Seasonal Migration and Child’s Schooling: A Survival Approach

Abstract

The article opts to investigate the long-term effects of parental seasonal migration on a child’s access to school education. The phenomenon of seasonal migration ‘leaving child at home’ or ‘accompanied by’ is a very common feature in the Purulia district where migration is the only viable option to sustain livelihood in lean-agricultural season. Although parents’ migration in such areas seems to be essential for the family economy, lack of parental care is found to be responsible for academic and psychological non-adjustment that affects a child’s education to a great extent. The Cox Regression Hazard Model and the Kaplan–Meier Estimator analysis of school participation have been employed to explore the survival probability of children at varying contexts, viz. migration status, gender, caste and age. The result shows the negative impact of parental migration on school participation of left-behind children leading to early dropout before the completion of the school education cycle.

Keywords

Seasonal migration left-behind opportunity cost survival function dropout

Introduction

India made a Constitutional commitment (86th Amendment Act, 2002; Article 21-A) to provide free and compulsory education to all children in the age group of 6–14 years, but the goal which was expected to be achieved by 1960, remains elusive, till today. Still large numbers of children, especially in specific groups remain excluded from schooling, jeopardizing equitable access to elementary education (Govinda & Bandyopadhyay, 2008). The reasons why children are unable to complete basic education in developing countries have been attributed to structural factors at household, school and societal levels (Colclough et al., 2003). Seasonal migration at the household and societal level usually shrinks the window of opportunity to access education of the children left behind. Seasonal migration leaving children behind at home may have adverse consequences for adolescent behaviours (Antman, 2012), especially educational attainment leading to frequent absenteeism and a high rate of school dropout (Antman, 2010). Moreover, the long absence of either of the parent may entail a psychological cost, switch the intra-household decision-making process and children become overburdened by performing parents’ extra duties during their absence. This effect is likely to be larger if the migration spell becomes longer. The present article opts to analyse the ‘left-behind’ characteristics of the children of migrated households and to investigate the impact of fathers’ migration on their educational and psychological adjustment.

Literature Review

History of Seasonal Migration in India

Seasonal migration has long been a major source of income and risk-coping strategy in many agriculture-based economies in the developing world. In locations where access to non-agricultural employment is very much limited or physiography and climate exert severe limitations for agricultural production throughout the year, seasonal migration is often an alternative key to add to the family income in rural India (Jha, 2007). Rao’s (1994) study of Palamur labour in Andhra Pradesh and de Haan’s (2002) historical inquisition regarding migration in Western Bihar explored that seasonal and circular (also called cyclical, oscillatory) migration has long been a part of the livelihood portfolio of poor people across India. It is now well-recognized that migration becomes a part of the normal livelihood strategy of the poor (McDowell & de Haan, 1997) and does not happen only during times of distress. The micro level study by Deshinkar and Anderson (2004) found that circular migration has been emerged as a dominant form amongst the poor groups of India. The National Commission on Rural Labour (NCRL) sets the number of seasonal inter-state migrants at around 10 million from rural areas, majority of them are usually engaged in cultivation, brick-kilns, quarries as well as in urban informal manufacturing, construction, service or transport sectors employed as casual labours, head-loaders and hawkers (Dev, 2002). Estimates of short-term migrants vary from 15 million to 100 million (Deshinkar & Akter, 2009). Most of them belong to socio-economically disadvantaged groups, such as scheduled castes (23.1%), scheduled tribes (18.6%) or other backward classes (39.9%) who have minimum educational attainment, limited assets and resource deficits (NSSO, 64th Round 2007–2008).

Seasonal Migration and School Education

Education is a critical issue to the seasonal migration as the window of opportunity that children of migrant families have is very small (Smita, 2006). It has long been a practise for sustaining alternative livelihoods of the family, with adult male members leaving the village in search of a paid casual job. Persistent drought and mono-cropping systems have led to the escalation of this trend. The NSSO 64th Round survey tells that more than 13 million out of 300 million migrants in India are seasonal in nature. At least half of these migrants have a very poor level of educational attainment, that is, below the primary level. This severe phenomenon perhaps continues through the human capital channel, as migrants pass on to the next generation, entrapping them in the migration cycle forever (Majumder, 2011). Hunt (2008) linked poverty and child labour and depicted how this vicious cycle pulls children out of school to work.These children are often compelled to support their mothers in both household and income-generating activities (Smita, 2006). The father’s absence may have a negative psychological impact on the child and may compel the youngster to spend more time with the family or work to make up for the absence of their parents. Last but not least, paternal migration may alter the balance of power in the family, so that decisions within the household are mostly made by mothers and other surviving family members. Children’s educational attainment may increase if these decision-makers show more concern for educational investments (Antman, 2012).

Gender Differences

McKenzie and Rapoport (2006) investigated the impact of belonging to a migrant household on educational attainment in Mexico using an instrumental variable censored ordered probit model and data from the National Survey of Demographic Dynamics (ENADID). The authors discovered that boys from migrant households were less likely to complete junior secondary school, while both boys and girls from migrant households were less likely to complete senior secondary school. Being born into a migrant family reduced the likelihood of completing secondary school by 13% for boys and 14% for girls. The reasons for the lower educational enrolment of migrant teenagers aged 16–18 years differ by gender; boys were more likely to migrate, whereas girls frequently left school for domestic chores (Christopher et al., 2000; Francisca, 2012). Santrock (1972) hypothesizes that children may be more negatively affected by their father’s absence if their fathers leave earlier in life (before the age of 6 years) as opposed to later since older children are able to compensate for their father’s absence with the peer attachments in school. Santrock (1972) also argued that left-behind boys are more negatively influenced by the father’s absence than girls. Thomas (1994) and Bertrand and Pan (2011) also found the same results in their research in the United States.

Opportunity Costs of Schooling

In a similar research on Mexico, 12–15-year-old males, Antman (2010) found comparable findings regarding the impact of migration. Her analysis of data from the Mexican Migration Project and the National Statistical Institute of Mexico (INEGI) was done using individual fixed effects and instrumental variables estimation model (MMP). She concentrated on paternal migration and discovered that when their fathers immigrated to the USA, boys spent less time in school and on academic activities and more time working outside the family. McKenzie and Rapoport (2011) find that migration reduces boys’ school enrolment between the ages of 16 and 18 years applying a disincentive effect on nearby youth. Children are compelled to reduce study hours and increase work hours in response to their father’s migration (Antman, 2012). Lack of parent’s monitoring, motivation may lead to grade repetition and ultimately dropout (Su et al., 2012). These disruptive effects on family structure can change the leadership in the family, giving more power to older males who are less educated and less prone to understand the importance of investment in human capital as regards to their grandchildren (Ginther & Pollak, 2004), thus the possible long-term effects of migration might cancel out the effects of a temporary improvement in household incomes (Giannelli & Mangiavacchi, 2010).

De Brauw and Giles (2008) compared the opportunity cost of high school attendance of children left-behind to the actual cost of attendance. This debate also looked at the relationship between domestic migration opportunities and the opportunity cost of continuing education. With the aid of an instrumental variables method, the authors showed that as the cost of relocating in search of work decreased, so did the likelihood that a rural child would graduate from high school. The following were proposed as potential pathways for this detrimental effect: (1) increased awareness of employment opportunities in other locations as the number of migrants from a community increases; and/or (2) the potential for an increase in employment prospects within a community as more of the working-age population migrates.

Psychological Costs

According to global research (Fang et al., 2006; Jeremy et al., 2010; Wen, 2008), parent-child communication is crucial for boosting children’s psychological development and meeting their developmental requirements. Due to the absence of their parents in their early years, children who are left behind may experience a variety of issues. Limited information on psychological adjustment in children who were left behind, however, has been conflicting. On the one hand, some studies suggested that left-behind children were likely to experience a wide range of mental health issues, including depression, anxiety, and loneliness (Fan et al., 2009; Ren & Shen, 2008); they may also have a poor quality of life (Jia et al., 2010), a low level of satisfaction and happiness (Fan et al., 2009; Gao, 2010; Liu & Ouyang, 2010). However, several researches (Hu et al., 2008; Zhang et al.,, 2006) found no differences between comparison and left-behind children in various areas (such as problem behaviours, school satisfaction, and contentment). In addition to the inconsistent results in the literature, the majority of past studies handled children who were left behind as a single group.

Researchers over a range of disciplines have focused on left-behind children who experience early separation from one or both of their parents (Fan et al., 2009; Li et al., 2010). The present study restricted its attention to migrant fathers since the number of migrant mothers is small in a rural setting. An effort has been done to explore a comparative view of academic and psychological adaptation between left-behind children and their non-migrant peers. Lastly, followed by Edwards and Ureta (2003), the ‘left-behind’ phenomenon has been represented by the Kaplan–Meier probability estimator of dropping out from school in varying socio-economic arenas. This is intended to unfold the dynamics of livelihood hardship, seasonal migration and survival of ‘left-behind’ migrant children in an educational scenario that can be achieved through the detailed investigation of the following questions:

How does physical setup determine the ways of livelihood in a hilly drought-prone area?

How long the ‘left-behind’ children survive in the arena of school education in various social settings?

Does any gap persist in academic and psychological adaptation between ‘left-behind’ migrant children and their non-migrant peers?

Against this backdrop, the present study begins with a brief outlook about the study area, stating the objectives and methods adopted in the analysis, then it explains the left-behind features of the children making a comparison with non-migrant and lastly estimates the survival of these children in different socio-economic context and concludes.

Significance of the Study

The significance of the study lies in its attempt to address a novel problem of barriers of access of schooling of left-behind children in migrated families of rural remote plateau tracks of eastern India and related less survival in school. Gathering information from the field survey in selected ‘resource-poor’ areas, the present study applies a survival approach to capture the varying school-level as well as the societal effect of seasonal migration for accessing education among the left-behind learners. The study focusses on the school-going learners from migrant families at the transition period from elementary to secondary specially to make the study unique. The educational access of a left-behind migrant school student in the context of remote rural settings wrapped by several barriers inside and outside the school wall is very much understudied in social science. In the developing country’s context, the predictive strengths of ‘disadvantaged’ socio-demographic situation, economic uncertainty, and depressive and discouraging home environment without parents (Mangiavacchi et al., 2014) keep the vital essence of the present study. The household and society level shocks induce the parents to migrate for an alternate job and also spur the school-going children to drop out of school before its completion (Antman, 2012).

In the developing countries like India, children from poor socio-economic backgrounds are disadvantaged concerning their ‘development, learning and attaining potential’ (Kamper, 2008; Powers, 1996). Children are more likely to achieve success in school if they come from a ‘certain type of family’ (Kay, 2000). Dependence on the subsistence economy and seasonal migration in search of an alternative job in the lean season may be the probable reason for this assumption. Parents of these families have very little time to spend with their children, due to the ‘lack of financial resources and the need to deal with day-to-day basic survival issues’ (Ren & Shen, 2008; Sampson, 2002). However, father’s absence may ‘High opportunity cost of school attendance’, ‘huge absenteeism’, ‘Social disapproval’, and ‘depressive home environment’ experienced by left-behind students of low socio-economic backgrounds might lead to early dropping out of school (Nichols & White, 2001).

Study Area

The rain-fed Purulia district represents an area of particularly low agricultural productivity and a high incidence of severity of poverty among the rain-fed areas in Asia. Purulia is one of the administrative districts of West Bengal Chronically affected by meteorological drought¹ conditions (Palchaudhuri & Biswas, 2013). The north-western and south-western portion can be labelled as extreme drought-prone areas of the district that includes both of the Jhalda blocks (I and II). High seasonality of rainfall (82% occurred in the monsoon period) confines the cropping period to only part of the year (mono-cropping). Available but not-sufficient moisture over the entire monsoon period limits the time window of opportunity for the various cropping systems practiced by farmers in the district. Irrigation potential during the dry season is relatively low. The small local population subsists from the extraction of local forest products and carries out very little agricultural activity like local rice, pulses, millets and vegetables.

Relevance of Physical Setup in Rural Livelihood

A glimpse of the ecological landscape of a region may be regarded as the sine qua non of the livelihood pattern of the people residing in that area. The district of Purulia is considered as one of the most backward districts (ranked 16th out of 17th districts in Human Development Report 2008) in the state of West Bengal due to the maximum concentration of backward villages (994 villages) within its boundary out of which the 102 villages (highest in district) are situated at Jhalda II block and the number is 53 in Jhalda I block (District Magistrate Office of Purulia District 2009).² In such a less favourable economic and agro-climatic condition, alternative income sources are one of the most prevalent household strategies of rural people for coping with risk and vulnerability. Ample peasants with marginal lands having limited or no irrigation facilities suffer from an erratic monsoon and thus their food supply becomes under severe strain (Mohapatra, 1985). The West Bengal District Gazetteer, Purulia uncovers the fact that the unremunerative character of agriculture largely failed to attract any sizeable immigration but rather caused a sizeable volume of emigration, thus offsetting the effects of immigration (1985:116). Most of the household labour force is, thus engaged in a wide variety of non-agricultural income-generating activities during the agricultural off-season, causing seasonal migration to sustain alternative livelihood in the neighbouring more fertile regions like Barddhaman (pube jaoa) expecting assured work (Census 1961).³

Livelihood exerts a smart impact on the educational scenario of any area. The Census reports in different decades reveal the fact that the rural literacy rate of the district has not changed significantly throughout the last 50 years. Various push-pull factors like poor accessibility, socio-cultural discontinuities and economic uncertainties contribute a great deal to the educational marginalization of the rural pockets of the Purulia district. The left-behind migrant children are among those at the highest risk of dropping out. The livelihood pattern reflects the educational scenario of the study area. The average literacy rate of the two blocks is 60.47% but the gender gap in education is still very high (36.24%) as per Census 2011. Illiteracy is strikingly very much high in these backward villages of the study area as 55.01% cannot read and even is able to sign. The figure is 75.49% in the case of women.

Hensahatu Fate Sing High School is the only school to feed the surrounding villages of the GP, but Rigid GP is devoid of any Higher Secondary School. Huge dropout is a common phenomenon of these two villages reflecting a very gloomy picture in all stages of the school education cycle. Nearly 11.11%, 22.30% and 20.21% of students have dropped out from 16 Primary, 2 Upper Primary and 1 Higher Secondary School at Hensahatu, respectively. The comparative figure is 17.62% and 30.71% at 11 Primary and 4 Upper Primary Schools of Rigid village (DISE, 2016).

Children especially from poor families in the villages under study enrol at the beginning of the academic year in school but leave by the middle of the term as a result of demands for their labour during harvest time as noted by Colclough (2000). The phenomenon is very much common in migrant households where seasonal migration of either of the parents clashes with the school calendar (Akyeampong et al., 2007) leading to irregular attendance (Guarcello et al., 2005). The high opportunity cost builds pressure on schooling time. Very often, students engaged in such activities terminate their schooling. The dropped-out respondents (82 out of 114 left-behind migrant children) of the selected villages informed that most of them (92%) were sporadically absent for at least 45 days or more whereas only 8% of children had a continuous absence⁴ of 45 days in school leading to drop out from school.

Methodology

The methodology adopted in the study has been precisely Quantitative as well as Qualitative.The villages were selected systematically based on Social Development Index (SDI)⁵ and migration intensity. The UNDP goalpost method has been fruitful for calculating the migration intensity of Gram Panchayats. Following the standardization of indices and then K-Means cluster analysis, the cluster of Hensahatu GP (SDI: -1.48, MI: 1.64), Rigid GP (SDI: -2.07, MI: 1.39) and Tatuara GP (SDI: -2.28, MI: 0.89) has been extracted, which are located in the lowest rung of Social Development ladder but highest rung in the Migration Intensity ladder. Out of the three from the selected cluster, Hensahatu from Jhalda I block and Rigid from Jhalda II block have been chosen for a close scrutiny.

To capture the migrated households from the selected villages specifically, households were grouped into two categories: migrant households and non-migrant households. A household is considered to be a migrant one if at least one adult member, preferably the head of the family/father of a school-going child/children migrate seasonally for more than 5 years during the field session.

To sort out the dropout candidates from migrated households, methodological help was taken from the Office of School Inspector as the houses of the dropped-out students are marked as ‘D’ by the ‘Siksha Bondhu’ of S.I. Office. In addition, the Village Information Centers and Village Education Communities of Jhalda blocks helped a lot to select the perfect time and to recognize the residence of the target groups from where the survey of ‘left-behind migrant’ and ‘not-migrant’ respondents as well as their parents made possible.

Direct open-ended interview was conducted through Questionnaire and Opinionaire to put a question to the individual migrated respondent to gain information about their left-behind atmosphere at home, socio-economic background, academic and psychological adaptation within and outside school, and so on. The study population in the enquiry comprised 114 children of migrated households, out of which 82 belong to the dropped out category and the rest were currently enrolled in school. Survey in Sishu Siksha Kendras is carefully avoided due to the below age of the little learners. To glean out the adverse effects of parental migration priority was given to include dropped-out candidates more in number than the currently enrolled. Another 90 non-migrated children were interviewed at the same time to have a comparative view of the academic and psychological adjustment they perceive.

In addition, various sophisticated quantitative tools like Kruskal–Wallis One Way Analysis of Variance, Cox Proportional Hazard model, and lastly Kaplan–Meier Survival estimation enriched the study a lot to fulfil certain objectives.

Results and Discussion

Villagers’ Livelihood

Jhalda blocks carry 45% of the total population living below the poverty level category.⁶ The Deprivation Index (measured as per literacy, infant survival and per capita income) is also high as Jhalda I Block and Jhalda II Block carry 28 and 32 deprived villages (District Magistrate Office, Purulia 2009). Subsistence farming is the mainstay of the villagers, but almost one-fourth of the rural households of Hensahatu (25.12%) possess no agricultural land of their own. The amount is 26.93% for Rigid. On the other, more than half of households (58.99%) possess only less than one acre of irrigated land at Hensahatu and the value is 53.7% for Rigid. The options for carrying livelihood depending on agriculture obviously is limited (Table 1).

Table 1.
Occupational Diversification of Sampled Households on Agriculture and Non-agricultural Sectors.

Gram Panchayat Name Hensahatu Rigid

Households (%) with farming and one more occupation 45.28 44.43

Households (%) with Farming and two more occupations 15.09 27.01

Households (%) with Farming and three more occupations 1.89 1.85

Households (%) without farming and only one occupation 20.75 5.56

Households (%) without Farming and two occupations 9.43 10.04

Source: Primary survey of migrated households, 2019.

Note: Bold characters reveals very low farm and non-farm based livelihood diversification in both villages.

The Table 1 reveals very low farm and non-farm based livelihood diversification in both villages with bold characters. In the study villages, enough dependence has been noticed in non-agricultural activities in migration-destination regions combined with agricultural activities in the native villages during monsoon season mainly; and the cumulated figures appear to be 62.26% and 73.29% at Hensahatu and Rigid, respectively. Moreover, households with pure non-agricultural activities with values of 30.18% and 15.6% are also note-worthy. All statistics reveal the over-dependence of non-agricultural activities as well as low diversification in livelihood adopted by the migrated households. They lack resources, assets, skill and education which constrain their diversification towards more enumerative activities and they are forced to diversify to low-return activities (Barret et al., 2001; Khatun & Roy, 2016). The poor well-being index bearing the same tune with the low diversification image of the sampled villages (Figures 1 and 2) uncovers the dismal condition of the livelihood status of the households.

Figure 1.
Source: The authors.

Figure 2.
Source: The authors.

The ternary diagram (Figure 3) explores the fact more barely that except for two villages, the rests are concentrated in an occupationally depressed region. The prospects of earning income from agriculture, in poorer households of rural areas still remain a gamble of monsoon and hence arises the importance of alternative jobs that can enable the maintenance of at least the subsistence-level earnings (Datta & Singh, 2011). Dry agro-ecological condition has made the rural livelihood of Hensahatu and Rigid very much challenging. In general, for poor households, livelihood diversification is the key survival strategy to cope with adverse livelihood shocks in developing countries and stabilize their incomes (Christopher, 2001; Dutt & Singh, 2011; Ellis, 1998). A significant portion of villagers of Tanasi, Garia, Brajapur, Madhupur, Muradih, Ichhatu of Hensahatu Gram Panchayet and Kanriyor, Tamakbera, Tahaddri, Rahan and Hesalber of Rigid Gram Panchayet migrates every year, they have no other alternative left than to engage themselves in agricultural labour in eastern agro-intensive districts of West Bengal like Bardhhaman or in non-farm/construction sectors, mines, factories, rickshaw pulling, road-building construction, and so on. in nearby towns like Asansol, Durgapur as well as other states of Orissa, Jharkhand or Tamil Nadu in India. Seasonal migration fills the occupational gap in the slack season, usually the migrants engage in low-skilled temporary or casual work for several months.

Figure 3.
Source: The authors.

‘Left-behind’ Features of Children of Migrant Households

To analyse the ‘left-behind’ features of respondent children of migrated households, descriptive statistics have been summarized in Table 2. Out of 114 ‘left-behind’ migrant children, 66 (57.89%) were boys and the rest were girls (42.11%). The average age of the respondents is 14.76 years with 1.449 years of standard deviation out of which the majority (66.1%) have dropped out from school in the mid-way, and only 25.8% ‘left-behind’ children are currently enrolled in school revealing an adverse impact of migration status of households on child’s education. The fact already stated earlier that as seasonal migration is a risk-coping strategy in a lean season of remote villages of Jhalda blocks, villagers irrespective of caste are being forced to migrate; 51.6% respondents belong to the scheduled social group category with a tiny standard deviation of 0.498. The literacy status of parents as shown in the descriptive statistics reveals a very gloomy picture, fathers on average enjoyed 0.9 years of school education, whereas it is only 0.18 years for mothers, though in the case of fathers (1.238 years) the standard deviation is a little bit better than mothers (0.39 years). The skewness of fathers (1.699) and mothers (2.312) education indicates very low years of educational attainment of both holding the maximum frequency. It is also noteworthy that parental education exerts a strong influence (father’s education: r²= 0.447; mother’s education: r²= 0.556) on children’s education. Keeping in mind the economic condition, it has been found that 61 (49.2%) migrated households belong to the poorest category and earn less than ₹24,000/per annum whereas 42 (33.9%) and 11 (8.9%) households belong to the ‘poor’ and comparatively ‘better’ category with a very low value of SD (0.666) and skewness (0.779). But the length of migration helps the family to improve its economic condition to some extent (r² = 0.583).

Table 2.
Descriptive Statistics: Left-behind Features of Children.

Variable Frequency Mean (SD) Skewness Kurtosis Description

N 114

Gender 0.58 ± 0.496 -0.324 -1.929 Boy (1) and girl (0)

Boy 66 (57.89%)

Girl 48 (42.11%)

Age 14.76 ± 1.449 0.106 0.589 Age (in years)

Social affiliation 0.56 ± 0.498 -0.251 -1.972 Status: Scheduled (1) and others (0)

Scheduled 64 (51.6%)

Others 50 (49.3%)

Educational status 0.28 ± 0.451 0.989 -1.04 Currently enrolled (1) and dropped out (0)

Currently enrolled 32 (25.8%)

Dropped out 82 (66.1%)

Father’s education 0.9 ± 1.238 1.699 2.893 Father’s year of schooling

Mother’s education 0.18 ± 0.39 2.312 5.106 Mother’s year of schooling

Economic condition 1.56 ± 0.666 0.779 -0.48 Annual Family income (in Rs.): Poorest: <₹24,000/ (1); Poor: ₹24,000 to ₹48,000/ (2) and Better: >₹48,000/ (3)

Poorest 61 (49.2%)

Poor 42 (33.9%)

Better 11 (8.9%)

Age when parent/s started the migration 1.46 ± 0.68 1.192 0.137 Age (in years) of the child

Length of migration (months) 1.36 ± 0.652 -0.623 -0.998 >4 months (1); 2 to 4 months (2) and <2 months (3)

>4 60 (48.8%)

2 to 4 44 (35.8%)

<2 10 (8.1%)

Primary caregivers 1.36 ± 0.566 1.317 0.792 Mother (3); Other older people (2) and grandparents (1)

Grandparent 78 (62.9%)

Other older people 31 (25%)

Mothers 5 (4.0%)

Education level of primary caregivers 1.43 ± 0.977 2.516 5.644 Illiterate (1); below primary (2); primary (3); Upper primary (4) and above (5).

Illiterate 89 (71.8%)

Below primary 13 (10.5%)

Primary 4 (3.2%)

Upper primary 4 (3.2%)

Above 4 (3.2%)

Source: Based on authors’ calculations.

The next concern of the study is to analyse the migration phenomena of the households as well as the education scenario of the children of migrated households. Generally, the head of the household migrates for several months leaving the child at a very early age (mean: 1.42 years). The ‘left-behind’ phenomenon has become worse (skewness: –0.623) when the majority of parents (48.8%) migrated for more than 4 months, whereas only 8.1% of parents stayed outstation for 2 months only. In 62.9% of cases, the grandparents performed the role of primary caregivers during the long absence of parents, mothers (4%) gaining the least opportunity to care of their own child. The more disappointing fact is that the majority of the caregivers (71.8%) are purely illiterate, only 3.2% has attained above upper primary school education level.

Parental education exerts a significant role in children’s access to schooling. In the present analysis, though the majority of the parents have a very low level of literacy, fathers’ and mothers’ education levels determine 45% (r = 0.447) and 56% (r = 0.556) of children’s educational attainment, respectively. During the absence of parents, literate caregivers like uncle and elder brother helped upgrading migrant children’s educational status (r = 0.607). Here it must not be overlooked that the child under the care of his/her mother during the absence of the father gained opportunity to carry on with his/her education (r = 0.398). Migrated households with ‘better’ economic conditions tried to continue their child’s education a lot as revealed in the correlation matrix (r = 0.590).

Academic and Psychological Adjustment of Left-behind Children

Let’s turn the focus into the academic and psychological adjustment of ‘left-behind’ children that carry ample relevance as these children usually face various difficulties in their early life due to the long absence of parents at home. Parental involvement in schooling includes attending school meetings, reading encouragement at home, participating in school governance, and so on. Children develop more psychological maturity and do better in school when they come from families in which parents monitor and regulate their children’s activities and at the same time, they provide emotional support and encourage independent decision-making (Astone & Mchanahan, 1994). In migrated households of selected villages of Jhalda blocks, almost illiterate caregivers do not realize the importance of education, and as a result, they do not encourage as well as participate in their child’s education. The problem becomes more vigorous due to the long absence of the father. Due to the lack of parent’s care and supervision for several months, left-behind migrant children were likely to suffer from depression, loneliness, low level of satisfaction (Fan et al., 2009) and poor school performance (Démurger, 2015). After explaining the left-behind atmosphere of migrated households it felt the necessity to explore the academic perfor- mance of left-behind children in school (Table 3). To achieve the purpose, Kruskal Wallis One Way Analysis of Variance by Ranks test acted as an instrument to test the academic score of these children compared to their non-migrated peer community in the school. The analysis further opted to investigate whether the academic performance of left-behind migrant children significantly differs from the academic performance of their peer ones or not at all. Kruskal–Wallis test was performed to compare academic scores in different aspects, that is, participation of parent/caregiver in child’s education by attending school meetings and helping in solving homework; emotional support given to the children for maintaining grades in school examination and expressing joy in child’s special educational achievement, educational aspiration that how long the child wants to carry on with his/her education and lastly engagement in co-curricular activities like school’s annual function, Saraswati puja,⁷ annual sports, and so on. In all significant differences were observed between two groups of children with a greater academic score performed by non-migrant children compared to left-behind migrant children. Frequent parent-child communications lead to good academic performance and psychological development, whereas those with less frequently reported poor outcomes (Li, 2006).

Table 3.
Academic and Psychological Adjustment of Left-behind Migrant Children: A Comparative Analysis with Not-migrant Children.

Variable Left-behind Not-migrant H value p value Description

Participation 153.96 .000 Academic participation and involvement of primary caregivers of left-behind children and parents for non-migrant children

Attending school meeting (Mean ± SD) 1.133 ± 0.034 2.209 ± 0.055 – Caregivers/parents were asked: ‘How many times he/she attended a meeting in the child’s school?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).

Help in doing homework (Mean ± SD) 1.088 ± 0.032 2.22 ± 0.514 – Caregivers/parents were asked: ‘How many times he/she helped your child in doing homework?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).

Emotional support 163.08 .000 Emotional support for encouraging studies

Support in maintaining grade (Mean ± SD) 1.0708 ± 0.258 2.222 ± 0.536 – Caregivers/parents were asked: ‘Did he/she give support in maintaining grades?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).

Happy in special achievement (Mean ± SD) 1.088 ± 0.315 2.244 ± 0.567 – Caregivers/parents were asked: ‘Does he/she express joy in the child’s special achievement?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).

Academic Aspiration (Mean ± SD) 1.35 ± 0.497 1.93 ± 0.832 32.01 .000 Children were asked: ‘Up to which grade you want to continue your education?’ Scale: Don’t know (1); Higher secondary (2); Graduate (3); and Professional degree (4).

Engagement in Co-curricular activities (Mean ± SD) 1.43 ± 0.564 1.97 ± 0.678 30.08 .000 Children were asked: ‘Are you engaged in co-curricular activities in school?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).

Source: Based on authors’ calculations.

Survival Analysis of ‘Left-behind’ Migrant Children in School Education

Survival analysis seeks to find a relationship between the dependent variable, that is. survival probability of a child in school and a set of independent time-constant as well as time-varying covariates (variables). In the study the Cox Proportional Hazard Model and Kaplan–Meier estimates have been employed has been applied to estimate the impact of the ‘left-behind’ migrant children and their family on the hazard of dropping out of school. The Cox proportional hazard model is given as follows:
$H_{i} (t) = \exp (β ’ x_{i}) h_{o} (t),$

where X_i = (x_1i,…, x_pi)’ is a vector representing of the explanatory variables for the ithindividual, β = (β₁,…, β_p)represents the regression coefficients, and H_o(t)is the dropout function for an individual for whom (x = 0).

Table 4.
Results of Cox Regression Model.

Variables B SE Sig. Exp(B)

Migration status of households

Gender 0.86 0.25 0.00 2.36

Age 0.24 0.06 0.04 0.99

Social affiliation 0.36 0.22 0.14 1.43

Father’s education –0.014 0.09 0.61 0.96

Mother’s education –0.134 0.18 0.80 1.05

Length of stay 0.22 0.16 0.017 1.25

Years of migration 0.14 0.14 0.031 1.15

Source: Based on authors’ calculations.

Notes: Omnibus Tests of Model Coefficients: 874.69 (2 log likelihood).

Chi-square: 21.25; Sig. 0.00.

The time-dependent factor includes the status of censoring (X₀), that is, either the child completed the school education cycle (0) or dropped out before its completion (1). On the other hand, the independent factors consist of dummy variables like migration status of the household (non-migrant: 0 and migrant: 1), gender (male: 1 and female: 0), social affiliation (scheduled: 1 and others: 0) and on the other, continuous variables like age, parental education, average length (months) of stay at the migrated site and lastly the approximate years the members of the family started migration in search of an alternate job. Thus, the Cox regression hazard model (Table 4) may be represented as follows:
$H_{i} (t) = \exp (β_{1} g e n d e r + β_{2} S o c i a l A f f i l i a t i o n + β_{3} a g e + β_{4} f a t e d u + β_{5} m o t e d u + β_{6} m i g r a t s t a y + β_{7} m i g r a t s t a r t) h_{o}_{} (t) .$

The model has been fitted for the ith student, given
$H_{i} (t) = \exp (0.86 g e n d e r + 0.36 c a s t e + 0.24 a g e - 0.014 f a t e d u - 0.134 m o t e d u + 0.22 m i g r a t s t a y + 0.14 m i g r a t s t a r t) h_{o} (t),$

where h_o(t) is the hazard rate for a child of a migrant family to succeed the school education.

From the Cox model, it can be interpreted that a boy from a migrant family with scheduled social affiliation with the poor parental educational background may prone to face more hazard in accessing school education compared to others. The more length of stay in the migration site as well as the initiation of this way of livelihood in one’s early childhood may increase more possibility of dropout of that boy from school. Lastly, increasing age may hamper access of schooling also. It is noteworthy that though parental education helps to decrease the dropout possibility but the matter is not significant in the present study.

In the Cox-Regression Hazard curve (Figure 4), the abrupt jump in Grade-IX depicts that most of the ‘Left-behind’ reading in class IX dropout from school without completion of the secondary education before appearing the Madhyamik Examination, the first board examination in student life. The result explores that parental migration has a negative effect on school attendance in the long term with higher hazards of school drop-outs for children left behind (Giannelli & Mangiavacchi, 2010).

It is perennial to say here that while estimating the causal effect of parental migration, children’s age at the time of parental migration (Antman, 2012) is very much crucial in the context of survival in school. The study follows the basic assumption of 14 to 18 years of age as the ‘threshold level’, the transition period of school education from elementary to secondary level. The younger siblings of this age limit may be adversely affected by parental absence at home leading to early drop-out from school. The above model (Figure 4) reveals the left-behind learners studying in elementary and secondary suffer from the maximum hazard of dropout from school. Therefore, the finding of the Cox Regression hazard model with increasing grade is seen to be consistent with Antman’s (2012) study.

Figure 4.
Source: The authors.

The next section of the present article opts to explore the survival probability of a child continuing school education at varying contexts with the aid of the Kaplan–Meier estimator. The estimator defines the probability of surviving a child given the XII grade of the school education cycle. Thus, the estimator of the survival function S(t) is given by:
$\hat{S} (t) = \prod_{i : t_{i} \leq t} (1 - \frac{d_{i}}{n_{i}})$

With t_i referring to the particular time (grade in the study) when dropout happens, d_i the number of dropout that happens at time t_i and n_i the individuals known to have survived up to time t_i.

Estimation Results

Akin to the previous studies (Sankar, 2017; Sylvie, 2015), the estimates show that seasonal migration on the part of the father adversely affects his child’s schooling in the long term. The probability of survival in the school education cycle decreases substantially for the left-behind children with the increase of grade, especially for Secondary and Higher Secondary level creating a widening gap between the two survival curves (Figure 5). The median value of survivals for a non-migrant child is 10 years whereas it becomes 6.75 years only for a left-behind migrant child. Being left-behind aggravates the probability of the child’s leaving school before the completion of the education cycle.

Figure 5.
Source: The authors.

Presenting Kaplan–Meier estimates (Figure 6) that a boy in a migrant family has a smaller probability of surviving in the school education cycle than a girl of that same category (Median gender-gap: 1.75 years). The survival probability of completing the primary education cycle is 0.98 for girls whereas it is 0.53 for boys. Gradually, it reduces in the upper primary level also, that is, 0.20 for girls and 0.05 for boys. The probability of boys in a migrant family to survive up to secondary education is almost nil as being seen in the study area. The deleterious effect of migration starts at the very beginning of the compulsory education stage for boys, and the disruptive effect persists throughout all primary and secondary educational stages for boys. But for girls, migration only has a significant effect at the secondary school stage (Deep, 2017). Adolescents’ responsibilities in farm and non-farm activities become substantial during the father’s long absence thereby hampering the continuous attendance of the child in school (Majumder, 2011). Survival largely declines at the transition of primary and upper primary grades for boys. The boys at the age of 11–12 years old get maturity for casual daily paid jobs to augment household income or to make company with their father for migration, thereby increasing the probability of dropout and delaying school progression (Mangiavacchi, 2016).

Figure 6.
Source: The authors.

In general, socio-economically disadvantaged groups have a greater propensity to migrate seasonally, which reflects its distress-driven nature. The need for survival is paramount in such cases and education becomes secondary. But in the present study, the Kaplan–Meier Estimation curve shows (Figure 7) that there exists little survival gap between children of scheduled social affiliation and others in accessing school. Repeated overlapping of the curves depicts that for tackling poverty most of the households in sampled villages of Jhalda blocks migrate seasonally in search of alternate livelihood irrespective of social affiliation, thereby the survival of children in the school education system declines not only in the scheduled group migrant family but for the general caste children also (caste wise survival gap: 0.75 years) that leads to the negligible educational attainment, limited assets and resource deficits.

Figure 7.
Source: The authors.

It is to note here that scheduled tribe people who usually have apathy to migrate seasonally outside due to their strong affinity to ‘jungle-life’, are restricted for the present study. Physical isolation, primitive subsistence livelihood, place attachment, illiteracy and poverty characterize this segment of the population. Field study observes that tribal children are part of the domestic activities in their family, like cattle grazing, collection of fuel and fodder, part-time casual job in Jhalda town; jungle play a crucial role in their daily-livelihood.

Conclusion and Policy Advice

With about 10 million seasonal and circular migrants (The National Council for Rural Labour) only in rural India, the relationship between migration and left-behind families must be an important policy question. In Economics, migration decisions are treated as individual optimization choices. In general, the human capital model of migration treats it as an investment that bears some positive returns in the future (Sjaastad, 1962). However, the parental absence might also have adverse consequences for left-behinds, be they children, the elderly, or a spouse.

India recognized the Right to Education as a fundamental right in 2005. India is also a signatory of the UN Convention on the Rights of the Child (1989). In spite of the positive scenario of endeavouring to achieve the Millennium Development Goals (MDGs) and Universal Primary Education by 2015, the more ambitious Sarva Shiksha Abhiyan (SSA) programme, there is hardly any attention to the plight of the children of migrants demanding proper attention from all levels.

In migration-prone villages of Jhalda blocks, migration implies an adverse effect on education or health status or more child labour. Though not surveyed, but it may be concluded that the surrounding villages also carry a more or less gloomy picture like villages of Hensahatu and Rigid. For the sake of the left-behind families, the potential long-term costs of migration need to be mitigated through appropriate home state/country policies.

The followings are some holistic approaches to bring the light of complete education of these left-behinds of the Purulia district:
Compensate for Opportunity costs: Offering Stipends and scholarships will increase left-behinds’ enrolment: Stipends covering tuition fees, textbooks, school supplies, uniforms, transport costs, etc. meet three criteria: (1) regular attendance and (2) achieving certain minimum pass marks. It will increase enrolment in school gradually.

Building schools close vicinity to households boosts enrolments: Distance matters, particularly for girls. Building more schools will achieve full enrolment in rural areas.

Flexible school calendar: Adjusting the school calendar to accommodate household child labour requirements will increase school attendance;

Satellite school and mobile teachers: For left-behinds, it will help in accessing education in remote rural corners and decrease drop-out.

Psychological Health Promotion Initiatives:Future health promotion and intervention initiatives should aim to draw more public attention to the well-being of children who are left behind and implement strategies to support their psychological adjustment. In order to help rural parents better appreciate the value of parent-child communication and to improve the effectiveness of their communication skills with their kids, the government and concerned community organizations should also offer training in this area. Additionally, children with psychological adjustment issues should be given the opportunity to volunteer in the community to help them cope with unpleasant feelings (such as loneliness) and increase life satisfaction and happiness, all of which appear crucial for their development.

Gram Panchayat Name	Hensahatu	Rigid
Households (%) with farming and one more occupation	45.28	44.43
Households (%) with Farming and two more occupations	15.09	27.01
Households (%) with Farming and three more occupations	1.89	1.85
Households (%) without farming and only one occupation	20.75	5.56
Households (%) without Farming and two occupations	9.43	10.04

Variable	Frequency	Mean (SD)	Skewness	Kurtosis	Description
N	114
Gender	0.58 ± 0.496	-0.324	-1.929	Boy (1) and girl (0)
Boy	66 (57.89%)
Girl	48 (42.11%)
Age		14.76 ± 1.449	0.106	0.589	Age (in years)
Social affiliation		0.56 ± 0.498	-0.251	-1.972	Status: Scheduled (1) and others (0)
Scheduled	64 (51.6%)
Others	50 (49.3%)
Educational status		0.28 ± 0.451	0.989	-1.04	Currently enrolled (1) and dropped out (0)
Currently enrolled	32 (25.8%)
Dropped out	82 (66.1%)
Father’s education		0.9 ± 1.238	1.699	2.893	Father’s year of schooling
Mother’s education		0.18 ± 0.39	2.312	5.106	Mother’s year of schooling
Economic condition		1.56 ± 0.666	0.779	-0.48	Annual Family income (in Rs.): Poorest: <₹24,000/ (1); Poor: ₹24,000 to ₹48,000/ (2) and Better: >₹48,000/ (3)
Poorest	61 (49.2%)
Poor	42 (33.9%)
Better	11 (8.9%)
Age when parent/s started the migration		1.46 ± 0.68	1.192	0.137	Age (in years) of the child
Length of migration (months)		1.36 ± 0.652	-0.623	-0.998	>4 months (1); 2 to 4 months (2) and <2 months (3)
>4	60 (48.8%)
2 to 4	44 (35.8%)
<2	10 (8.1%)
Primary caregivers		1.36 ± 0.566	1.317	0.792	Mother (3); Other older people (2) and grandparents (1)
Grandparent	78 (62.9%)
Other older people	31 (25%)
Mothers	5 (4.0%)
Education level of primary caregivers		1.43 ± 0.977	2.516	5.644	Illiterate (1); below primary (2); primary (3); Upper primary (4) and above (5).
Illiterate	89 (71.8%)
Below primary	13 (10.5%)
Primary	4 (3.2%)
Upper primary	4 (3.2%)
Above	4 (3.2%)

Variable	Left-behind	Not-migrant	H value	p value	Description
Participation			153.96	.000	Academic participation and involvement of primary caregivers of left-behind children and parents for non-migrant children
Attending school meeting (Mean ± SD)	1.133 ± 0.034	2.209 ± 0.055	–		Caregivers/parents were asked: ‘How many times he/she attended a meeting in the child’s school?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).
Help in doing homework (Mean ± SD)	1.088 ± 0.032	2.22 ± 0.514	–		Caregivers/parents were asked: ‘How many times he/she helped your child in doing homework?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).
Emotional support			163.08	.000	Emotional support for encouraging studies
Support in maintaining grade (Mean ± SD)	1.0708 ± 0.258	2.222 ± 0.536	–		Caregivers/parents were asked: ‘Did he/she give support in maintaining grades?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).
Happy in special achievement (Mean ± SD)	1.088 ± 0.315	2.244 ± 0.567	–		Caregivers/parents were asked: ‘Does he/she express joy in the child’s special achievement?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).
Academic Aspiration (Mean ± SD)	1.35 ± 0.497	1.93 ± 0.832	32.01	.000	Children were asked: ‘Up to which grade you want to continue your education?’ Scale: Don’t know (1); Higher secondary (2); Graduate (3); and Professional degree (4).
Engagement in Co-curricular activities (Mean ± SD)	1.43 ± 0.564	1.97 ± 0.678	30.08	.000	Children were asked: ‘Are you engaged in co-curricular activities in school?’ Scale: Never (1); seldom (2); often (3); almost always (4); and always (5).

Variables	B	SE	Sig.	Exp(B)
Migration status of households
Gender	0.86	0.25	0.00	2.36
Age	0.24	0.06	0.04	0.99
Social affiliation	0.36	0.22	0.14	1.43
Father’s education	–0.014	0.09	0.61	0.96
Mother’s education	–0.134	0.18	0.80	1.05
Length of stay	0.22	0.16	0.017	1.25
Years of migration	0.14	0.14	0.031	1.15

Footnotes

Acknowledgements

The authors are grateful to the respondents for their kind cooperation in completing the study.

Author’s Contributions

SKG and SS designed the study and developed the methodology. SS processed the data, analysed and mapped the data and wrote the manuscript. SKG and SS aided in interpreting the results and in drafting the manuscript. They both critically reviewed the manuscript. All authors read and approved the final manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Notes

References

1.
Akyeampong K. , Djangmah J. , Oduro A. , & Seidu A. (2007). Access to basic education in Ghana: The evidence and the issues. Consortium for Research into Educational Access Transition and Equity (CREATE), Country Analytic Review. University of Sussex.

2.
Antman F. M. (2012). Gender, educational attainment, and the impact of parental migration on children left behind [IZA Discussion Paper No. 6640]. Institute for the Study of Labor (IZA). https://hdl-handle-net.web.bisu.edu.cn/10419/62534

3.
Antman F. M. (2010). The Intergenerational effects of parental migration on schooling and work: What can we learn from children’s time allocations? Journal of Development Economics, 96(2), 200–208.

4.
Astone N. M. , & Mclanahan S. S. (1994). Family structure, residential mobility, and school dropout: A research note, Demography, 31(4), 575–584.

5.
Bhatty K. , Saraf R. , & Gupta V. (2017). Out-of-School children: Some insights on what we know and what we do not. Economic & Political Weekly, 52(49), 69–76.

6.
Bertrand M. , & Pan J. (2011). The trouble with boys: Social Ináuences and the gender gap in disruptive behavior [National Bureau of Economic Research Working Paper No.17541].

7.
Christopher B. (2001). Non-farm income diversification and household livelihood strategies in rural Africa: Concepts, dynamics and policy implications. Food Policy, 26(4), 315–331.

8.
Christopher C. , Rose P. , & Tembon M. (2000). Gender inequalities in primary schooling: The roles of poverty and adverse cultural practice, International Journal of Educational Development, 20, 5–27.

9.
Colclough C. , Al-Samarrai S. , Rose P. , & Tembon M. (2003). Achieving schooling for all in Africa. Ashgate.

10.
De Brauw A. , & Giles J. (2008). Migrant opportunity and the educational attainment of youth in rural China [World Bank Policy Research Working Paper: No. 4526].

11.
De Haan A. (2002). Migration and livelihoods in historical perspective: A case study of Bihar, India, Journal of Development Studies, 38(5), 115–142.

12.
Deshinkar P. , & Akter S. (2009). Migration and Human Development in India [UNDP Human Development Research Paper 2009/13].

13.
Deshinkar P. , & Anderson E. (2004). People on the move: new policy challenges for increasingly mobile populations. Natural Resource Perspectives, 92, 1–4.

14.
Dev S. M. (2002). Pro-poor growth in India: What do we know about the employment effects of Growth 1980–2000? [Working Paper No. 161]. Overseas Development Institute.

15.
District Information System for Education (DISE). (2016). Educational statistics at a Glance. Ministry of Human Resource Development, Department of School Education and Literacy. Government of India. http://mhrd.gov.in/statist

16.
Dutta S. K. , & Singh K. (2011). Livelihood diversification:: Case study of somebackward regions in India. International Journal of Current Research, 3(2), 139–151.

17.
Edwards A. C. , & Ureta M. (2003). International migration, remittances, and schooling: Evidence from El Salvador (NBER Working Papers No. 9766). National Bureau of Economic Research, Inc. https://ideas.repec.org/p/nbr/nberwo/9766.html

18.
Ellis F. (1998). Household strategies and rural livelihood diversification, Journal of Development Studies, 35(1), 1–38.

19.
Fan X. , Fang X. , Liu Q. , & Liu Y. (2009). A social adaptation comparison of migrant children, rear children, and ordinary children. Journal of Beijing Normal University (Social Sciences), 215, 33–40.

20.
Fang X. , Dai L. , Fang C. , & Deng L. (2006) The relationship between parent-adolescent communication problems and adolescents social adjustments. Psychological Development and Education, 22, 47–52.

21.
Francisca M. A. (2012). Gender, educational attainment, and the impact of parental migration on children left behind, Journal of Population Economics, 25, (4), 1187–1214.

22.
Gao J. (2010). Analysis on the self-consciousness of rural migrant workers’ left-behind children and the influencing factors. China Journal of Health Psychology, 18, 58–62.

23.
Ginther D. K. , & Pollak R. A. (2004). Family structure and children’s educational outcomes: Blended families, stylized facts, and descriptive regressions, Demography, 41(4), 671–696.

24.
Govinda R. , & Bandyopadhyay M. (2008). Access to elementary education in India: Country analytical review. National University of Educational Planning and Administration 17-B, Sri Aurobindo Marg.

25.
Giannelli G. C. , & Mangiavacchi L. (2010). Children’s schooling and parental migration: Empirical evidence on the ‘Left Behind’ generation in Albani. Department of Economics, University of Florence.

26.
Guarcello L. , Lyon S. , & Rosati F. C. (2005). Impact of children’s work on school attendance and performance: a review of school survey evidence from five countries. Understanding Children’s Work Project and SIMPOC, UNICEF.

27.
Hu X. , Liu X. , Shen J. , & Fan X. (2008). The effect of social support and coping styles on left children’s well-being. Psychological Research, 1, 34–38.

28.
Hunt F. M. (2008). Dropping out from school: A cross country review of literature. Consortium for Research into Educational Access Transition and Equity (CREATE) [Working Paper No. 16]. University of Sussex.

29.
Jeremy W. L. , Tilda F. , Ronald J. I. , & Bruce G. S. (2010). Parent child communication and substance use among adolescents: do father and mother communication play a different role for sons and daughters? Addictive Behaviors, 35, 426–431.

30.
Jha J. (2007). A workshop report on migration, education and socio-economic mobility. The School of Development Studies, University of East Anglia Norwich.

31.
Jia Z. , Shi L. , Cao Y. , Delancey J. , & Tian W. (2010). Health-related quality of life of ‘left-behind children’: a cross-sectional survey in rural China. Quality of Life Research, 19, 775–780.

32.
Kamper G. (2008). A profile of effective leadership in some South African high-poverty schools. South African Journal of Education, 28(1), 1–18.

33.
Kay K. (2000). Uniquely gifted: Identifying and meeting the needs of the twice exceptional student. Avocus Publishing Inc.

34.
Khatun D. , & Roy B. C. (2016). Rural livelihood diversification in West Bengal: Nature and extent. Agricultural Economics Research Review, 29(2), 183–190.

35.
Li X. , Yuan J. , Gao W. , Luo J. , & Du Y. (2010). Research on negative emotions, behavior, personal relationship caused by left-behind experience of rural children after growing up. China Journal of Health Psychology, 18, 92–94.

36.
Liu B. , & Ouyang W. (2010). A survey of subjective well-being on rural left behind children. Journal of Longdong University, 21, 119–120.

37.
Majumder R. (2011). Paternal migration, child labour and education: a study in brickfield areas of West Bengal [MPRA Paper]. University Library of Munich.

38.
Mangiavacchi L. , Federico P. , & Piccoli L. (2014). Intrahousehold distribution in migrant-sending families [Ecineq Working Paper No. 344].

39.
McDowell C. , & De Haan A. (1997). Migration and Sustainable Livelihoods: A Critical Review of the Literature [IDS Working Paper 65]. Institute of Development Studies, University of Sussex.

40.
McKenzie D. , & Rapoport H. (2011). Can migration reduce educational attainment? Evidence from Mexico. Journal of Population Economics, 24, 1331–1358. 10.1007/s00148-010-0316-x

41.
McKenzie D. , & Rapoport H. (2006). Can migration reduce educational attainment? Evidence from Mexico? [World Bank Policy Research Working Paper 3952].

42.
Mohapatra P. P. (1985). Coolies and Colliers: A study of the agrarian context of labour migration from Chotanagpur 1880–1920. Studies in History, 1(2), 297–298.

43.
Nichols J. D. , & White J. (2001). Impact of peer networks on achievement of high school Algebra students. Journal of Educational Research, 94(5), 267–273.

44.
NSSO 64th Round (2007–2008). Employment and unemployment situation in India. Report no. 531 (64/10.2/1). The Ministry of Statistics and Programme Implementation, Government of India.

45.
Palchaudhuri M. , & Biswas S. (2013). Analysis of meteorological drought using standardized precipitation index-A case study of Puruliya District, West Bengal, India. International Journal of Environmental and Ecological Engineering, 7(3), 167–174.

46.
Powers D. A. (1996). Social background and social context effects on young men’s idleness transitions. Social Science Research, 25, 50–72.

47.
Rao U. (1994). Palamoor labour: A study of migrant labour in Mahbubnagar District. C. D. Deshmukh Impact Centre. Council for Social Development.

48.
Ren N. , & Shen L. (2008). A survey on loneliness status of children left in rural areas in primary school. China Journal of Health Psychology, 16, 754–756.

49.
Sampson W. A. (2002). Black student achievement: How much do family and school really matter? The Scarecrow Press.

50.
Sankar D. S. (2017). Seasonal migration and exclusion of children in school education, International Journal of Advanced Educational Research, 2(4), 1–7.

51.
Santrock J. W. (1972). Relation of type and onset of father absence to cognitive development. Child Development, 43, 455–469.

52.
Sjaastad L. A. (1962). The costs of returns of human migration. Journal of Political Economy, 70, 80–93.

53.
Smita, (2006). Locked homes, empty schools. Zubaan.

54.
Su S. , Li X. , Lin D. , Xu X. , & Zhu M. (2012). Psychological adjustment among left-behind children in rural China: the role of parental migration and parent-child communication. Child: Care, Health and Development, 39(2), 162–170.

55.
Sylvie D. (2015). Migration and families left behind. IZA World of Labor, 144, 1–10. 10.15185/izawol

56.
Thomas D. (1994). Like father, like son; like mother, like daughter: Parental resources and child height. Journal of Human Resources, 29(4), 950–988.

57.
Wen M. (2008). The effect of family structure on children’s health and well-being: evidence from the 1999 National survey of America’s families. Journal of Family Issues, 29, 1492–1519.

58.
Zhang L. , Tang R. , Hu Y. , & Xu D. (2006). A study on subjective well-being of children stayed in the country and parenting styles. China Journal of Health Psychology, 14, 368–370.