Homeless People in West Bengal: A District-Level Study

The phenomenon of homelessness has often been explained as an obvious result of rapid urbanisation along with the limited housing facility in the urban areas (the destination region), but undoubtedly, it is interlinked with multifaceted issues which have been deeply rooted in the socioeconomic and political set up of the rural areas and small towns (the source region). Ruralurban development disparity, rural-urban migration, poverty, intraregional unequal human development and housing shortage are some of the factors that have intensified the problem of pavement dwelling in urban areas. According to the Census of India (2011), the state of West Bengal holds the 7th position among the states of India with 134040 homeless people. The development disparity between Kolkata and other parts of the state has resulted a definite district-wise spatial distributional pattern of homeless people with the greatest concentration in Kolkata city. The study attempts to probe into the spatio-temporal distribution of homeless population and strives to ascertain the instrumental factors of the phenomenon in a contextual framework with the help of available and acquired data and information. Simple statistical methodslocation quotient, Z-score, bi-variate correlation, Sphereman’s Rank Correlation Coefficient and Student t test have been used for the analysis. The findings suggest that the less developed districts, as well as the states, surrounded the Kolkata act as the source region that has pushed the destitute community towards the city and compelled them to reside on the footpaths of the city without a roof.


Introduction
The shelter is one of the necessities of humankind for comprehensive physical, psychological, economic, socio-cultural and intellectual development (Springer, 2000) as 'home' protects from the external adversities and opens multiple windows for selfimprovement. It personifies the ideas of comfort, belonging, identity, security etc. Thus the 'right to shelter' has become intently associated with the issue of human survival and well-being. Adequate housing is recognised as a basic human right by the Universal Declaration of Human Rights (article 25, para1) (UDHR, 2015). However, millions of people of the world in various circumstances are compelled to spend their night under the open sky and continue their lives in front of the public without any security and privacy. Irrespective of the variation of definition, these unfortunate people may be regarded as 'homeless' or 'pavement dwellers' (on the basis of the night time residence).They do not live in 'census houses' (a structure with roof) rather live on the pavement or footpath, under the bridge, flyover, in hume-pipes, places of worship, bus stops, railway stations or yards, parks or any public place in the metros without any 'residence' number or municipal address (Mukherjee, 1975;Census of India, 2011;Homeless Pavement Dwellers Welfare Bill, 2016). They are unable to maintain fixed, regular, safe, secure and adequate domicile (Allred, 2006;Yonge, 2007) and reside at a public or private place not designed for or ordinarily used as a regular sleeping accommodation in the night (Sattar, 2014). This 'poorest of the poor' section of the society is jobless, insolvent, functionless, support less (Lipton and Sabatini, 1984) and even identityless. Bhattacharya (2012) has argued that homelessness is not only having nowhere to sleep, instead it carries implications of belonging nowhere. So, it is more than rooflessness as it "involves deprivation across a number of different dimensions -physiological (lack of bodily comfort or warmth), emotional (lack of love or joy), territorial (lack of privacy), ontological (lack of rootedness in the world, anomie) and spiritual (lack of hope, lack of purpose)" (Somerville, 2013: 384). They are dwelled in a messy and unhygienic condition, without access to the basic needs; have to compromise with their privacy and often uprooted from their place in the name of development. Along with the pavement dwellers (floating homeless), the 'Situated Homeless' (living at semi-permanent shelter in illegally occupied public land) and the 'Potential Homeless' (live in substandard housing in slum areas with scanty facilities) must be taken into consideration as all such makeshifts are precursor of shelterlessness (Sivramkrishnan and Mondal, 2014). Hence, the growth of slum dwellers can be considered as an indicator of the increase in the magnitude of homelessness (Sattar, 2014).
Probably the phenomenon of homelessness is as old as the history of human settlement. However, it attained a global dimension in the 20th century (Springer, 2000) with the changing socio-political and economic situation of the world. It has become a live issue in the cities, both in the industrialised and technology-rich developed nations as well as developing countries (Ali, 2012). United Nations Habitat Report (2015) has estimated that nearly 100 million people of the world are homeless and another 1.6 billion people 'lacked adequate housing' (UNHR, 2015). In reality, there is no country and virtually no city in the world without the presence of 'sidewalk dwellers' (Mukhopadhyay et al. 1994). However, it is almost impossible to trace the actual magnitude of homelessness as large section of the population remains invisible due to their shifting nature, coupled with the fear of eviction by the civic authorities (Roy Chowdhury, 1999). Many countries have no official concern for homeless people, even to the extent of denying their existence (UNCHS, 2000). At present (2018), the estimated numbers of homeless population in some large cities of the world are as follows (figures are in thousands): Manila 3100, New York 74, Greater Mumbai 60, Los Angeles 58, Jakarta 50, The multidimensional phenomenon of homelessness is closely associated with ruralurban migration and deeply rooted in the socio-cultural, economic, political, infrastructural set up of the source region of the migrants. Increasing disparity in the development of rural and urban areas generates an inflow of poor, impoverished, unemployed people into the city that enhances the complexities of the problem of homelessness in the urban areas. This study endeavours to throw light on the spatiotemporal dynamics of homelessness in West Bengal along with the instrumental forces and factors behind the issue. The article begins with a review of literature. This is followed by a background to the study area. Thereafter, the article discusses the methodological issues, and the findings of the study.

Review of Literature
The gathering of homeless people in the streets of almost every urban centre of the world has drawn the attention of the administrators and academicians. Several scholars of different disciplines like Patel (1990), Baker et al. (1997), Springer (2000), Tribhuvan and Andreassen (2003), Tipple and Speak (2004), Vázquez et al. (2005), Koehlmoos et al. (2009), Patra andAnand (2008), Morrow (2010), Nooe and Patterson (2010), Ali (2012 and (Das, 2013;Mustaquim and Ismail, 2013;Chaudhuri, 2013;Adhikari, 2014;Ghosh, 2014;Mukherjee, 2014;Shivramkrishnan and Mondal, 2014;Dey and Majumdar, 2015;Bagchi, 2016;Basu, 2016;Kar, 2016). Bela Bhattacharya (1996) in her book titled "Slums and Pavement Dwellers of Kolkata Metropolis" presents a detailed ethnographic enquiry concerning the sociodemographic structure, living conditions, economic and occupational characteristics, social associations and relations of the homeless people. In "Homeless Pavement Dwellers: A Case Study of their Social Awareness in Kolkata from Human Rights Perspective" (2012) she has presented an extensive disquisition over the issue from the human rights perspective. In their comprehensive study, 1989) also have discussed the issues with particular reference to family characteristics, occupational patterns, income status, rural-urban linkage and mobility. Almost all the previous studies have predominantly stressed upon the socioeconomic and demographic status of the community in Kolkata, whereas, the state level scenario is invariably overlooked or somehow denied. The causal factors, in many cases, do not adequately explain and hence become uncertain, murky. Therefore, the present study focuses on the district level analysis of homelessness in West Bengal and strives to interlink the responsible factors with the issue.

The Study Area
West Bengal has been taken as the unit of study. The state (21°30'N to 27°14'N, 86°30'E to 89°53'E) is one of the eastern states of India with a total geographical area of 88,752 sq. km. It accommodates nearly 91.3 million (rural 68.11% and urban 31.89%) people with a population density of 1029/ sq. km (Census, 2011). It is the 14th largest state in terms of area (occupies 2% area of the country) and the 4th largest state in terms of population (comprises 7.6% of the total population). The state has 19 districts, divided into three administrative divisions (2011). The capital city of the state is Kolkata (the third largest city of India). Bangladesh and Assam border the state in the east; Bihar, Jharkhand, and Odisha states on the west; Nepal, Bhutan and Sikkim on the north and the Bay of Bengal in the south ( Figure 1).

Figure 1: Location Map of the Study Area Source: Prepared by the Authors Objectives and Methodology
The principal objectives of the study are:  to present a detailed account of the spatiotemporal distribution of the homeless population in the districts of West Bengal;  to identify the source areas of homeless people in the area and investigate the causes of homelessness in the source region as well as the destination region The major part of the study has been framed on the basis of available secondary data and information collected from various sources like the Census of India Relevant literature has been intensively reviewed to develop a conceptual framework for the study and also to identify the key factors responsible for homelessness. The collected information has been analysed with the help of simple statistical techniques to examine the interrelationships among the various causal factors. Bi-variate analysis has been used in this context. Z-score is computed to assess the spatial pattern of distribution of homeless population among the states and districts using the formula: X -Mean Z-Score = S.D. Location quotient has been calculated to detect the degree of concentration of homeless population using the formula: Where, e i = total homeless population in each district, e= total population in the district, Ei= total homeless population of the state and E= total population of the state.
Concurrently, a primary survey of 203 homeless households (having a population of 632; 460 males and 172 females) has been taken from different parts of the Kolkata city (especially in the Sealdah rail station and Barrabazar area, etc.) during March-June 2017 and 2018 with a semi-structured questionnaire to acquire the primary data. Purposive random sampling technique is used to select the respondents. The conclusion is drawn through a review of vast literature and analysis of collected data. The significant findings have been explained through vivid and informative description along with various maps and cartographic diagrams. The following section discusses the homelessness landscape at the national level.

Homelessness at National Level: A State Level Analysis
Though the Article 21 ('Right to Life and Personal Liberty') of the Indian Constitution encompasses the 'Right to Shelter' by signifying 'shelter' as the fundamental necessity for a happy and healthy life with human dignity; and free from any exploitation. However, nearly 1.77 million people in India (comprising 0.15% of the country's total population that is 1210.19 million) have lived without any shelter, keeping themselves off from all safeguards of human rights. Of them, 52.9 % are found in the urban areas whereas 47.1% are found in the rural areas. Nearly, 0.94 million people in the urban areas (0.25% of the total urban population of 377.11 million) and 0.83 million (0.1% of the total rural population of 833.08 million) people in rural areas live without any shelter (Census, 2011). Therefore, even after six decades of independence, it is one of the biggest challenges in front of the policymakers and the administrators of India. Notwithstanding, the number of the total homeless population has been decreased since 1981 in response to the reduction of rural homelessness, but the urban homelessness has recorded a continuous increment since 1961 ( Figure 2). The country shows 8.76% decrease in the homeless population during 2001-2011(Census, 2011. With rapid urbanisation (growth rate of 2.76% per annum during 2001-2011), the urban areas have reported a 20.51% increase in homeless population during the last decade.
It clearly indicates the interrelationships between the phenomena of urbanisation and homelessness. On the contrary, the rural areas show 28.34% decline in homeless population during the stipulated period as the consequence of the shift of the rural poor towards the urban areas that increments the volume of the homeless people in the cities (Sattar, 2014).
Generally, the bigger states in terms of size and population comprise the large share of the homeless population. Nearly 65.3% of the homeless population is spread over six states of India: Uttar Pradesh, Maharashtra, Rajasthan, Madhya Pradesh, Gujarat and Andhra Pradesh (Kumuda, 2014)   West Bengal, with 134040 houseless people (above the national average of 52148), comprises 7.78 % of the nation's total homeless population. 68.52 % homeless population in the state are males while, 31.48% are females. Kolkata, the smallest district and the largest urban centre of the state has recorded the highest number of pavement dwellers. The city has accommodated nearly 12.250 houseless households with a total houseless population of 69798 in 2011. It comprises 52.07% of the state's total followed by North 24 Parganas (6.81%), Uttar Dinajpur (6.16%), Haora (4.25%), South 24 Parganas (4.06%) and Bardhaman (3.77%). District Cooch Behar has recorded lowest concentration of houseless population (0.52%), followed by Dakshin Dinajpur (0.65%), Darjeeling (1.16%), Purulia (1.18%) and Purba Medinipur (1.28%) (Census, 2011) ( Figure 5).As stated above, Kolkata has recorded the highest concentration of homeless population (LQ value: 10.57), followed by Uttar Dinajpur with moderate concentration (LQ value 1 -2). The districts like Darjeeling, Malda, Birbhum, Haora and North 24 Parganas, however, show a relatively less concentration. Other districts represent an insignificant concentration of homeless population (< 0.50) ( Figure 6). Z-score map traces the spatial pattern of distribution of the homeless population. (Figure 7) Naturally, Kolkata shows the highest z-score value followed by North 24 Parganas and Uttar Dinajpur(z-score value : 0 -0.2). On the contrary, Cooch Behar and Dakshin Dinajpur show lowest z-score values (< -0.4). The maximum number of districts of the state shows low to moderate z-score value.

(Compiled by the Authors) Distribution of Rural and Urban Homelessness
Homeless people are unevenly spread over the state with maximum concentration in the urban areas compared to its rural counterparts. Nearly 21.69% shelterless population are found in the bucolic part of the state, whereas, 78.31% of them have reside in the footpaths of the large towns ( Figure 8). In case of rural homelessness, South 24 Parganas shows the highest concentration; followed by North 24 Parganas, Murshidabad, Bankura, Paschim Medinipur, Birbhum and Bardhaman with moderate concentration (Figure 9). In case of urban homelessness, undoubtedly the state's capital has recorded the highest (Z-Score value 4.09). North 24 Parganas and Uttar Dinajpur have recorded moderate concentration, whereas, rest of the districts record an inconsiderable concentration of urban homeless people ( Figure 10).

District-Wise Decadal Growth Rate of Homelessness, 2001-2011
The state has experienced a moderate rate of increase (21.26%) in homeless population during 2001-2011. The rural homelessness shows a comparatively lesser rate of increase of 15.59%, whereas the urban homelessness has increased up to 47.38% during the same period. Four districts, namely Darjeeling (-54.46%), Jalpaiguri (-27.04%), Cooch Behar (-19.35%) and Hugli (-1.83%) have reported negative growth in homeless population during 2001 to 2011. Except for these four districts, all other districts show a positive growth rate. Malda (273.79%) shows the highest growth in homeless population followed by Bankura (259.81), South 24 Parganas (192.10%), Nadia (152.09%), Uttar Dinajpur (133.67%) and Birbhum (129.32%) ( Figure 11). Darjeeling district has experienced highest negative growth rate both in rural and urban homelessness (-53.96% and -54.9% respectively), whereas, Bankura has recorded highest positive growth (270.37%) in rural homelessness and Malda shows highest growth rate in urban homelessness (1940%). Acute poverty and failure of Govt. housing policy seems to be the prime causes behind such massive increase in rural homelessness in Bankura. On the other hand, high increase of urban homeless people in Malda District is the outcome of the massive urbanisation (129.07% growth in urban population in  caused by the incoming of rural uprooted people from the river-bank erosion affected areas. The capital city Kolkata shows 3.14% increase in urban homelessness (Table 1).

Discussion
The causal factors of homelessness have been widely perusal and evaluated by scholars of different disciplines (Bhattacharya, 2012). Most of the previous studies have advocated the difficulties of establishing any linear relationships between one specific factor and homelessness due to the multifaceted and complex character of the phenomenon where several overlapping risk factors influence the phenomenon. In reality, the correlation between each factor and homelessness is typically intricate and can be understood only in the context of people's life histories (Somerville, 2013). The significance of the elements has changed drastically with the changing space, time as well as socio-economic and political perspectives of the society. Historically, the study of homelessness in the western world shows a typical switch in its approach over time. Gowan (2010)  A 'new orthodoxy' (May, 2000;Pleace, 2000) has been transpired at the beginning in the 1980s (Brandon et al. 1980) in which homelessness is explained in terms of a specific amalgamation of structural factors and individual vulnerabilities. 'Structural' factors create the environment within which homelessness occurs, and 'individual' factors conclude the likelihood of becoming homeless in those conditions (Fitzpatrick et al. 2000;Pleace, 2000).

Causes of Rural Homelessness
Although homelessness is a common concern both in rural and urban areas (Morrow, 2010) the rural homelessness almost remain unnoticed in policy debates as well as in the study of the academicians due to distinguishing hidden nature accompanied with the migrating trend of rural poor towards the urban centres (UNCHS, 2000). The rural homelessness remains veiled as the rural poor often cope with the miserable condition through temporary makeshift arrangements (Wardhaugh, 2012;Sattar, 2014). Dupont et al. (2000) have asserted the interconnection of rural and urban homelessness as rural poverty pushes the male members of households to migrate to urban areas in search of job or income. Consequently, the rural homelessness has been reduced and the occurrence of homelessness in urban areas has increased its magnitude. The migrated persons send money to their family residing in the native village in poor and insecure circumstances. Poverty, landlessness and unemployment are the prime economic factors of rural homelessness (UNCHS, 2000;Wardhaugh, 2012). Disasters, either natural or human-made, are other critical factors of rural homelessness (UNCHS, 2000). For example, in the year 2009, 21782 rural houses fully collapsed, and 44849 houses were partially damaged in 24 Parganas (North and South) by cyclone Aila (Situational Report, 2009). Up to 2005, more than 700,000 people of Malda and Murshidabad districts have been displaced due to bank erosion of the river Ganga (Khatun et al. 2018). Such massive displacement by natural calamities generally increases the quantity of rural homeless people. It has been estimated that annually, around 30 million people are affected and 2.34 million houses are being lost by disasters (Chauhan, 2009). Natural calamities like cyclones, floods and river bank erosion, mostly evident in some districts like South 24 Parganas, Murshidabad and Malda, have not only destroyed the domicile of the rural poor but also affected their source of livelihood. Physical loss of land or alteration of its nature compelled them to leave their habitat of the ancestors as assurance of continuing the life system is denied and rush towards the nearby urban areas (especially towards Kolkata) in destined condition because in this circumstance, begging in the streets is acceptable to these groups rather than starving for long days.
Displacement by any developmental activity raises the immensity of homelessness in rural areas (Tripathi, 2017 , 1985). Such significant displacement events have incremented the number of rural homeless people, who generally move towards the nearby urban centres and has enlarged the volume of urban homelessness.
It has been estimated that only 25.48%of displaced persons were rehabilitated in the national level during 1947-1989(Fernandes and Rao, 1989Bharati, 1999) (Table 2). From 1947 to 2000, nearly 47 lakh acres of land has been acquired for development projects that have affected 70 lakh persons in West Bengal. Of them, 36 lakh were displaced persons (DP), and 34 lakh were deprived of livelihood without being physically relocated projected affected persons (PAP) (Fernandes, 2007). The state has resettled only 9% of the displaced population (Fernandes et al. 2006). The vast percentage of the non-settled population generally has positively influenced the volume of homeless people.
Besides, some political forces (local political pressure, political conflict, unresponsiveness etc.) and socio-cultural causes (ethnic isolation, social unrest/communal riots, eviction, social marginalisation) may also have caused widespread displacement and long-term homelessness (Seshadri, 2008). This is evident in developing countries but not widely discussed due to the inadequacy of data or information and socio-political complexities (UNCHS, 2000). The urban dimension of homelessness is the most visible and prominent phenomenon. In a simple framework, homelessness of urban areas may be addressed as an outcome of exponential urbanisation coupled with the existing gap between the demand and supply of housing units. However, in reality, the pathway towards homelessness has been characterised with many complexes, multifaceted and overlapping interconnected issues that need further investigation for complete understanding of the role of factors and their connections with homelessness (Somerville, 2013). Nooe and Patterson (2010) have developed an ecological model of homelessness encompassing almost all the biopsychosocial risk factors (both individual and structural factors) as well as the social and individual outcomes. The model recognises the complexities of the interplay between the individual factors and social factors on different systems levels.
It has been revealed from the primary survey (2017-18) that 89.72% of the respondents came to the city for economic objectives. 78.33% of them does not possess any land in their native places, whereas, 58.62% has reported their inability to earn sufficient to sustain life in their area in their native place. Rest 21.67% respondents possess small landed properties in their villages, which is insufficient to meet the daily needs of their family. So, the male member of the family had to shift to Kolkata for working at least as labour. 26.39% of respondent households (mainly intra-district migrants) have lost their houses and agrarian land due to some natural calamities. As a result, they have rushed towards the city of Kolkata to take shelter in the streets.
The survey has also pointed out that 11.33% of the respondents have shifted towards the city for non-economic purposes. 8.37% of the families have been led by women, either a widow or abandoned by husband and have shifted to the city for feeding their children. Another 2.96% homeless women have reported that they have left their houses to get rid of domestic violence and 1.48% pavement dwelling women come to the city through trafficking. 4 young boys (1.97%) lived in the station premises and engaged in rag picking reported that they were orphans and had come to the city to get rid of the torture of their local guardians.
Urbanisation, associated with rural-urban migration is one of the prime causes of urban homelessness. Bhattacharya (2012) has pointed out that uncontrolled urbanisation in the Third World has intensified the gathering of destitute people on the street of the urban centres. The cities/towns of West Bengal are not exceptions (Table 3).
The state was one of the most urbanised states of India before independence (Ghosh and Chakma, 2014) and currently, 31.87% people live in urban areas which is marginally higher than the national average (31.16%) (Census of India, 2011) ( Figure 12). The urbanisation pattern of West Bengal is characterised with 'urban primacy,' that is the excessive concentration of the urban population in and around Kolkata with the very low level of urbanisation in the interior districts (Giri, 1998;Chakraborty et al. 2015). The city of Kolkata (formerly Calcutta) has been exponentially developed as the administrative core and a leading economic hub of eastern India by the colonial rulers as well as the administrators in the post-independence period. The other parts of the state remain backward and neglected. Thus improper regional planning guided by wrong administrative decisions has formed an imbalanced regional development. Consequently, a wide disparity in the socioeconomic and infrastructural development has been created between KMC (Kolkata Municipal Corporation) and its adjoining areas and the rest of the bucolic parts of the state (Anisujjaman, 2015).
The overwhelming role of the primate city along with the increasing development disparity has generated a flow of migrants towards the city (Anisujjaman, 2015).A chunk of poor people from the less developed neighbouring states like Bihar, Jharkhand, etc. have come to the city as cheap labours that have escalated the homeless scenario. In addition, the massive influx of refugees from the eastern part of Bengal (now Bangladesh) contributed to urban concentration immediately before the independence (Giri, 1998). The state, especially the city of Kolkata has witnessed an immense flow of cross-border immigrants (especially the people belonging to the Hindu religion) over four decades for its locational proximity to Bangladesh owing to the political instability during the partition of Bengal and the liberation war of Bangladesh (1971) (Chatterjee, 1990). According to Som (1987), nearly six million refugees have arrived in West Bengal from East Pakistan in different waves up to 1971 and nearly 2 million have resided in Calcutta Municipal District Area in 1973 which constitutes almost 26% of the total population (CMDA, 1976).
Primary survey (2017) has revealed that a bulk of homeless people in Kolkata has arrived from the backward states like Bihar (37.18%), Jharkhand (17.41%), Uttar Pradesh (14.23%) and Odisha (3.48%) whereas only 2.53% of homeless people have migrated from other states of the country. On the contrary, 7.12% of homeless people have migrated from South 24 Parganas, followed by Murshidabad (5.38%) and Paschim Medinipur (3.96%). Rest 6.81% has shifted from other districts of the state. Only three families comprising 13 members (occupying 1.90% of the respondents) were found in the station areas, who have originated from Bangladesh (Figure 13). Only 24.14 % homeless households among them have migrated during the last 20 years, and rest of them live in the streets from 20 to 60 years. It has also been also found that 43.83% of the homeless people are born in the streets of the city, and don't have any connections with their native villages. The survey clearly reveals that the bulk of the homeless population of the city has originated from the less developed adjoining areas that serve as the source region. It has also been evident that the percentage of inter-state migrants is higher in the Barabazar area, whereas the Sealdah station area is primarily occupied by the intra-state migrants (Primary Survey, 2017). Local governance of the source region has failed to provide essential amenities to them for their continuation of life. As stated earlier, 89.72% of the respondents have migrated towards the city for economic purposes, with the aspiration of a better job, income and lifestyle (Primary Survey, 2017). Conversely, the Kolkata city generally dispenses opportunity of all sorts, from begging to scavenging, low paid jobs and even illegal works that support the minimum nutrition of the dwellers. Thus certain 'pull factors' have transmuted the city as the shelter (destination region) for the survival of rootless immigrants.
In many cases, the city with limited accommodation and infrastructural facilities, have failed to compensate the incoming immigrants. Moreover, those people have failed to get proper jobs and are incapable of affording a house for survival owing to their insufficient economic strength. Nearly, 97.15% of the respondents have reported their incapability to manage a shelter due to the lack of income. Only, 2.85% of the respondents have admitted that they can rent a home in sharing basis, but they are unwilling to expend any money for this purpose as they have to send the maximum share of the income to their families living in the native place. So, they prefer to spend their night under the open sky to minimise expenditure. Naturally, the footpaths of Kolkata metropolitan city have witnessed the assemblage of large quantities of homeless people, either single or with family, struggling to continue their life on the streets without security. A different trend of urbanisation has been observed during the last decade (2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011) where the process of urbanisation began to spread into the interior districts away from Kolkata like Malda, Murshidabad, Nadia, Birbhum, Purulia and Jalpaiguri. Consequently, these districts have recorded higher growth rate of the urban population compared to highly urbanised districts of North 24 Parganas, Hugli and Bardhaman located around Kolkata (Chakraborty et al., 2015). The small and medium towns of the state have experienced lopsided immigration of rural people with varying magnitude. As a consequence, considerable population growth has occurred in the medium and small urban centres with the assemblage of 'sidewalk dwellers' in almost all cities or towns of the state. Naturally, the growth of urban population and urban homelessness has shown a positive relationship. Figure 14 has clearly shown the moderate correlation (r2= 0.407) between the phenomenon of urbanisation and urban homelessness. It may be concluded that the gradual shifting of rural people towards the urban centres has positively influenced the urban homelessness scenario in the towns of West Bengal. Surprisingly, as an exception, Kolkata has shown negative growth (-1.88%) in urban population during the last decades but has recorded an increasing trend in the homeless population (3.14%). On the contrary, despite the fast rate of urbanisation (38%), Darjeeling shows negative growth in concentration of urban homeless people (-54.9%).
It has been observed that inter-district disparity in the level of Human Development has close inter-linkages with the magnitude of homelessness. The district with higher HDI value has recorded a higher percentage of homeless people that have reflected the moderate positive correlation (r2) between the phenomenon of homelessness with the level of Human Development of the District (Figure 15). Spearman's Rank Correlation Coefficient is used as a method to compute the degree of association between HDI and concentration of the homeless population. The result shows that there is a moderate optimistic relationship between the HDI and homelessness of the districts (Table 4). To test the significance, Student's t-test has been computed with the help of following formula: t=√r2 (N-2)/ 1-r2 The calculated t value scores to 1.96, which is higher than the critical value of t at 0.05 significance level with 17 degrees of freedom that is, 1.74 (one-tailed test). Thus, as t0.05<1.96 the value of r is significant. Data Source: Census and DHDR (2010-11) It is accepted that the region with less HDI (Human Development Index) generally have a considerable number of homeless population, but in reality, the situation is reverse. The movement of poor people from less developed areas (having low HDI value) towards the areas of better social and economic condition (having high HDI value) with the aspiration of a better lifestyle has explained the positive relationship. It may be said that the less developed areas (districts) have become the 'source areas' and have pushed a bulk of population towards the nearby towns without any security of food or shelter that generally increases the volume of homeless population in different pockets of the cities. The primary survey has pointed out that large share of the homeless population of Kolkata has come from South 24 Parganas (HDI Value 06), Murshidabad (0.46) and Paschim Midnapore (0.57) districts which has recorded much lower HDI value than Kolkata (0.78).
Similarly, it is interesting to observe that the homelessness shows a moderate negative relationship with the poverty level of the districts of West Bengal (Figure 16). Poverty is one of the prime causes as well as the result of homelessness (Johnsen and Watts, 2014). It is a condition in which a person or community lacks the financial capacity and essentials to meet the basic needs to maintain a minimum standard of life and well-being normally followed in society. It deprives people of the freedom to decide over and shape their own lives (Tennerfeldt and Ljung, 2006). Poverty is viewed as a universal risk factor of homelessness, as "people who are not poor can usually avoid homelessness even if they experience personal crises" (Fitzpatrick et al. 2000: 28).
The result of Spearman's rank correlation coefficient (r= -0.47) shows that there is a moderate inverse relationships between the poverty level and homelessness scenario of the districts (Table 5). The calculated t value scores to2.13, which is higher than the critical value of t at 0.05 significance level with 17 degrees of freedom, that is, 1.74 (one-tailed test). Thus, as t0.05 is far less than 3.43 the correlation is significant. The relationship may be analysed with the help of the phenomenon of migration. Poor areas (with high Human Poverty Index value) have pushed the insolvent people to move towards the areas with low HPI value. The gradual shifting of insolvent people towards the areas of less poverty has lowered

Conclusion
Homelessness depicts the most apparent and severe manifestation of denial of the human rights to adequate housing. The maximum concentration of poor and homeless human groups is evident in Kolkata as it extends them the provisions of earning for survival. It has been revealed in the analysis that the main factors for leaving their roots have been multifarious, but economic factors are most dominant. Lopsided development in the urban areas and relatively less development in the source areas have attracted the jobless, homeless people to the towns and cities. Besides social isolation or marginalisation and political instability in source areas, the destruction of habitats and sources of livelihood by natural calamities pushed the poor rural folks towards the nearby urban areas (especially towards Kolkata). Thus, the city has evidenced the booming of homeless people, whereas, the rural homelessness has recorded a declining trend.
The economic and socio-political reasons for leaving the rural roots and migrating to the urban pockets are still active in the source areas of migration. Thus, assurance of socioeconomic and political security along with self-provisioning in those source regions may be a suitable policy to reduce the flow of the uprooted migrants.