Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia Yana van der Meulen Rodgers Rutgers University January 17, 2012 Gender CoP and Health CoPAsian Development BankThe views exp ressed in this paper are the views o f the author and do not necessarily reflect t he views or p olici es of the Asian Development Bank (ADB), or it s Board of Governors, or the go vernments they represent. ADB does no t guarantee the accuracy of the data included in this paper and accepts no responsibilit y for any conse quence of their use. The countries list ed in this paper do n ot imply any view on ADB's p art as to sovereignty or independen t status or necessarily conform to ADB's terminology.
27
Embed
Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
Women's Employment and Children'sNutritional Status:
New Evidence from South and SoutheastAsia
Yana van der Meulen Rodgers
Rutgers University
January 17, 2012
Gender CoP and Health CoP
Asian Development Bank
The views expressed in this paper are the views of the author and do not necessarily reflect the views or policies of the
Asian Development Bank (ADB), or its Board of Governors, or the governments they represent. ADB does not guarantee
the accuracy of the data included in this paper and accepts no responsibility for any consequence of their use. Thecountries listed in this paper do not imply any view on ADB's part as to sovereignty or independent status or necessarily
conform to ADB's terminology.
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
• Socioeconomic status and time spent with children operatethrough intermediate variables (“proximate determinants”)that directly influence child birth outcomes and children’snutritional status– Environmental hazards (e.g. mothers with higher income more likely
to live in residences with improved facilities that protect children
from infectious agents)– Fertility patterns (e.g. higher socioeconomic status mothers more
likely to have longer birth intervals and fewer children, associatedwith more favorable nutritional status for children)
– Utilization of health services (facilitated by higher income, but lack of time could be a constraint)
– Feeding practices (also influenced by income and time)
• Primary contribution of this book is to provide newempirical evidence on the potentially competing effects of maternal employment on children’s health using householdsurvey data from 9 Asian countries
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
• New evidence for developing countries using recent (2005-2009) waves of the Demographic and Health Survey for:
– South Asia: Bangladesh, India, Maldives, Nepal, Pakistan
– SE Asia: Cambodia, Indonesia, Philippines, Timor-Leste
• Focus on a single region makes the execution of theempirical analysis more manageable
– Also ensures that sample encompasses countries with broadlycomparable geographical, historical, and cultural contexts(vs. using data that cover countries from different developingregions)
• Approach: Bivariate analysis and a set of multivariate
nested models• Also test for differences by sex of the child
– A priori, expect greater birth size for boy children (biologicalreasons), and greater likelihood of stunting and wasting for girlchildren in countries with son preference (especially India)
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
• Empirical, theoretical work suggest complex set of channelsthrough which maternal employment affects child health status– Use a “proximate determinants” framework that links socioeconomic
• DHS Surveys for 9 Asian countries, 2005-09– Large nationally-representative samples of women between
the ages of 15 and 49, matched with children ages 0-59months
– Part of collection of DHS datasets on population, health, HIV,and nutrition for 85+ countries (http://www.measuredhs.com/)
– Surveys conducted and coded by national governments andprivate agencies
– Use the children’s recode (merge in variables from householdmember recode)
– Limit sample to children with recorded measurements forchildren’s birth size, height, and weight, and no missingvalues for other variables. Sample size ranged from 2,347children in Maldives to 40,676 in India
– Eight countries in birth size sample (all except Bangladesh);six countries in stunting/wasting sample (all except Indonesia,
• Measurement of Small at Birth– DHS has subjective measure of birth size and objective
measure of birth weight
– Subjective: mothers asked if child was very small, smallerthan average, average, larger than average, or very large atbirth
– Small at birth = very small or smaller than average
– Subjective measure included in DHS b/c majority of babies inlow-income countries not weighed at birth
– Lack of objective birth weight measures correlated withsocioeconomic status, in project sample and in other studies
– Subjective birth size correlates fairly well with objective birthweight for most countries, in project sample and in largercomparison of 62 DHS datasets by Blanc and Wardlaw (2005)
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
• Next 3 figures illustrate prevalence of poor nutritional statusin Asian countries compared to other regions– Data points for other countries from published DHS final reports
for individual countries that used the new 2006 WHO ChildGrowth Standards
– On average, prevalence of small birth size in S/SE Asia is
comparable to SSA (about 16.5% on average), less than LatinAmerica/Caribbean, and higher than N. Africa/W. Asia/E. Europe
– On average, prevalence of stunting in S/SE Asia is substantiallyhigher than the averages for other regions
– On average, prevalence of wasting in S/SE Asia is somewhathigher than the averages for other regions
– In all regions, incidence of wasting is considerably lower thanincidence of stunting, consistent with previous studies
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
Implementing the Framework: Data & Method• Key independent variable: maternal employment
– Consistent with DHS coding, women’s employment categorized as:1. Not employed at all in past year (reference category)
2. Employed in past year but not currently
3. Currently employed
– More finely tuned measure of maternal employment. Did tryrobustness tests with binary measures (currently vs. not currentlyemployed; employed now or in past year vs. not employed at all)
– Survey does not ask if mother employed while pregnant; next bestalternative for birth size analysis is to do estimates for sub-sample of children <1 yr old
Estimation Results: Small at Birth• Employment in the past year but not currently has beneficial
effect for 1 country and adverse effect in 2 countries, andcurrent maternal employment has beneficial effect in 2countries and adverse effect in 2 countries
• Estimations for sub-sample of children under the age of one(to better ensure that maternal employment covers period of
pregnancy) yields two stat. significant results:– India: current maternal employment 4% reduction in likelihood of
small birth size relative to children whose mothers did not work at all inthe past year. Implies that mothers’ earned income during pregnancyhas benefits beyond SES variables and contributes to improved prenatalhealth and healthy birth outcomes.
– Nepal: current maternal employment 11% increase in likelihood of
small birth size. Extremely high proportion of women perform physicallytaxing work in agricultural jobs with compensation that is commonly non-cash based. Results imply negative effects on prenatal health and birthoutcomes.
• In all but 2 countries, boys substantially less likely to be smallat birth than girls, even after adding all control variables
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
• Bivariate results point to a greater prevalence of stuntingamong children of employed mothers– In low-income countries, women of lower socioeconomic status
have little choice but to engage in market-based work in order tosupport their households
– Apparent association between greater stunting and maternalemployment could be explained by the relatively lowersocioeconomic status of employed women rather than women’stime spent away from children
• Multivariate results support this assertion: once full set of SES and household characteristics are included, maternalemployment loses its statistically significant association with
risk of stunting. See figure for case of current employment• Risk of stunting is slightly higher for boys than girls in 2 of
6 economies: Cambodia and Timor-Leste– Hence no evidence of son preference in this long term indicator
of childhood malnutrition, not even in India
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
Estimation Results: Children’s Wasting• Multivariate results: current maternal employment leads to
roughly a 3% increase in the likelihood of child wasting inBangladesh, Maldives, and Timor-Leste, even after includingthe full set of control variables
• Wasting, an indicator of short-term nutritional deprivationthat tends to reflect recent illness or trauma from
catastrophic events, is more likely to cut across householdsof different socioeconomic status– Bangladesh: severe flooding
– Timor-Leste: lingering after-effects of long-term violent conflict
• Adverse effect of women’s employment in these threecountries may be capturing the opportunity cost of women’stime away from children during times of heightened need
• Two countries have a higher risk of wasting for boyscompared to girls: Timor-Leste and India
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
• Overall, once SES is controlled, household composition andenvironmental factors play relatively small roles in explainingthe association between mother’s employment and childnutritional outcomes
– None of the other variables related to hh composition andenvironmental factors appear to mediate the association betweenmaternal employment and children’s nutritional outcomes as muchas socioeconomic status
• The most important SES indicators for all 3 measures of children’s nutrition are maternal education, mother’sinformal access to information (especially reading
newspapers), father’s education, and household wealth– Providing mothers with nutrition knowledge outside of classroom
can contribute to child health even in areas where women have lowlevels of formal schooling
– Consistent with other studies in the literature
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia
Take Home Points• No evidence of son preference in the indicators of long-term
and short-term nutritional deprivation
– Consistent with earlier conclusion of a discrepancy betweenevidence of son preference and lack of bias against girls inanalyses of child nutrition surveys. Possible explanations:
• Potential lack of country-wide discrimination against girls
• Data covers surviving daughters; translation of son preferenceinto selective abortions not captured by data
• Girls may cope with food deprivation differently than boys
• Favoring boys with long breastfeeding duration, at the expense of solid foods, may potentially hurt them
• Son preference may be translated into health care seekingbehavior that favors boys, rather than feeding practices
• Mothers may favor young boys by carrying them around in thekitchen longer, with longer exposure to cooking fuel toxins
8/3/2019 Women's Employment and Children's Nutritional Status: New Evidence from South and Southeast Asia