TRIP LENGTH CHARACTERISTICS IN LAGOS

METROPOLIS

Wale Alade
Department of Urban and Regional Planning,
University of Lagos,
Lagos, Nigeria
+2348052600359
[email protected]

ABSTRACT
The pattern of urban transportation have been recognized as a major factor affecting work
and living condition of many urban residents in third world cities. In particular, the trip
length between home and work is a major concern in large cities where the transport
system is not efficient. Also, efforts aimed at improving urban mobility have been difficult
due to absence of current and adequate data on household travel demand. Findings have
shown that the socio-economic status of individuals and households affects intra-urban
travel demand and trip length. Consequently, this study examines the trip length
characteristics of 1785 household heads in relation to their socio-economic characteristics
in selected residential areas of Lagos metropolis using systematic sampling technique. The
objective is to determine respondents’ socio-economic attributes that provide explanation
for the observed variation in trip length using descriptive and inferential statistics.
Analysis of the socio-economic characteristics revealed that 97.6% of respondents have
formal education, 70.6% are employed, 65.9% are low income earners, 57.0% are from
medium sized households, 81.7% have at least one car in the family and 98.1% have
access to at least one mobile phone service (GSM). The study also showed that trip length
has significant and positive correlation with education, income and use of GSM and
negatively correlated with car ownership, gender and employment at P≤0.05. Based on the
multiple regression analysis, the study concludes that education, car-ownership, income,
gender and telecommunication in that order are the dominant socio-economic
characteristics that affect the trip length of households in Lagos and that policies aimed at
reducing urban trip length must take cognizance of these factors.
KEY WORDS: Trip length, Travel Behaviour, Residential Density, Travel Demand

2
Introduction
The function of transport in cities is to enable goods and services move from one point of
demand to that of supply between various buildings within a locality and from one place
to another. Apart from the movement of goods, urban centres require the movement of
people to their different parts to carry out socio-economic and political activities either
voluntarily or out of necessity. Wide separation of activities in urban centre often imposed
by rigid zoning practices creates a demand for relatively long trips among modes of
transport and unbalanced capacity of networks, create congestion and cause excessive use
of energy and environmental pollution. (Adeniji, 1993, Oyesiku, 2002). The relatively
long urban trip length usually experienced by urban residents also emanates from poor and
uncontrolled urban growth. For instance the city of Lagos covered about 3.97km 2 with a
population of 25,083 in 1886. The total land area increased to about 70km 2 in 1950, giving
an increase of about 1750 percent over a period of about 85years. By 1976, the city had
enveloped about 271.2 km2 and increased to 355 km2 by 2000. By the year 2002, the
lateral growth of and expansion of Lagos have increased to 845 km 2 within 24years,
giving an increase of about 323 percent (Oyesiku, 2002).

This situation implies that more people than ever before are living and working in cities
and more people and more goods will be making more trips in urban centres, often over
long distances (Zegras, 1997). With expansion of urban population and land area,
movement of people over long distance became a primary transportation problem within
the city. One of the ways of meeting this challenge in many cities of the world is the
expansion of the road transport infrastructure, a situation which further encourages
sprawling of urban centres and make intra-city movement longer and more complex. This
also makes accessibility of residents in the suburb to the city centre more difficult.
Significant efforts have been made in different parts of the world to manage urban
transportation especially in many developed nations. This has led to several studies on
travel behaviour of urban households with a view to determining the magnitude and
dimension of the problems associated with urban travels. This paper argues that the socio-
economic characteristics of the households are significantly related to their trip length. It
therefore examines the socio-economic characteristics households in relation to their trip
length in Lagos metropolis.

Literature Review
The study of travel behaviour over the last half century has yielded critical insights into
the choices that individuals and households make about their daily travel (Clifton and
Handy, 2001). These insights have contributed to the development of more studies in
America, Europe and Asia with increasingly sophisticated methods by researchers and
transport experts to understand and predict travel behaviour. The outcome of many of
these studies have influenced to a great extent several transport planning decisions and
policy issues in many countries of the world (Fadare, 1989, Mokhtarian, 2002, Srinivasan,
2005). Several studies have been carried out to examine the socioeconomic characteristics
of individuals and households as it affect travel behaviour. Scholars (Fadare, 1987, 1989,
Ogunjumo, 1986, Pucher and Renne, 2003, and Fujiwara et al, 2005) have identified
household size, car ownership, income, age, gender, number of employed people in the

3
family, occupation among others as major socio-economic attributes of households that
influence their travel behaviour.

Income is regarded as a primary determinant of travel behaviour and car ownership and
the higher the income the higher the number of vehicles per household. In the US, three-
quarter of even the poorest households owns a car. Increase in number of cars per
household has however been observed to have minor additional impacts on travel
behavior. Pucher and Renne (2003) observed that high income households make more
trips per day than low income households and travel longer distance about twice that of
low income households. Srinivasan and Rogers (2005) noted that women conducted more
trips and tended to use the least expensive mode in Chennai. Women in both developed
and developing countries tend to make more trips characterized by complex chains than
men (Brun and Fagnani, 1994, Astrop, 1996, Helling, 1996). Apart from gender,
Srinivasan (2005) found that number of vehicles in the household and the income level
were significant socio-economic factors influencing travel behaviour in Chennai, India.
Fadare (1989) observed that household socio-economic attributes (age, sex, and income
and car ownership) influenced the number and purpose of trips which household members
make in a weekday in Ibadan, Nigeria.

The study by Fadare (1989) on Ibadan further revealed that household size, employed
members of household, car ownership and monthly income in that order appeared to be
the socio-economic variables explaining the trip rates by the various trip types. These
variables however seem less useful in describing the discretionary (social/recreation) trip
rates particularly with respect to households living in high density areas. This is due to
the variation in the socio-economic and cultural background of the people in the different
residential density areas. Ogunjumo (1986) observed that travel time rather than travel
distance influenced work trips at Ile-Ife, Nigeria and of the original nine independent
variables considered, six were found to be significant. These include workers per
household, household size, number of vehicles, travel time, income and age of household.
These variables seem to be consistent with what is found in literature.

Hillman (2003) in a comparative study of travel pattern in the UK and The Netherlands
observed that 86% of all trips are less than 16km in the UK and that public transport
accounts for 10% of all trip in this distance band. The study further revealed that about
one third of all trips are within the 6.5km and are suited for cycling while trips beyond
40km account for about 4% of all trips. In the Netherlands, cycle trips were observed to be
more than public transport for distance up to 15km. In general cycle, the ratio of car
journeys to cycle journeys is about 2 to 1, whereas in the UK, the ratio is about 32 to 1.
The study concluded that high level of cycle use in the Netherlands than in the UK has
much to do with its transport policy over the last 30 years, which has led to the bicycle
playing such a significant role in spite of a continuing rise car ownership.

Pucher and Renne (2003) noted that trip length varies between income groups in the US.
The study revealed that high income groups make long trips than the low income groups
with the average trip length of low income households being 2.4km shorter than for the

4
highest income households (9km vs 11.4km) for all modes. The differences in car trip
lengths are not very large, about 1.6km between the top and bottom income groups, while
the transit trips of low income households are about half that of most affluent households.
Variations however exist by type of transit. While there is a slight difference in metro trip
length among income groups, the rich make bus trips that are almost twice as long as
those made by poor households and make commuter rail trips that are four times longer.
The bicycling and walking trips of the rich are also found to be longer than that of the
poor household. This might be due to higher incidence of recreation, walking and cycling
for exercise or relaxation by the rich while the shorter trips by the low income households
may be due to the central location of their residential areas in a more compact mixed-use
neighbourhood (Pucher and Renne 2003).

It is very clear from various studies above that travel behaviour issues have gained
prominence in both developed and developing nations and that several factors affect the
travel demand of households. In particular, studies on trip length have shown that the
factors that affect urban trip length are beyond the households’ socio-economic
characteristics. However, the changing pattern in the lifestyle of urban residents requires
that data on households’ characteristics be collected and analysed on regular basis with a
view to identifying the effects of these changes on the travel behaviour in urban centres
especially in large cities like Lagos. It is against this background that subsequent sections
of this paper is devoted to understanding the context of Lagos and the socio-economic and
trip length characteristics of households.

The Study Area

Lagos Metropolis is located in Lagos State in the South-Western part of Nigeria (See
Figure1). It is the largest metropolitan area in Nigeria (Ayeni 1968, 1979) and consists of
eighteen local government areas. Lagos metropolis lies generally on low lands, with
about 17500 hectares of built-up area. According to the 2006 national population census,
Lagos State has a population of about 9million out of a national estimate of 140million.
Of this population, metropolitan Lagos, an area covering 37% of the land area of Lagos
State is home to 85% of the state population (Lagos State, 2004b). The projected average
population density of the built-up area of Lagos metropolis is about 20,000 people per
square kilometer.

The choice of Lagos metropolis for this study is based on the fact that it is the largest
urban conurbation in Nigeria. Lagos is at present the largest metropolis in Nigeria, with an
estimated population of about 9 million people (NPC, 2006). By 2025, Lagos will
become the third largest global city with an estimated population of 24 million people
(Lagos State, 2004a). Lagos has the highest vehicular density in Nigeria (222
vehicles/km) as against the national average of 11vehicles/km. These scenarios have great
implications for future travel demand in Lagos. From 2001, Lagos Metropolitan Area
Transport Authority (LAMATA) has carried out some studies on general travel pattern in
the city but the focus was on economics and usage of public transport rather than
households’ travel demand. This study therefore seeks to examine the households’ socio-
economic factors that influence their trip length.

5
 Fig.1: Map of Nigeria Showing Lagos State

Sokoto
Niger
Katsina Republic Yobe
Jigawa
Zamfara
Borno
Kebbi Kano

Republic of Kaduna Gombe

Benin
Bauchi
Niger

Adamawa
Plateau
FCT
Kwara
Nassarawa
Oyo
Taraba
Ekiti Kogi
Osun
Benue
Ogun
Ondo
Lagos Edo Enugu
Anambra Ebonyi
Gulf of Camer
Legend
Imo Abia CrossRiver Other States and the FCT
Guinea Delta

Source: Federal Bayelsa Rivers AkwaIbom oon Lagos State

K ilom etr es

Survey 0 20 40 60 80 100

2005

Research Approach

A total of 157 residential areas which have been stratified into three homogenous
residential densities (low, medium and high) based on 1991 population census were
identified in this study. From the 157 residential areas, thirty four of them (5 low, 11
medium and 18 high) were randomly selected and consist 49,130 (6,538 low, 24,811
medium and 17,781 high) density residential buildings (See Table 1). In the low density
residential areas every 15th house was systematically selected while every 40 th and every
20th house were systematically selected in the medium and high density areas respectively.
A total of 1785 duly completed questionnaires administered on household heads were
used for data analysis. Data collected through the questionnaires were focused on the
socio-economic characteristics and trip length of households. Both descriptive and
inferential statistics were used for data analysis as discussed in the subsequent sections.

Table 1: Distribution of Sample in Residential Density Areas

No of residential No of households
Residential Density No of Buildings
Areas Selected Selected
Low 5 6,538 376
Medium 11 24,811 574
High 18 17781 835
Total 34 49,130 1785

Socio-Economic Characteristics of Respondents

In all, data were collected on eight socio-economic attributes of respondents were
analysed in this section. These include gender, age, education, employment, income,
household size, car ownership and use of Global System of Mobile Telecommunication
(GSM).

Gender of Respondents
Gender analysis in this study as presented in Table 2 indicated that 62.0% of the
respondents are males and 38.0% are females. The relevance of gender in this work is due

6
to the prevalence of female household heads in a city like Lagos. The result obtained from
gender analysis in this study does not truly reflect the gender ratio in the study area; but
indicates that male respondents were more available for interview than the females during
the field survey.

Table 2: Gender of Respondents

Gender Frequency Percentage
Male 1105 61.9
Female 680 38.1
Total 1785 100.0
Source: Author’s Field Survey, 2009.

Age of Respondents
As summarized in Table 3, 70.8% of the respondents were not more than 40 years in age.
Respondents within the age bracket of 41 and 50 years were 18.8% of the sample size.
Similarly, 10.3% of the respondents were above 50 years. Further, the table indicated that
respondents who were in their productive years (20-60 years) constituted 94.0% of the
sample size. Respondents below 20 years and above 60 years represent 3.8% and 2.2%
respectively.
Table 3: Age of Respondents
Age Frequency Percentage
18-30Yrs 733 41.1
31-40Yrs 532 29.8
41-50Yrs 336 18.8
51-60Yrs 145 8.1
Above 60Yrs 39 2.2
Total 1785 100.0
Source: Author’s Field Survey, 2009.

Education Status
Analysis of education status of respondents in Table 4 reveals that majority (97.6%) of the
respondents has formal education while those without formal education were only 2.4%.
Further, 5.7% of the respondents have Bachelors degree and above. This implies that
majority of the respondents were relatively educated.

Table 4: Education Status of Respondents

Education Frequency Percentage
Informal 42 2.4%
Primary 57 3.2%
Secondary 324 18.2%
Diploma 315 17.6%
Bachelors 821 46.0%
Postgraduate 226 12.7%
Total 1785 100.0%
Source: Author’s Field Survey, 2009.

7
Employment Status
Table 5 shows the distribution of employment status of respondents and revealed that
70.6% were employed (44.7% formal and 25.9% informal), 7.3% were unemployed, and
18.3% were students while 3.9% were retired. This implied a relatively high level of
employment among the respondents and this may be due to the fact that many of them
were relatively educated. The availability of job opportunities in cities like Lagos may
also be responsible for high level of employment recorded among the respondents.
However, the relatively high proportion of students (18.3%) among the respondents could
be explained by the opportunities for part-time studies provided by institution of higher
learning in Lagos.

Table 5: Employment Status of Respondents

Employment Status Frequency Percentage
Formal 798 44.7
Informal 462 25.9
Retired 69 3.9
Unemployed 130 7.3
Student 326 18.3
Total 1785 100.0
Source: Author’s Field Survey, 2009.

Monthly Income
Within the context of prevailing income in Lagos metropolis, three income groups may be
identified. These are low income (less than N50, 000), medium income (N50, 000 – N100,
000) and high income (above N100, 000). Analysis of the estimated monthly income of
respondents as presented in Table 6 indicated that 65.9% of respondents were low income
earners, 21.2% were middle income earners while 12.8% were high income earners. In
general terms, the majority of the respondents are within the low-income group.

Table 6: Income Status of Respondents

Monthly Income Frequency %
Less than 50000 (low income) 1177 65.9%
50000-<100000 (middle income) 379 21.2%
>100000 (high income) 229 12.8%
Total 1785 100.0%
Source: Author’s Field Survey, 2009.

Household Size
Table 7 categorized the household size of respondents into three (small, medium and
large). Households with 4 members and below were termed “small sized” whereas the
medium sized household” contains 5 to 8 members. The “large sized household” group is
any household with more than 8 members. Table 6 indicated that respondents from small
sized households represented 35.4% of the total respondents. The respondents with
medium sized households represented 57.0% of the total households interviewed.
Respondents from large sized households were 7.7% of the total sample size. This implies
that medium sized household is prevalent in Lagos metropolis.

8
 Table 7: Household Size of Respondents
Household Size Frequency Percentage
1-2 134 7.5
3-4 497 27.8
5-6 724 40.6
7-8 292 16.4
Above 8 138 7.7
Total 1785 100.0
Source: Author’s Field Survey, 2009.

Car Ownership
Households’ car-ownership pattern as presented in Table 8 indicated that 19.3% of the
households have no car or any other type of vehicle. Respondents with just one vehicle
accounted for 33.6% of the respondents while owners of two vehicles represented 29.0%
of the sample size. Households with three vehicles and above represented 18.0% of the
respondents. In general low car ownership was found to be prevalent among households in
Lagos metropolis. This may be due to the prevalence of low income households in the
city.
Table 8: Car-Ownership of Respondents
Car -Ownership Frequency %
None 345 19.3%
1 599 33.6%
2 518 29.0%
3 215 12.0%
Above 3 108 6.0%
Total 1785 100.0%
Source: Author’s Field Survey, 2009.

Use of GSM
Respondents’ access to Global System of Mobile telecommunication (GSM) services is
presented in Table 9 and showed that 2.9% of them do not have personal access to GSM
services while 97.1% has personal access to at least one GSM service. The analysis
further revealed that 54.3% of the respondents have personal access to just one GSM
service, 34.2% has two, and 7.7% has 3 while just 1.0% has more than three services. This
suggests a very high accessibility of respondents to telecommunication.

Table 9: Use of GSM by Respondents

Use of GSM Frequency %
None 52 2.9%
1 969 54.3%
2 610 34.2%
3 137 7.7%
>3 17 1.0%
Total 1785 100.0%
Source: Author’s Field Survey, 2009.

9
Trip Length of Respondents
For the purpose of analysis, trip length in this study is defined as the distance covered for
a journey made within the city to a destination for a particular purpose notwithstanding
the number of modal interchange. As shown in Table 10, 50.4% of the respondents made
trips lesser than 5km per day, 18.3% made trips of between 5 and 10km in a day while
15.7% made trips between 11 and 20km per day. The remaining 15.6% made trips above
20km per day. The mean trip length from this analysis is 12.0km per person per day and
this suggests a fairly short trip length among residents although about a little above one-
third of respondents travel above 10km per day. This may be due to the fact that people
tend to live very close to their place of work and activity centres in order to minimize cost
of travel within the city since it has been observed earlier that majority of respondents in
Lagos are low income earners. This view is corroborated by evidences from the 2001
NHTS in the US. Further, those who made daily trips above 10km are likely to belong to
the high income earners who could afford to live far away from their place of work.
Table 10: Trip Length of Respondents
Trip Length Frequency Percentage
<5km 900 50.4
5<10km 326 18.3
10<20km 280 15.7
20km+ 279 15.6
Total 1785 100.0
Source: Author’s Field Survey, 2009

Household trip length and their determinants vary in many cities of the world due to
differences in their socio-economic and cultural background, government policies and
level of transport development. Consequently, this paper attempts to identify the major
socio-economic attributes of households that provide explanation for their daily trip
length. This was achieved through correlation and multiple regression analysis the result
of which is presented subsequently.

Bivariate and Multivariate Analysis of Factors Influencing Trip length of

Households in Lagos

To determine the major households’ socio-economic factors that influence trip length,
eight socio-economic variables were subjected to correlation and stepwise multiple
regression analysis. The dependent variable was trip length and the socio-economic
variables were the independent variables. These are gender, age, education, employment,
income, use of GSM, car ownership and household size. The results of the correlation
analysis as presented in Table 11 show that six of the socio-economic variables are
significantly correlated with trip length at p≤0.05 and these are car ownership (r = -.166),
gender (r = -.166),
Table 11: Correlation Coefficients of Trip Length and Respondents’ Socio-economic
Variable
Household Car Gender Age Education Employment Monthly Use of
Size Ownership Income GSM

10
 .068 -.166** -.154** .074 .360** -.114** . 162* .148**
The regression coefficients of trip length as presented in Table 12 reveals that all the
independent variables are significantly related to trip length at p<0.05 with education
having the highest beta value of .675 and household size with the lowest beta value
of .075. The model summary in table 11 indicates that the R² for the combined
independent variables is 0.799 (79.9%). Education provided the highest contribution of
71.2% to the value of R² which increased to 74.1% when income was included in the
model. The value of R² further increased to 76.6% and 77.8% when car-ownership and
gender were respectively included in the model. Further increase in the value of R² was
not too significant with the inclusion of other independent variables (household size,
employment, gender and age) in the regression model. This analysis therefore shows that
education, income, car- ownership and gender are the major socio-economic
characteristics of households that influence trip length in Lagos metropolis.

Table 11: Model Summary for Trip length Regression

Std. Error of the
Model R R Square
Estimate
Education .844(a) .712 .394
Car-ownership .861(b) .741 .365
Income .875(c) .766 .340
Gender .882(d) .778 .326
Use of GSM .889(e) .790 .313
Employment .891(f) .794 .307
Household Size .892(g) .795 .304
Age .894(h) .799 .302

Conclusion
The socio-economic and trip length characteristics of households in Lagos Metropolis
have been examined in this study. The major socio-economic variables that influence trip
length have also been identified and education, car ownership, income, gender and use of
GSM have been recognized as major households’ variables that affect trip length.
Consequently, this study concludes that these variables have significant relationship with
households’ trip length and should be considered in travel demand forecast in Lagos and
for other similar metropolis in developing nations. For future studies, it may be necessary
to disaggregate households’ data on the basis of residential density to identify the
variations that may exist in the identified factors between and within the three
homogenous residential densities used for data collection in this study.

11
References
Akinbami, F. K. and Fadare, S.O. (1997): Strategies for Sustainable Urban and Transport
Development in Nigeria, Transport Policy, Vol 4, No 4, Elsevier Science Ltd.
Great Britain. Pp 237-245
Astrop, A. (1996) “Urban Travel Behaviour in Low Income Households”. Proceedings
from Second National Conference on Women’s Travel Issues. Baltimore, Oct.
1996.
Auclair, C. (1999). “Measures of travel time in Cities” Urban Age Spring. 1999. World
Bank.
Ayeni, B. (1979). Concepts and Techniques in Urban Analysis. Croom Helm Ltd.
London.
Ayeni, M. B. (1968): The Built-up Area of Ikere: Past, Present and Future, A Simulation
Study. BSc Dissertation, Department of Geography, University of Ibadan.
Brun, J and Fagnani, J. (1994) Lifestyles and Location Choices – Trade Offs and
Compromises: A Case Study of Middle-Class Couples Living in the Ile-de
France Region, Urban Studies 31 (6). Pp 921-934
Clifton K.J. and C.L. Handy (2001) “Qualitative methods in Travel Behaviour Research”
Paper presented at the International Conference on Transport Survey Quality
and Innovation, Kinger National Park, South Africa.
Fadare, O. (1989) “Analysis of factors affecting Household trip length in the Residential
Area of Ibadan”. Ife Research Publications in Geography, Dept of Geography,
O.A.U., Ile – Ife, 3.
Fadare, S.O. (1987) “Intra-Urban travel characteristics: A study of Socio-economic
attributes in Ibadan”. Ph.D Thesis submitted to University of Sheffield,
England.
Fujiwara, A., Soehodho, S., Hyodo, T., and Montalbo, C. (2005): “Urban travel behaviour
characteristic of 13 cities based on Household interview survey data”. Journal
of the East Asia Society for Transportation studies. Vol. 6 pp 23-38.
Helling, A. (1996): “The Effects of Residential Accessibility to Employment on Men and
Women’s Travel”. In Women Travel Issues, Second National Conference,
Baltimore, MD.
Hillman, M. (2003): The Future of Public Transport: The Dangers of Viewing Policy
through Rose-tinted Spectacles. In World Transport: Policy and Practice. John
Whitelegg and Garry Haq (Eds). Earthscan Publication Ltd., Pp 139-145.
Lagos State (2004a): Perspectives on Lagos Economy Lagos State Ministry of Economic
Planning and Budget, Ikeja, Lagos.
Lagos State 2004b): State of Lagos Mega city and Other Nigerian Cities. Lagos State
Ministry of Economic Planning and Budget, Ikeja, Lagos.
Mokhtarian, P. I. and Meenakshisundaram, R. (2002): “The patterns of
Telecommunicating engagement and frequency: A Cluster Analysis of
Telecenter Users”. Prometheus, 20 (1). Pp 21-37
Morikawa, T. Yamoto, T., Dissanayake, D., Sanko, N., Kurauchi, S., Maesoba, H.,
Ohashi, S., Tiglao, N., Rubite, C. and Rivera, M. (2001): Travel Behaviour
analysis and its implication to Urban Transport Planning for Asian cities. Case
studies of Bangkok, Kuala Lumpur, Manila and Nagoya. International

12
 Cooperative Research Activity (ICRA) Report, Graduate School of
Environmental Studies, Nagoya University, Japan.

Ogunjumo, A. (1986): “The pattern of Trip current at Ile-Ife” Journal of the Institute of
Town Planners. Vols. VI & VII, Pp 99-113
Olanrewaju, S.A., Fadare, S.O., Akinlo, A.F. and Alawode, A.A. (1995): Urban Passenger
Transport in Lagos, (Nigeria). Research Monograph, IDRC Urban Transport
Project, Department of Economics, Obafemi Awolowo University Press, Ile-
Ife.
Olaore, G.O. (1990) “Transport and Land use policies and practices in Nigeria” in
Transport and National Development in Nigeria. Adegbeyeni, O. and O.J. Rapu
(Eds) Federal Ministry of Transport, Lagos.
Pucher J. and Renne J. (2003) “Socio Economics of Urban Travel: Evidence from the
2001 National Household Travel Survey”. Transportation Quarterly, Vol. 57.
No. 3. Pp 49-77
Srinivasan, S. (2005): “Influence of Residential Location in travel Behaviour of Women
in Chennai, India” Division of Engineering and Applied Sciences, Harvard
University.
Srinivasan S. and P. Rogers (2005):” Travel Behaviour of Low Income Residents:
Studying two locations in the city of Chennai” Journal of Transport
Geography. Vol. 13 Pp 265-274
Zegras, C. (1997) “Urban Transportation” in World Resources: The Urban Environment
Habitat.

13