Geographies of poverty in Kano State: The role of GIS in identifying and mapping multidimensionally deprived households

Abstract

The broadest approach to wellbeing and
multidimensional approach to poverty was the one
articulated by Amartya Sen (1987). Since then, efforts
being intensified to find out a suitable measure that
satisfy some basic properties expected with each poverty
measure. Alkire and Foster (2007) proposed a family of
measures which provides an alternative measurement of
multidimensional poverty. The method allows the
identification of households as poor or otherwise using
two steps analysis. To critically examine
multidimensional poverty in three senatorial districts of
Kano State, this paper adopts this method. Poverty
identification and measurement is based on 10
dimensions (education, materials of housing, livelihood,
sanitation, assets and durables, food security, health,
cooking fuel, electrification, source of drinking water).
The overall results shows that at different levels of k,
Kano North is the poorest of the three. At k=4, 98% of
households in Kano North falls below poverty line while
only 97.1% in the Kano South and 86.9% in Kano
Central falls below the threshold. It is observed that
Kano Central contributes 27.6% to the overall poverty,
whereas, South and North contributes 34.3% and 38.1%
respectively. This paper will provide policy makers with
information on poverty situation in the three districts for
urgent intervention.

Full text

Ph ton
200
The journal of Geography and Geology. Photon 119 (2015) xxx-xxx
https://sites.google.com/site/photonfoundationorganization/home/the-journal-of-geography-and-geology
Research Article. ISJN: 4396-3839: Impact Index: 5.56
The journal of Geography and Geology Ph ton
Geographies of poverty in Kano State: The role of GIS in identifying and
mapping multidimensionally deprived households
M.M.
Sanusi
*
, D.M.
Denis
Department of Soil, Water, Land Engineering & Management, Remote Sensing & GIS unit, Sam Higginbottom Institute of
Agriculture, Technology & Science (SHIATS), Allahabad, UP, India
Article history:
Received: 29 September, 2015
Accepted: 01 October, 2015
Available online: xx xxx, xxxx
Keywords:
Multidimensional poverty, geographic targeting, poverty
mapping, Kano Central, Kano North, Kano South
Abbreviations:
AFM- Alkire and Foster Family of Measures, MPI-
Multidimensional Poverty Index, GIS- Geographic Information
System
Corresponding Author:
Sanusi M. M.*
Research Scholar
Email: comrademunji ( at ) yahoo ( dot ) com
Denis D.M.
Professor
Abstract
The broadest approach to wellbeing and
multidimensional approach to poverty was the one
articulated by Amartya Sen (1987). Since then, efforts
being intensified to find out a suitable measure that
satisfy some basic properties expected with each poverty
measure. Alkire and Foster (2007) proposed a family of
measures which provides an alternative measurement of
multidimensional poverty. The method allows the
identification of households as poor or otherwise using
two steps analysis. To critically examine
multidimensional poverty in three senatorial districts of
Kano State, this paper adopts this method. Poverty
identification and measurement is based on 10
dimensions (education, materials of housing, livelihood,
sanitation, assets and durables, food security, health,
cooking fuel, electrification, source of drinking water).
The overall results shows that at different levels of k,
Kano North is the poorest of the three. At k=4, 98% of
households in Kano North falls below poverty line while
only 97.1% in the Kano South and 86.9% in Kano
Central falls below the threshold. It is observed that
Kano Central contributes 27.6% to the overall poverty,
whereas, South and North contributes 34.3% and 38.1%
respectively. This paper will provide policy makers with
information on poverty situation in the three districts for
urgent intervention.
Citation:
Sanusi M.M., Denis D.M., 2015. Geographies of poverty in
Kano State: The role of GIS in identifying and mapping
multidimensionally deprived households. The journal of
Geography and Geology. Photon 119, xxx-xxx.
All Rights Reserved with Photon.
Photon Ignitor: ISJN43963839D8185xxxxxxxx
1. Introduction
In recent years, there was shift in paradigm upon
which poverty is seen, define and analyse. This
shift forced many development organizations, such
as World Bank, UNDP, to mention few, to analyses
poverty in such a way that all aspects of human
deprivation like self-determined lifestyles, choice,
assets, capabilities, social inclusion, inequality,
human rights, entitlement, vulnerability,
empowerment and subjective wellbeing are
considered (Gonner, et al., n.d; Sen, 2004). Yet, the
gap between multidimensional and unidimensional
indicators of poverty is too disturbing, despite the
association amongst economic growth, income,
subjective wellbeing and non-material aspects of
poverty (World Bank, 2001).
As per international poverty line of less than $1.25
a day, around 1.8 billion people in 1990, 1.4 billion
people in 2005 and 920 million people in 2009 all
over the world were living below the poverty line,
with varying impact across the regions and
countries (UN, Millennium Development Goal’s
Report, 2010). Based on 2010 population data,
Sub-Saharan Africa (SSA) home 91 percent of the
region’s population (UNDESA, 2013). In 2014, a
total of 462 million people, i.e. 58.9 percent of all
people living in these countries in SSA sub-region
are living in multidimensional poverty. Nearly 30
percent of total Multidimensional Poverty Index
(MPI) poor of the world (out of 108 countries
analysed) live in SSA (Alkire, Conconi and Seth,
2014). Of these 462 million people, 36.3 percent

Ph ton
201
lives in West Africa, 36.0 percent in East Africa,
14.5 percent in Central Africa and 13.3 percent in
Southern Africa. With 85.8 percent living in rural
areas (Alkire and Housseini, 2014). In West Africa
for example, Nigeria alone is home to 15.4 percent
of total number of SSA MPI poor. That is to say
71.2 million people in Nigeria are identified as
poor based on MPI. While Niger republic is the
country with the highest percentage of MPI poor
people (ibid).
Though the use of Geo-information (GI) in
addressing poverty problems such as identifying
poverty, its spatial distribution and reduction is still
in its infancy. Yet, it plays a vital role in that
wherever it is being employed in poverty related
activities, marvellous results are being experienced
(Sachs, 2005). Applying GIS in poverty analysis,
mapping and reduction allow the utilisation of a
multitude of environmental and community
characteristics such as rainfall amount and timing,
evapotranspiration, land cover and land use, and
geographically derived indicators such as distance
and physical accessibility measures (Akinyemi,
2007). Moreover the overlay function in GIS
enables the analysis of various poverty indicators in
order to understand the spatial association existing
between these indicators. Explanatory and
dependent variables for use in multivariate analysis
as determinants of poverty are spatially generated,
including natural capital and infrastructure, and
access to public services, product and labour
markets (Bigman and Deichmann, 2000).
With rising public shortfalls and dwindling public
resources geographical targeting may be a viable
way to allocate resources for poverty alleviation in
developing countries. Efficiency can increased and
leakage to non-poor reduced substantially by
targeting increasingly smaller areas (Bigman &
Fofack, 2000)
This paper briefly reviews the literature on
multidimensional measures of poverty and their
application. It adopts the Alkire and Foster family
of measures – which is an alternative methodology
– to estimate multidimensional poverty in the study
area and to identify poor households and their
aggregation. The measures are robust and
specifically design for cardinal/ordinal data. It first
begins by selecting dimensions, indicators for each
dimension and their cut-offs and then, observation
is made on whether household is deprived in each
dimension. In the next step, it aggregates the
number of dimensions a household is deprived. The
final step is deciding the aggregate cut-off; that is
the minimum number of dimensions that a
household need to be declared as multidimensional
poor. Household is identified as multidimensional
poor if he/she falls below the threshold. This
method is flexible and also sensitive to change in
weight.
1.1 Multidimensional poverty
Studies of the problems of poor people and
communities, and of the impediments and
opportunities to improve their situation, have led to
an understanding of poverty as a complex set of
deprivationsthat encompasses not only the material
deprivation (measured by an appropriate concept of
income or consumption). This notion changes the
way poverty seen in the eyes of many development
scholars. The concept of poverty is than seen as a
human condition that reflects failures in many
dimensions of human life – hunger;
unemployment; homelessness; illness and health
care; powerlessness, voicelessness and
victimisation; social injustice; low achievements in
education; and other forms of vulnerability and
exposure to risk; all add up to an assault on human
dignity (Fukuda-Parr, 2006; Edward, 2006;
Kakwani, 2006; World Bank, 2001),
Though multidimensional approach to poverty
gives more insight on so many aspects of human
deprivation, yet, it raises question on how to
measure overall poverty and how to compare
achievements in different dimensions, in that, each
and every dimension has its unique characteristic
(World Bank, 2001). One dimension might move in
a different direction from another. For example,
health could improve while income worsens. Or an
individual might be “income poor” but not “health
poor.” Or one country might show greater
improvement in health than in education, while
another shows the converse (Alkire, 2009). In order
to be able to compare across countries, households,
or individuals and over time, the simplest way is to
assign weight to the different dimensions of
deprivation base on their importance and then
measure poverty intensity accordingly (World
Bank, 2001).
1.2 Geographic targeting
There are pockets of poverty in both urban and
rural areas. The reason why some areas remains as
pockets of poverty while some remains to be
islands of prosperity as asserts by (Bigman &
Fofack, 2000; Davis, 2002) includes differences in
agro-climatic conditions, “feminisation” of
poverty, the low quality of public services
(particularly education and health), the poor
condition of rural infrastructure (which retards
local investment and affects trade), the low level of
social capital in poor communities which slows
diffusion and adoption of new technologies
(especially farm innovation), endowments of
natural resources, or geographic conditions –
particularly distance to a sea outlet and to centres
of commerce – and biases in government policies.

Ph ton
202
Certainly, in many developing countries the
differences in living standards between regions are
often larger than the differences within regions.
Disparities in living standards may persevere
because of obstacles to internal migration, which in
some countries are the results of thoughtful
government policies and in all countries are the
result of economic, demographic, and cultural
factors (Bigman & Fofack, 2000).
Geographic targeting is furthermost operative when
geographic units are pretty small, such as village,
ward, or district (Bigman & Fofack, 2000). But the
conclusion of (Epprecht, Minot, Dewina, Messerli,
& Heinimann, 2008) is that small area estimation
(which make use of census and household surveys)
is not suitable for poverty mapping that need to be
updated annually in that census is conducted once
in every decade. As such an alternative method
need to be use. Bigman, Dercon, Guillaume, &
Lambotte, (2000) concludes that geographical
measurement of poverty and targeting at province
or region level may be an effective approach for
reaching poor in countries where there are
considerable disparities in living conditions
between geographic areas and also where
administering poverty programmes is somewhat
forthright for the reason that local administration is
already in place.
1.3 Poverty mapping
The concept of mapping poverty encompasses
measuring the incidence of poverty, food security
and other forms of human deprivations by some
predetermined area. Poverty maps are important
tools that provides information on the spatial
distribution of poverty within a country (Davis,
2002). They are used to affect different kinds of
decisions, ranging from poverty alleviation
programmes to emergency response and food aid
(Hentschel, Lanjouw, Lanjouw, & Poggi, 2000;
Petrucci, Salvati, & Seghieri, 2004; Rogers,
Emwanu, & Robinson, 2006; Ballini, Betti,
Carrette, & Neri, 2008. Yet, it is evident that the
use of poverty maps alone does not provide an
estimate of the causal linkage between poverty and
the variables influencing it; such maps furnish only
“visual” advice. In attempt to find out the possible
empirical relationship between environmental
factors; socio-economic indicators and poverty,
researchers make use of statistical methods such as
econometric model, that combines census and
survey data as applied in many countries – like
South Africa; Ecuador; Vietnam; Mexico to name
few – in order to get better result (Hentschel, et al,
2000, Davis, 2002).
Poverty maps can be made at different levels of
aggregation ranging from global, national, regional
or sub-regional (Davis, 2002). The type of poverty
map to produce depends largely on the conceptual
approach used to define poverty and indicators of
poverty used as they have significant implication
on the sort of data to be compiled (Henninger,
1998). Considering issues relating to degree of
disaggregation at which poverty maps can be
produced (Henninger, 1998; Hentschel, et al, 2000;
Davis, 2002); emphasise that the following factors
need to be considered. Firstly, the precise purpose
of the map (whether it is aimed at identifying poor
at household level or community level). Secondly,
whether we can assume that the parameter
estimates from a regression model estimated, say,
at the regional level, apply at sub-regional levels.
Thirdly, availability of other sources of information
(possibly local sources), on poverty level of
individuals. Finally, believe that other methods of
local targeting, such as self-targeting, will become
more important and effective at certain levels of
disaggregation. Akinyemi, Sester, & Balogun,
(2012); Baschieri, Falkingham, Hornby, & Hutton,
(2005); Elbers, Lanjouw, & Lanjouw, 2003);
Hentschel, et at., (1998), point out that
development of detailed poverty maps in many
settings is hampered due to data constraints.
There are various methods used for spatial location
of poor. Some of these methods are under
continuing development while some already reach
their maturity stage and in use for different
purposes. The methods includes small area
estimation which applies parameters from a
predictive model to identical variables in census or
auxiliary database, assuming that the relationship
defined by the model holds for the larger
population as well as for the original sample. This
technique was first used in U.S.A. and later
adopted in many developing countries (Davis, 2002
& 2003). The method is applied to household level
or community level data depending on what is
availability at the moment. Other methods are;
multivariate weighted basic need.They are
statistical technique – principal components, factor
analysis, ordinary least squares, cluster analysis
and multiple correspondence analysis – with
weighting scheme and others with no weighting
scheme (Davis, 2002 & 2003). Detail on the
remaining methods can be found in (Henninger,
1998; Davis, 2002).
1.4Case studies
In literature on poverty maps it was found that
small area estimation method applied in Lao by
Epprecht, et al., (2008) to observe poverty rate;
Cuong, Truong, & Weide, (2010) in Vietnam to
observe poverty changes over the rural and urban
regions between the period of 1999 and 2006,
Okwi, et al. (2007) in Kenya to explore the effect
of geographic factors on poverty (see Henninger,
2008; Davis, 2002 & 2003)

Ph ton
203
With multivariate methods, it was found that
Mexican Government used principle component
analysis to create marginality index for policy
planning purpose and as part of her targeting
mechanism in PROGRESA anti-poverty
programme (Henninger, 1998; Davis, 2002 &
2003). FAO in conjunction with Columbia
University applied same method in Costa Rica to
examine the relationship between poverty and
deforestation over four decades (Henninger, 1998;
Davis, 2002 & 2003). Erenstein, Hellin, &
Chandna, (2010) uses district level household’s
livelihood assets – natural, physical, human, social
and financial – to construct poverty map for the
Indo-Gangetic Plains based on principal
components. CIAT envisions in Honduras mapped
poverty using manifold measures of wellbeing
based on household expenditures (and a designated
poverty line) and amalgam of indexes measuring
unsatisfied basic services and human needs with
the best available data (Henninger, 1998). Factor
analysis with rotation was applied to 1996
population census data in South Africa by
Alderman, (detail can be found in Hirschowitz, et
al, (2000)).
Centro Internacional de la Papa (CIP) in Bolivia
and Peru uses remotely sensed data to produced
poverty maps to analyse the relationship between
rural poverty and environmental degradation.
International Centre for Agricultural Research in
the Dry Areas (ICARDA) measures the severity
and distribution of rural poverty, identified natural
resource constraints, and relate these indicators to
the production value of ICARDA’s mandate
commodities, while the GIS staff integrates
relevant national data sets related to poverty,
human wellbeing, natural resources, agricultural
production, and labour force, and upon request,
provides thematic maps – of higher resolution –
based on these data to ICARDA’s staff for further
action
2. Materials and Methods
2.1 The study area and dataset
The study area (Kano State) lies betweenlatitudes
10°30′00″ to 12
o
45′00″ north of the equator and
longitudes 7°10′00″ to 9
o
20′00″ east of the
Greenwich Meridian. The area borders with four
states, namely: Jigawa to the north, northeast and
east, Bauchi to the southeast, Kaduna to the
southwest and Katsina to the west and northwest. It
has an approximate land area of 21,276.87 km
2
.
The area is divided into three senatorial districts
(Kano Central, Kano North and Kano South), that
further subdivided into 44 local government areas
with 484 wards (Figure 1& 2).
Kano State has a total population of 9,401,288 (out
of which 4,947,952 are males while the remaining
4,453,336 are females) (NPC, 2006).The dominant
socio-economic activities are mainly primary
activities such as farming, local crafts, trading and
livestock rearing among others. Majority of the
farming and livestock breeding populations
practice small scale subsistence farming with little
surplus for market (Momale, 2010).
Figure 1: The study area showing districts
Figure 2: The study area showing LGAs
This research used data from 2006 Nigerian
Population and Housing Census conducted by
Nation Population Commission (NPC); and
National Core Welfare Indicator Questionnaire
Survey (NG-CWIQ) of 2006. The survey conducts
by National Bureau of Statistics and is national as it
covers all the states of the federation and the
federal capital territory (Abuja), and the entire 774
local government areas in the country. The survey
also covers 77, 400 households in both rural and
urban areas. From each local government, total
number of 100 households made the sample. This
survey is chosen in line with the guideline for

Ph ton
204
mapping poverty which states that the survey and
the census should be of the same year or nearly
same period of time depending on the economic
changes experience in the country regarding
inflation and others.
The analysis of poverty was carried using
Distributive Analysis Stata Package (DASP version
2.3), a Stata extension, developed in 2013 by
Abdelkrim Araar and Jean-Yves Duclos to assist
researchers and policy analysts interested in
conducting distributive analysis with Stata (Araar
&Duclos, 2013). The results are presented in the
form of tables, charts and graphs where
appropriate. Finally, ArcGIS 10.3.1 is utilised for
map production.
Figure 3: Flow diagram of method
2.2Setting deprivation cut-off
The base information in multidimensional poverty
measurement is typically represented by × 
dimensional achievement matrix , where 

is the
achievement of person  in dimension . It is
assumed that achievements can be represented by
non-negative real numbers (i.e. 

∈ℝ
+
) and that
higher achievements are preferred to lower ones.
Based on the deprivation profile, each person is
assigned a deprivation score that reflects the
breadth of each person’s deprivations across all
dimensions. The deprivation score of each person
is the sum of her weighted deprivations. Formally,
the deprivation score is given by:
Where: = censored deprivation score; =
deprivation values attached to dimension ; =
censored deprivation matrix of person i in
dimension j (such that whenever
and otherwise); = weighted censored
deprivation matrix of person i in dimension j (such
that if and if )
The score increases as the number of deprivations a
person experiences increases, and reaches its
maximum when the person is deprived in all
dimensions. A person who is not deprived in any
dimension has a deprivation score equal to 0.
2.3Setting poverty cut-off
In addition to deprivation cut-off

, the AF
methodology uses a second cut-off or threshold to
identify the multidimensionally poor. This is called
the poverty cut-offand is denoted by . The poverty
cut-off is the minimum deprivation score a person
needs to exhibit in order to be identified as poor
(Alkire, et. al 2015, chp.5). This poverty cut-off is
implemented using an identification function ,
which depends upon each person’s achievement
vector 

; the deprivation cut-off vector , the
weight vector , and the poverty cut-off . If the
person is poor, the identification function takes on
a value of 1; if the person is not poor, the
identification function has a value of 0.
Notationally, the identification function is defined
as (

; ) = 1 if 

≥  and (

; ) = 0
otherwise.
2.4 Alkire and Foster (2007) family of measures
2.4.1 Multidimensional Headcount Ration (H)
Multidimensional Headcount Ratio shows the
proportion of households that are poor. With H as
headcount ratio; (X;z,w,k) asproportion of
population that is poor (it is defined as H = q/n,
where q is the number of persons identified as poor
using the dual-cut-off approach); as
identification function which takes the value of 1 if
the indicated condition of is true for the
person, and 0 when otherwise, denoting censored
deprivation score (or more appropriately weighted
deprivation score), k as dimension cut-off and as
population size, headcount ratio is estimated using
the following notation as:
As pointed by (Alkire, et. al., 2015: chp.5), the
term ‘adjusted’ in their multidimensional poverty
measures refers to the fact that all measures
incorporate the intensity of multiple deprivation.
The Adjusted Headcount Ratio (
0
) reflects the
proportion of weighted deprivations the poor
experience in a society out of the total number of
deprivations this society could experience if all
people were poor and were deprived in all
dimensions.
2.4.2 Adjusted Headcount Ratio (M
0
)
The adjusted headcount ratio is the total number of

Ph ton
205
deprivations experienced by the poor divided by
maximum possible number of deprivations
experienced by all the people (Alkire and Foster,
2008). It is the mean (µ) of an appropriate vector
built from the original data and censored using the
poverty line , the Adjusted Headcount Ratio
is the mean of the censored deprivation score
vector obtained using:
Or more preferably:
Where: = achievement of person i in dimension
j; = number of dimensions; All others same as
above.
2.4.3Average deprivation score/Average poverty
(A)
This is also referred to as poverty intensity. It
incorporates the information on the depth of
poverty. With censored deprivation score
representing the share of possible deprivations
experienced by a poor person i, the average
deprivation score across the poor is given by:
. Like the poverty gap
information in income poverty, this partial index
conveys relevant information about
multidimensional poverty, in that persons who
experience simultaneous deprivations in a higher
fraction of dimensions have a higher intensity of
poverty and are poorer than others having a lower
intensity (Alkire, et. al 2015, chp.5).
Thus, M
0
is given by
2.4.4 Subgroup Decomposition
Overall poverty is a population-share weighted sum
of subgroup poverty levels and this proved that
poverty can be analysed by regions, by ethnic
groups, and by other subgroups defined in a variety
of ways (Alkire, et. al, 2015, chp.5). Using
population subgroup decomposability property of
Alkire and Foster measures, it is possible to
monitor and understand the subgroup M
0
levels and
compare them with the aggregate M
0
. In addition,
the subgroup decomposability provide us with
information on the contribution of each subgroup to
overall poverty, in that, the contribution of
subgroup to overall poverty depends both on the
level of poverty in subgroup and on the population
share of the subgroup. As such, whenever the
contribution of a region or some other group
greatly exceeds its population share, this suggests
that there is a seriously unequal distribution of
poverty in the country, with some regions or groups
bearing a disproportionate share of poverty (Alkire,
et. al, 2015, chp.5). Having population share and
the achievement matrix of subgroup ℓ denoted
by and , respectively. The overall
subgroup M
0
is expressed as
The contribution of each subgroup to overall
poverty is computed using the additive form of
equation --, to do so, the contribution of subgroup ℓ
to overall population is denoted by formulated
as:
(7)
2.4 Justification of research
Poverty (urban and rural) poses the problems of
housing and shelter, water, sanitation, health,
education, social security and livelihoods along
with special needs of vulnerable groups like
women, children and aged people. Poverty
mapping is increasingly becoming a useful tool for
targeting these programmes, which usually consist
of cash transfers to the poor as well as providing
improved access to education and health services
(Skoufias et al., 2001; Henninger and Snel, 2002).
Advances in geospatial analysis and availability of
spatial data make feasible the mapping of a
combination of agro-ecological and socio-
economic variables, such as poverty incidence
(Byerlee, 2000, in: Bellon, Hodson, Bergvinson,
Beck, Martinez-Romero, & Montoya, 2005).
This research will provides donor agencies and
policy makers in Nigeria at large and Kano State in
particular with information about poverty level,
geographical location of the poor, the kind of assets
owned by the poor, available assets within different
communities, economic and infrastructural services
in the study area and their level of accessibility. It
will also provide information on the need of the
poor and poor communities and channels through
which government and others concern can target
resources to the poor and poor areas directly in
order to avoid spill over and leakage of policies;
interventions and programmes
3. Results and Discussion
3.1 Decomposition of Alkire and Foster (2007)
MDI by Senatorial district
Figure 4 gives the percentages of households
deprived in various dimensions. Over 80 percent of
the households in Kano North and South districts
and 63.7 percent in Kano Central are deprived in

Ph ton
206
housing. More than 43 percent of households in
Kano Central, 59.6 percent of those in northern
district and 63.6 percent in southern district are
deprived in drinking water. Households from all
the three districts are seriously deprived in
sanitation as no district with less than 90 percent of
deprived household on this dimension. As such, we
can say that most of the households are living in an
unhygienic environment. This will have
repercussion on health condition of the household
members as it can lead to outbreak of different
diseases such as cholera.
Table 1 presents the estimates of multidimensional
poverty (H
0
) at district level and the relative
contribution of each region to Alkire and Foster
(2007) multidimensional poverty indices. Other
estimates of M
0
families of Alkire and Foster
(Adjusted headcount ratio and Average poverty
gap) are presented on maps. The southern district
records higher poverty incidence (99.9 percent)
when compared to North (99.7 percent) and Central
(97.8 percent) when the threshold levelled at k=2.
With k=3, poverty incidence is 99.7 percent in
Kano South, amounting to 99 percent in the North
and 93.1 percent in Kano Central. District level
poverty estimates also follows the same trend of
decreasing with increasing k-value. This tally with
what Alkire and Seth (2009); Batana (2008);
Naveed and Islam (2010) (to mention few) found in
their various studies.
Figure 4: Poverty incidence in the 3 senatorial districts
The reasons for low poverty incidence in Kano
Central is attributed to the fact that the centre
serves as the hub and all development initiatives
begins there. Households in the Kano Centralhave
access to education, health services, electricity and
other social and economic services such as
security. They also have chance to engage in so
many activities that will improve their resilience
and reduce their vulnerability to poverty.
Table 1: incidence of poverty and relative contribution of each district to MDI
Sector (Senatorial District)
Central North South
Population Population Population
1,478 ( 33.82) 1,297 (29.68 ) 1,595 (36.50) 4370 (100)
Aggregate
cut-off point
(k) H % Contribution H % Contribution H % Contribution Total %
Contribution
k=2 0.978 0.284 0.997 0.340 0.999 0.377 1.00
k=3 0.931 0.281 0.990 0.341 0.997 0.378 1.00
k=4 0.869 0.276 0.971 0.343 0.980 0.381 1.00
k=5 0.758 0.264 0.931 0.350 0.930 0.386 1.00
k=6 0.604 0.250 0.824 0.361 0.797 0.389 1.00
k=7 0.358 0.219 0.600 0.381 0.565 0.400 1.00
k=8 0.139 0.184 0.311 0.419 0.263 0.397 1.00
k=9 0.026 0.132 0.096 0.494 0.065 0.375 1.00
Each of the three (3) districts contributes
differently to overall poverty. It is evident from
Table 1 that at k=2, Kano South contributes highest
quota (37.7 percent) to overall multidimensional
deprivation followed by Kano North (with 34
percent). The lowest contributing district is Kano
Central (28.4 percent). The relative contribution
changes with k=3 in that Kano South contributes
37.8 percent, whereas, Kano North and Central
contributes 34.1 & 28.1 percent. With k=4, North
contributes 34.3 percent, South 38.1 percent and
Central 27.6 percent.
Critical observation of table 1 provide us with
some salient issues: (a) that households in Kano
Central are better-up then households in Kano
South and Kano North in that at all levels of k,
incidence of poverty and its severity is lower in that
district when compared to other two (2) remaining
districts; (b) at k=2, incidence of poverty in Kano
South is higher than that of Kano North. This
means that the number of deprived households
using two dimensions as deprivation cut-off is large
in the South than in the north; and (c) the most
deprived or the extreme poor household (those
deprived on eight or nine dimensions) in the study

Ph ton
207
area are residing in the North. Hence, we can say
that Kano Central is better-up than Kano South and
North. And also Kano North is the poorest of the
three (3).
3.1.1 Adjusted Headcount Ratio
Figure 5 to 12 presents the Adjusted Headcount
Ratio while figure 13 to 20 presents that of
Average Poverty Gap for the three senatorial
districts. Figure 5 shows that at k=2, the highest
incidence of poverty from the Adjusted Headcount
Ratio is found to be in Kano North where 67.4
percent of the household are below poverty line,
followed by Kano South with 66 percent and Kano
Central with 0.564 percent. When k=3, the ratio
drops for both districts. Estimates on figure 6
shows that Kano North records 0.672 percent,
while Kano South and Kano Central records 0.658
and 0.555 percent respectively.
Figure 5: Adjusted Headcount Ratio for the 3 Senatorial
districts (k=2)
Figure 6: Adjusted Headcount Ratio for the 3 Senatorial
districts (k=2)
Figure 7: Adjusted Headcount Ratio for the 3 Senatorial
districts (k=4) Adjusted Headcount Ratio for the 3
Senatorial districts (k=4)
Figure 8: Adjusted Headcount Ratio for the 3 Senatorial
districts (k=5)
Figure 9: Adjusted Headcount Ratio for the 3 Senatorial
districts (k=6)

Ph ton
208
Figure 10: Adjusted Headcount Ratio for the 3
Senatorial districts (k=7)
Figure 11: Adjusted Headcount Ratio for the 3
Senatorial districts (k=8)
Figure 12: Adjusted Headcount Ratio for the 3
Senatorial districts (k=9)
With k=4, the Adjusted Headcount Ratio for Kano
North is 0.667 percent, while that of South is 0.655
and 0.536 for Kano Central (Figure 7). If k=5,
figure 8 shows that based on adjusted headcount
ratio, Kano North have poverty incidence of 0.651
when it is 0.635 percent in South and 0.492 in the
Centre. With change in the value of k to 6, the ratio
is 0.597 in the North are found to be below poverty
line, 0.568 in the South and 0.415 in Kano Central
(Figure 9). Figure 10 presents the Adjusted
Headcount Ratio at k=7, from the figure, the
incidence of poverty is 0.463 in Kano North, while
Kano South have 0.429 and Kano Central 0.267. At
k=8 and k=9, the ratio drops significantly. With
k=8, it is about 0.267 in Kano North. It is just
0.218 in the South and 0.114 at the Centre (Figure
11). Finally, levelling the cut-off to k=9, figure 12
shows that the Adjusted Headcount ratio is 0.089 in
Kano North, 0.059 in Kano South and 0.024 in
Kano Central.
3.1.2 Poverty Depth
The depth of deprivation suffered by poor
household at k=2 is0.576 percent in Kano Central,
0.661 percent in Kano South and 0.676 percent in
Kano North (Figure 13). This indicates that using
this cut-off, poor households in Kano Central suffer
57.6 percent of the total dimensions, whereas,
Kano South and North, it is 66.1 and 67.6 percent
respectively.
Figure 14 shows that at k=3, poor households in
Kano North are on average deprived in 67.8
percent of the total dimensions, whereas, the
Southern poor are deprived in 66.2 percent and
their Central counterparts are deprived in 59.6
percent of the total dimensions. By using four
dimensions as cut-off, it is observed from Figure 15
that poor households in Kano North are deprived in
68.7 percent of the total dimensions and their
Southern and Central counterparts in 66.8 and 61.7
percent respectively.From the estimates presents in
Figure 16, Northern poor are deprived in 70 percent
of the total dimension, when poor in the South and
Central are deprived in 68.3 percent and 64.9
percent.
With k=6, the level of deprivation in the North is
72.5 percent. It is however 71.3 percent in the
South and 69 percent in Kano Central (figure 17).
It is observed from Figure 18 that at k=7, the depth
of poverty in Kano North is 0.772 percent which
means that poor households in the district are
deprived in 77.2 percent of the total dimensions.
For South and Central, it is 76 and 75 percent
respectively. The most deprived households in
Kano North are deprived in 83.6 percent (k=8) and
92.7 percent (k=9). In the South, it is 83 percent
(k=8) and 92.3 percent (k=9) and 82.4 (k=8) and
91 percent (k=9) at the centre (Figure 19&20).

Ph ton
209
Figure 13: Poverty depth at the 3 Senatorial districts
(k=2)
Figure 14: Poverty depth at the 3 Senatorial districts
(k=3)
Figure 15: Poverty depth at the 3 Senatorial districts
(k=4)
Figure 16: Poverty depth at the 3 Senatorial districts
(k=5)
Figure 17: Poverty depth at the 3 Senatorial
districts (k=6)
Figure 18: Poverty depth at the 3 Senatorial districts
(k=7)

Ph ton
210
Figure 19: Poverty depth at the 3 Senatorial districts
(k=8)
Conclusion
Results in the three (3) senatorial districts showed
that the resulting models are dissimilar and confirm
the known fact that there is wider inequality
between the three senatorial districts. However,
The multidimensional indicators that describe the
three areas for both the poorest households and the
least poor are often different, which is an important
finding for policy purposes.
Furthermore, the estimated Alkire and Foster
poverty indices are subject to the number of
dimensions of deprivation well-thought-out and
that measured poverty decreases with the number
of dimensional cut-offs, that is, the weighted sum
of the deprivations (k), and the weight assign to
individual dimensions. The most contributing
factors to multidimensional poverty are sanitation
(as 90.4%of the household in Kano central, 95.6%
in Kano North and 98.2% in Kano South are
deprived in this dimension) and fuel used for
cooking (that 82.5% of the households in Kano
Central, 95.6% in Kano South and 98.2% in Kano
North deprived in). Others are child health, housing
condition and education of the household
head.These results therefore corresponds generally
to the opinions of many scholars (particularly
development scholars) in defining poor and poverty
as it confirms results elsewhere. A basic
challengingissue that arises is how to eliminate
poverty in the area. It is justifiable to conclude that
the landscape of poverty in Kano Stateis not evenly
distributed over space and households do not have
equal access to economic and infrastructural
services.
Research Highlights
1. Poverty is multidimensional;
2. Geographic targeting is the only solution to
allocate limited resources to areas of need; and
3. Poverty situation is not uniform in the 3
senatorial districts of Kano State.
Limitations
One of the key limitations to this study is having
data on the geographical characteristics of the study
area in that this serves as a barrier to relate
geographical variables and other social variables in
the area.
Recommendations
All statistical agencies (at both national and local
levels) should be considering poverty as from
multidimensional view point instead of
unidimensional perspective. They should also
involve geospatial technology in identifying and
mapping multidimensional poverty in Kano State
in particular and Nigeria at large. This is because
resources are available, the only problem is how to
allocate the resources. Efficient resource allocation
will lift many people out of poverty.
Funding and Policy Aspects
Resource allocation formula – which is either based
on population or derivation – need to be revisited in
order to first give more emphasis to areas with
higher need as well as those aspects of human
wellbeing that contributes more to
multidimensional deprivation.
There is need to increase the share of annual budget
allocating to education in that, this sector alone
contributes 11.02% to multidimensional poverty in
the area, but for many years, at both national and
state level, this sector’s allocation never reaches
30% of the total annual budget. In many local
governments, the number of schools are not
adequate to serve the existing population (needless
to think of growing population). People need to be
empowered through education and vocational
trainings so that they can be able to understand and
utilises the gifts of nature available in the area. This
will provide them with many chances to move out
of poverty. In that education open-up possibilities
to people.
Health is also another dominant contributor to
multidimensional poverty in the state (as it
contributes 12.92%). Most of the household do not
have access to health establishments, either because
they are not available in their locality or because
they do not have access to the established facilities.
Government should improve people’s access to
health facilities by constructing at least one facility
in every village, one in every ward, one in every

Ph ton
211
local government and one in every senatorial
districts (as it is the case in India), this will add to
availability of health facilities. There is also need to
link the health facilities through good
transportation network. By so doing, health poverty
will decrease in the satate.333
Sanitation contributes 14.96% to multidimensional
poverty in Kano State. There is need to device
means to improve sanitation especially in urban
areas where squatter settlers dominates. With
proper sanitation, many people will move out of
poverty as their living standard will increase. This
will also reduce other problems associated to
contamination such as cholera.
Most of the households depends on firewood as a
major source of energy for cooking. In Kano State,
cooking fuel contributes 14.59% to the overall
multidimensional poverty. This is harmful to
environment at large and individual households in
particular. Nigeria is blessed with natural gas. The
amount of gas missing through gas flaring believed
to serve the timing population. Government should
find a way to make efficient use of this resource.
They should also make a policy of subsidising gas
in order to give poor people adequate access to it.
This will help environment as the rate at which
trees are cutting down will decreases and hence the
health status of the poor improved.
Poor housing condition adds 11.68% to
multidimensional poverty in Kano State as such,
there is need for the authorities concerned to make
use of available resource to relocate and resettle
people living in the informal settlements
(especially in urban areas)
Poor people need to be involved in any attempt
aimed at poverty eradication in the area.
Government should also consider urban poverty as
part of challenges facing humanity in urban areas.
That is to say, governments and NGOs need to
include urban poor in their attempt to eradicate
poverty in that, most of poverty eradication
programmes targets only rural poor.
Direction for Further Research
This research tried to look at poverty in Kano State
from multidimensional perspective, in this
research, emphasis was made on measuring poverty
levels using multiple dimensions of poverty with
five different approaches so there is need to make
another study that will make use of all available
methods in the literature in order to see how the
derive estimates will vary. Though this research
attempted to relate some geographical variables
with poverty, but this was not explicit as data is not
available on some of the environmental factors, so
there is need to carry out another study that will
relate environmental variables with poverty level –
as done in many studies that made use of poverty
mapping – so that the relationship between poverty
and other variables will be revealed. There is also
need to incorporate resource allocation into poverty
analysis in order to see whether resource allocation
is putting into consideration the areas with higher
need. However, it will be good to look explicitly
into the underlying causes of clustering of poor.
References
Akinyemi F.O., Sester M., Balogun O.Y.,2012.
Communicating Information for Development: A
Cartographic Approach. The Electronic Journal on
Information Systems in Developing Countries, 53(6), 1-
20. Retrieved August 23, 2013, from
http://www.ejisdc.org
Akinyemi F.O., 2007. Spatial data needs for poverty
management: Research and theory in advancing spatial
data infarstructuring concepts, 33 – 54.
Alkire S.,Foster J., 2007, rev. 2008. Counting and
Multidimensional Poverty Measurement‘. OPHI
Working Paper 7, Oxford University.
Alkire S., 2009. Multidimensional Poverty Measures:
New Potential. The 3rd OECD World Forum on
Statistics, Knowledge and Policy Charting Progress,
Building Visions, Improving life. Busan, Korea.
Alkire S., Seth S.,2009, Determining BPL Status: Some
Methodological Improvements, Indian Journal of Human
Development, 2 (3), 407-424.
Alkire S., Conconi A., Seth, S.,2014. “Multidimensional
Poverty Index 2014: Brief Methodological Note and
Results”. OPHI Briefing 19, Oxford University.
Retrieved August 12, 2015.
http://www.ophi.org.uk/wp-content/uploads/Global-MPI-
2014-Brief-Methodological-Note-and-
Results.pdf?f68d3a
Alkire S., Housseini, B.,2014. “Multidimensional
Poverty in Sub-Saharan Africa: Levels and Trends”.
OPHI Briefing 81, Oxford University. Retrieved August
12, 2015.
Alkire S., Foster, J.E., Seth S., Santos M.E., Roche J.M.,
Ballon P.,2015. Multidimensional Poverty Measurement
and Analysis.Oxford University Press, Chapt.5.
Retrieved April 30, 2015, from
http://siteresources.worldbank.org/INTPOVERTY/Resou
rces/335642-1124115102975/1555199-
1124115201387/cry.pdf
Araar A. Duclos, J., 2013. User manual: DASP version
2.3, Université Laval PEP, CIRPÉE and World Bank
Ballini F., Betti G., Carrette S., Neri L., 2006 & 2008.
Poverty and Inequality Mapping in the Commonwealth
of Dominica. Estudios Econ´omicos, 123-162. Retrieved
June 12, 2014

Ph ton
212
Baschieri A., Falkingham J., Hornby D., Hutton C.,
2005. “Creating a Poverty Map For Azerbaijan” World
Bank Policy Research Working Paper 3793
Batana Y.M., 2008, August. Multidimensional
Measurement of Poverty in Sub-Sahara Africa. OPHI
Working Paper.
Bigman D., Fofack H.,2000. Geographic Targeting for
Poverty Alleviation: An Introduction to the Special Issue.
(F. Bourguignon, Ed.) The World Bank Economic
Review, 14(I), 129-145. Retrieved November 6, 2014
Bigman D., Dercon S., Guillaume D., Lambotte M.
2000,. Community Targeting for Poverty Reduction in
Barkina Faso. (F. Bourguignon, Ed.) The World Bank
Economic Review, 14(I), 167-193. Retrieved November
6, 2014
Cuong N.V., Truong T.N., Weide R.V., 2010. Poverty
and Inequality Maps for Rural Vietnam. Policy Research
Working Paper 5443. Retrieved November 19, 2013
Davis B.,2002. Is it Possible to Avoid a Lemon?
Reflections on Choosing a poverty Mapping Method.
UN, FAO. FAO. Retrieved October 31, 2013
Davis B.,2003, Choosing a method for Poverty Mapping.
Rome, Italy: FAO.
Edward P., 2006. The Ethical Poverty Line: A Moral
Definition of Absolute Poverty. (D. Ehrenpreis, Ed.)
Poverty in Focus, 14-16. Retrieved October 27, 2014
Elbers C., Lanjouw J.O., Lanjouw P., 2003. Micro-Level
Estimation of Poverty and Inequality. Econometrica,
71(I), 355-364. Retrieved August 24, 2013
Epprecht M., Minot N., Dewina R., Messerli P.,
Heinimann A., 2008. The Geography of Poverty and
Inequality in the Lao PDR.Swiss National Center of
Competence in Research (NCCR) North-South,
University of Bern, and International Food Policy
Research Institute (IFPRI), Bern: Geographica
Bernensia.
Erenstein O., Hellin J., Chandna P.,2010. Poverty
Mapping based on Livelihood Assets: A M. Applied
Geography, 30(2010), 112-125.
doi:10.1016/j.apgeog.2009.05.001
Filmer D., Pritchett L.Z., 1998. ‘Estimating wealth
effects without income or expenditure data – or tears:
Educational enrolment in India’, World Bank Policy
ResearchWorking Paper, No. 1994, Washington DC:
World Bank (also published in Demography (2001),
38(1).
Fukuda-Parr S., 2006. December. The Human Poverty
Index: A Multidimensional Measure. (D. Ehrenpreis,
Ed.) Poverty in Focus, 7-9. Retrieved October 27, 2014
Gonner C., Haug M., Cahyat A., Wollenberg E., Jong
W., Limberg G.,… Becker M., (n.d). Capturing Nested
Spheres of Poverty: A Model for Multidimensional
Poverty Analysis and Monitoring. CIFOR Occasional
Paper(4). Retrieved June 2, 2014
Henninger N., 1998. Mapping and geographic analysis
of poverty and human welfare – Review and assessment,
Report prepared for the UNEP/CGIAR Initiative on GIS,
World Resources Institute, Washington, D.C. Retrieved
October 31, 2013
Hentschel J., Lanjouw J.O., Lanjouw P., Poggi J.,2000.
Combining Census and Survey Data to Trace the Spatial
Dimensions of Poverty: A Case Study of Ecuador.
(Bourguignon F., Ed.) The World Bank Economic
Review, 14(I), 147-165. Retrieved November 6, 2014
Hirschowitz R., Orkinand M., Alberts P., 2000, “Key
baseline statistics for poverty measurement,” in Statistics
South Africa, Measuring poverty in South Africa
StatisticsSouth Africa. Pretoria.
Kakwani N., 2006. Poverty and Well-being. (Ehrenpreis
D., Ed.) Poverty in Focus. 20-21. Retrieved October 27,
2014
Momale S.B,. 2010. Postoralists’ Mobility and Access to
Resources in Northwest Nigeria. PhD Proposal,
presented to Geography Department, BUK
Naveed A., Islam T., 2010. Estimating Multidimensional
Poverty and Identifying the Poor in Pakistan:
AnAlternative Approach. RECOUP Working Paper No.
28
Okwi P.O., Linderhof V., Hoogeveen J.G., Emwanu T.,
Begumana J., 2006. Welfare and Environment in Rural
Uganda, Panel analysis with small-area estimation
techniques. PREM Working Paper, PREM 06/03,
Amsterdam, The Netherlands.
Petrucci A., Salvati N., Seghieri C., 2004. Autologistic
Regression Model for Poverty Mapping and Analysis.
Metodoloˇski zvezki, I(1), 225-234. Retrieved October 4,
2014
Rogers D., Emwanu T., Robinson T., 2006. Poverty
Mapping in Uganda: An Analysis Using Remotely
Sensed and other Environmental Data. Pro-Poor
Livestock Policy Initiative Working Paper 36. Retrieved
October 31, 2013, from http://www.fao.org/ag/pplpi.html
Sachs J., 2005. The End of Poverty: Economic
Possibilities for Our Time. New York. Penguin Press.
Sen A.K., 1987. Commodities and Capabilities.
Amsterdam: North-Holland.
Sen A.K., 2004. Dialogue: Capabilities, lists and public
reasons: Continuing the conversation. Feminist
Economics 10(3) 77-80
UN, 2010. The Millennium Development Goals Report
2010. New York
http://www.un.org/millenniumgoals/pdf/MDG%20Repor
t%202010%20En%20r15%20-
low%20res%2020100615%20-.pdf

Ph ton
213
UN Department of Economic and Social Affairs
(UNDESA). 2013. World Population Prospects: The
2012 Revision. New York: UN. Retrieved August 12,
2015.
http://www.fn.no/content/download/18853/160781/file/
World%20Population%20Prospect%202012%20revision
.pdf
World Bank, 2001. Poverty Trends and Voices of the
Poor, Washington, DC: World Bank.
For publications/ Enquiries/ Submissions:
Email: photonjournal@yahoo.com