1.0 Introduction
A debate about the structure of the city is currently underway in the academic and popular press and this is partially connected with the future of urban and industrial society. Much of the debate revolves around the pervasive use of the automobile. In particular, the effects on human well being which result from automobile dependent cities. Automobiles are cited for causing a number of harmful effects including contributions to greenhouse gases, urban sprawl, and leisure time lost to the daily commute. This later concern has a number of consequences. Long commutes of perhaps one hour each way in an automobile mean less time for family, community and political engagement as well as lost time for exercise and other forms of leisure. Even for individuals who travel shorter distances, the cost of maintaining an automobile uses a significant portion of a middle-class person’s disposable income.
This dissertation will examine cycling as a
possible solution to the urban transport problem of how to move large numbers
of people over the often short concentrated distances of the city. The research question I have set for myself
is, “Can the bicycle address some of the
problems associated with auto dependency in modern cities?”. Traditionally, public transit is cast as the
alternative to private automobiles in the city. This dissertation attempts to shine some light on cycling as a
lower cost and often overlooked form of urban transport.
The case for cycling can be made on a number
of grounds. Cycling is relatively
inexpensive, it emits little or no pollution, and it is quiet. Cycling moves at a slower pace than cars and
public transport travelling at high speed, but it can be the fastest mode of
transport over short distances (Tolley and Turton 1995: 211). Bikes are relatively cheap and easy to
maintain and repair. They are easy to
store and require less room to park in the city centre. Moreover, bikes are a source of
exercise. Even in situations where
someone travels 30 or more minutes by bike to get to and from work, the
commuting time is combined with moderate aerobic activity.
In this paper, I propose to examine some of
these advantages for cycling although my discussion in the first part of the
dissertation focuses more on setting the stage for cycling by examining
problems with current urban transport.
In particular, I will review some of the theories framing debate on the
concept of sustainability. I will also
look at the pollution argument for cycling by examining some of the issues
surrounding greenhouse gas emissions. I
will then examine the role that the automobile plays in contributing to
pollutants such as carbon dioxide and many others, as well as considering some
of the criticisms levelled against it as undermining public and private
life.
In the later two thirds of the paper, I begin
examining cycling directly. I will
touch on some of the community and health benefits of cycling, and I will look
in some detail at the historical role of cycling as a way of demonstrating that
at one time it played a far greater role in urban transport than it does
today. More significantly, I will carry
out a comparison of cycling at both the local and the national level by
comparing cycling in the English City of York to cycling in the UK as a
whole. This comparison will be an
effort to investigate whether cycling in its modern form is primarily a fringe
activity or whether it is a broadly based and popular means of travel with
further room for expansion.
Two subtexts run through this
dissertation. The first is that there
is a potential threat to human well being and our environment posed by our
dependence on automobiles for so many of our travel journeys. Generally, there appear to be three
categories of response to the perceived threat to the environment posed by our
industrialised modern life. These responses
can be categorised as the market response,
the sustainable development response,
and the deep ecology response. In addition to defining these concepts, I
will try to put these responses in context, first by examining them in greater
detail and then by examining how well they hold up against real world data for
energy consumption in the transportation sector and in greenhouse gas
emissions. The second subtext is that
the current mix of urban public transportation options available is not the
most optimal. At times I may appear to
be critical of public transportation; however, my intent is merely to place the
spotlight on cycling.
2.0 Literature Review
2.1 Differing views about the nature of sustainability
The concept of sustainability is
controversial since it is not clear what it is we wish to sustain. There are a number of candidates: material
wellbeing, economic growth, or the maintenance of the ecosystem. In concepts such as sustainable development,
a combination of all three goals is usually sought. Sustainable development is
an intermediate environmental framework which, I will argue, does not trust the
market alone to solve environmental problems, but rather suggests that by
government intervention the goals of ecological sustainability and economic
growth can be achieved. Overlapping at
one extreme with sustainable development advocates are those who trust the
market to solve environmental problems and who would resist putting
restrictions on growth for ecological goals.
Overlapping on the other side of the sustainable development spectrum is
the deep ecology movement. This group
appears to be less concerned or enamoured with the state of modern industrial
society and would see ecological sustainability as the primary goal over and
above economic considerations.
A review of sustainability could be framed
against a backdrop of water pollution, the loss of bio-diversity of plants or
animals or the management of agriculture resources. I propose to discuss sustainability in the context of the built
environment with a particular emphasis on climate change (considered one of the
central battlegrounds of the environmental debate ( Lomborg 2001: 258)) and
transportation. I will also briefly
discuss personal transportation in light of the challenges to civil society and
population health. This detour is
important for examining the argument that efforts to alter society in response
to environmental concerns do not simply involve costs, but also convey
benefits.
The classic villain in this narrative is
often said to be the automobile and the high output of greenhouse gasses
associated with it as a means of travel. Increased public transport is often
put forward as the solution. I hope to
demonstrate that cycling, of all modes of transport, has the best combination
of permitting travel of up to 10 kilometres while producing few or no emissions
of harmful greenhouse gases (presumably some emissions result from the
production of bicycles in factories).
Commuting to and from work is only one
contributor to overall greenhouse gas emissions. National greenhouse gas reduction initiatives such as those
planned in Canada (Government of Canada 2002) propose multi-sector approaches
to the problem. They recommend reducing emissions by private automobiles, the
trucking and airline industries, as well as in non-transportation sectors of the
economy such as industrial and commercial activities, home heating, energy
production as well as agricultural activities.
2.2 Deep Ecology, Sustainable Development and Market solutions.
Although I will use the deep ecology movement
as a catchall phrase, the group is hardly uniform. Within the category, I will include the Green movement as well as
those who use the phrase eco-socialist.
All of these groups share in common a willingness to make deep cuts in current industrial
activities in their efforts to ensure that
society functions in balance with the ecological systems of the planet. My conception of deep ecology is therefore broader than the more
common definition which is often associated with “New Age” philosophies (Sessions
1995:291) and therefore involves an overlap with some proponents of
“sustainable development”. I will have
more to say about this overlap later.
Saral Sarkar, as an eco-socialist, explicitly
calls for a reduction in the scale of economic production in Western developed
countries, arguing that only a strong centralised socialist state could produce
such a necessary effect. The
prescription he proposes is indeed a radical one. “For conventional economists, the contraction process would be
similar to an ever-worsening recession, and the low-level steady state a great
crisis without end. There can be no
doubt that in such a situation the whole economy would have to be socialised,
beginning with nationalisations.” (Sarkar 1999: 214). Arguing that our current lifestyle cannot be sustained at the
current rate of production, he suggests a future less dependent on oil and mass
production and more dependent on human labour.
In effect, his program calls for sustained recessions as we shrink our
economies down to a more sustainable level.
Sarkar is an economist and speaks in the
language of GDP. The deep ecology
movement is also represented by a social and political theory. An example of this is found in the writing
of Arne Naess. Naess tends to characterise
the environmentally concerned community as being composed of deep ecologists
and shallow ecologists (Naess 1995: 71).
Naess lumps the growth oriented proponents who trust in the market for
solutions to environmental problems together with others who argue for a more
interventionist sustainable development.
Discussing what he sees as the shallow ecology response to pollution
Naess says, “Technology seeks to purify the air and water and to spread pollution
more evenly. Laws limit permissible pollution.
Polluting industries are preferably exported to developing countries.” He goes on to say that, “The deep approach … calls for a high
priority fight against the economic conditions and the technology responsible
for producing the acid rain.”
Naess and Sarkar can perhaps be criticised
for taking too narrow a view of the state of the ecological movement. They
characterise the industrial society as requiring a radical change befitting the
crisis that they see the world facing.
The sustainable development proponents would more likely see themselves
as occupying a middle and moderate position, recognising the need for change
which is consistent with modern industrialised society. Moreover, the deep ecology and green
movement is criticised often as being utopian and having a too uncritical view
of the presumed benign behaviour of pre-industrial society. Roy Allen (1986) attempts to demystify the
ecologists program by providing examples of areas of potential conflict between
proponents of a deep ecology platform and the actual behaviours of the
aboriginal and pre-industrial groups on whom they model their visions of a
sustainable balance between people and environment.
Perhaps more mainstream among individuals
concerned with the environment are proponents of sustainable development. One way to understand the sustainable
development movement is to look at one of its key objectives – achieving the
targets set out in the Kyoto agreement.
Kyoto is an international agreement whereby developed nations agree to
limit their green house gas emissions to targeted amounts linked to 1990 levels[1]. Proponents argue that such moves are
possible without damaging economic growth.
The Canadian government, with a target of reducing emissions to 96
percent of 1990 levels, claims, “By acting now Canadian companies and
individuals can get ahead of the curve and create a sustainable competitive
advantage. The development of new
products and services, in turn, will stimulate economic growth, expand exports
and create jobs for Canadians.”
(Government of Canada 2002: 2).
In this example sustainability is defined as sustaining the advances of
the industrial revolution, while development involves the ongoing creation of
wealth at the same time as reducing the level of pollution from human
activities.
Agreement about the definition of sustainable
development is not universal, however.
Sharachchandra Lélé points out that the development aspect of the
sustainable development project need not be sustained growth of material consumption,
but he does argue that proponents are hard pressed to define the concept
without expressing it in terms related to continued economic growth. (Lélé 1991: 609).
Bouts of enthusiasm by national governments
such as Canada’s for sustainable development perhaps overplay the optimistic
possibilities of a sustainable future.
Lélé is certainly more cautious. He suggests that development may need
to be divorced from growth in GDP.
Throughout his paper he cites examples to suggest that development is more
about increased social welfare rather than pure economic growth. Lélé points out the difficulty of defining
even such a term as social welfare (Lélé 1991: 609). Given the divergent ideas about the nature of sustainable
development it is clear that the term is an umbrella for a broad spectrum of
views, that at the margins overlaps with deep ecology and market approaches to
reductions in environmental harm. I
would argue that sustainable development is distinguished by a commitment to
economic growth as it is currently defined in economic terms such as GDP,
combined with an interventionist approach by government to meet sustainable
ecological objectives.
Distinct, and critical of both the deep ecology and sustainable development arguments is the market approach to dealing with adverse industrial effects. The market approach is characterised as seeing continued economic growth as the most likely means of solving environmental problems. Ultimately the argument appears to be that the threat to economic growth by costly efforts to regulate greenhouse gases is of far greater concern than the threat of greenhouse gas emissions to civilisation. By allowing the economy to grow naturally, they suggest society is more likely in the future to be able to afford to deal with the ramifications of global warming than by spending money now on dubious solutions to a very real problem ( Lomborg 2001:322-323).
Spelling out their position on global warming and greenhouse gas emissions, the American oil company Exxon argues that although they are concerned about greenhouse gases, a clear picture about how to proceed does not exist (Exxon Mobile Corporation 2001). They state, “Although there is no consensus about long-term climate trends and what causes them, it is clear that the weather is always changing.” Further they oppose efforts such as the Kyoto agreement stating “Like many other companies, we do not believe Kyoto is the right approach.” Exxon’s criticism includes concerns that the most rapid growth in the future will occur in developing countries which are not required to accept constraints on their growth or greenhouse gas emissions (Exxon Mobile Corporation 2001). Opposition also comes from government leaders who suggest that the Kyoto agreement threatens economic growth with little environmental benefit (Klein 2003).
It seems, therefore, that the Kyoto accord
illustrates a dividing line which separates the various schools of thought
concerning climate change. The deep
ecologists put forward arguments which suggest that they do not feel that it
goes far enough and is therefore likely to be ineffective. Industrial expansion proponents also feel
that it will be ineffective but see it causing more harm than good, while sustainable
development proponents argue for targeted goals.
The market approach to dealing with the
problems of pollution appears to recommend a hands-off strategy. Accordingly the issue in question is whether
the goals of sustainable development and deep ecology advocates are attainable
and which group more accurately advocates the most appropriate solutions. By examining cycling as one activity
championed by deep ecologists and sustainable development advocates I hope to
provide some insight into the effectiveness of these two group’s recommended
changes for alleviating social problems.
Western cities populated by masses of cyclists going about their
business is no doubt a nightmare scenario for oil, and automobile producing
provinces and companies, and likely dismissed as too optimistic and simplistic
a solution by many less radical proponents of sustainable development. As such, it seems an ideal testing ground
for examining how to deal with some of the unfortunate side effects of
industrialisation and modernity.
2.3 Greenhouse Gas Emissions
Some (Lomborg 2001:258)
argue that global warming due to climate change, “has become the overriding
environmental concern since the 1990s.”
Certainly, if world CO2 levels were too high in 1990 then the situation
seems destined to get worse. Figure 1 shows
the growth of CO2 emissions[2]
as a time series chart. The figure
compares Global CO2 levels for all countries where data is available to growth
in CO2 emissions in wealthy Organisation for Economic Co-operation and
Development (OECD) countries. The data
are drawn from the 2003 World Bank World Development Indicators (World Bank
2003). There are a number of
observations which can be drawn from this chart. The first is that output of carbon dioxide has been tending to
grow at a much slower rate in the Western developed world than in the whole
world. This suggests that, although the
West has traditionally been the principal producer of greenhouse gases, the
development process underway in the rest of the world is rapidly reducing its
share of the total output. The graph
also demonstrates the threat posed from the developing world adopting a
development path similar to that followed in the industrialised countries.
It is also apparent that a
levelling, and even slight reductions, in human CO2 output levels have been
occurring since the early and mid 1990s.
While this might seem like grounds for optimism, it should be remembered
that during this period, massive re-orderings were taking place in the global
economy. In the early 1990s, the Soviet
Union collapsed with the closure of many state run factories, and a decrease in
economic activity. Japan spent much of
the 1990s in a state of recession and sluggish growth. Latin America underwent a number of crises,
and Africa struggled with the AIDS epidemic.
In North America and Europe, high tech and telecommunications became
major industries.
Therefore, while the
levelling of CO2 emissions may suggest room for optimism, I would argue instead
that we have merely reached a temporary plateau. While growth of CO2 may continue to slow in the West in response
to our relatively stable built environments, the restructuring in the developed
world suggests that CO2 emissions are only temporarily halted. Many countries are restructuring inefficient
economies at present, with the prospect of returning to market based growth,
with higher wages and greater individual wealth and consumption. It is at this point that CO2 output levels
threaten to rise dramatically. If the
same patterns of automobile and highway usage, and the same patterns of the
built environment are adopted by rising numbers of middle-class consumers in
the developing world as was done in the OECD, then CO2 levels will surely rise
further.
Figure 2 examines the CO2 emissions that
would be necessary if there were a concerted effort to maintain total world CO2
emissions within the levels established as part of the Kyoto agreement but on
an equalised and global basis. This
agreement seeks to establish CO2 emissions among Western developed nations at
often reduced levels relative to their 1990 levels while allowing the
developing nations to expand into the created emission space (United Nations
1992). World emission levels at or below the 1990 amounts are recommended as
being necessary for stabilising CO2 levels (Houghton et. al. 1997: 151).
Therefore, in the table I have used the 1990 world level (approximately 16
billion tonnes) as a hypothetical levelling point for total world emissions. However, it should be understood that CO2
levels were a concern long before 1990, and their effects were being felt for
much lower levels. This would suggest
that should a stable level of allowable CO2 emissions be articulated, it is
likely to be less rather than more than the 1990 level.
As the chart shows, for a world population of
6 billion people, the average CO2 emissions should be 2.73 tonnes per
capita. This is a startling
amount. Figure 3 shows Organisation for
Economic Co-operation and Development (OECD) countries’ per capita emission
levels. Note that in 1999 the United
States emitted nearly twenty tonnes per capita. In fact every OECD country emits more CO2 gas than might be
allotted to them if CO2 emissions were distributed on an equal basis. It is
therefore difficult to envision a single developed country which we could visit
in order to observe the type of structure needed to live within the constraints
recommended for slowing greenhouse gases.
Even Sweden emits more than twice the amount of greenhouse gas that is
recommended despite its best efforts to produce a sustainable society. Considering the projected growth of world
population to levels of perhaps 8 billion or more, this suggests that significant
changes will need to be made to the energy consumption of society if we wish to
reverse the possible consequences of CO2 emissions. If European countries with well developed transit systems are
unable to meet the necessary CO2 emission targets, then more drastic structural
changes to our transport systems may be necessary.
In fairness to the automobile industry, all the blame for CO2 emissions cannot be laid on the use of cars for transport. Comparing Figures 3 and 4 (in section 2.4) provides some balance to this picture by showing that there is wide variation in automobile ownership despite the fact that all listed countries emit carbon dioxide at levels above the sustainable level. A simple correlation of these two variables suggests an R squared value of .283 significant at the .007 level. Clearly there is a small but significant relationship between the level of automobile ownership and carbon dioxide levels. Efforts to reduce carbon emissions cannot rest on changes to transportation alone since it has been estimated to contribute only 25% of emissions (Pickering and Owen 1994:80), but will also require alternatives to coal fired power plants and poorly insulated houses and buildings. However, the significant role that transport does play in reaching pollution targets, certainly justifies a greater focus on ways to reduce the harmful side effects associated with it.
2.4 Urban Life, Industrialisation and the Car
Having examined carbon dioxide levels, it is
important to sharpen the focus on transportation. While only one contributor to green house gases, the choice of
transportation mode seems to demonstrate the contrast between the differences
of deep ecologists, sustainable development advocates and those who believe
that the market will deal with environmental problems naturally. Moreover, it is the sustainable development
and deep ecology advocates who must demonstrate that the market approach is
producing unsustainable urban structures, since for the most part, it is the
market which has been responsible (at least in the OECD) for our current
transportation systems.
Economic development and urban life seem to
be strongly linked. Carlo Cipolla in
his classic text - The Economic History
of World Population demonstrates that industrialisation in a country tends
to be associated with urbanisation (Cipolla 1962). Britain, which urbanised before the automobile was invented,
developed extensive rail lines while the cities of the United States matured at
the same time as the automobile and built cities around the new
technology. As Figure 4 demonstrates,
citizens of the industrialised countries appear to have exercised their right
to consume transport, and by and large, they have chosen automobiles.
The
danger is that the same pattern will occur during the industrialisation of the
high population developing countries.
It is possible that new models will be found for economic development
without mass migrations to cities; however, such a change in the nature of what
it means to be a developed country would break with the past patterns. Some encouragement may be found in the
development of rail and public transit in Europe during the same time period as
automobiles were being relied on in the United States. Europeans plotted a different course. However, Figures 3 and 4 demonstrates that
even the relatively recently industrialised country of Korea is in the middle
of the pack of carbon dioxide emissions.
Clearly Korea’s story does not bode well for hopes that newly
industrialising countries will avoid Western pollution patterns.
It is possible that the changing nature of technology
will have some impact in reducing the amount of carbon dioxide emissions. The rise of the Internet allows for changes
in the way that we work and interact.
Broadband connections promise to give access to unprecedented
information while allowing individuals to remain in their own community,
travelling only as far as their local school or library or even living room
while accessing a plethora of government, educational and vocational
information (Timms, S. 2003). Balanced
against these developments is the fact that many of these technologies require
significant capital investment to install.
It seems likely that more citizens of developed nations will reap the
benefits of these new expensive technologies than citizens of developing
nations who may yet inherit the industrial patterns so common in the
pre-internet West. Ultimately, the
evidence does not look hopeful that left to their own devices, consumers will
make sustainable choices in their transportation choices without some form of
collective action agreed to at the political level.
2.5 The Automobile - Public Transportation Mix
There are a number of factors which affect the predominant choice of transportation mode in a country. Peter Newman and Jeffrey Kenworthy in their book entitled Sustainability and Cities compare American, Australian, Canadian, European and Asian cities in their dependence on automobiles. In general the authors report that geography appears to plays a key role in determining car ownership. Figure 5, which is adapted from their book (Newman and Kenworthy 1999: 80), shows car ownership and public transit usage in various regions.
Both measures of transport usage show
passenger kilometres per person, and indicate an inverse relationship between
car usage and transit usage. One
encouraging sign from this chart is the low levels of automobile dependence in
Asia compared with all other regions of the world. The Asian results include data from both wealthy cities such as
Tokyo and Singapore as well as cities in developing countries including Seoul
and Bangkok. Figure 5 gives some room
for optimism that the high density developing countries of the world will tend
to place a much higher emphasis on transit and as a result not consume North
American levels of gasoline.
The public transport advantage is more
visibly suggested in comparing the energy consumption for private and public
transit of highly developed cities. Los
Angeles is often seen as a particularly problematic city from an urban
development perspective. Kenworthy and
Noble (1999; 70-71.) provide a table which allows us to compare Los Angeles’
energy use per capita in both private and public transportation to a number of
its international peers. Figure 6 shows
the usage of gasoline, diesel and electrical power (measured in Mega Joules per
capita) for both types of transport.
Clearly all of these cities differ in terms of
their population, and the results would be carry much more weight if we knew
the numbers of people travelling by various modes in each group for each city.
However, that factor is partially controlled by the use of per capita numbers. Each city is a large metropolis with a
public transit system, yet there is a large difference between the energy
consumed in each city. Los Angeles
consumes nearly seven times the per capita energy from gasoline that Tokyo
does, this despite the fact that both are wealthy cities in countries
considered to be among the most technologically advanced in the world.
The second interesting point evident in this
figure is the comparisons between energy consumption in private automobiles and
that in public transportation. Even in
public transport reliant cities like Paris and Tokyo, the energy required to
run their systems of trains and buses is a fraction of that used to power their
automobiles. The comparison is
particularly telling in a comparison between these two cities and Los
Angeles. Tokyo’s public transit system
transports its citizens using sixty-one times less energy than Los Angles’
private transport. Clearly, there are
vast savings in energy consumption to be had in increased use of public
transport. Moreover, to the extent that
developing nations follow Tokyo’s model instead of Los Angeles’ there is
considerable room to accommodate their development towards a more sustainable
form of urbanism.
Although Figures 5 and 6 suggest vast energy
savings obtainable from greater use of public transport, the reader should not
forget the discussion about greenhouse gases earlier in this paper and the fact
that no developed country currently appears to have a sustainable level of
carbon dioxide emissions. Despite the
benefits of public transit, the discussion so far suggests that even greater
economies in fuel consumption are needed.
Intuitively we recognise that many urban
trips are of relatively short distance.
While automobiles are ideal for transporting individuals to remote
locations, public transit is more suited for inter-city and intra-urban
travel. However, as we have just seen,
even public transit consumes some energy, which suggests greater room for
cycling to reduce energy needs in the city even further.
2.6 Additional Criticisms of automobile Dependency
The built urban environment has long been a
battleground for proponents and opponents of a car culture, quite independent
of the polluting nature of cars.
Automobile based cities are criticised for limiting personal
interactions, and imposing a garish and impersonal structure on the face of the
city (Gehl 1996:72,73). Moreover, there
appear to be an idealised form in our minds of community as rural or as an
urban village, in which we feel linked to neighbours and to services accessible
by foot (Rudlin and Falk 1999:106,107) as opposed to suburban forms dependent
on automobiles.
For critics, automobiles appear to have
wrecked the face of the city. While public
transportation has advantages over automobile focused cities many of these
advantages seem amplified for a cycling based city. The choice of automobiles over cycling is argued to have a
detrimental effect on human health (British Medical Assocition 1992:
10-28). Not only do cyclists not
produce pollution, but they obtain valuable exercise in the course of their
travel. Cycling is also credited as
providing economic advantages to individuals who no longer have to maintain an
automobile (European Commission 2000:
15). Moreover, the vast suburban
landscape made possible by automobiles has been linked with a decline in civic
engagement with cities of over one million and their surrounding suburbs
experiencing a marked decrease in involvement in local political activities
compared to their smaller counterparts (Putman 2000:206)[3].
Putman considers both mobility and sprawl as
partial explanations for the decline in civic engagement. He quickly rules out mobility as the cause
for the decrease by pointing out that mobility has not increased in fifty
years. Rather, he points to the low
density suburbs and the car saying, “far from seeking small-town connectedness,
suburbanites kept to themselves, asking little of their neighbours, and
expecting little in return.” (210), and “it is difficult to overstate the
symbiosis between the automobile and the suburb” (212).
For Putman, part of the story of civic
disengagement is the time costs of travelling the distances involved in a low
density lifestyle, as well as the isolating effect travel by car has on
interactions between fellow city residents.
“The car and commute, however, are demonstrably bad for community
life. In round numbers, the evidence
suggest that each additional ten minutes in daily commuting time cuts
involvement in community affairs by 10 percent …” (213). He sums up the isolation of the car, “one
inevitable consequence of how we have come to organize our lives spatially is
that we spend measurably more of every day shuttling alone in metal boxes among
the vertices of our private triangles.” (202).
It seems clear that it is not so much the car
which fragments our society, but the urban form demanded by the car. Putman
clearly indicates that smaller urban centres have greater civic engagement. However, small cities and towns in the US
are not dominated by public transport, but rather are also dependent on
autos. What makes them different no
doubt is the shorter distances involved in travel. It would seem that if we wish to increase the benefits of civic
engagement of these smaller centres, we should evolve the structure of the city
to replicate village and town life by reducing travel times to work and
services. At the same time reducing the
dependence on auto travel for making short trips in favour of public transport,
walking and cycling might serve to increase a sense of community within the
cities and suburbs so many people seem to choose to make their homes. Perhaps the model for such efforts can be
seen in Stockholm, the capital of Sweden.
Throughout the 20th century the government has been
deliberately trying to plan the city away from private automobiles and towards
public transit as population boomed (Sidenbladh, 1968).
Such arguments at times seem like isolated
calls from the wilderness when compared to the actual economic and market
forces shaping modern urban life.
Nowhere is the argument for automobile centred cities more forcefully
made than in urban geographies such as greater Los Angeles. “Fifteen million strong and one of the
world’s largest economies (twelfth if it were a country, above Mexico but below
Spain), Greater Los Angeles includes an almost unique diversity of talents,
problems and opportunities.” So begins
one defence of the sprawling metropolis form of city (Gordon and Richardson
1999: 575).
Gordon and Richardson in their essay set out
to rebut a characterisation of cities such as Los Angles as the epitome of the
evils of the modern urban form. Two
particularly relevant points are made.
In their defence of the automobile as the dominant means of
transportation in such cities, the authors suggest that in fact the automobile
has evolved out of consumer choice.
Contrary to E. Soja who argues that there is a “… huge population
stranded up to sixty miles from their place of employment.” the authors point out that according to
national statistics many individuals in outlying suburbs actually have one-way
commutes of less then 45 minutes (575).
Moreover, they argue against the notion that public transit was dealt a
death-blow by auto companies[4]. The authors argue that rather than
conspiracies to destroy public transit in the city, the investigator need look
no further than the consumer. Citing
rider-ship numbers for public transit and the overwhelming choice to use autos
as the means of transport, the argument is that the consumer has chosen and
automobiles are the winners.
This argument appears to be supported by
Glenn Yago. He summarises the
historical development of American transportation by suggesting that the
victory of the car was due to a complex set of circumstances. These included rising labour costs among
transit workers, corruption in transit companies, rising incomes among workers
and gains in productivity and available capital in all sectors of the auto
industry (Yago 1983: 178,179). Perhaps
this set of factors partially explains why trams continue to be popular in
Europe but not the United States.
In his paper, Yago is certainly not an
apologist for the automobile, but rather sees the choice of transportation mode
as an area requiring far greater study.
He points out gender and wealth factors prove important for the types of
transport that are available economically to urban residents. He also argues that historical and political
analysis is necessary to understand why certain transportation modes such as
tram lines or cross city rail lines are available in one centre and not in
another. As a consequence Yago argues
that a city’s transportation network is not completely predetermined by the size
of the city or its history of development, as I have suggested in the previous
section on the links between urbanisation and industrialisation.
Stockholm and Los Angeles seem to stand as examples of opposite sides of a debate in urban form. Both are wealthy urban metropolises which reacted differently to a modern economic and population boom. Their differences in responding to the influx of new residents are strong evidence for Yago’s thesis that the pressures of urbanisation and population explosion can be channelled away from automobiles. Furthermore, they demonstrate the complexities of the factors which shape city development. What is uncertain is whether building extensive public transport as Stockholm has done will be sufficient to channel the economic growth occurring in the developing world into non-polluting forms of urban life.
2.7 The Bicycle as Transport
Up to this point, I have discussed transport
primarily by car and public transit and made little mention of cycling. I have briefly discussed some of the
societal health benefits to be gained from cycling, but not its possibilities
as a mature transportation mode. Given the inability of any OECD countries to
operate to date at a level considered sustainable at global levels, cycling appears
to be a transport technology which is far under-utilised in urban life.
One objection to cycling as transport is that
some countries in Europe have currently well established cycling traditions
(European Commission 2000), and still
produce far too many carbon emission.
Denmark and the Netherlands, despite their high usage of cycling,
contribute far higher levels of CO2 than is desirable. Clearly, however, cycling is not the culprit
since bikes contribute no emissions.
More likely it is the manner in which power is generated in those
countries to provide power for homes, industry and public transport. A country such as Denmark with its high
cycling levels and plans to generate much more of its electricity from
non-carbon sources may become a pioneer for the future of urban sustainable
living.
While the next two sections of this paper specifically focus on a quantitative examination of cycling as a commuting mode, it is helpful to look the state of cycling in the period prior to the growth of the automobile. Cycling had a brief glory period prior to the development of the automobile which demonstrated its ability to serve as a pillar in the urban transport system of cities. Sidney Aronson (1952) summarises the state of cycling at the turn of the last century. He notes first that cycling had a significant impact on women’s role in society. Cycling opened up the way for women to wear more comfortable clothing, including trousers. More significantly, cycling ushered their way into a wider communication outside of the home. Informal conversations between single and married women and the men who would sometimes help fix flat tyres when their paths crossed on the road led to a new informality of contact between the sexes. Aronson cites several sources suggesting that the suffrage movement was indebted to women’s adoption of cycling.
Equally significant is the effect that cycling had on the
economy. Aronson estimates that over
17,000 people were involved in the manufacturing alone of bicycles, thousands
more participated in repair, retailing and other affiliated industries. In total, he estimates that 10 million of
the United State’s 76 million strong population participated in cycling at the
turn of the century.
Aronson, using an American football metaphor,
argues that cycling’s key revolutionary impact lay in what he termed running,
“interference … for the automobile.” (310).
Many early automobiles were manufactured and repaired in bike shops, and
even the efficient manufacture techniques developed to mass produce bikes
further enhanced the production of cars.
Moreover, it was cyclists and their national organisations which lobbied
for and began the development of early road systems in the US together with
proper lighting and signage. In short,
the image of the bicycle industry appears to be that of an incubator of the
coming automobile age.
Ultimately, it is beyond argument that the
automobile provides far greater mobility than a bicycle ever could. A combination of modern cars and the highway
system has permitted the spread of the modern metropolis over vast tracks of
land; however, it should not be overlooked, that the bicycle was a form of
transit which supported early urban mobility.
Aronson notes that it opened up the countryside and urban travel to the
middle class and that it allowed people to travel distances of 25 to 50 miles
in a single day. Cycling opened up new
vacation possibilities as families set out on road trips together, and
permitted early suburban development for less well to do families which
permitted them to move from urban tenements.
2.8 Prior Work on Cycling in the United Kingdom
So far, I have attempted to frame the
question of whether an expanded role is possible for cycling by looking at
issues which either suggest that the current transportation system is not
sustainable, or more positively, by looking at problems which increased levels
of cycling may solve. Much of the rest
of my paper will involve a quantitative investigation of cycling by looking at
census data. In doing this, I am
building on work that has been done before but in a limited way.
In order to understand the potential of
cycling as an increased mode of transport, it is essential to understand the
underlying demographics of cyclists.
Some work has been carried out by the UK Department of Transport (The Department of Transport 1996). Their publication provides a brief overview
of the state of cycling in the United Kingdom, using the 1991 UK Census as well
as various transportation surveys. The
report provides some excellent basic statistics which permits the reader to
identify cities and regions where cycling is more popular. The report also indicates trends in the
growth of cycling to work (from 696 thousand in 1991 to 823 thousand in
1995). Finally, the report identifies
useful UK data sets with cycling information.
While the report serves as a useful first step for someone who wants to
understand cycling in the UK, it does not go into great depth in terms of
allowing policy makers to understand better who cyclists are. By focusing on one data set and one city as
well as the whole country, I hope to seek a more detailed picture of the urban
cyclist than is provided in that useful report.
2.9 Conclusion
So far I have reviewed a wide range of
literature which at times seems disparate.
I have focused on issues such as CO2 levels, civil society and the
diminution of urban civic engagement. I
have also examined some of the health benefits of cycling and looked briefly at
historical evidence that suggests that cycling has been a more widely used form
of transportation which once played a vital role in the life of turn of the
century America. The common theme of
this review is that cycling is a technology which seems greatly under-utilised
in contemporary life. Barring the development of a full-scale hydrogen economy
to power fuel cell cars, cycling may be the necessary technology able to
supplement public transit in supporting urban living at reasonable levels of
CO2 emissions.
In the next two sections of this report, I
propose to examine cycling in action.
Using the city of York and the UK as a whole, I will examine the
characteristics of actual cyclists.
Having shown the benefits of cycling in the literature review, it is now
necessary to demonstrate that cycling is a robust modern commuting mode used
for transport by individuals entirely in the mainstream of contemporary
society.
3.0 Methodology for Studying Cycling – York and the UK
3.1 Introduction
Social science uses a number of different strategies for investigating phenomena. Generally these can be classified as either a quantitative or a qualitative approach. Quantitative research generally works with data-sets and carries out statistical analysis, while the qualitative approach generally involves unstructured interviews, focus groups, and/or ethnographic approaches. Quantitative approaches are often associated with deductive strategies designed to test theories while qualitative approaches focus more on inductive theory generation (Bryman 2001: 20). Since my research is intended to test the hypothesis that cycling can play a much greater role in urban transportation, a quantitative approach seemed, at this stage, the best methodology to employ.
Having, in the first two sections, built a theoretical case for cycling as a sustainable transportation mode, the next step appears to be to examine cyclists using empirical data. In particular, it is important to find out whether cyclists differ significantly from the general population, and whether those differences are such that cycling could only ever serve as a fringe transportation mode. In order to determine this, it seemed best to work with secondary data sets for reasons of cost and avoidance of data collection problems (Kiecolt and Nathan 1985:11). Specifically, the use of UK individual census data gives me access to high quality data that allows me to be reasonably confident that the characteristics I observe for cyclists are as representative of the cycling community as I can make them.
3.2 Overview of data to be employed and why.
The UK census is a convenient data set to use
in this instance because it asks questions about the transportation choices of
respondents with respect to their journeys to work. In order to examine
cyclists, I make use of chi-square techniques to analyse the 2 percent Sample
of Anonymised Records (SARs) for the 1991 UK census (Census Microdata
Unit). The analysis is intended to
investigate variables associated with the choice of cycling as a transport
mode. In order to do this, I have
looked for relationships with the variables which might be associated with
transportation choice.
The 2% sample is exceptionally useful because
of its great size. It contains over one
million records, and in particular over 14,000 records for bicycle
commuters. There are, however,
weaknesses in the dataset when studying cyclists. Most students are excluded, since they would answer questions
about travel to work as being non-applicable unless they worked part-time as
well as studying. Other potentially
excluded groups are the unemployed, the retired, stay-at-home parents and other
individuals outside the workforce.
Finally, individuals who work at home are also excluded since they are
treated as a separate commuting class (presumably their commuting tool is a
pair of slippers on the way to the home office!) All of these exclusions to some extent limit the ability to study
cycling as transport, since it is possible that some excluded groups use their
bicycles for daily tasks. While work is
a major destination for travel, shopping, entertainment, and family
responsibilities are also potential trip generators for the bicycle.
Despite this limitation, the journey to work is
interesting to study, because of its importance. Travel to work is not an optional choice for most people, and it
is a significant place for citizens to engage with the economy. While focusing on the journey to work
excludes some cyclists and underestimates the use of cycling as transport, it
also demonstrates cycling’s role as serious transport.
3.3 Key Variables
The following are the variables I used in my analysis.
Dependent Variable[5]:
Tranwork
–
Mode of Transport to work. This is a categorical variable indicating the means
used to travel to work and including various forms of public and private
vehicles as well as cycling and walking.
Independent Variables:
For my analysis, I examined the choice of
cycling as a mode of travel to work over against a number of socio-demographic
variables including age, gender and ethnicity.
I also looked at the distance a person travels to work as a means of
determining whether the mode of travel they have chosen could be substituted
for an alternative. I also attempted to
examine social class and economic well being by looking at the data available
for social class in the census as well as tenure of household.
Most of the variables I looked at are
relatively self-explanatory with the possible exception of social class. Social class is a variable used in the
census which has been defined by the Office of Population, Censuses and Surveys
for the United Kingdom’s Registrar General.
Peter Clavert summarises this system of identifying social class in his
book, The Concept of Class:An
historical introduction (Clavert 1982:173,174). The five primary class are defined as:
Social Class 1: Composed of professional
occupations like doctors, lawyers and other professionals.
Social Class 2: Composed of intermediate
occupations such as managerial work, artists and journalists.
Social Class 3N: composed of
non-manual skilled occupations including police officers, shop assistants and
sales representatives.
Social Class 3M: Composed of
the skilled manual workers including bus drivers, cooks and printers.
Social Class 4: Composed of
partly skilled workers.
Social Class 5: Unskilled
workers.
Clavert (174) is somewhat
critical of this division of workers into arbitrary classes. Certainly, he does not tend to look at the
classification as being a hierarchy since he recognises that it is possible
that an unskilled worker might in fact earn far more than workers in higher
classes.
Partly in anticipation of
this difficulty, I have included a second variable of tenure of housing. Given the potential of a plumber categorised
as a skilled manual worker in social class 3M to earn significantly more than
individuals in even the professional classes, the tenure variable seems best
able to capture some of the nuance involved.
Paul Balchin (1995: 1) identifies the success of post-war Labour and
Conservative governments to develop policies which appealed to the electorate’s
desire for access to adequate housing as being key to their electoral
success. Home ownership seems to
provide a great deal of status and stability in the United Kingdom. By looking at the tenure of housing of
cyclists, I hope to make some comment on whether cyclists are building
financial resources or are a more marginalised group.
The benefit of using the
variables I have chosen is that they are well established in social science
literature. As such, they serve as a
useful starting point for examining the phenomena of cycling . Although variables such as gender and social
class cannot be considered to be independent (Walby, 1986), the purpose of this
paper is merely to explore the data and determine the suitability of cycling as
a full-fledged transportation option. The usefulness of studying transportation
in light of the types of variables I have selected was demonstrated by Swedish
researchers Carlsson-Kanyama, Lindén. and Thelander. They examined the distance travelled by various Swedes by gender
and found a large difference between men and women as well as among different
ages and social classes. Their study
suggested that the basic socio-economic variables are a useful starting point
in investigating individual differences in greenhouse gas emissions.
3.4 Chi-square Analysis
Often when a researcher is faced with a large
number of variables, some means of data reduction is needed to reduce them to a
more manageable group. This is
justified by the notion that a number of variables may be highly associated
with each other, particularly in the case of census data where information
about type of housing might be related to the type of job a person holds.
Traditionally, one uses factor analysis
(Hutcheson and Sofroniou 1999:217) to reduce the variables to a smaller number
or to create a set of independent factors.
The factors can then be used in regression analysis to attempt to
predict an independent variable, in this case cycling as a transportation
choice.
This sort of analysis is not in fact possible with many of the census variables used in my dissertation because I was working with categorical data. Factor Analysis requires ratio or interval data in which, “the distances between the categories are identical across the range of categories”( Bryman 2001: 219) While this rule is sometimes bent, it is complicated to do so with categorical census data. This is particularly obvious for the dependent variable of transportation choice – one is either a cyclist or an auto-driver or a motor cyclist. Although the dependent variable can be reduced to that of cyclists versus non-cyclists, and thus open up logistic regression as a possibility, the independent variables used here are themselves categorical and therefore make logistic regression overly complicated. If I were working with ordinal or ratio data, the logical approach might have been to do a factor analysis with a number of the census variables to obtain a set of factors. Those factors could then be used in a regression in order to create a model which predicted the choice of cycling from a set of related variables. I have opted for a simpler examination of the dataset, by looking for bivariate relationships with my selected key socio-demographic variables. Given that there is a lot of data investigation involved in my research, the simpler approach seems like the best use of time at this stage.
Using categorical data forces an alternative
method for looking for relationships between the variables. Cross tabulation using chi-square statistics
can be quite useful in identifying variables which appear to be associated with
the dependent variable. By using
Chi-square analysis, it is possible to peer past the absolute number of people
engaging in an activity and examine instead proportions and thereby begin to
look at causality (Hellevik 1984)
In my study, the use of chi-square analysis
of cross-tabulated data is particularly important as a way of examining
populations of different sizes. This is
most obvious in my examination of cycling by gender. Since more men work outside the home than women, reporting the
absolute number of cyclists by gender might suggest that men are more likely to
be cyclists. Chi-square analysis
examines the expected level of cycling in a given population (here men and
women) given the prevalence of cycling and compares it to the actual number of
cyclists to see if the proportion differs significantly from the expected level
for the general population. A numerical
example may make this more clear.
If 1 percent of workers are cyclists and 1000
men work outside the home while only 500 women do, but 80 male workers are
cyclists and only 70 women cycle to work, then it might seem that men are more
likely to be cyclists than women. By
looking at the expected level of cycling, however, chi-square analysis would
suggest that the expected number of cyclists should be 100 men and 50 women. The analysis, then suggests that women are
in fact more likely to cycle to work then men despite the absolute numbers.
The use of chi-square analysis to look for
bivariate relationships is certainly not a complex statistical methodology, but
it does have the benefit of keeping this analysis relatively simple. Analysis of interactions with gender,
ethnicity, age and other basic socio-demographics is a well established
process, and a natural starting ground for mapping out the territory for future
more detailed investigations into the possibilities for expansion of cycling as
a transportation option.
3.5 Understanding Cycling at Local and National Levels
My analysis was conducted at two levels. Because there are more than a million
records, doing initial data sorting seemed best done with a smaller data set
and I chose to investigate York data.
Appendix A discusses some of the alternative options I considered for my
analysis. The city of York has a high
percentage of cyclist relative to its population and yet only about a thousand
records in its data file. Multiple
cross tabulations were possible with this data without tying up heavy processor
time. Also, using York data allowed me
to compare cycling at two scales, urban, and nation wide. As Figure 7 shows,
York is an excellent place for studying cycling because of its diversity of
travel modes. In particular, cycling is
well represented, but so to are other modes.
Comparison of York data with that of the UK
is beneficial as a means of examining the potential outcome for changes in
policy which would allow the United Kingdom to replicate the success of cycling
in York. By studying York as a microcosm
of the United Kingdom, policy makers can consider possible benefits of
increases in cycling by replicating the types of policies which have made York
attractive to urban cyclists.
Approximately 13 percent of York commuters cycle to work, suggesting that
13 percent might serve as an upper bound
on how much of the United Kingdom’s population might be inclined to cycle given
similar policies enacted elsewhere which resemble those in York.
Likely there are unique geographic variables which
encourage cycling in York which do not exist elsewhere. York is a relatively flat and prosperous
city which is compact and has a medium sized population. Replicating the infrastructure of York in a
hilly city with cold wet winters would likely not lead to the same levels of
cycling. However, this argument can
only be taken so far. The United
Kingdom as a whole tends to have a temperate climate and, while hilly, is certainly
not mountainous. UK cities (and particularly London) likely have more in common
with York than would chilly Stockholm or mountainous San Francisco. Moreover, if a cycling culture were
instilled in urban residents from a young age, presumably they would have
plenty of time to develop the type of fitness required to cycle on hills with
the aid of multi-geared modern bikes.
Finally, it is instructive to look at the
possibility of increased cycling in the United Kingdom through the lens of York
because of York’s relatively mixed economy.
It possesses a relatively new university that is more typical of UK
universities cities than say Oxford or Cambridge. Also it possesses a strong tourist trade, religious institutions,
as well as being a regional centre with markets and retail shopping for local
residents. While there is no such thing
as a typical city, York may serve quite well as one which is sufficiently
diverse socially and economically to model the opportunities of a cycle
friendly city.
3.6 Working with Sample Data
In the analysis which follows, I will work with UK sample data for both York and the United Kingdom. My reported statistics will use the sample data as it was presented in the census data file I was provided. The sample data for York contains 204 cyclists, while the sample for the UK as a whole contains 14, 252 cyclists. Appendix B discusses some of the issues involved in extrapolating the sample data to full populations. Intuitively, you would just multiply each number by 50 in order to transform the 2 percent sample numbers into population numbers, thereby suggesting that York has 10,200 cyclists and the United Kingdom has 712,600 cyclists. There complications in doing this which I deal with in the appendix. For simplicity purposes, I felt that it was more appropriate to leave the numbers as they were.
A second issue in dealing with sample data was variation in responses. While there may have been 204 cyclists in York, transformations I made in creating my revised variables sometimes reduced the number of cyclists because of the way that they were coded. For example, when I created my social class variables, I simplified the categories down to the five main social classes that account for the bulk of all records. I chose to exclude members of the armed services and as a result reduced the number of cyclists from 204 to 202. I provide more discussion about this issue in appendix C.
4.0 Results
4.1 Introduction
So far we have seen that there appears to be
a problem with urban transportation. We
have seen that the transportation system we have now plays a significant part
in contributing to greenhouse gases. I subsequently reviewed some aspects of
the urban transportation system, and the results seem to suggest that private
automobiles are a pervasive part of modern urban transportation, and that they
are very resource heavy in terms of energy consumption. Furthermore, I have pointed out that in
addition to contributions to greenhouse gases, automobiles are associated with
other modern ills such as the disengagement many observers feel is occurring
between citizens and their communities.
In response to this review, I have examined
cycling and found that a possible case exists for suggesting that cycling could
play a more important role in alleviating some of the strain current transport
patterns place on the social and natural environment. Historically, I have demonstrated that cycling has played a much
larger role in urban transportation in the past. I have also pointed out some of the benefits that come from
cycling, and cited data which suggests that for trips of under 4 kilometres
cycling is perhaps the fastest travel mode.
What remains to be done, is to determine whether cycling is a broadly
based mode of transportation, or is a niche activity limited to a narrow
spectrum of the population.
4.2 Gender
Initial examination of the association of gender with cycling for the York data suggested that there was not a very strong gender effect. Figure 8 shows the expected number of cyclists given that approximately 12 percent of all commuters are cyclists as well as the actual numbers observed for both genders. Remember that the expected number of cyclists is taken from the number of cyclists in the general population of York or the United Kingdom depending on the unit which is being analysed. For all the charts that follow, the expected number of cyclists is around 12 percent (with some slight variation as is explained in appendix C. For the charts comparing cycling by gender, I have included percentages as well as the absolute numbers of cyclists to make the chi-square results more explicit. For the remaining tables, I only show the expected number of cyclists to those observed.
Figure 8 suggests that once the different sizes of the populations are controlled for, the expected number of cyclists for both men and women is nearly identical with the observed amount. Accordingly, the Pearson chi square score does not begin to approach significance. The size of the sample and the small variation of deviation from expected scores suggest that this is an extremely robust result. It would appear that there is no gender difference in York for the use of bicycles for travel to work.
The story is quite different in examining the
data for the UK as a whole. Figure 9
shows a pronounced difference in the likelihood of cycling for men versus for
women.
Nearly 8000 sampled men were expected to
choose cycling in the sample data compared with 6258 expected among women. Again, this number is based on the
population percentage of cyclists in the United Kingdom which is about 3
percent. As the example shows, however,
chi-square analysis suggests that men are more likely to cycle to work than
women at the national level are.
4.3 Age
Initially, one would expect a significant
interaction between age and the decision to cycle to work. If bicycle commuting were primarily a
function of financial constraints, or if it were principally an activity of
vigorous and healthy individuals who incorporated it as part of a sports based
lifestyle then cycling should be more prevalent among the young. In addition, young people are most likely to
have limited financial resources at the start of their working lives since it
is at that period that they are establishing themselves in new careers, paying
off student loans and making furniture and working wardrobe purchases. All of
these expenses occur while they are starting out at the bottom pay grades of
their employers. Additionally, one
would expect that it is at this initial stage of their career when status
concerns about driving a car, safety concerns about having easy transport for
infants, or the need for transportation between suburban housing and inner city
office jobs are less important.
Among older workers, in addition to having
more time to accumulate resources for automobiles and housing, there are the
additional issues surrounding age and wear and tear on the body. The expectation should be that cycling
steadily drops off as a commuting option as age increases.
The following table presents bicycle
commuting in York for different ages.
Contrary to expectations, there appears to be no real interaction
between age and the choice of cycling as a commute mode. The chi-square analysis of the cross
tabulation shows a complete absence of significance. Although the numbers of cyclists appear to be higher than
expected for workers under 35 and over 44, it is impossible to rule out that
this is a random effect of the drawn sample.
While the slight age interaction for the York
data may be due to random sample effects, the pattern is to some extent
continued at the national level in Figure 11 but with a significant chi-square
value. Similar to the information for
York, cyclists appear to be more likely at the extremes of the working age
distribution. Both the young workers
aged 15 to 24 and the older workers aged 55 to 65 are slightly more likely to
cycle than workers in the middle years.
Perhaps the results for older workers can be in part explained by the
interpretation that less financially secure workers are those most likely to
work until their pension begins at 65, while the more financially stable
workers have retired by this point and therefore moved out of the studied
population. However, the age
interaction for all ages is quite small.
Certainly there are no strong trends, but rather a slight indication that
the oldest and youngest workers are more likely than expected to cycle.
Age was one variable in which a logistic regression could have been conducted instead of relying on a more simple chi-square. In fact I did carry out a logistic regression and failed to find a significant result. Therefore for reasons of consistency, I chose to report the chi-square results.
4.4 Ethnicity
It is difficult to assess whether an
interaction exists with ethnicity and cycling due to small sample sizes. The York sample includes only 13 records for
individuals who are coded as being non-white, rendering any analysis
meaningless. At the national level, an
interaction between cycling and ethnicity does exist, which suggests that
cycling is less likely to be pursued by non-white workers. The results are in fact striking as Figure
12 demonstrates. Again, it is useful to
be reminded that the level of cycling to work at the national level is three
percent so given that 6294 people of Indian descent are included in our sample,
we would expect approximately 188 people.
That in actual fact only 22 Indians identified themselves as cyclists
emphases the ethnicity interaction with the cycling variable.
Although, the sample sizes are quite small
for the non-white population, the expected numbers of cyclists are in many
cases several multiples of the actually observed numbers of cyclists. This holds in cases where the sample size of
cyclists is quite small as well as for much larger ethnic communities. Ultimately these results must be seen as
robust given that the total samples even for the smallest ethnic group
(Bangladeshi) are still reasonably large (N=463).
4.5 Social Class
Figure 13 examines the relationship between
social class and cycling. As has been
mentioned, class is divided up into 5 classifications. The results of the
chi-square test indicate that there is a significantly higher than expected number
of cyclists who work in occupations lower down the skill scale. Unskilled and partly skilled workers are
much more likely to cycle than expected.
Moreover, cyclists make up a significant proportion of the people in
these two classifications. 28 of 113
unskilled workers cycle to work (nearly 25 percent), while 51 of 288 partially
skilled workers (18 percent) are bicycle commuters. Among those in the second and third highest social classes, the
observations are reversed. Here we see
that significantly fewer than expected of these two classes are cyclists. Among professionals, the observed and
expected numbers nearly match. Although
the sample size is quite small and likely not significant, it suggests that
professionals are more likely to cycle than are those in the next two classes,
although certainly not as much as those in the lower occupational classes.
This relationship appears to be the same for
the UK data. Broadly speaking, Figure
14 demonstrates the same shape of relationship as is seen for the York
data. Like the York data the Pearson
Chi-square score is highly significant.
As in the case of York, higher than expected cycling results are found
for those in the lower skilled classes, with a strong reversal for highly
skilled workers and managers and technical workers, with the number of
professional cyclists slightly lower than expected. For both York and U.K. data it is individuals at the lower scale
who are the most highly likely to be bicycle commuters. As in the case of York, the actual number of
cyclists among the unskilled workers is double the number expected. However, the percentage of cyclists among
this occupational class is significantly lower. Cyclists make up only approximately 6 percent of unskilled
workers compared to 25 percent for York.
The York and the UK results point to a possible relationship between occupational social class and the choice of cycling.
4.6 Housing Tenure
The results reported in Figure 15 examine the
relationship of cyclists and non-cyclists to their housing tenure,
specifically, the extent to which they have developed an asset base in real
estate. Not only does this variable
serve as an indication of financial wellbeing, but also to some extent
complements the previous measure of social class. Ignoring for a moment a class system based on status of career,
it could be said that an alternative system could be based on one’s
relationship with real estate.
Certainly, the ownership of real estate has been associated with status
in the past. In modern terms, owning
one’s home outright permits an individual to use money that would otherwise go
toward the mortgage for savings or increased consumption. More money opens up options for early
retirement and as such movement away from identification by employment.
For the purpose of this research, I am
interested in determining relationships of cyclists with their property. One of the benefits of cycling is the
ability to save significant amounts of money on auto expenses, or even on bus
or rail passes. Once the up-front cost
of a bike has been made, and maintenance costs factored in, a bike is a free
mode of transportation. Such savings
can be used for other expenses, such as paying a mortgage more quickly.
In Figure 15 it is clear that in York
cyclists exceed their expected levels significantly only in the case of those
owning a property free and clear. Fewer
than expected are still paying mortgages, and the observed and expected number
of renters is nearly equal. These
results are somewhat difficult to interpret.
Previously it was seen that cyclists were most present among unskilled
workers. However, Figure 15 suggests that
cyclists are more likely to own their homes outright than to have a mortgage or
to rent. These results provide
encouraging support to the notion that cyclists are able to build up
significant financial assets as a result of their transportation choices. However, further research is needed to
determine the extent that non-skilled and semi-skilled workers are counted
among the homeowners.
The trend is strengthened when we look at the
national situation as shown in Figure 16.
The UK data lends support to the hypothesis that some cyclists are able
to build up significant equity in their homes through their saving from auto
and public transport expenses, while also showing that there is a tendency for
cyclists to be renters as well. That
being said, the majority of cyclists either own outright or are paying off a
home. Appendix E looks at this
phenomena in more depth at the regional level.
Looking at that data seems to suggest that the cycling – outright home
ownership relationship particularly seems to be associated with the wealthier
South and Midlands region of the UK. In
the North, in Scotland and Wales, cyclists are less likely to own their own
homes than expected. This relationship
seems to be a fruitful area for further research.
4.7 Travel Distance
In section 1.0, I mentioned that cycling may
be the fastest mode of travel for distances under 4 kilometres. Figure 17 examines the likelihood of
cycling over various distances. Not
surprisingly, beyond a certain distance, the cycling drops off as a commuting
mode choice. Beyond a certain distance,
a bike ride becomes a serious workout.
Figure 17 suggests that this critical value is approximately 5 km.
In Figure 18, similar results are shown for
the United Kingdom as a whole. Again,
the trend seems to be strong and also similar to that suggested in the York
data. Distances longer than five
kilometres seem to represent the cut off point beyond which significantly fewer
cyclists are present. It is once again
important to recognise the robustness of the York data in predicting national
data. The relationships themselves are
therefore similar at the local and national level.
4.8 Commuter Efficiency
Figures 19 and 20 attempt to examine a
concept I would term commuter efficiency.
Given the previous results indicating that cyclists are more likely to
travel distances of up to 4 km, the next set of tables looks at the types of
travel chosen over this optimal cycling distance. Given the ease of getting around by bike over such short
distances, it is instructive to examine the modes of travel to work actually
chosen by commuters. It is no surprise
that walking and cycling are much more likely than expected. Most striking is the number of individuals
driving their cars over this short distance.
Commuting to work in a car is by far the largest and most popular means
of commuting short distances. Walking
is the next most common mode, followed by cycling. It is surprising that the bus should be the fourth most popular
mode.
These results suggest that even in a walking
and cycling friendly town such as York, automobiles seem to dominate. While there are alternative means of getting
to work quickly and with some exercise thrown in, a majority of the population
opts to use their automobile to get to work.
The story is the same at the national
level. Once again, cycling and walking
are much more likely than is expected over short distances. However, more individuals drive these short
distances than all other modes combined.
The data in Figure 20 is an aggregate for the whole country, which means
that short trips in London are combined with short trips in small communities. In practice this means that doctors driving
to jobs in London who choose to drive rather than take a bus or the metro are
lumped together with country vets who drive from their homes in the countryside
to their surgeries perhaps in a neighbouring village. Not all individuals have public transport options as alternatives
to their automobiles; however, it is still significant that nearly half of all
commutes to work in the United Kingdom are within easy cycling distance. Tolley
and Turton (1995: 213) support this by reporting that 60 percent of all car
trips in the UK are less than 8 kilometres. Terrain issues, inadequate and
unsafe cycling paths due to either high urban traffic on roadways or high-speed
rural traffic on country motorways may deter cyclist over such short
distances. However, the need to travel
long distances to work does not appear to be a major obstacle for large numbers
of residents of the United Kingdom.
5.0 Conclusions
5.1 Stabilising Greenhouse Gases Requires Significant Action
There is wide-spread disagreement about the
most adequate response to the environmental situation facing the planet. While there are many opponents to the Kyoto
agreement, there does appear to be widespread agreement that 1990 levels
represent a reasonable target. We have
seen that significant challenges exist if the global community intends to reach
1990 levels. More importantly, on a
global basis, emission levels in the developed world will have to be reduced
dramatically in order to stabilise global emissions at this level once
developing nations take on Western development patterns.
Conclusion One: It appears that significant cuts in carbon emissions are needed in OECD countries in order to stabilise emission levels at sustainable levels. In order to accommodate growth in emissions among developing nations, wealthy countries will have to reduce emissions significantly.
5.2 Urban Travel in the Developed World Contributes to Carbon Emissions
From an examination of carbon levels, the
analysis proceeded to examine some of the problems with private
automobiles. It was shown that private
vehicles are particularly associated with demanding energy requirements, much
more so than public transport. In
addition, the dominance of automobiles was shown to have additional adverse
effects on urban life. While public
transit was shown to have many benefits compared to private automobiles for
improving urban life, even in countries or cities like Japan and Paris with
high levels of public transport, CO2 emissions are far higher than those
associated with sustainable levels. A
non-polluting form of transport would help alleviate some the pollution
problem.
Conclusion Two: The current urban transport system causes too much pollution. Movement toward non-polluting modes of travel holds the promise of lowering greenhouse gas emissions.
5.3 Cycling Could Alleviate Some Problems Created by Automobiles
At present, it appears that efforts to
stabilise CO2 emissions with current forms of transport are failing. What is needed is a non-polluting form of
transport capable of providing mobility over medium range distances of a few
kilometres. An analysis of cycling
demonstrates that historically it appeared to be poised to play a major role in
increasing mobility in the early twentieth century, but was replaced by the
automobile as the dominant mode of transportation for the twentieth
century. In addition to allowing city
travel, cities friendly to cycling also have the potential of increasing their
citizen’s and city’s health through greater exercise. It is worth investigating whether it might also lead to a greater
involvement in the local community.
Conclusion Three: The bicycle holds the potential to play a much greater role in the urban transportation system as well as improving the quality of life in cities.
5.4 Demographically, Urban Cyclists are not a Fringe Group
In the analysis of York and UK cycling
statistics, it has been demonstrated that cyclists come from all walks of
life. While there do tend to be segments
of society who are more likely to cycle, no particular group is dominant. As a result, it is impossible to dismiss
cyclists as a fringe group. Clearly,
cycling is not a particularly popular commuting mode, compared to automobiles,
or even public transit. That being
said, it does benefit from diversity.
While some interactions exist, particularly for women at the national
level, it appears likely that greater study into the concerns women have might
reduce the disparity as seems to be the case for York.
Conclusion Four: Cyclists as a sub-population appear to look reasonably like the general population. Although some socio-economic variation exists, one particular group does not appear to dominate.
5.5 Cycling Appears to be Consistent with Economic Wellbeing
Cyclists were seen to be somewhat more likely
to be drawn for working class and unskilled backgrounds than were clerical
workers; however, the trend reversed slightly among professional at the top of
the worker prestige categories.
Moreover, cyclists were much more likely to own their own home outright
than expected in both York and the UK.
While greater research into this relationship would be necessary to draw
definitive conclusions, the trend seems to suggest the possibility that cyclists
are a financially stable group.
This point is driven home further by looking at the top three centres of cycling in the UK. Oxford, Cambridge and York are listed at the top of the cycling scale (The Department of Transport 1996). All three of these cities are known to be economically strong cities, and in the cases of Oxford and Cambridge, major contributors to the high value added knowledge economy. Furthermore, the two countries most associated with cycling among the developed nations are the Netherlands and Denmark. Both countries rank high in terms of standard of living and national wealth (United Nations 2003). While it would be overstating the case in the extreme to attribute their wealth to cycling, it can certainly be said that wealth and dependence on cycling as a major transportation mode are not mutually exclusive.
Conclusion Five: High levels of cycling as transport is consistent with the operation of wealthy high value added economies.
5.6 Cycling is not the Preserve of the Young
My analysis suggests that cycling can serve
as effective transport throughout one’s working life. While there are limitations on cycling among the very young and
presumably the very old, cycling certainly appears to be popular among workers
well into their sixties. Not only has
this study shown that older workers continue to cycle right up to the point at
which they begin to collect their pensions, but that they do so in greater
numbers than their population size would suggest. While this is an impressive result in and of itself, the main
conclusion of the analysis of age and cycling was of the prevalence of cycling
throughout the lives of workers.
Clearly cycling is not primarily a young person’s activity, nor a
stopgap transport mode before a person can purchase a car; rather, cycling is
chosen as a commuter mode by young workers and those in their middle working
years as well as workers preparing to retire.
Conclusion Six: Cycling has the potential to provide transport to city residents of all ages, and is not age constrained.
5.7 Cycling Appears to be Appropriate for Many Urban Commuters
The analysis in this paper appears to suggest
that urban transportation is chosen in a very inefficient manner. While we have seen that cycling is most
popular over short distances of up to 4 kilometres, it is striking that
automobiles continue to be the preferred means of transport at these
distances. Cycling even appears to have
benefits over public transport over short distances.
Clearly, trains, buses and trams play a
significant role in transporting people over city distances. For city residents with mobility problems,
extensive public transit and availability of automobiles can make the difference
between being able to engage in civic life, and being marginalised. Moreover, variation in climate and terrain
as well as personal preference ensures that the bicycle is not the solution for
every city trip. However, given its
many advantages, the bicycle as a mode of city travel appears to be greatly
under-utilised. When trips by
automobile or even occasionally public transit are chosen over cycling, the
potential exists that the travel is choosing their travel mode inefficiently.
Conclusion Seven: An efficient transportation system is one which uses the most appropriate mode for the distance consistent with personal and environmental conditions. Cycling in concert with greater use of public transportation and walking appears to have the potential in some situations to improve the urban transport system.
5.8 The UK Appears to Have Room to Grow in Cycle Usage
Finally, the York data can be seen as an
indication of the possibilities for UK cycling. Certainly York does not represent the upper boundary for what is
possible for urban cycling. There must be potential cyclists who choose transit
or private cars due to lack of convenient cycle lanes or other safety
concerns. However, approximately twelve
percent of commuters in York choose to cycle to work. The city has set out an official policy to encourage cycling
(York City Council 2003). If the UK
were to adopt similar cycle friendly policies to attract thirteen percent
cycling levels on a national scale the number of bicycle commuters (multiplying
the 2 percent sample numbers by fifty) could be estimated to rise from just
over 700,000 to nearly 3 million.
Conclusion Eight: With cycling friendly strategies in suitable contexts, cycling as transport appears to have the potential to make up a much larger proportion of the UK transport system.
5.9 Cycling and Urban Planning Are Inter-dependent
Looking beyond UK examples and thinking
globally, there are a number of challenges in allowing cycling to grow as a means
of transport in urban settings beyond merely building transit lanes. Many of these likely pertain to climactic
factors of rain and snow, geographic factors of terrain, and security factors
resulting from inadequate cycling infrastructure which separates the cyclist
from traffic. One additional factor
comes from a finding in Kenworthy and Noble (94-95). Comparing Los Angeles and Paris, their analysis shows that in the
respective Central Business districts of both cities, the density of population
is 28.2 in Los Angeles as compared with 179.7 for Paris, while the density of
jobs is 506.1 for Los Angeles and 369.2 for Paris.
These two cities are extreme examples of a
trend for many American and European cities.
Americans tend to not populate the areas where the business of the city
is done to the same extent that Europeans do.
Clearly, cycling will not be a commuter option for forty kilometre
commuting distances. While cycling
might be a means of travelling to commuter rail stations which take workers
into the central city, a better solution seems more likely. That is that the built environment, its
geography, land use, activities and energy consumption are intimately
connected. Building separate bike
lanes, or seeking other ways to reduce the reluctance of travellers to use
cycling for their daily trips is only part of the battle for a bicycle friendly
transportation system. Equally
important is the creation of a built environment which reduces the need for
private automobiles. Clearly, the location
of services, homes and workplaces will play an enormous role in reducing our
energy consumption. A better
co-ordination of land use and the location of services and perhaps a general
movement toward compact cities appears to be necessary to unlock the potential
of energy savings.
Conclusion Nine: In addition to cycle friendly strategies, changes to the structure of the built environment are necessary in order to develop an efficient urban transport system.
5.10 Summary
This paper has sought to examine the case for
cycling playing a larger role in the search for a sustainable urban
transportation system. The consensus
that greenhouse gas emissions are too high, and the automobile’s role in
contributing to this form of pollution as well as some of the other
consequences of an automobile dominated city have been examined. In choosing to look at the bicycle as a
possible replacement for automobile trips, I have tended to side with arguments
which are more popular with deep ecologists and advocates of sustainable
development rather than supporters of market solutions to the rise in
greenhouse gases. In doing so, it is
impossible to overlook a subtle criticism of the market solution to dealing
with environmental concerns. One of the
most striking observations has been the number of short commutes to work that
are taken by automobile. Given the
obsession by economists for efficiency and productivity, money spent on these
trips appears to be poorly invested.
Given the availability of cycling as a cheep and healthy alternative,
short car rides to work might be better classified as a luxury good. However, contrary to the deep ecologist
argument that cuts in economic well-being may be necessary, there is some
indication that a city of cyclists might actually lead to greater wealth. My analysis has revealed that cyclists tend
to be homeowners, and moreover, that they are more likely to own their home
outright than would be expected. My
analysis hints that rather than costing society a great deal to move away from
an automobile based society, major sources of wealth may be able to be unlocked
by a move toward cycling, the case is even stronger if we choose to measure
wealth by the possession of good health and strong communities.
5.11 Next steps.
This dissertation is only a first step in
examining the prospects for greater use of cycling for transportation. It suggests that cycling has the potential
to play a much larger role in modern transportation systems in the UK. To strengthen this argument, several
follow-up studies would be helpful to confirm this research over time and
internationally and to create a strategy for increasing the role played by
cycling in urban transport.
Step One: Repeat the socio-economic analysis
of York and the UK using the soon to be released 2001 UK census data. By confirming the structure of the cycling
community in a mutli-year study, it would be possible to assess whether there
is stability in its make-up or whether growth or decrease is focused in certain
groups.
Step Two: Make use of Geographic Information
System (GIS) technology to examine cycling from a spatial perspective. Since cycling is most popular over short and
medium distances, an important next step would be to examine the spatial
distribution of cyclists in order to understand their geography and the
characteristics of cycling friendly neighbourhoods. By looking at cycling from a neighbourhood perspective, barriers
to cycling as well as factors encouraging cycling might be more easily
identified.
Step Three: Repeat the analysis in a
multi-country study. Not only is it
beneficial to determine if there is time stability in the cycling community,
but also the extent to which cyclists are similar or different as a group in
other developed countries. If the social
composition of cyclists are similar, the city and national programs aimed at
encouraging cycling are probably more likely to be effective across
jurisdictions. If not, it would suggest
that each country will need to take a unique approach in order to encourage
cycling within their city or country.
Step Four: Make use of national transport
surveys as well as social surveys which are conducted on a more frequent basis
than census data. Although these
surveys lack the sample sizes of census data, they do have a greater breadth of
detail about cyclists, and also permit cycling patterns to be tracked annually,
thereby making patterns more obvious.
Step Five: Once a clearer picture of the
cycling community has been established, qualitative studies of cyclists and
potential cyclists will be particularly valuable to clarify motivational
aspects of their transportation choices.
This is a useful step before making expensive policy and infrastructure
decisions, as it should help to clarify which will have the greatest impact on
encouraging residents to choose cycling as a transportation mode.
6.0 References
Allen, R. 1986. “What Black Elk Left Unsaid: On the Illusory Images of Green Primitivism”. Anthropology Today, Vol. 2, No. 6, pp. 8-12.
Aronson, S. 1952. The Sociology of the Bicycle. Social Forces, Vol. 30, No. 3, pp. 305-312.
British Medical Association 1992. Cycling: towards heath and safety. Oxford: Oxford University Press.
Balchin, P. 1995. Housing Policy: An Introduction. London: Routledge.
Carlsson-Kanyama, A., Lindén. A.and Thelander, Å. 1999. ‘Gender Differences in Environmental Impacts from Patterns of Transportation – A Case Study from Sweden’. Society and Natural Resources, 12, pp. 355-369.
Census Microdata
Unit. 1991 Sample of Annonymised Records. Manchester University. http://www.ccsr.ac.uk/sars/
Census Microdata
Unit. 1991 Sample of Annonymised Records - codebook. Manchester
University. http://www.ccsr.ac.uk/sars/
Census Microdata
Unit. 1991 Sample of Annonymised Records - Glossary. Manchester
University. http://www.ccsr.ac.uk/sars/
Cipolla. C. 1962. The Economic History of World Population. Middlesex: Penguin
Clavert, P. 1982. The Concept of Class: an historical introduction. New York: St. Martins Press Inc.
Dale, A. Fieldhouse, E. & Holdsworth, C. 2000. Analysing Census Microdata. London: Arnold.
ExxonMobile Corporation 2001. Media Statement – Global Climate change. Available from http://www.exxonmobil.com. Accessed on August 1, 2003
European Commission 2000. cycling: the way ahead for towns and cities. Luxembourg: Office for Official Publications of the European Communities.
Gehl, J. 1996. Koch, J. (trans.) Life Between Buildings: Using Public Space. Bogtrykkeriet: ArkitektensForlag.
Gordon, P. and Richardson, H. 1999. “Review Essay: Los Angeles, City of Angels? No, City of Angles”. Urban Studies, Vol. 36, No. 3, pp. 570-591.
Government of Canada 2002. Climate Change Plan for Canada http://www.climatechange.gc.ca Accessed July 1, 2003
Hellevik, O. 1984. Introduction to Causal Analysis: Exploring Survey Data by Crosstabulation. London: George Allen & Unwin (publishers) Ltd.
Houghton. J., Hilho, L. Bruce, J. Hoesung, L. Callander, B. Haites, N., and Maskell, K. 1997. “Climate Change: Radiative Forcing of Climate – Executive Summary” in Owen, L. and Unwin, T. (eds.) Environmental Management: Readings and Case Studies. Oxford, Blackwell.
Hutcheson, G. and Sofroniou, N. 1999. The Multivariate Social Scientist. London: Sage.
Humphries, S. 2001. ‘Who’s Afraid of the big, Bad Firm: the Impact of Economic Scale on Political Participation’. American Journal of Political Science, Vol. 45, No. 3, pp. 678-699.
Kiecolt, K. and Nathan L. 1985. Secondary Analysis of Survey Data. Beverley Hills: Sage.
Klein, R. 2003. Steady as We Grow: The Government’s Plan for 2003. Speech to the people of Alberta, January 27, 2003. http://www.gov.ab.ca/premier/address2003/content/page_complete.cfm Accessed August 23, 2003.
Lélé, S. 1991. “Sustainable Development: A Critical Review”. World Development, Vol. 19, No. 6, pp. 607-621.
Lombourg, B. 2001. the skeptical environmentalist: Measuring the Real State of the World. Cambridge: Cambridge University Press.
Naess, A. 1989. Rothenberg, D. (Trans.) Ecology, Community and Lifestyle: outline of an ecosophy. Cambridge: Cambridge University Press.
Naess, A. 1995. “The Deep Ecology Movement”. Sessions, G. (ed.) Deep Ecology for the 21st Century: Readings on the Philosophy and Practice of the New Environmentalism, Boston: Shambhala. pp. 64-84.
Newman, P. and Kenworthy, J. 1999. Sustainability and Cities: Overcoming automobile dependence. Washington: Island Press.
Pikering, K. and Owen, L. 1994. an introduction to global environmental issues. London: Routledge.
Putman, R. 2000. Bowling Alone. New York: Simon and Schuster.
Rudlin, D. and Falk, N. 1999. Building the 21st Century Home: The sustainable urban Neighbourhood. Oxford: Architectural Press.
Sarkar, S. 1999. Eco-socialism or eco-capitalism?; a critical analysis of humanity’s fundamental choices. London: Zed.
Sessions, G. 1995. “Deep Ecology and the New Age Movement” in Sessions, G. (ed.) Deep Ecology for the 21st Century: Readings on the Philosophy and Practice of the New Environmentalism. Boston: Shambhala. pp. 290-310.
Sidenbladh, G. 1968. ‘Stockholm: a Planned City’ in Scientific American’s Cities. New Yor:, Alfred A Knopf inc.
Span, G.2003. “Paving the Way for Buses: the Great GM Street Car Conspiracy.” San Fransisco Bay Crossings.Com Available at: http://www.baycrossings.com/Archives/2003/03_April/paving_the_way_for_buses_the_great_gm_streetcar_conspiracy.htm Accessed September 8, 2003.
The Department of Transport 1996., Cycling in Great Britain. London: HMSO.
Thoreau, H.D 1985. Walden; or, Life in the Woods. Cambridge: The Library of America
Timms, S. 2003. Minister’s Speech to the London E-commerce conference July 10, 2003. Available from http://www.dti.gov.uk/ministers/speeches/timms100703a.html. Accessed August 26, 2003.
Tolley, R. and Turton, B. 1995. Citing Burton 1988) Transportation Systems, Policay and Planning: A Geographical Approach. Essex: Longman Scientific & Technical
United Nations Human Development Report 2003. Available at http://www.undp.org/hdr2003 Accessed August 27, 2003.
World Bank 2003. World Development Indicators. CD-ROM Version
Walby, S. 1986. ‘Gender, Class and Stratification: Towards a New Approach’ in Crompton and Mann’s (ed.) Gender and Stratification. Cambridge: Polity. Pp. 23-39.
Yago, G. 1983. ‘The Sociology of Transportation’. Annual Review of Sociology, Vol. 9. pp. 171-190.
United Nations Kyoto Agreement 1992. Available at http://unfccc.int/resource/docs/convkp/kpeng.html Accessed August 27, 2003.
York City Council Cycling Strategy Available at: http://www.york.gov.uk/environment/transport/excellence/strategy.html Accessed August 27, 2003.
Appendices
Appendix A Alternative Cities to York for Investigation
Although I decided to look at York data for
my investigation, I considered a number of other cities. Both Oxford and Cambridge have higher
cycling populations but I ruled them out because both are unusual university
towns which I did not feel would be appropriate for generalising to the whole
population. Because of their unique
history, and their heavy dependence on knowledge based industries, cycling in
both cities may be grounded on different foundations.
Similarly, I was tempted to make use of
Glasgow or Edinburgh data for my analysis.
Being located in Edinburgh would have made the selection of that city a
natural choice. However, cycling is a
much less popular transport choice in Edinburgh at around 2 percent of the
population, and even less popular in Glasgow at below 1 percent of the population. In fact, the low incidence of cycling in
most cities made it difficult to choose any exemplar city to examine cycling in
a context in which it carried a significant part of the transportation
load. The Department of Transport
Document Cycling in Great Britain
(1996) lists a number of other candidate cities, but few seemed to meet my
criteria. Although York is not a
perfect choice, it seemed to offer the best set of characteristics to study
cycling in a natural setting. Figure 20
shows cycling levels for a cross section of Urban centres in the UK, and as the
reader will note, despite its smaller size, York has largest total number of
cyclists in its sample.
Appendix B The Two Percent Sample
All data reported in the results which follow
are drawn directly from the SARs two percent sample of census data. Weighting factors are present in the data,
however (Dale, Fieldhouse & Holdsworth 2000: 80-82). As these authors point out, most weights in
the sample sum to one, however for small samples there can be variation which
should lead to caution about extrapolating results to the whole
population. The authors point out that
weights represent, “the ratio of usual residents in each SAR area, by age and sex
categories, to the corresponding number of persons in the Registrar General’s
mid-1991 estimate for that area” (81)
Rough estimates of the true population
numbers can be obtained by multiplying a particular cells numbers by 50. And the larger the sample, the more accurate
the population estimate. So for
example, the reported sample of UK cyclists is listed in my tables as being
14252. The actual population of cyclist
who commuted to work in the UK in 1991 when the census was taken was
approximately 700,000.
A
more formal treatment would require using the weightings. Figure 21 demonstrates the difference if the
simple versus formal method of calculating population totals were used for the
18 cyclists in Chelsea. As is obvious
to the reader there is some variation in the two estimated populations, and the
figure is a good illustration of some of the pitfalls for making accurate
estimates of small subgroups of the SARs population.
Appendix C Variations in the Population of Cyclists
While the sample of cyclists in the United
Kingdom is based on 14,252 cases, some of my reported results are actually
based on a smaller sample. This is due
to transformations I made in the data to make the output of the charts more
meaningful, and to deal with small sample size concerns. The best way to see this is to consider my
results for gender. The chi-square
analysis of this variable employs the full population because gender is an
unambiguous variable. There is no
non-response rate to this variable and there are only two acceptable
categories. As a result, my reported
numbers for gender are representative of the entire sample population. The gender tables which follow indicate that
there are 204 cyclists in the data set for York and 14,252 in the dataset for
all of the United Kingdom.
The same does not hold true for social
class. For this variable I purposely
excluded small occupational class variables because of the small sample size
for York. For the purposes of
consistency, I did the same at the national level despite the fact that the
larger samples rendered mute the small sample size effects. As an example, I excluded the category of
armed service members from social class because of the small sample of this
group in York.
Consistent across all variables other than
gender were nonsense categories such as not-applicable or no-response. I tended to remove these cases from my
samples because of difficulties in interpreting the results. A final category of transformations I used
was aggregation. For example, the mode
of travel to work is broken down into a number of similar categories. For completeness sake, I could have recorded
separately individuals who drive to work in an automobile and individuals who
travel to work as passengers in automobiles.
However, it seemed more reasonable to combine the two categories into
one category of automobile commuters.
While in theory carpooling is a different class of commuting than
driving to work in one’s own car, in practice there was too little information
to study this phenomenon with that data available. The same holds true for users of British Rail (as it was in 1991)
and other forms of rail travel. While
the use of a city tram or underground is qualitatively different from full size
commuter rail and inter-city rail, in practice the sample sizes were often so
small, and the inability to interpret the different types of rail so difficult,
that in the end, it seemed more appropriate to merely differentiate between
rail and bus users.
Similarly, I simplified the 10 categories of
tenure of household into 3 categories of house ownership – outright, house
ownership with mortgage, and renters.
Certainly I lost a large amount of detail about renters, particularly
pertaining to their accommodation in public housing versus private rental
accommodation, but I hope I gained something in terms of clarity of
presentation.
A final note should be made of the prevalence
of individuals who work from home.
Clearly, working from home was not a focus of my research. As a result, I tended to exclude such
workers from my analysis. While working
at home may be very desirable as a way of avoiding traffic congestion and reducing
greenhouse gases through avoiding travel, my primary focus was on the characteristics
of travel. Where it was necessary to
compare cyclists to other modes of travel, I excluded home-workers for the
natural fact that they do not in fact travel to their place of work (if we can
exclude the commute down the stairs to the home office as travel).
Appendix D Efficiency and Public Transportations
The modern obsession with the automobile is
only the most recent round in the environmentalist battle for appropriate
transportation. The nineteenth century
philosopher Thoreau (1985: 365) long before the development of the automobile
identified the train as an eyesore. In
his extended essay entitled Walden, he identifies them as a questionable good:
“I have learned that the swiftest traveller is he who goes by foot. I say to my friend, supposes we try who will get there first. The distance is thirty miles: the fare ninety cents. This is almost a days wages … Well, I start on foot, and get there before night … You will in the mean while have earned your fare, and arrive there some time to-morrow … instead of going to Fitchburg, you will be working here the greater part of the day. And so, if the railroad reached round the world, I think that I should keep ahead of you; and as for seeing the country and getting experience of that kind, I should have to cut your acquaintance altogether.”
And,
“This spending of the best part of one’s life earning money in order to enjoy a questionable liberty during the least valuable part of it, reminds me of the Englishman who went to India to make a fortune first, in order that he might return to England and live the life of a poet. He should have gone up garret at once. “What!” exclaim a million Irishmen, starting up from all the shanties in the land ‘is not this railroad which we have built a good thing?’ Yes, I answer, comparitively good, that is, you might have done worse; but I wish, as you are brothers of mine, that you could have spent your time better than digging in this dirt.”
While Thoreau was an idealist and a dreamer, he does serve to remind us that transportation choices we make have profound impacts on the quality of our daily lives. Over dependence on automobiles and even public transportation comes at a cost. While both forms of transportation have their place in the modern transportation system, a simpler urban form built around cycling and other forms of human powered transportation promises to transform the quality of our urban life while allowing us to enjoy many of the benefits of modern technology. By applying the correct mode of transportation for the particular journey requirements, we have the potential to reap a healthier and more sustainable future.
Appendix E Home Tenure by Region
Although I focused
on the local and National levels of cycling, I also toyed with the possibility
of examining cycling at a regional level.
One advantage would be that I would gain a larger sample size, but the
disadvantage is that I would be working with cyclists as a much smaller
percentage of the population.
Perhaps an even
more significant advantage is the prospect of comparing some of the
relationship I found at both the York and National level in several other
geographic contexts. Figure 22 shows
Tenure of ownership by region. As was
mentioned in the text, the tendency of cyclists to own their own home is
somewhat confirmed in this table, but with the qualification that the phenomena
is not consistent throughout all regions of Britain. Of the 12 regions shown, observed home ownership is higher than
expected in 7 of the regions and lower in 5.
Moreover, the trend is strongest in the wealthier South.
