell
Products
Make
Travel
Reservations
Use
Bank
Services
Buy
Products
Source: OxIS 2013, n = 2657
Figure 2.17:
Working from Home by Urban, Shallow Rural and Deep Rural
Never Seldom Often Do All Work
60
% of employed Internet Users
Frequency of employed internet users working from home
45
15
30
0
Deep Rural
Shallow Rural
Urban
Source: OxIS 2013, n=989

31
TWO-SPEED BRITAIN REPORT
mobile Internet services, which leads to a reliance
upon home-based Internet provision when they are
not at their place of work. Vignettes number 8 and
10 illustrate how poor Internet connectivity at home
makesitdifcultforanemployeewhoisfrequently
‘oncall’tofullthedemandsandexpectationsofan
employer. In contrast, vignette number 12 reports what
a good Internet connection at home can allow a remote
rural home-worker to achieve.
Some people use the Internet at home for work
related activities. For most respondents, having home
Internet has not changed the amount of work they
do from home. However, as shown in Figure 2.18,
employed Internet users in deep rural areas were
the most likely to often use the Internet at home for
work, and those living in urban areas were the most
likely to never use it. The geographic differences
reportedinFigure2.19arestatisticallysignicant
(x2 = 44.53, p=0.005). As reported in Figure 2.19,
most respondents reported that having access to
email and the Internet at home had not changed
the amount of work they did at home. Deep rural
respondents were the most likely to report that it had
increased the amount of work they did at home, which
couldbeareectionofgradualimprovementsinIT
infrastructure making home working more feasible
than before.
Figure 2.18:
Frequency of Use of Internet at Home for Work Related Activities
Never
Deep Rural
100
% of Employed Internet Users
80
60
40
20
0
Seldom Often
Shallow Rural
Urban
Figure 2.19:
Does Having Access to Email and Internet Change the Amount of Work You Do at Home?
Increase
Deep Rural
100
% of Employed Internet Users
80
60
40
20
0
No Difference Decrease
Shallow Rural
Urban
Source: OxIS 2013, n=989
Source: OxIS 2013, n=989

TWO-SPEED BRITAIN REPORT
32
With respect to the use of the Internet for social
communication online, the most common activity
for all Internet users, regardless of where they live,
is checking email. Geographical differences were
observed for this very common activity: 2% of urban
Internet users did not use email, compared to 4% in
shallow rural and 7% in deep rural areas (x2 = 13.22,
p = 0.02). Other modes of communication also show
geographical differences – see Figure 2.20. Residents
ofurbanareasaresignicantlymorelikelytouse
the Internet for making or receiving phone calls than
those in rural locations (x2 = 33.63, p = <0.00). This
islikelytobeareectionofgeographicallyvariable
infrastructure: more urban Internet users live where
there is capacity in the ICT network to support
applications such as Skype. Urban Internet users are
alsosignicantlymorelikelytouseinstantmessaging
than those in shallow rural areas or deep rural areas
(x2=45.96,p=0.00),possiblyareectionofthe
fact that urban residents are more likely to be users
of social networking sites with embedded instant
message services than deep rural residents are, as
noted in Figure 2.20.
Internet users in deep rural areas are the least likely
to use social networking sites (Facebook, LinkedIn,
Twitter, Pinterest, Instagram etc.), with 70% doing so
compared with 75% of users in shallow rural areas
and79%inurbanlocations.Thismightbeareection
ofageproles:theproportionofolderpeopleis
highest in deep rural areas and this age group is less
likely to use social networking than younger age
groups(OfceforNationalStatistics,2013).Amongst
those who use social networking sites, there are few
differences in the number of sites that were reported
as being used by residents in the three types of area.
The Internet has opened up new ways of
communicating with friends and family. OxIS 2013
respondents from urban, shallow and deep rural
areas reported no notable differences in their modes
of communicating with family and friends who live
nearby. However, as shown in Figure 2.21, there are
differences in the ways that people in urban, shallow
rural and deep rural areas communicated with friends
and family who lived far away. Deep rural dwellers are
more likely to use text messages and the telephone
to keep in touch with relatives who live far away than
those living in urban locations. Examination of the
frequency of contact with family and friends who live
far away using different modes of communication
found that phone, text and email were the most
commonly used methods and that while email use
varied little by geographical location, deep rural
respondents were the most likely to have weekly or
daily contact by phone and by text message.
Maintain
Personal
Website
Write
Blog
Post
Messages
Internet
Phone
Call
Post
Photos
Send
Email
Attachment
Instant
Messaging
Check
Email
100
% of Internet Users
80
60
40
20
0
Deep Rural
Shallow Rural
Urban
Figure 2.20:
Communication Online by Urban, Shallow Rural and Deep Rural
Source: OxiS 2013, n=1839

33
TWO-SPEED BRITAIN REPORT
This section of the report has described some aspects
of Internet use and has shown where Internet use in
urban, shallow rural and deep rural areas is similar
or different. Most striking are differences between
urban and shallow rural Internet users, and deep
rural Internet users, and it is likely that they can be
explained, at least in part, by the variations in ICT
infrastructurenationwide.Keyndingsinclude:
• Non-use of the Internet is most common in deep
rural areas and least likely in shallow rural areas.
• Deep rural Internet users are the most likely to
think that their Internet connection is ‘too slow’
and are over 5 times more likely than urban
Internet users to think that their connection is
‘too slow’.
• Deep rural Internet users are more likely to be
in the upper middle and middle socio-economic
groups than Internet users who live in shallow
rural and urban areas.
• Next Generation Users are least common in deep
rural areas, even when age is controlled for.
• The use of Internet-enabled features on mobile
phones (e.g. email, browsing the Internet) is lowest
in deep rural areas.
• Deep rural Internet users are the least likely to
access the Internet from multiple locations and
are least likely to use their mobile phones to
access the Internet. Deep rural Internet users are
thus the most likely to be limited to home-based,
xedbroadbandorpublicallyavailablyWi-Fi
connections for their Internet connectivity.
• Urban and shallow rural Internet users are much
more likely to use the Internet for entertainment
(e.g.streaminglms)thandeepruralInternet
users.
• The increase in home working facilitated by having
email and Internet access is most pronounced in
deep rural areas.
2.vi Key Findings
100
80
60
40
20
0
Figure 2.21:
Communication with Family and Friends who Live Far Away by Urban, Shallow Rural and Deep Rural
Deep Rural
% of Internet Users
Shallow Rural
Urban
Letter Email Text MessagingVisit Phone
Mode of Communication with Family and Friends Who Live Far Away
Source: OxiS 2013, n=1839

3. ICT infrastructure:
Internet Use,
Broadband Speeds
and 3G and 4G
Availability
34 TWO-SPEED BRITAIN REPORT

35
TWO-SPEED BRITAIN REPORT
The most recent Ofcom data reported in Maps 1.1, 1.2 and 1.3 clearly
show that the provision of telecommunications infrastructure that
supports Internet access is variable across the United Kingdom. The
areas worst served by 3G coverage (which supports mobile Internet
access), and those areas most likely to have a high proportion of
households connected to a xed broadband connection with a speed
of 2.2Mbit/s or less, are concentrated in peripheral and remote rural
areas.
These areas are the least likely to have superfast
broadband available and they have no 4G coverage
at present. A recent Ofcom report that published an
analysis of download speeds by geography (Ofcom,
2014a, p1) notes:
“ …the average urban download speed in November
2013 was 31.8Mbit/s, a 21% increase since May 2013;
the average suburban download speed in November
2013 was 21.8Mbit/s, a 22% increase since May 2013.
The research also suggests that average speeds in
rural areas increased from 9.9Mbit/s to 11.3Mbit/s
between May and November 2013”.
Whilst the increases in average download speeds
in urban and suburban areas between May and
November2013werestatisticallysignicant,the
increase reported for rural areas was not (Ofcom,
2014a and b). The difference between average
urban and rural download speeds increased over the
six month period of May to November 2013, from
16.5Mbit/sto20.6Mbit/s(Ofcom,2014b).Asbre
broadband (which supports superfast broadband)
availability increases in rural areas (c.f. the BDUK
supported roll out of superfast broadband) this
differential is expected to contract but, in the short
term, the differences may increase.
Average download speed data hide considerable
variations, notably the extent to which download
speeds are affected by network contention such as
that which occurs at peak times. Peak time speeds,
on all types of broadband connection, are reported
by the UK communications watchdog to be lower
than average maximum speeds and the 24 hour
average speed (Ofcom, 2014b). ADSL connections are
particularly badly affected by peak time contention:
‘for ADSL connections capped at 10Mbit/s or less,
the peak-time download speed was 3.2Mbit/s, 86%
of the average maximum speed, and 98% of the 24
houraverage’(Ofcom,2014bp5).Cableandbre
connections are, to date, far less common in rural
areas than in urban areas. In consequence, the peak
time contention experienced on ADSL connections
has a considerable impact on the broadband speeds
available to much of the rural population and the
impact of contention is, arguably, of more importance
in rural areas because it reduces download speeds
suchthatsomeonlineactivitiesbecomedifcultifnot
impossible. Recent Ofcom reports have not included
information about satellite broadband in their
download speed analysis.
Less data about upload speeds are published than
is the case for download speeds. The importance of
upload speeds is recognised by Ofcom: “ … upload
speedsmattertothosesharinglargeles,usingreal-
time two-way video communications and for some
online gaming” (Ofcom, 2014b, p6). Across the UK
average upload speeds increased in 2012 and 2013,
but the rate of improvement was most pronounced
for households with an ‘up to’ 30Mbit/s and higher
broadband connection. In other words, the increase
in upload speeds is largely due to increased take-up of
superfast broadband services which support higher
upload speeds than broadband connections provided
via older technologies, such as the ADSL connections
serving many remote rural areas.
Opposite: Photo courtesy of Dr Lorna Philip;
not to be reused without prior permission.

TWO-SPEED BRITAIN REPORT
36
Broadband speed data is published by Ofcom for
local authority areas and at unit postcode (i.e. AB24
3UF) level. The unit postcode is the smallest of the
geographical units represented by UK postcodes and
in 2011 there were 1.7 million unit postcodes across
the UK. A unit postcode represents a group of adjacent
premises:theUK’sOfceofNationalStatisticsnotes
that “A single small user postcode may contain up
to 100 addresses, but 15 is a more typical number”
(http://www.ons.gov.uk/ons/guide- method/geography/
beginner-s-guide/postal/index.html). The sampling
strategy for OxIS was based on unit postcodes (see
Oxford Internet Institute, 2013 for full details of the
sampling procedure).
Data made available at unit postcode level such
as the broadband speed data, therefore represent
micro-scale geographical units. The most recent unit
postcode level broadband speed data were published
in December 2013 (available at http://maps.ofcom.org.
uk/broadband/). These data were matched to the unit
postcode of each respondent to the Oxford Internet
Survey (there were 965 unit postcodes in total) and
variables from the OxIS dataset were analysed against
speed data. Our analysis shows that respondents to
the OxIS 2013 survey lived in unit postcode areas that
experienced a wide range of broadband sync speeds3.
The average broadband sync speed available to OxIS
2013 respondents varied by the type of location in
whichrespondentslived.Selectedndingsfromthis
analysis are presented in Table 3.1. Speeds experienced
by the deep rural sample were the lowest, whilst the
highest speeds were available to respondents living in
urban areas. Average sync speeds were highest for the
urban sample, which was twice as high as the shallow
rural average and three times as high as the deep
rural average. The highest sync speed for any deep
rural sampling unit postcode, 17.4Mbit/s, was lower
than the average sync speeds for the urban sample.
As already noted in this report, broadband speeds
directlyinuencewhatcanandcannotbedoneonline.
Low speeds make ‘data heavy’ download and upload
activities either very slow or impossible.
3.i Broadband Speeds for Respondents to the Oxford Internet Survey 2013
3 Also known as Sync rate, Downstream rate or DSL Connection rate, Sync speed is “the ‘physical’ speed of the connection between your router
(or modem) and your local telephone exchange. It’s determined by the characteristics of your line e.g. line length and quality” (Plusnet Broadband
terminology guide, available at http://www.plus.net/support/broadband/speed_guide/broadband_terminology.shtml).
Table 3.1: Selected speed by location data for OxIS 2013 sample
Urban Shallow Rural Deep rural
Average sync speed (Mbit/s) for postcodes of sampling
points
19.2 10.5 6.2
Minimum and maximum sync speeds (Mbit/s) for postcodes
of sampling points
1.5 – 30+ 0.8 – 30+ 0.6 – 17.4
% of respondents living in an area with superfast broadband
available
86% 30% 0%
Source: Based on Ofcom UK broadband speed data at postcode level 2013 and all OxIS 2013 responses

37
TWO-SPEED BRITAIN REPORT
Figure3.1displaysave-foldaveragesyncspeed
(each category was determined by an evaluation of the
distribution of average sync speeds for the entire OxIS
2013 sample of 965 unit postcode data points) for all
respondents to the study by the type of area they lived
in (urban, shallow rural and deep rural). Whilst 53.2%
of the deep rural sample lived in unit postcode areas
where the average sync speed was less than 6.3Mbit/s,
only 4.9% of the urban sample lived in an area with this
speed. Conversely, whilst almost a half of the urban
sample lived in an area with average speeds of at least
20.6Mbit/s, only 12.6% of shallow rural respondents
lived in an area with speeds of this level and none of
the deep rural sample lived in areas with this speed.
The area based differences in speed were statistically
signicant(x2 = 860.32, p = 0.000). These speed data
clearly show there is a ‘Two speed Britain’ in that the
lowest speeds are most commonly found in rural areas
whilst the highest speeds are most commonly found
in urban areas. It also highlights the difference within
rural Britain, further demonstrating the usefulness of
the ‘rural boost’ to the OxIS 2013 survey.
Unfortunately, 3G coverage data is not available at unit
postcode level in a standardised format from Ofcom so
we have been unable to link the survey sampling points
with mobile services data. However, as shown in Map
1.1 in Section 1, the geographical coverage of a reliable
outdoor mobile signal being available from any operator
is highly variable, with the best coverage corresponding
to densely populated areas and the worst coverage
corresponding to less densely populated areas. It is thus
reasonable to infer that the OxIS 2013 respondents
least likely to live in an area with a reliable 3G signal are
those who live in deep rural areas.
Towhatextentdoesbroadbandspeedinuence
responses to questions about online behaviour
andexperiences?Twospeedvariableswereused
to explore this: (i) average connection speeds that
exceeded the minimum requirements to watch the
BBCiPlayerinHighDenition–thatis,3.5Mbit/s(as
stated on the BBC website) and (ii) whether or not
superfast broadband was available at that postcode.
In the absence of robust data about upload speeds,
variable (i) was created to represent a proxy for the
minimum speed that would allow a user the ability
to download and upload photographs, stream movie
clips etc. although we recognise that these data
heavy online activities would be very slow at this
speed. The effect of peak time contention on speed,
as noted above, is likely to mean that respondents
whose average connection speed was 3.5Mbit/s or
less are likely to experience times when their speed is
much lower than the average. Figure 3.2 shows that
the proportion of the differences between the three
differenttypesofareawasstatisticallysignicant(x2
= 434.24, p = 0.000). Respondents living in rural areas
are much more likely than those living in urban areas
to live in a unit postcode with the lowest speed.
Note: 30+Mbit/s sync speed available either via Virgin Media cable, Openreach Fibre-To-The-Cabinet or Digital Region networks
Figure 3.1:
Average Sync Speeds (Five-Fold Grouping) by Urban, Shallow Rural and Deep Rural
20 40 60 80 1000
% of Internet Users
Urban
0 - 6.3Mbit/s
Shallow Rural 6.4 - 13.8Mbit/s
20.6 - 29.9Mbit/s
Deep Rural
13.9 - 20.5Mbit/s
30+Mbit/s
Source: OxIS 2013 all responses, n=2657, WUDS weighting applied, and Ofcom UK broadband data postcode level 2013

TWO-SPEED BRITAIN REPORT
38
Next generation users and low broadband speed
Slow Internet connections are unsuitable for
multiple, simultaneous use (by multiple individuals
in a household being online at the same time and/ or
multiple devices being connected to a single Internet
connection simultaneously). Our analysis shows
that speed only has a very weak association with the
likelihood of an Internet user household being a next
generation user household (Figure 3.3). This suggests
This section attempts to explore some of the place
baseddifferencesinInternetbehaviouridentied
and discussed in Section 2 of the report to establish
whether they are associated with broadband
speed. In so doing we attempt to explore whether
infrastructureattributesareinuencingbehaviour
(bearing in mind that infrastructure variations are
directly linked to geography).
3.ii Broadband Speed, Location and Internet Behaviour
Figure 3.3:
Next Generation Users and Low Broadband Speed
% of Internet Users
100
0
60
80
40
20 Not a Next Generation User
Next Generation User
Average Sync Speed
Is Less Than 3.5Mbit/s
Average Sync Speed
Exceeds 3.5Mbit/s
Figure 3.2:
Connection Speed by Location: Connection Exceeds Minimum Required to Watch the BBC iPlayer
Connection Does Not
Exceed BBC iPlayer
Minimum Requirements
Deep Rural
% of All Respondents
100
80
60
40
20
0
Connection Exceeds
BBC iPlayer Minimum
Requirements
Shallow Rural
Urban
Source: OxIS 2013 all responses,
n=2657, WUDS weighting applied
Source: OxIS 2013 Internet Users,
n=1839, WALL weighting applied

39
TWO-SPEED BRITAIN REPORT
that households that want to make use of a variety
of technologies to use the Internet try to do so.
However,thesendingssaynothingaboutwhetheror
notallnextgenerationusersaresatisedwiththeir
Internet connection when multiple use of it is being
made. Vignette number 7 in Section 4 illustrates the
challenges faced by next generation households with
low broadband speeds.
‘Data heavy’ online activities and low broadband speed
In Section 2 it was postulated that some online
activities, such as listening to music online, watching
movies online etc, were not as common in deep rural
areas as elsewhere because Internet speeds in many
deepruralareaswerenotsufcienttosupportthese
activities. As shown in Figure 3.4, Internet users with
the lowest low sync speed were less likely than those
with sync speeds exceeding 3.5Mbit/s to report that
they participated in ‘data heavy’ activities including
watching TV programmes on the Internet, posting
videos including music videos, and downloading
musiconline.Statisticallysignicantdifferenceswere
observed for listening to music online (x2 = 4.53, p
= 0.03) and downloading videos (x2 = 4.59, p =0.05),
both activities that require large quantities of digital
data to be streamed in real time and are beyond the
capabilities of a slow broadband connection.
Respondents who lived in areas where superfast
broadband was available were more likely than those
without access to superfast broadband to participate
in ‘data heavy’ online activities, as illustrated in Figure
3.5. On a fast connection, Internet users can undertake
these ‘data heavy’ activities far more quickly and
reliably than on slower connections, even at peak times
whencontentioncancreatedifcultiesforthoseusing
theInternet.Statisticallysignicantdifferenceswere
observedfor‘watchmoviesorlmsonline’(x2 = 8.85, p
= 0.03) and ‘download videos’ (x2 = 11.35, p = 0.02).
The comparison of slow speeds and fast speeds
and ‘data heavy’ online activities could mean that,
irrespective of broadband speed, if someone wants
to participate in ‘data heavy’ online activities they
do so. However, it must be noted that the Oxford
Internet Survey respondents were not asked how
often they undertook these activities or questioned
about whether they found undertaking these online
activities problematic in any way (e.g. had continuity
or buffering problems, or found the activity too
slow).
Figure 3.4:
Selected ‘Data Heavy’ Online Activities and Sync Speeds of Less than 3.5Mbit/s
100
% of Internet Users
0
60
80
40
20 Yes
Never
Average
Sync Speed
is Less Than
3.5Mbit/s
Listening to
Music Online
Watch Movies or
Films or Videos Online
Download
Videos
Average
Sync Speed
is Less Than
3.5Mbit/s
Average
Sync Speed
is Less Than
3.5Mbit/s
Average
Syncs Speed
Exceeds
3.5Mbit/s
Average
Sync Speed
Exceeds
3.5Mbit/s
Average
Sync Speed
Exceeds
3.5Mbit/s
Source: OxIS 2013 Internet Users,
n=1839, WALL weighting applied

TWO-SPEED BRITAIN REPORT
40
Social networking sites have opened up new modes
of keeping in touch with friends and family, developing
new personal relationships and keeping in touch with
special interest groups. Their use may be associated
with the speed of broadband connection available.
Internet users from areas with average sync speeds
of less than 3.5Mbit/s did not make use of social
networking opportunities such as instant messaging
or chat rooms any differently from respondents with
higher connection speeds. However, use of social
networking and associated applications was observed
to be more common amongst those Internet users who
lived in areas where superfast broadband was available,
asillustratedinFigure3.6.Statisticallysignicant
differences were observed for ‘participate in chat rooms’
(x2 = 12.70, p = 0.02), and ‘use MySpace’ (x2 = 18.05, p
= 0.001). Internet users living in an area with superfast
broadband available were also more likely to use the
social networking site Bebo than those without access to
superfast broadband; however, it should be noted that
across the UK there are regional patterns associated
with the use of different social networking sites, such
that whilst Bebo might be commonly used in one area, its
use in another is uncommon.
Social networking and broadband speed
Source: OxIS 2013 Internet Users,
n=1839, WALL weighting applied
Figure 3.5:
Selected ‘Data Heavy’ Online Activities and Superfast Broadband Availability
100
% of Internet Users
0
60
80
40
20
Yes
Never
Superfast
Broadband
Available
Listening to
Music Online
Watch Movies or
Films or Videos Online
Download
Videos
Superfast
Broadband
Available
Superfast
Broadband
Available
Superfast
Broadband
Not Available
Superfast
Broadband
Not Available
Superfast
Broadband
Not Available
Source: OxIS 2013 Internet Users,
n=1839, WALL weighting applied
% of Internet Users
Figure 3.6:
Social networking and Superfast Broadband
100
0
60
80
40
20
Yes
Never
Superfast
Broadband
Available
Participate in
Chat Rooms
Use Instant
Messaging
Use
Myspace
Superfast
Broadband
Available
Superfast
Broadband
Available
Superfast
Broadband
Not Available
Superfast
Broadband
Not Available
Superfast
Broadband
Not Available

41
TWO-SPEED BRITAIN REPORT
Figure 3.7 suggests that online behaviour associated
with working lives has a relationship with the
speed of an employed Internet user’s broadband
connection. It appears that the likelihood of
respondents using the Internet to read or send
work email or other electronic messages is highest
for those with the slower sync speed. Likewise,
those with the slower sync speed are the most
likely to report that having access to email and the
Internet has increased the amount of work they
do at home. Those most likely to report that they
often used the Internet at home for work related
activities were in the lowest sync speed group,
as were those who always worked from home.
However, none of the relationships presented in
Figure 3.4 show differences large enough to be
statisticallysignicant.Nostatisticallysignicant
differences between the online behaviour associated
with working lives and the availability of superfast
broadband were observed, but this could well
beareectionofthefactthatthenatureofan
individual’s work varies (whether or not they are
ofceordeskbound,travelregularly,workindoors
or outdoors etc.).
The illustrations of Internet behaviour and speed
presented above presented some rather puzzling
ndings.Whileitcouldbeassumedthatslowspeeds
would prevent people from undertaking data heavy
online activities this does not appear to be the case.
Respondents appear to be engaging in online activities
despite the problems their low speed connections
bring: vignette number 6 illustrates that despite
having to wait minutes for a web page to load the users
persevered because they wanted the information to
which the Internet facilitated access. People want
to be online, they want to participate in the types
of activities seen as being increasingly ubiquitous
across the UK, and if they have never had personal
experience of a faster broadband connection they do
not know how poorly their broadband compares with
connections found elsewhere. This would certainly
explain why some next generation users are found in
low speed areas. Some Internet users have no choice
but to be online. For example, some farm regulatory
paperwork must be submitted online, and many small
businesses feel that they must have an online presence.
In such cases, if an activity has to be conducted online,
then it must be done, regardless of how long it takes.
Working lives, online behaviour and broadband speed
Figure 3.7:
Selected Work Related Online Activities and Speed
% of Employed Internet Users
Average sync speed is less than 3.5Mbit/s
80
100
60
40
20
0
Average sync speed exceeds 3.5Mbit/s
No
Do You Read or
Send Work Email
or Other
Do Access to Email and the
Internet Change the Amount
of Work You Do At Home
How Often Do You Use the
Internet At Home for Work
RelatedActivities?
Do You Work All Or Some
OfTheTimeFromHome?
Yes
Seldom
Increases
Often
Never
No Difference
Seldom
Decrease
Often
Never
Always
Source: OxIS 2013 Internet Users, n=1839, WALL weighting applied

TWO-SPEED BRITAIN REPORT
42
In section 3.ii it was established that, regardless of
broadband speeds, Internet users in urban, shallow
rural and deep rural areas are engaging in a wide
variety of online activities. Use does not, however,
provide any evaluation of user satisfaction. Figure
2.3 in Section 2 clearly shows that deep rural
Internet users were the most likely to think that
their Internet connection was ‘too slow’ to allow
them to do the things they wanted to do online.
We hypothesise that this view is directly related
to the low speeds commonly found in deep rural
areas. The following section considers whether this
observation is borne out, i.e. if speed of a broadband
connectioninuenceswhetherornotInternetusers
think they can do what they want to do online.
Figure 3.8 reports users’ perceptions of the Internet
speeds they experienced, using the terms ‘too slow’,
‘it depends’ and ‘fast enough’, and distinguishing
whether or not their average speed exceeded the
minimum for using the BBC iPlayer. Responses
are shown by location in urban, shallow rural or
deep rural areas. For users with speeds exceeding
3.5Mbit/s, satisfaction with their Internet speed was
statistically different for deep rural, shallow rural, and
urban respondents. (x2 = 47.31, p = >0.001). Deep rural
and shallow rural respondents with speeds of less than
3.5Mbit/s were the most likely to report that their
Internet connection was not fast enough to do what they
want to do online: 48% of deep rural and 36% of shallow
rural respondents with this connection speed expressed
this view. In urban areas, those with a perceived need for
greater speed are likely to have switched away from <3.5
Mbit/s, since they have more options, leaving only users
with minimal needs who might be content with slower
speeds, such as for handling email. Conversely in deep
rural areas, many users with an interest in applications
for which higher speeds would be very valuable, such as
video conferencing, might be stuck on <3.5Mbit/s, unable
to upgrade to a higher speed connection. Therefore, they
are most likely to consider their connections as ‘too slow’.
But deep rural users on >3.5 are also more likely to say
that their connection is ‘too slow’, than are shallow users,
orurbanuserson>3.5.Thismayreectweakerandless
reliable connections in deep rural areas when compared
to shallow rural and urban areas.
3.iii Does Broadband Speed Inuence Internet Users’ Ability to do What They Want Online?
Figure 3.8:
Is Your Internet Connection Fast Enough to do what You Want Online by Urban, Shallow Rural and
Deep Rural, and Speeds Greater and Lower than 3.5Mbit/s
Fast enough
100
% of Internet Users
0
60
80
40
20 It depends
Too slow
Average
Sync Speed
is Less Than
3.5Mbit/s
Deep Rural
Internet Users
Shallow Rural
Internet Users
Urban
Internet Users
Average
Sync Speed
is Less Than
3.5Mbit/s
Average
Sync Speed
is Less Than
3.5Mbit/s
Average
Syncs Speed
Exceeds
3.5Mbit/s
Average
Sync Speed
Exceeds
3.5Mbit/s
Average
sync speed
exceeds
3.5Mbit/s
Source: OxIS 2013 Internet Users,
n=1839, WUDS weighting applied

43
TWO-SPEED BRITAIN REPORT
It could reasonably be expected that those with
access to superfast broadband would report that
they could do what they wanted online all the time.
ThendingsreportedinFigure3.9donotsupport
this assumption. Approximately two thirds of urban
Internet users and half the shallow rural Internet
users who lived in a unit postcode area where
superfast broadband was available thought their
connection was ‘fast enough’. This leaves a sizeable
minority of urban and shallow rural Internet users
reporting that, despite access to superfast broadband,
their Internet connection was not fast enough.
Shallow rural respondents were the most likely to
consider that their Internet connection was ‘too
slow’. These area-based differences were statistically
signicant(x2=23.79,p=0.03)andcouldreect
the fact that consumers always want more from the
services they pay for: with so many daily activities
now being conducted online, some people may have
unrealistic expectations of what even the most up-to-
date digital infrastructure can support.
Is there is a speed threshold above which perceptions
that an Internet connection is not fast enough to
allowpeopletodowhattheywantonlinedeclines?
Figure 3.10 considers Internet users’ perceptions of
thespeedoftheirInternetconnectionbyave-fold
speedclassication.Acomplexpictureofrespondent
opinions is evident. Note that none of the deep rural
respondents lived in a unit postcode where Ofcom
reported that average sync speeds greater than
20.5Mbit/s were available and very few lived in areas
where speeds in excess of 13.8Mbit/s were available.
This limited the statistical analysis that could be
performed on the data reported in Figure 3.10.
Shallow rural Internet users appear to be the most
critical of their broadband speeds even when they live in
unit postcodes with access to the highest speeds: they
are much less likely than their urban counterparts to
think that superfast broadband is ‘fast enough’. Perhaps
thisisareectionofthesocio-economiccomposition
of suburban areas (home to many who live in areas
classiedas‘shallowrural’)whicharecharacterisedby
being home to more households in the higher income
groups than are found in urban and deep rural areas. It
couldalsoreectthefactthatsinglepersonhouseholds
are less common in shallow rural areas than elsewhere
(the suburbs in close proximity to large urban areas
are popular locations for families) and in households
where even a fast Internet connection may suffer a drop
off in speeds when multiple users want to be online
simultaneously resulting in the households’ connectivity
being viewed unfavourably.
Figure 3.9:
Is Your Internet Connection Fast Enough to Do What You Want Online by Urban, Shallow Rural and Deep
Rural, and Speeds Greater and Lower than 30Mbit/s (‘superfast broadband’)
Fast enough
100
% of Internet Users
0
60
80
40
20 It depends
Too slow
Superfast
Broadband
Not Available
Deep Rural Shallow Rural Urban
Superfast
Broadband
Not Available
Superfast
Broadband
Not Available
Superfast
Broadband
Available
Superfast
Broadband
Available
Superfast
Broadband
Available
Source: OxIS 2013 Internet Users,
n=1839, WUDS weighting applied

TWO-SPEED BRITAIN REPORT
44
In deep rural areas, a speed of up to 6.3Mbit/s was
considered ‘too slow’ by 38% of Internet users but
48% thought it was ‘fast enough’. Contrast this with
the 34% of deep rural Internet users who though that
a speed in the 13.9Mbit/s – 20.5Mbit/s range was
‘too slow’ whilst 66% thought this speed was ‘fast
enough’.Onereadingofthisndingisthattheslowest
speeds are satisfying deep rural Internet users’ needs.
However, we caution against such an assumption.
Many people who live in deep rural areas only have
personal experience of being online in deep rural
areas, and they may never had personal experience
of using an Internet connection faster than the one
they have at home or work: they can thus only base
their speed satisfaction ratings on what they know
and what they are able to use the Internet for. The
fact that so many urban and shallow rural Internet
users living in areas with speeds of 20.6-29.9Mbit/s
and 30Mbit/s and above (superfast broadband) do not
think that their broadband connection is fast enough
isevidencethatthedeepruralndingsshouldnotbe
read as tacit acceptance of the current Government
target of 2.2Mbit/s broadband being a realistic speed
to offer remote communities.
Figure 3.10:
Is Your Internet Connection Fast Enough to do what You Want Online by Urban, Shallow Rural and Deep
Rural and Five-Fold Speed Classication
% of Internet Users
80
100
60
40
20
0
Deep Rural
0 - 6.3
Mbit/s
6.4 - 13.8
Mbit/s
13.9 - 20.5
Mbit/s
20.6 - 29.9
Mbit/s
30+
Mbit/s
Shallow Rural
Deep Rural
Urban
Shallow Rural
Urban
Deep Rural
Urban
Shallow Rural
Shallow Rural
Shallow Rural
Urban
Urban
Fast enough It depends Too slow
Source: OxIS 2013 Internet Users, n=1839, WUDS weighting applied

45
TWO-SPEED BRITAIN REPORT
Our analysis demonstrates that infrastructure
capabilitydoesinuencewhatpeopledoonline.
Clear urban – shallow rural – deep rural variations
inbroadbandspeedwereidentied.However,the
ndingsalsosuggestthatInternetuserswithpoor
connectivity engage in online activities despite the
limitations of a low speed broadband connection. Key
ndingsinclude:
• Superfast broadband was not available to any of
the deep rural respondents.
• The highest sync speed for any deep rural
sampling unit postcode, 17.4Mbit/s, was lower
than the average sync speeds for the urban
sample.
• Broadband speeds exceeding the minimum
required to use the BBC iPlayer (3.5Mbit/s) were
most common in urban areas.
• Only 4.9% of the urban sample lived in a unit
postcode area where the average broadband sync
speed was 6.3Mbit/s or less, compared to 46% of
shallow rural respondents and 53% of deep rural
respondents.
• Low broadband speeds do not appear to prevent
next generation use of the Internet (but the user
experiencewillbeinuencedbytheavailableICT
infrastructure).
• Internet users living in unit postcode areas with
thelowestbroadbandspeeds(≤3.5Mbit/s)were
less likely to participate in ‘data heavy’ online
activitiesthanthosewithspeeds≥3.5Mbit/s.
• Internet users living in unit postcode areas
with superfast broadband were more likely to
participate in ‘data heavy’ activities than those in
areas without superfast broadband.
• Social networking was less common amongst
Internet users living in unit postcode areas with
thelowestbroadbandspeeds(≤3.5Mbit/s)than
thosewholivedinareaswithspeeds≥3.5Mbit/s.
• Social networking was more common amongst
Internet users living in unit postcode areas with
superfast broadband than amongst those living in
areas without superfast broadband.
• The analytical limitations of using self-reported
views of broadband speeds are illustrated by the
fact that Internet connections were considered
to be ‘too slow’ by respondents who lived in areas
with the lowest broadband speeds and in areas
with superfast broadband available.
• 48% of Internet users in deep rural areas who
livedinareaswithbroadbandspeeds≤3.5Mbit/s
considered their Internet connection was ‘too
slow’.
3.iv Key Findings
The next section presents a selection of vignettes that
effectively illustrate the facts that many rural people
want to be online, many have to make do with a slow
connection,andmanyarenotatallsatisedwiththeir
online connectivity.
The vignettes are of intrinsic interest, but also
illustrate how improving Internet connectivity in the
currently‘difculttoreach’areasisimportantfor
individuals and businesses.

4. Does Being
Digitally Connected
Matter in Rural
Britain Today?
46 TWO-SPEED BRITAIN REPORT

47
TWO-SPEED BRITAIN REPORT
Whilst survey data is very useful in
providing statistically representative
overviews of attitudes and opinions it, like
all data sources, has some limitations. The
Oxford Internet Survey 2013 did not set out
to explore, for example, why respondents’
online behaviour was as they reported. Nor
did it seek to explore any of the challenges
and compromises that individuals make
in their online activities, or explore
whether different types of people (e.g. by
age or place of residence) have different
expectations of their connectivity that
would,inturn,inuencetheirbehaviour.It
is these factors that concern us when we
consider the urban-rural digital divide, or,
more accurately, the deep rural – all other
areas digital divide in the UK.
In an attempt to overcome some of the
limitations of the survey data illustrative
vignettes4 drawn from some of the research
projects undertaken in the University of
Aberdeen’s dot.rural Rural Digital Economy
Hub are now presented. Often using
the voices of people who live in remote
rural areas of Britain, they illustrate the
perceived importance of being online, and
thefrustrationanddifcultiesexperienced
by rural residents in accessing and using the
Internet in rural Britain today.
4 Note: the names of all respondents have been anonymised and marked with an asterisk. The vignettes illustrate recent research undertaken in
remote and rural northern and north-eastern Scotland, the North West of England and the Welsh Marches.
Evan* is a third generation hill farmer. He runs the
farm business in partnership with his parents. Apart
from a period at University, Evan, in his 40s, has lived
on the farm all of his life. His wife Vicky* moved to
the farm seven years ago. At this time the couple
attempted to get broadband at the farmhouse. Vicky
spoke of the challenges:
“… it was really dodgy wasn’t it and I used to spend
virtually every Saturday on the phone to [providers] to
try and get them to x it. … And then they re-laid the
whole cable […] and it got even worse after that, after
they laid new cable. And they said sorry there’s nothing
1. Difculties in securing a home broadband connection
we can do to get you Internet please don’t phone us
again.”
In the absence of an alternative, Evan and Vicky use
a dongle to access a broadband service via a mobile
signal, a means of access that is proving increasingly
ineffective. At the time of interview, Vicky loaded
a Sheep Society page – this took 4 minutes and 49
seconds.
Source: Interview conducted by Fiona Williams with
participants from the dot.rural Rural PAWS project,
2014.
Julie* is 19 and lives in a remote rural area. Like most
young adults today she uses her mobile phone to
communicate with others via Facebook, texting, email
and YouTube. Her home Internet service is poor and
she often goes to a café in a nearby village to use the
Internet. She complains that the 3G phone coverage
where she lives is sporadic and this inhibits her using
her smart phone.
“Oh, you mean, like, on your mobile phone? That, yeah,
in [town] it’s absolutely rubbish. It’s awful. There’s some
streets where you can’t get it at all and there’s some
streets where you can’t really get it in the middle of the
house; you just have to like go up to the windows and put
it against the windows. And that’s… pretty much [town]
in a nutshell. Because everyone’s like ‘oh, I’m going to
[town], so I probably won’t speak to you tonight because
I won’t get any signal’ [laughs]”.
Source: Interview conducted by Claire Wallace as part
of ‘Communities and Culture Network+’ research,
2013.
2. Young people feel excluded

TWO-SPEED BRITAIN REPORT
48
Sheila* and her husband farm in a community which
developed its own broadband access. The telephone
lines where they lived did not have the capacity
to support traditional broadband. Before the
community service was installed they had relied on a
dial-up connection which was then withdrawn by the
service provider. She said:
“Internet, yes. Well, for business, I need it for registering
calves; when calves are new-born, they have to be
registered within 28 days, which has to be done online
with BCMS [British Cattle Movement Service]. So I use
it for that, and for tax purposes, doing my tax work
online, my VAT returns have to be done online now:
you’ve no option, now they have to be done online, so
I’m grateful that we’ve got it. Other things... Personally I
do a lot of my shopping online; quite a lot. Not so much
my food shopping, but household goods and things, and
clothing”.
Sheila’s comments indicate how important a reliable
Internet connection is for their farming business:
without the community broadband service their
business would be compromised.
Source: Interview conducted by Fiona Ashmore
for her PhD research entitled “Exploring superfast
broadband provision in rural UK: A qualitative study
of community-based broadband development and use
and the potential for community resilience”, 2012.
3. Challenges faced by a farming family
As part of research to explore the relationship
between accessibility and social exclusion, 62
older people living in rural Aberdeenshire were
questioned about their relationship with the
Internet. 48% said they had never used the
Internet. 45% (28 people) said they do use the
Internet. They learned how to do so in different
ways: most worked it out for themselves and some
went on a training course. A local IT training social
enterprise closed recently and this may make it
difcultfornon-Internetusersinthefuturetolearn
how to use this now ubiquitous technology. Older
rural non-Internet users may become increasingly
disadvantaged as, for example, government services
transition to online delivery and other service
providers assume the entire population is digitally
connected.
Source:Selectedndingsfromasurveyconducted
as part of Rob Craig’s PhD research entitled
“Accessibility and the Capabilities Approach: Towards
an Aid to Decision Taking,” 2014.
4. Internet use amongst rural older people
Living rurally meant, that, for John* and Sarah*, they
felt “badly served” with their Internet connection, and
wanted to sign up with the B4RN (Broadband for
the Rural North) service since, “the way the world is
developing everything is being connected electronically”.
Sohowdoesthedividehithome? “It’s the rural
communities that miss out...hardly equal is it?”
Source: Interview conducted by Fiona Ashmore for
her PhD research (as above), 2012.
Matthew* lives and works as a dairy farmer outside
Lancaster and is connecting to the new B4RN 1 gigabit
per second service. Currently, however, his ability to
access the Internet is non- existent: a consistent lack of
digitalaccessibilityltersintohispersonallifeaswell:
“I sometimes feel we’re excluded from certain aspects of
what you might call ‘modern life’ because things come
on iPlayer…lots of things, they put ‘want to know more,
go to our website’. You sort of feel a bit excluded from
things that a lot of people take for granted”.
Source: Interview conducted by Fiona Ashmore for her
PhD research (as above), 2012.
5. Feeling that you are ‘badly served’ and missing out

49
TWO-SPEED BRITAIN REPORT
In the Technology to Support Older People’s Personal
and Social Interaction (dot.rural TOPS) project, a
patient and her spouse who lived on a very remote
island in Scotland were asked whether they used the
Internet to look for information about their medical
condition. Problems with the speed of their Internet
connection were mentioned.
Patient 1’s spouse commented: “I’ve got a couple of
websites that actually come up automatically every so
often, one is an American thing and they are very much
into the things to help <specific medical condition>,
it’s a particular <medical condition> site and it is
interesting. But again, a problem here is that the
Internet is so slow so you’ve got to have time to sit and
let it – it can take two or three minutes for a page to
load but there’s quite a bit of information with that.”
Source: Interview conducted by Anne Roberts for the
dot.rural TOPS project, 2013.
6. Challenges of using the Internet to source factual information
John*, Fran* and their two teenage daughters live in
a small village. John is home-based as a maintenance
electrician for a large utilities company and Fran
works at the local primary school. The family attempt
to operate two laptops, two iPads, two mobile phones,
a desktop, an iPod and their satellite television
(recordings) off their broadband service. The demands
made on the “half a Meg to a Meg” service cause
tensions within the household:
“When we are all on our devices it’s so slow isn’t it?
And then things start crashing. You know. I mean I’m
only getting my emails and doing my online shopping
and sometimes it just takes so long I may as well have
just gone over to [the nearest town]. … The biggest bug
bear I hear is ‘God this is so slow, why’s it going so slow,
oh it’s buffering, oh it’s dropped out’ and well, you’re
saying, ‘there’s too many of us online now’. …”
One of the teenage girls complains: “It drops out quite
regularly and I’ve got friends complaining that I have
such bad Internet – I shouldn’t be on the Internet with
such bad Internet.“
The situation for the family is compounded by the
absence of a mobile signal in the village where they live.
Source: Interview conducted by Fiona Williams with a
participant household from the dot.rural Rural PAWS
project, 2014.
7. ICT infrastructure makes it difcult, if not impossible, to be a next generation user
household
Edward* lives and works as a business consultant in
the B4RN region outside Lancaster. Not having yet
connected to the 1 gigabit per second service, he
found that “at the moment the speed is pretty useless
for anything”. This had a impact on his work: “from
a business point of view it means you can’t effectively
download videos, transmit video clips, it’s just not
practical”. As a business in an increasingly digitally
connected economy, “we’re dealing with suppliers
online a lot more than we ever did” making the lack of
connectivity even more apparent.
Source: Interview conducted by Fiona Ashmore for
her PhD research (as above), 2012.
8. Reections on how a business suffers from poor connectivity

TWO-SPEED BRITAIN REPORT
50
Marian and Jack
Marian* lives in the village with her husband Jack*.
They are both in their late 60s and have not used
computer technology or had broadband in their home.
John takes the view that “If you’ve never had it, you
don’t miss it” but Marian is aware that a number of her
friends are online and regularly access information –
sometimes for her. She relays:
“…there might be the odd occasion like when I couldn’t
find this hotel in Bournemouth, I couldn’t find the
number so [friend] did it and looked and could see that
there were only two rooms left or something.
You know – I can be on the phone with [friend] or
somebody and she’ll talk about something and say ‘Oh
just hang on’ and she’s on the phone and she’s checking
something and then she’s telling me about it, perhaps
something we’ve just been talking about and she’s
‘Just hold on a minute and I’ll have a look. [Friend]
has looked up a lot of things [health-related] for me,
because there’s been a lot as you know and then she’s
read it all out over the phone.”
Marian is very cautious about buying and using
services online, particularly submitting personal or
businessnancialinformation:
“I don’t know I’d have to think about that one, I mean
because you know like when [family members], when
that happened with them and the bank and that, it just
worries you. I had something, for instance, the other
day, they ring up occasionally and see if you want to buy
something for [a charity] and she always says do you
want to pay over the phone, and I always say no I’ll send
it and I sent a cheque the next day, but maybe that’s a bit
old fashioned I don’t know.”
Richard and Linda
Richard* and Linda*, both in their 60s, moved from
their farmhouse to the village some years ago.
Richard talked about an opportunity to undertake
some computer training in the local community
centre, nearly 15 years ago, but stated that he had
encouraged his wife and son to attend because “I
thought it would be more applicable to them than me”.
More recently however, Richard has sought out a
beginner’s computer course. When asked why he had
changed his mind, the couple’s response:
R: Yeah well it’s widespread now isn’t it – everyone.
L: No matter what you’re watching. If you’re watching
the weather on television, they’ll say look it up on
such and such and then there’s this little thingy at the
bottom that you can – it is on everything now isn’t it?
R: Well it’s about spoilt conversation … well they just
go, get their pad out of their pocket.
L: They haven’t got to think anymore.
R: You don’t have a debate because it’s there in
black and white – I suppose I’m just talking about
pub discussions and that sort of thing you know. It’s
certainly changed. And I thought it was about time I
could do the same you see!
Source: Interviews conducted by Fiona Williams with
participants from the dot.rural Rural PAWS project,
2014.
9. Contrasting views of older generation non-Internet users
John’s* work mobile operates off the broadband and he
accesses his work schedule for the day online via a work
laptop. He is often ‘on call’. The household’s broadband
connection and speed cause problems for John:
“I’m meant to be anywhere within two hours so when
they ring me I’ve got to be where I’ve got to be within
two hours […] So to go from here to [large conurbation]
in two hours you don’t want to spend three quarters
of an hour trying to get the job down on your laptop.
… like I was called out last night, I stopped on the way
home to send my job back … because I knew that when
I came home I’d have to try and log on via the Internet
at home and I may or may not get it in which case if I
haven’t sent my job back they don’t know I’ve left site.
Well the job is still there so eventually they’re going to
turn around and say the job is still there and try and
send it out again. So I do end up cancelling my lone
worker forty-five minutes drive away because if I don’t
I’ll forget to do it and when I get home I can’t do it.”
Source: Interview conducted by Fiona Williams with
a participant from the dot.rural Rural PAWS project,
2014.
10. Poor connectivity makes it difcult to full the demands and expectations of an employer

51
TWO-SPEED BRITAIN REPORT
Alongside the link to download a pdf property for sale
schedule the Dumfries and Galloway Solicitors Property
Centre web page noted: “As a guide, on a 56K modem, a 200K
pdf le may take up to 4 minutes [to download]”. The size of
eachpropertyscheduleleisalwaysstated.
Source: Dumfries and Galloway Solicitors Property Centre
Website, October 2013
11. Catering for a rural clientele with slow Internet connections
But it was snowing like crazy. I deemed it unsafe to
try and get to work. But I could quite easily work from
home because the speed of the service is pretty good
quality. Ahm... we use Skype quite a bit. And that’s
pretty decent. I think it relies a lot on what the person
at the other end’s got…”
Source: Interview conducted by Claire Wallace as
part of ‘Communities and Culture Network+’
research, 2013.
James* is a professional who lives in a small village in
remote rural Scotland. He sometimes works from home
and is able to use many wireless applications.
“What do I use it for? Yeah. The usual: surf Internet type,
access for web browsing, information, booking holidays,
all the usual stuff. We have a TV connected wirelessly
through Apple TV, so we use that for movies and things
like that… But, you know, I got snowed-in last winter, so in
December I couldn’t get to work. I didn’t try very hard to
be honest.
12. What good Internet access in remote rural areas enables

TWO-SPEED BRITAIN REPORT
52
5. Conclusion

53
TWO-SPEED BRITAIN REPORT
c. 11.6 million people live in rural Britain, and experience the socio-
economic and infrastructure conditions that go with rural dwelling.
Of this total, c. 1.3 million people live in deep rural areas, and c. 10.3
million in shallow rural areas. Deep rural areas cover approximately
half the total land mass of the UK, and include resources associated
with agricultural production, tourism and recreation, biodiversity, and
creative and entrepreneurial activity and potential.
ByexploringruralInternetuseforthersttimein
such detail, and by distinguishing between people
in deep and shallow rural areas, we have been able
to uncover major differences between the ways in
whichurbanandrural–specically,deeprural–
dwellers make use of the Internet. These include
the recognition that, as expected, online behaviour
generallyreectsconstraintsontheconnectionsto
the Internet. The effects of these include an overall
limitation on what people are able to do online
compared with what they want to do. Deep rural
dwellersaresignicantlylesslikelytobe–andtobe
able to be – Next Generation Users.
Theoverallndingspointtoageographicallydened,
excluded group, who by implication are less likely
than other groups in Britain to be able to engage
online with the creative, social, commercial, and
civic life regarded as normal in other areas. Previous
research masked this effect due to the research
obstaclestogainingasufcientsampletodiscern
them, but also due to the degree that patterns of use
in shallow rural areas tend to compensate for and
hidethedecienciesinaccesswithinthedeeprural
areas when analysis does not discriminate between
different types of rural area.
Aparticularlydifcultissueforpolicyiswhether
it is preferable to aim at deploying low speeds
universally and rapidly, or to systematically plan to
proceedmoreslowly(say,overave-yearperiod)
and achieve a higher universal speed. The rapid growth
of high-demand services and content delivery suggest
that unless the latter strategy is adopted, universal
low speed broadband is not future-proofed, and could
be obsolete by the time it is achieved. The context
of ‘digital by default’ in the provision of Government
services is especially pertinent, as the constraints
of low-speed connection highlighted in this report
question the viability of a universal model of online
service delivery.
Somecommunity-ledbroadbandprojectsconrm
this view: for example B4RN (Broadband for the
Rural North) in northern England, have committed
to providing speeds of 1 Gbp/s (1,000Mbit/s).
Furthermore, technical issues mean that the
installationofhigherspeedinfrastructureviabre
or copper is unlikely to reach the most remote
households, and a mix of technologies including
wireless and satellite will be required: some of these
arelesslikelytoprovidesuchhighspeedsasbre-
based systems.
Clearly, there are many policy issues raised by the deep
rural divide discovered and documented in this study,
and reinforced by related research on communication
infrastructures (such as in reports published by the
telecommunications regulator, Ofcom). We hope this
study provides additional evidence of this divide and
stimulates and informs serious debate over the policy
and regulatory responses necessary to address it.

References
54 TWO-SPEED BRITAIN REPORT

55
TWO-SPEED BRITAIN REPORT
Cloke,P.andThrift,N.(1994)Introduction:Reguringthe‘rural’inPCloke,MDoel,DMatless,NThriftandMPhillips
eds Writing the Rural Five Cultural Geographies Chapman, London, p1-5.
Department of Culture, Media and Sport and Department for Business Enterprise and Regulatory Reform (2009), and
DigitalBritain.TheInterimReportJanuary2009.OfceofPublicSectorInformation,Cm7548.Availableathttp://www.
ofcial-documents.gov.uk/document/cm75/7548/7548.pdf
Dutton, W. H., and Blank, G. (2011), Next Generation Users: The Oxford Internet Survey 2011. Oxford: Oxford Internet
Institute, University of Oxford.
Dutton, W. H. and Blank, G. with Groselj, D. (2013) Cultures of the Internet. Oxford Internet Survey 2013 Report.
Available at http://oxis.oii.ox.ac.uk/wp-content/uploads/2014/11/OxIS-2013.pdf
Furuholt,B.andKristiansen,S.(2007)Arural-urbandigitaldivide?RegionalaspectsofinternetuseinTanzania.
Proceedings of the 9th International Conference on Social Implications of Computers in Developing Countries, Sao
Paulo,Brazil,May2007.Availableathttp://www.ipwg94.org.br/fullpapers/R0090-1.pdf
Hindman, M. (2000), The Myth of Digital Democracy Princeton University Press, Princeton, New Jersey.
NationalAuditOfce(2013)DigitalBritain2:Puttingusersattheheartofgovernment’sdigitalservices.Reportbythe
ComptrollerandAuditorGeneral,HC1048,Session2012-13.TheCabinetOfce.
Ofcom (2013a) Average UK broadband speed continues to rise, Media release, 7th August 2013, accessed 19th August
2013: http://media.ofcom.org.uk/2013/08/07/average-uk-broadband-speed- continues-to -rise/
Ofcom (2013b) The availability of communications services in the UK, 16th May 2013, accessed 19th August 2013:
http://stakeholders.ofcom.org.uk/binaries/research/markets-infrastructure/economic- geography.pdf
Ofcom (2014a) UK experiences superfast broadband surge but challenges remain to address speed mismatches Media
release 15th April 2014, accessed 30th April 2014: http://media.ofcom.org.uk/2014/04/15/uk-experiences-superfast-
broadband-surge-but-challenges- remain-to-address-speed-mismatches/
Ofcom(2014b)UKxed-linebroadbandperformance,November2013–Theperformanceofxed-linebroadband
delivered to UK residential consumers published 15th April 2014, accessed 30th April 2014: http://stakeholders.ofcom.
org.uk/market-data-research/other/telecoms-research/broadband- speeds/broadband-speeds-nov2013/
OfceforNationalStatistics(2013)StatisticalBulletinInternetAccess–HouseholdsandIndividuals,2013availableat
http://www.ons.org.uk/ons/dcp171778_322713.pdf
Pateman,T.(2011)Ruralandurbanareas:comparinglivesusingrural/urbanclassicationsRegionalTrends432010/11
Philip, L.J., Brown, D. and Stockdale, A. (2012), Demographic Ageing in Rural Areas: insights from the UK and US in M
Shucksmith, D Brown, M Warner, J Vergunst and S Shortall editors “Rural Transformations and Policies in the UK and
US” Routledge pp58-78
RoyalSocietyofEdinburgh(2010)DigitalScotlandavailableathttp://www.royalsoced.org.uk/cms/les/advice-papers/
inquiry/digitalscotland/Digital%20Scotland%20%28med%20res%29.pdf
Scottish Government (2012) Rural Scotland Key Facts 2012 available at http://www.scotland.gov.uk/
Publications/2012/09/7993
Scottish Government (2014) Scotland’s People Annual Report: Results from 2013 Scottish Household Survey Chapter 8
Internet, available at http://www.scotland.gov.uk/Resource/0045/00457570.pdf
Wilson, R. (2012) Personal communication.

Appendices
56 TWO-SPEED BRITAIN REPORT

57
TWO-SPEED BRITAIN REPORT
Appendices
IntheUKdifferenturban-ruralclassicationsare
in place for the four constituent nations. (Pateman,
2011) provides a very useful overview of the different
‘ofcial’(i.e.government)classicationscurrentlyin
use. The OxIS 2013 report referred to two of these
classicationswhichapplytoScotlandandtoEngland
and Wales respectively. The Scottish Government’s
urban/ruraldenition,basedondatazones,can
be expressed as two-fold, three-fold, six- fold or
eight-fold area types (see http://www.scotland.
gov.uk/Topics/Statistics/About/Methodology/
UrbanRuralClassication)andtheEnglandandWales
rural/urbandenition,atmiddlelayersuperoutput
areas,isasix-foldclassication(http://www.ons.
gov.uk/ons/guide-method/geography/products/area-
classications/rural-urban-denition-and-la/index.
html).TheScottishclassicationisbasedaround
settlement size and proximity (expressed as drive
time) to a sizeable urban centre.
TheEnglandandWalesdenitionisbasedaround
settlement type and a population density (sparsity)
variable. Settlement and population distribution
patterns vary considerably across the UK and these
differences have informed the development of these
classicationsandareinlargepartthereasonwhy
asingle,UK-wideurban/ruralclassicationisnotin
use. In Scotland, for example, 82% of the population
live in an urban area, 8% live in accessible rural, 6%
in remote rural and 4% in very remote rural areas.
The urban Scottish population occupies c. 6% of the
Scottish land area: 18% of the population lives in the
remaining 94%, and very remote rural areas cover
half the Scottish land area. In England, by contrast,
20.9% of the land area is urban and only 1% of the
population live in a ‘sparse’ area, concentrated in
the northern and south-west fringes of the country
(Pateman, 2011).
APPENDIX 1 :
Urban and rural classications in Scotland, England and Wales
Fiona Williams is a dot.rural Post-doctoral Research
Fellow (in Geography) at the University of Aberdeen.
She is currently working on the Rural Public Access
Wi-Fi project (Rural PAWS) which explores means
of enabling digital inclusion in rural areas. In her
previous employment with the Institute of Rural
Health Fiona led the Welsh component of the dot.
rural TOPS project
Claire Wallace is Professor of Sociology at the
University of Aberdeen. Her research is about the
impact of digital communications on social life and the
quality of society in international comparisons.
Anne Roberts is a Research Assistant at the Centre
for Rural Health, University of Aberdeen. Her current
research includes an evaluation of the Paediatric
Unscheduled Care Telehealth pilot where paediatric
consultantsusevideolinktosupportA&Eclinical
staff in making decisions about paediatric patients in
remote hospitals across the north of Scotland. Anne
was a researcher in the dot.rural funded TOPS project,
responsibleformostoftheScottisheldwork.
Rob Craig is a Postgraduate Research Student
in the University of Aberdeen’s Rural Digital
Economy Research Hub, dot.rural. He is interested
in the concepts of, and the relationship between,
accessibility and social exclusion in the context of
social justice. His interest in social justice also extends
to his philosophical and methodological approach
to his work, and in particular the notion of action
research.
Fiona Ashmore is a Postgraduate Research Student
in Geography at the University of Aberdeen. Her
PhD research explores community-based superfast
broadband organisations and the extent to which
superfast broadband development and use enhances
communityresilience.Withindot.ruralsheisafliated
with the Digital Engagement and Resilience (DEAR)
project.
APPENDIX 2:
Contributors of case vignettes and the research projects referred to Contributors
Opposite: Photo courtesy of Dr Lorna Philip;
not to be reused without prior permission.

TWO-SPEED BRITAIN REPORT
58
The Technology to support Older adults – Personal
and Social Interaction (TOPS) project is one of the dot.
rural healthcare theme projects. With a focus on older
adults with chronic pain in rural areas, this project
has explored personal and social interaction between
older adults and their health and social care providers
within the context of widespread upscaling in the use
of electronic healthcare technologies (sometimes
known as telehealthcare) to support patients in their
own homes.
The Rural Public Access Wi-Fi Services (RuralPAWS)
project is focused on enabling access to broadband
services in rural areas by developing technology
that will pave the way for new access methods that
willallowforcommerciallyviable,‘t-for-purpose’
Internet services in traditionally hard to reach areas.
Rural PAWS is funded by dot.rural as a partnership
project with the Horizon Digital Economy Hub
(University of Nottingham), MLAB (University of
Cambridge) and industrial partners including BT and
Avanti.
An EPSRC networking grant supported the
Communities and Culture Network+ under the
auspices of which research under the theme of
‘Everyday life and cultural communities’ has been led
byresearchersafliatedtodot.ruralattheUniversity
of Aberdeen.
Projects

Professor John Farrington,
Emeritus Professor
geo134@abdn.ac.uk