Page 25 - IJPS-3-2

Upadhyay AK et al.

2.3 Key Exposure Variables
The key exposure variables of interest were household forced migration (yes versus no) and the social support (low versus
medium/high) received by the households in Wave 1 (children aged 1 year).
The variable of forced migration was created using two questions asked in the first wave of YLS. The survey asked “Since
pregnancy of the index child, have there been any big changes or events (natural disaster; moved/migrated/fled; decrease
in food availability; deaths of livestock; crop failure; deaths of household members; severe illness/injury and victim of
crime) that decreased the economic welfare of your household? If the answer was yes, then the survey further asked “What
did the households do in response to the big changes/events?”. Out of the 2,011 households/children recruited in Wave
1, the information on the migration status was available only for 1,913 households/children in wave 3. A total of 6.2%
households (n = 119) reported that they had migrated since the pregnancy of the index child due to the aforementioned
catastrophic events in Wave 1. It is worth noting that the present study used the term forced migration as defined by the
International Organization for Migration (IOM) (Laczko and Aghazarm, 2009).
The information on social support (economic support and emotional support or assistance) was also collected in the
survey. The survey asked the women in the previous 12 months they had received any kind of economic or emotional help
or assistance from a work-related/trade union (yes, no), community association/co-op (yes, no), women’s group (yes, no),
political group(yes, no), religious group (yes, no), credit or funeral group (yes, no), sports group (yes, no), family (yes,
no), neighbourhood (yes, no), friends (yes, no), community leaders (yes, no), religious leaders (yes, no), politicians (yes,
no), government officials/civil service (yes, no), charitable organizations/NGOs (yes, no), and any other group (yes, no).
If a woman had received any kind of help, her reply was coded as ‘1’; otherwise, it was coded as ‘0’. Help received from
each group/person was added so that the amount of help received ranged from 0 to 16. If a woman reported no help or
assistance, it was regarded as ‘low’ social support. Support ranging from 1 to 4 was considered as ‘medium’ social support
and from 5 to 16 as ‘high’ social support. The details of the social support measurement are presented elsewhere (Galab et
al., 2003).

2.4 Other Variables

Past studies on the subject of child cognitive well-being have included a number of other socio-economic, demographic
and residence-related variables. Accordingly, we included birth size (below average, average and above), preterm birth (full
term, preterm), sex of child (male, female), ever breastfed (no, yes), serious illness/injury in Wave 1 (no, yes), stunting
at Wave 1(not stunted, stunted), child immunization (no/partial, full), pre-schooling(no, yes), type of school (private,
government), mother’s height (in cm), mother’s education (below primary, primary and above), mother’s age at birth of
child (<18 years, 18–24 years, 25–29 years, ≥30 years), mother’s working status (not working, agricultural work, other),
household head education (below primary, primary and above), sex of household head (male, female), drinking water
(improved, non-improved), toilet facility (improved, non-improved), wealth index in Waves 1 and 3 (poor, middle, rich),
religion (Hindu, Muslim, other), caste (Scheduled Tribe (ST), Scheduled Caste (SC), other backward classes (OBC),
other), and place of residence (rural, urban).
The survey collected data on the respondent’s (mother/caregiver) perception of the size of the baby at birth. The YLS
asked the respondent if the child was very small, small, average, large or very large at birth? Very small or small size at
birth was coded as ‘below average’, and average, large or very large size at birth was coded as ‘average and above’.
During the first wave of the survey, the mothers/caregivers were asked whether the child had suffered from any serious
illnesses or injury since birth which led the mother/caregiver to think that the child might die (Yes/No/Don’t know). If the
mother replied with a ‘yes’, then serious illness was coded as ‘yes’; otherwise it was coded as ‘no’.
In the second wave of the YLS, the survey asked to the mothers whether the child had received vaccinations for
tuberculosis (BCG); diphtheria, whooping cough (pertussis) and tetanus (DPT); poliomyelitis (Polio); and measles.
Children who had received all the afore-mentioned vaccines were coded as ‘fully immunized’. The remaining children
were coded as ‘no/partially immunized’.
The wealth index was calculated using the wealth scores, which are already computed and given in the YLS dataset.
The wealth scores were generated through principal component analysis conducted on a set of variables including
household assets (including radio, refrigerator, bicycle, television, motorbike/scooter, car, pump, sewing machine, mobile,
phone, landline telephone, fan, almirah, clock, table, chair, sofa, bedsheet and animals), household quality (including

International Journal of Population Studies | 2017, Volume 3, Issue 2 19