PROTOCOL: The effect of interventions for women's empowerment on children's health and education: A systematic review of evidence from low- and middle-income countries

Abstract

Background

The problem, condition or issue

While numerous facts and figures point towards the increasing opportunities for women and their growing participation in economic, political and public decision-making processes, women remain in disadvantaged positions compared to men in many places of the world (World Economic Forum, 2015). Gender bias is deeply embedded in cultures, economies, and political and social institutions around the world, denying women an equal part in society and decision-making. Gender based discrimination in access to economic opportunities, lack of representation and participation in economic and social spheres, and limited opportunities to accumulate resources could perpetuate vulnerability to poverty among women, limit human capital accumulation and restrict economic growth.

Women's economic involvement, their political participation, and their empowerment in the two crucial domains of education and health, are often seen as the four fundamental factors to create just and equitable societies where both women and men have control over their lives and exert similar, if not equal, influence in society. The Global gender gap report of 2015 uses the aforementioned four domains as the supporting pillars of their gender gap index, which they established as a tool to proxy for the status of women's empowerment since 2006. This index utilizes the information on the opportunities to equal economic and political participation, as well as access to health and education facilities and resources, to determine the gap between males and females in each country, within six different geographical regions. They find that in all aspects, school enrolment and attendance rates, political participation and earnings and labour force participation women are lower than men (World Economic Forum, 2015).

Although there has been a large improvement in the overall gender gap, regional trends vary vastly. Where most countries have improved, some have stagnated and a few have even deteriorated. On average, North America seems to have performed the best at reducing the gender gap as shown in the gender gap index, followed by Europe and Central Asia, while the Middle East and North Africa region seems to be the worst performing. Although this overall ranking also seems to persist in economic participation and opportunities gap, but in terms of political empowerment, the Asia and pacific region performs much better than Europe and central Asia. Disparities in health and survival have been reduced massively or eliminated entirely in all regions, besides those in the Asia and pacific region where nonetheless more than 90 per cent of the gap has been closed. Sub Saharan Africa performs the worst in terms of educational attainment, far behind any of the other regions.

There appear to be different forces that drive these gaps and their persistence within each region, and these are not limited to low and middle income countries. For example, compared to the gender gap index of 2006, even within the Nordic countries, which have the smallest gender gaps, Iceland has overtaken Sweden by reducing the gap at the rate of 10 per cent, while Sweden has been moving at a glacial pace of around one per cent. On the other hand, in Asia, Nepal and in Latin America, Nicaragua and Brazil are all low income countries that have reduced these gender gaps at outstanding rates, comparable to those of Iceland. Simultaneously though, this swift decline in the gap is not observed across all low and middle income countries. Sri Lanka, Jordan, Mali, Croatia and Slovak Republic are all countries that have widened their gender gaps over time, while Iran has not moved from its 58 per cent gender gap since 2006.

When examining each of the four components of the gender gap, there appear variations between the components that have made most progress in terms of reducing this gap. Gender gaps in economic participation and opportunity are the widest, indicating the wide gap in wages and unequal professional, skilled and technical participation between men and women. Despite the improvements in higher education, where female participation is higher than males in 97 of the 109 countries, it is men who make up the majority in skilled workers in over 60 per cent of the countries. Even if women's political participation has been rising the sharpest since 2006, their economic participation has arrested since 2009/10, where women in 2015 finally achieved the same level of earnings as men did nearly in 2005, 10 years ago.

The Global gender gap report is a clear indication of the intention of all international and national entities towards the measurement of gender equal goals and overall women's empowerment. Another convincing indication of its gaining prominence is visible in Goal 5 of the new Sustainable Development Goals (SDGs). This Goal sets out to “achieve gender equality and empower all women and girls” (United Nations, 2015). As a result, there have been many measures that researchers and politicians are following, and critically evaluating, to determine the gaps between females and males. These may be the aforementioned gender gap index, the Gender Development Index (GDI), the Gender Empowerment Measure (GEM), the Gender Inequality Index (GII) (UNDP, 1995, 1996) or the Social Institutions and Gender Index (SIGI) (Branisa, Klasen, & Ziegler, 2009). While the former rely on the gender inequalities within the four (or fewer) domains that comprise the gender gap index (apart from the standard of living component within the GDI and GEM), the SIGI differentiates itself by incorporating the legal and societal norms and practices that are the underlying cause of the self-same inequalities. Therefore, via widespread literature and research on the same, the concept of women's empowerment is not only gaining prominence and relevance in academic articles and growth related debates, but also in the overall international development and policy agenda.

As Malhotra et al. (2002) point out, women's empowerment has certain unique characteristics that distinguish it from the disempowerment of other disadvantaged or socially excluded group. One of these elements is that women are not just any other socially disadvantaged or excluded group among society, but rather a category of individuals that have a large overlap with each of these other disempowered and disadvantaged groups. Secondly, it is often household and interfamilial relations that are what disempower women in the first place and form the crux of the discrimination towards them. Therefore, familial relations are generally one that accentuate the problem, and this domestic disparity is another key point that differentiates this group from the others. Lastly, while most discriminations are borne of particular institutions prevalent in society, this discrimination is fundamentally driven by those supporting patriarchal structures. This inherently means that any move towards female empowerment would not only require an overhaul of the official institutions, as is important for empowerment of any marginalized group, but particularly those that foster patriarchal structures. These three crucial differences emphasize the nature and magnitude of the problem and the formidable measures that are required for any positive change.

While gender equality is an end in itself, there is also the expectation that empowerment of women in all sectors of life will build stronger economies and propel growth, improve quality of life and help to achieve equality of rights and actions for half of the world's population. One potential channel through which women's empowerment might affect growth, which is a key determinant of economic growth in itself, is child development, particularly child health and education. While both of these are important channels in and as of themselves, to contribute towards a rising growth trajectory, within this review, we intend to link the two. The importance of child development outcomes for economic growth itself has been discussed in several strands of literature, thereby stressing its role in promoting growth. David Barker posited the foetal origins hypothesis in epidemiological studies, which connects chronic, degenerative conditions of adult health, including heart disease and type 2 diabetes, with health circumstances decades earlier, especially in utero nutrition. Economists have expanded on this hypothesis and investigated a broader range of shocks in early life that impact later-life outcomes, including test scores, education attainment, income and health (Almond & Currie, 2011). James Heckman, Paul Gertler and others also published several studies investigating the impact of early learning and development programmes for disadvantaged children. Results suggest a significant impact on skill levels, health outcomes, overall workforce strength and economic growth over a longer time period (Campbell et al., 2014; Conti, Heckman, & Pinto, 2015; Cunha & Heckman, 2009; Doyle, Harmon, Heckman, Logue, & Moon, 2013; Doyle, Harmon, Heckman, & Tremblay, 2009; Gertler et al., 2013). The goal of this review is to systematically assess the evidence that investigates the effect of interventions that empower women on child development outcomes. Therefore, given the prerequisite of women's empowerment, we will assess exactly how child development outcomes are thus affected.

The intervention

For the purpose of this review, the definition of empowerment is one that we adopt from Kabeer (1999). It states empowerment as “the expansion in people's ability to make strategic life choices in a context where this ability was previously denied to them; a process that entails thinking outside the system and challenging the status quo, where people can make choices from the vantage point of real alternatives without punishingly high costs.” This inherently includes several measures, such as women's improved access to health and education, their financial, political and economic empowerment and overall legal and institutional amendments.

Programs that address women's economic empowerment are very diverse both in terms of intervention and in terms of the institutional framework in which they are implemented (Buvinic, Furst-Nichols, & Pryor, 2013). For instance, interventions have focused on promoting women's ability to secure decent jobs, improved access to education, provision of health care, support of women as entrepreneurs (i.e. financial assistance, training), promotion of women to participate in the labour market, changing laws that give women access to land, inheritance and property rights and those give women political decision power. Those interventions that empower women and that directly or indirectly have a second-generation impact are those that we are interested in. The latter might be represented by an intervention that increases women's exposure and access to part time work, job sharing and consequently access to childcare options. This intervention would directly affect women but also indirectly have an impact on the next generation.

How the intervention might work

There are several long-term and short-term interventions possible to arrive at better child development outcomes. Increases in economic and livelihood options for women can be a consequence of improvements in labour rights and laws, leading to improvements in women's household income, increased food intake and schooling consumption for children. Access to microcredit or conditional cash transfers and other such governmental subsidies could lead to increasing access to resources and credit, both of which are forms of intervention which have been shown to lead to a higher allocation of resources onto children's health and education (Blumberg, 1988; Duflo, 2003; Haddad & Hoddinott, 1994; Hoddinott & Haddad, 1995; Lundberg, Pollak, & Wales, 1997; Pitt, Khandker, Chowdhury, & Millimet, 2003; Qian, 2008; Quisumbing & Maluccio, 2003; Rangel, 2006). This increased access might also affect women's agency itself, implying increased autonomy, and self-sufficiency, translating the same values onto the girl child, and affecting child development outcomes (Ghuman, 2003; Kabeer, 1999; Shroff, Griffiths, Adair, Suchindran, & Bentley, 2009; Smith, Ramakrishnan, Ndiaye, Haddad, & Martotell, 2003; World Bank, 2014). These, however, are more long terms consequences of an improved political, ideological and economic environment for a girl child to grow in. Increased political opportunities for women could also lead to improvements in infrastructure and capacities related to children, such as more schools, better roads, closer access to water and increased supply of electricity and other such infrastructure and facilities (Chattopadhyay & Duflo, 2004; Duflo, 2005). A lot of these facilities might not only affect women's role and aspirations in rural villages and households, but also young girls, leading to an overhaul in the institutional structure and hierarchies existent within the society (Beaman, Chattopadhyay, Duflo, Pande, & Topalova, 2009; Beaman, Duflo, Pande, & Topalova, 2012). Better health for the mother may swiftly improve child health outcomes already from in utero stages (Almond & Currie, 2011; Osmani & Sen, 2003). Although these interventions might be implement over a short or medium term time frame, their impact can even expand over longer time frames by creating an enabling environment on account of sustained improvements in schooling, health and sanitation facilities, improved access to resources and relevant technology and a dissolution of the patriarchal mesh of society.

On the other hand, these improvements in social, economic and political opportunities could result in increased tensions within the households leading to a larger incidence of domestic violence (Eswaran & Malhotra, 2011; Heath, 2014; Macmillan & Gartner, 1999). A number of studies (also meta-analysis) have already examined the impact of child exposure to domestic violence, and its harmful psychological and cognitive effects (Fantuzzo & Mohr, 1999; Kitzmann, Gaylord, Holt, & Kenny, 2003; Osofsky, 2003). Moreover, with more economic or political participation, there might be lesser time allocated towards child rearing and caring activities, which has been shown to affect cognitive abilities of children and child development in the early years after birth (although these are generally studies in high income country context) (Baum II, 2003; Baydar & Brooks-Gunn, 1991; Brooks-Gunn & Duncan, 1997; Desai, Chase-Lansdale, & Michael, 1989; James-Burdumy, 2005; Ruhm, 2004, 2008; Waldfogel, Han, & Brooks-Gunn, 2002). It might also cause changes in the household and community structures, perhaps regressively (Luke & Munshi, 2011). Moreover, taking up work may expose the mother to health hazards (occupational injuries and diseases), which could in turn affect the child adversely, given that women's health is often synonymous with maternal health (Paul-Majumder, 1996; Pick, Ross, & Dada, 2002). Other channels could be via access to credit and financial services, which could result in increased indebtedness of the household, harmfully affecting child development (Mader, 2013; Schicks, 2014; Taylor, 2011). Education and health spending on children might be spread disproportionately, for e.g. female child could receive a smaller share of the investment than a male child. Furthermore, social norms regarding adequate gender roles, as well as the policy and resource environment, might in general limit the impact of the programs.

The theory of change that guides our analysis has been attached within the Appendix. It is based, to some extent, on the UNICEF framework that has been adopted by Ruel (2008), although some modifications have been made to include the education outcomes that are also of interest to us, besides the child health outcomes.

Why it is important to do the review

Despite the large attention that women's economic empowerment has as a policy objective, to date, this research appears largely scattered and the few existing literature reviews have focused on limited number of articles, a limited time span, or a non-systematic assessment of available published and unpublished studies (Bandiera & Natraj, 2013; Burger, 2010; Buvinic & Furst-Nichols, 2014; Duflo, 2012; Yoong, Rabinovich, & Diepeveen, 2012).

We intend to cover the gaps in these studies in three key ways:

To our knowledge, there is very little evidence cataloguing the link between women's empowerment and child development outcomes. So far, the literature examines the deficits that are present within adults, given the differential circumstances during their birth as well as growth period. These studies have also addressed various channels and potential mechanisms where female empowerment would affect child development outcomes.

Reviews so far have focussed on maternal and child nutrition and the relation or co-incidence between the two (Bhutta et al., 2013; Bold, Mara, Quisumbing, & Gillespie, 2013; Cunningham, Ruel, Ferguson, & Uauy, 2015), and we intend to extend this by including other health as well as education related outcomes for children. Some reviews have added onto the literature via a particular intervention and its impact on child health: participatory learning activities, for instance (Prost et al., 2013). Other reviews have focussed explicitly only on the review of the literature and disregarded any meta–analysis (Laverack, 2006). This review will attempt to examine the relative importance of each channel and how effectively can policy address the issues that inhibit child development outcomes due to particular female empowerment policies.

Another contribution is the methodological approach of a systematic review. This work would entail collection, appraisal and synthesizes of the evidence based on a systemised process. This implies a more rigorous screening of the evidence, more systematic and demanding in terms of the exact conditions surrounding the intervention, the outcomes, the time period and the country coverage.

The Campbell Collaboration has published systematic reviews and meta-analyses of international development interventions which explicitly aim to empower women - for example microcredit (Vaessen et al., 2014) and self-help groups (Brody et al., 2015) - or those which may have consequences for women - for example, land reform (Lawry et al., 2014). However, this review is the first systematic review to examine the comparative effects of programmes aiming to empower women on outcomes for the subsequent generation. The review is being conducted alongside a review by Ibanez et al. (2017) examining the effectiveness of community-level women's empowerment interventions on household wellbeing.

The third point where we intend to contribute to the literature on the particular relation between female empowerment and child development outcomes is that this study will conduct a meta-analysis with the data extracted from the selected quantitative studies. The meta-analysis will comprise a risk of bias assessment and synthesis of the effect sizes to provide a quantitative estimation of the effect.

Objectives

The primary objective of this review is to answer the following three research questions: 1.

What is the quantitative impact of interventions aimed at mothers on child development outcomes? Given our outcome variable of child health and education, the intergenerational impact of all interventions, which were specifically targeted at mothers, will be studied.

What exact channels are relevant to translate women's empowerment into improved outcomes for child's health and education? This intervention might have been intending to empower women, via changes in women's bargaining power, household decision making, labour force participation, time use, income, education and health status, level of domestic violence incurred etc. Alternatively, it might lead to a lower amount of time spent on child rearing activities, or decreased bargaining power, thereby leading to increasing domestic violence. Therefore, it is important to establish the link between these interventions and women's empowerment, before determining their role on child development outcomes.

What are the institutional barriers and facilitators that might prevent or enable a transformation of these increasing opportunities for women into greater improvements for child outcomes? These institutions refer to those which are national, community level and household level. For instance, any legal reforms that affect the polito-economic structure at the national level, or might affect any changes in the community and familial structures, such as paternal involvement in childcare or delayed age of marriage, are all expected to bring changes in child development outcomes.

Methodology

In this section we list the study design inclusion criteria, the search methods and the approach to data collection, critical appraisal and synthesis.

Inclusion criteria

Types of studies

Based on the research design, we categorize the studies into two major groups:

a. Controlled experimental designs and natural experiments

These type of studies specifically use a random assignment of the intervention to the treatment group, and evaluate the effect by comparing the outcome with the control group and by using an appropriate methodology.

b. Quasi-experimental designs

In cases whether the assignment of the treatment is not random, various quasi-experimental designs are used to evaluate the treatment effects. These methods include regression discontinuity design (RDD), instrumental variable (IV), difference-in-differences (DID), fixed effects analysis, propensity score matching, and interrupted time series (single country) designs. In the absence of randomization, the main challenge is that the treatment can be endogenous, i.e. it can be confounded with unobservable factors correlated with both the treatment and the outcome.

We include studies that address this potential endogeneity issue through any of the aforementioned methods. We include natural experiments and quasi-experimental designs to account for instances where it is impossible to randomize (e.g. a country-wide policy change) or unethical (e.g. certain types of outcomes like child mortality), and due to the inclusion of both intended and unintended (including adverse) outcomes. The includable non-randomized studies will use credible methods that take into account confounding.

We exclude studies that do not attempt to evaluate a causal effect of the intervention on the outcome using these study design. Thus, we exclude any qualitative study or any report that has purely descriptive analysis. We also exclude studies that use data on multiple countries for cross-country time series or panel analysis.

Types of participants

We include interventions where the participants are women and girls (irrespective of age) from low- and middle-income countries as defined by the World Bank categorization at the time the data were collected. Thus, we exclude studies of interventions in high-income countries. Programs where men also participated, but the intervention was aimed at promoting women's economic empowerment, are also included.

Types of interventions

We will include studies on interventions that are either solely aimed towards women's empowerment, or have a gender-differential component that can potentially improve women's empowerment. We provide examples for each kind of interventions. However, this list is not exhaustive and we shall consider any other intervention that is instrumental in enhancing female economic empowerment and falls under the scope of this review. a

. Enhance women's labour market opportunities This category will include programs like subsidies, vouchers, child care, elderly care, training programs, job placements, etc. which create better opportunity for women's participation in the labour market. This will also include programs that indirectly create an environment conducive for women's participation in the labour market. For instance, a boom in horticulture exports in Senegal created a dramatic increase in female off-farm wage employment, which resulted in greater bargaining power of women in the household and hence children's schooling (Maertens & Verhofstadt, 2012). Another example is India's National Rural Employment Guarantee Act (NREGA), which is primarily a public works program, but has a gender component through targeting of at least one third of its beneficiaries to be women, and equal wage rates for male and female. This program has been found to give women greater labour market opportunities, helping women to have higher decision making power within household, which subsequently improved children's education (Afridi, Mukhopadhyay, & Sahoo, 2016).

. Reform the law to give women access to land and inheritance rights These interventions can include giving women equal inheritance rights through a change in the law, or they can be due to land registration/redistribution rights and land titling programs to include women farmers as owners or co-owners along with men. Some of the examples in this type of interventions are Vietnam's 1993 Land Law that gave land titles to women and thus had a positive effect on children's health and education (Menon, van der Meulen Rodgers, & Nguyen, 2014); and in the Indian context, changes in the Hindu Succession Act, which granted daughters equal coparcenary birth rights in joint family property that were denied to them in the past (Deininger, Goyal, & Nagarajan, 2013).

. Increase participation of women in decision making in the household, firms or politics This type of intervention would aim at promoting participation of women in the decision making process in various socio-economic spheres such as firms, households, or politics. This can include policies such as gender based affirmative action to create more opportunities for women to take up leadership positions.

. Give better access to information regarding women's rights This includes interventions such as providing information about land rights and entitlements to women farmers, generating awareness about any gender based affirmative actions that may be in place, or existing laws related to women's rights such as prevention of domestic violence and other gender issues.

We will not consider studies that might improve children's health or education outcomes but do so through a channel other than women's empowerment. Therefore, we exclude interventions targeted directly on children's education or health outcomes, unless they have an effect on the second generation through the channel of women's empowerment (girls in this case). Thus, interventions such as cash transfers conditional on school enrolment/attendance, school level interventions towards raising effectiveness of teaching or teacher attendance etc. will not be included. Similar restrictions apply for interventions targeting children's health, such as deworming programs, provision of supplementary nutrition etc.

Note that some of these programs may indeed have a gender component; for instance, interventions that target girls’ education (e.g. provision of bicycles to girls in secondary schools) may indeed lead to some form of female empowerment. However, unless a study reports the effect of such program on the second generation, it will not be considered for our systematic review.

Some interventions may have unintended effects on women's empowerment even though the design of the program has neither any objective for women's empowerment nor any gender based component. We will exclude such programs, unless the intervention specifies any component which differentially targets men and women beneficiaries, the female empowerment effect will be considered as unintended consequence of the intervention. Hence such intervention will not be eligible for inclusion in our study.

Comparisons

We include studies that have a clearly defined comparison group for evaluation of the treatment effect. When the intervention aimed at female empowerment is targeted towards a set of households, then the comparison group would consist of households which are not exposed to this treatment. Since the intervention can also be at various other levels, such as individuals, communities, or regions, therefore the comparison will be made with the corresponding group that has not been affected by this specific intervention. The comparison group is considered to be not affected by the intervention, either when there is no intervention, or when they are still in pipeline while the study was conducted, or they are in a business as usual situation. The idea of comparison group is to capture the counterfactual scenario, therefore the comparison group has to be sufficiently similar with the treatment group. This will be reflected through baseline comparison (pre-test differences) of observable characteristics (including the outcome variable) between the treatment and the comparison groups. We shall also include direct comparison of different interventions (active controls), as it constitutes an important part of the comparative systematic review. Where a study includes more than two intervention arms, we will include in the review only the intervention and control/comparison arms that meet the eligibility criteria.

Types of outcome measures

a. Primary outcomes:

We focus on the effect of these interventions on child development, particularly on children's health and education.

Health: The outcomes are some indicator of child health, e.g. mortality, morbidity, stunting, underweight, wasting, immunization, etc. More specifically, we shall consider child level measures such as height-for-age z-score (stunting), weight-for-height z-score (wasting), weight-for-age z-score (underweight/overweight), body mass index, etc. In some cases the child level observations can be aggregated at the household level and the reported measure can reflect the percentage of children who are stunted/wasted/underweight/immunized. We shall also include them as outcome variables.

Education: We include outcomes related to children's human capital, including investment that improve children's human capital (e.g. expenditure related to child's education). Specific outcomes related to children's education to be considered are: •

Schooling outcomes: school enrolment, attendance, grade progression, primary (or other levels) completion rates, dropout, years of schooling etc.

•

Cognitive outcomes: test scores (e.g. reading, math, etc.), literacy, cognitive test (e.g. IQ test), performance in board exam etc.

The investment in children's health and education, both in terms of material resources and time, is also considered a primary outcome for the purpose of this review.

It should be noted that the effect of intervention can have both intended (positive) and unintended (negative) effects on the outcomes stated above. For example, an intervention that raises the mother's income can lead to better education and health of children due to greater investment on children. However, in absence of childcare facilities, an intervention that provides labour market opportunity to mothers can actually lead to worse education or health outcomes of children because the mother may not be able to devote enough time for child rearing activities. Studies that capture unintended consequences reflected through the outcomes described above are also included.

b. Secondary outcomes:

Moreover, it is important to mention that the channels through which the interventions specified above may have an effect on children's human capital comprise of another set of intermediate outcomes (mediators) such as women's intra-household bargaining power and time use. In the context of adverse outcomes, it is especially relevant to consider the change in mother's time use pattern due to her improved economic participation. If the mother is unable to spend time in child rearing, then it can have a negative effect on younger children's health and cognitive development, and older children may have to substitute for mother's time. Such mediator outcomes may include various indicators of women's empowerment, such as decision making within household, labour force participation, income, or education level.

While these intermediate outcomes are important, the objective of this study is to focus on the final outcome variables that reflect child's human capital formation, therefore we will include studies irrespective of whether they report intermediate outcomes. However, we shall also collect these intermediate outcomes from those studies that do report them, and include them in the review.

Duration of follow-up

We will include studies of any follow-up duration.

Other inclusion criteria

We shall include studies that are conducted within the time span of 1990–2016. The main reason for this selection is that the study designs that we considered started to be produced/appearing in this field mostly after 1990. As Cameron et al. (2016) points out, beginning mid-1990s, impact evaluation evidence was being published at a steady rate. They find that after the year 2000, experimental and quasi-experimental studies became even more popular and were frequently published in social science journals, and outlets of international agencies, governments, universities, and other institutions. Therefore, given our objective of focusing on experimental and quasi-experimental evidence, it is plausible to restrict inclusion to studies conducted during 1990–2016.

We will include studies irrespective of whether they are published in journals/books, or available as working papers, reports or other.

Search strategy

Electronic search

The search strategy used here is adopted from Hammerstrøm, Wade, & Jørgensen, 2010 and it aims to find both journal publications as well as other publications including working papers. A three-step search strategy will be utilized in this review.

In the first stage, studies will be screened for the text in the title and abstract and the keywords in the article description. The following databases will be searched: •

3ie database of impact evaluations

•

Econlit

•

SSRN (http://papers.ssrn.com/sol3/displayabstractsearch.cfm)

•

EconPapers (http://econpapers.repec.org/scripts/search.pl)

•

Scopus, ASSIA, IBSS

•

MEDLINE, PUBMED

•

Global Health

•

EBSCO Gender Studies Database

•

World Bank - enGENDER IMPACT

•

ERIC

•

Google Scholar (Only for forward and backward citations)

The second stage will consist of searching backward and forward citations, where reference lists of included studies will be searched and reviewed for additional studies. The reference lists of relevant systematic reviews will also be searched, along with important non-systematic literature reviews such as in Handbook of Development Economics. A record will be maintained, describing the databases searched, the keywords used, and search results from each search engine. The search results will be presented using the standard PRISMA (Preferred Reporting Standard for Systematic Reviews and Meta-Analysis) flow diagram, including the reasons for exclusion at full-text stage.

Search terms

For searching the above databases, our team will use search terms based on inclusion criteria, developed through test-searches. The search terms used to search the selected databases are presented in the appendix. These will be modified to fit the search format for each database.

Selection of studies

Titles and abstracts will be reviewed by a team member using the specific inclusion/exclusion criteria to determine whether the study should be included in the analysis. The selection will be done independently by one team member. An assessment of a random sample of studies by each assessor will be carried out as quality assurance.

Description of methods used in primary research

The selection of studies will be done in accordance with the types of interventions and outcomes discussed in the objectives section. For instance, we will include the following studies: “Old-age pension and intrahousehold allocation in South Africa” and “The Impact of Mother Literacy and Participation Programs: Evidence from a Randomized Evaluation in India”. Duflo (2013) and Banerji et. al. (2015) focus on middle income (South Africa) and low income countries (India) respectively, which are a part of our regions of interest. Second, these two studies estimate the quantitative impact of interventions aimed at mothers on child health and education outcomes.

We will exclude studies which do not meet the criteria discussed in the objectives. For instance, “Comparative Study: Impact of Family, School, and Students Factors on Students Achievements in Reading in Developed (Estonia) and Developing (Azerbaijan) Countries” (Shukakidze, 2013). This study is excluded, as it firstly focuses on a high income country, which would not be eligible for inclusion. Second, even for the middle income country, there is no direct intervention for mothers’ empowerment.

Criteria for determination of independent findings

There are a variety of different issues that give rise to issues of dependent effect sizes. For example, there are several publications from one study, or several studies based on the same data set, or studies with multiple treatment arms with only one control group, or studies may report outcome measurements from several time points, or use multiple outcome measures to assess related outcome constructs. In such cases, we cannot treat all outcome estimates as independent of each other (Borenstein, 2009).

Where studies report different outcomes, these will be pooled in separate meta-analyses. Else we will use synthetic effects (weighted mean and standard error) to calculate a single effect size for studies reporting multiple outcomes measuring the same construct or multiple follow-ups. Some more details on this issue are described below.

If there are several publications reporting on the same study we will use effect sizes from the most recent publication. And if there are several studies using the same data set or multiple outcomes are reported within the same study, we will select the study or specification which provides the lowest risk of bias in attributing impact.

Where studies include multiple outcome measures to assess related outcome constructs, we will select the outcome that appears to most accurately reflect the outcome construct of interest, following Macdonald et al. (2012).

For studies with outcome measures at different time points we will separate and synthesize outcomes measured in different time points, following De La Rue et al. (2014). If multiple time points exist within these time periods, we will use the most recent measure.

If the interventions are ongoing programmes, the follow-up will reflect duration spent/being part of in a program, rather than time since intervention. Outcome measures from such studies are reported at different time points, then we will identify the most common follow up period and include the follow up measures that match most to the meta- analysis.

For studies having multiple treatments with only one control group, where the treatments might represent separate treatment constructs, we will calculate the effect size for each pair of treatment versus control separately, and use sub-sample meta-analyses depending on the treatment construct. For different treatments representing the same treatment construct we will follow the procedures outlined in Borenstein et al. (2013, chapter 25) to calculate the weighted mean and standard error for treatments before calculating the effect size for the merged group versus control group.

Details of study coding categories

Data extraction and management

Two team members working independently will extract information from each study included in the review. In this step, data will be extracted and summarized using a pre-piloted extraction form (see appendix), by two team members. The coding sheet would collect information on the types of interventions, targeted population, moderator variables, study design, and outcome variables. Again, disagreements in coding will be resolved through discussion and third-member involvement. The full list of coding categories is provided in Appendix 3 in the Data Extraction Form.

Assessment of risk of bias in included studies

Using a published list of criteria developed by 3ie, the quality of the studies will be assessed by two independent team member. 3ie has developed a tool to assess risk of bias in social experiments and quasi-experiments, which are widely used in systematic reviews (Hombrados and Waddington 2012). The ROB tool is presented in the appendix.

Statistical procedures and conventions

Measurement of treatment effect

We will calculate standardized mean difference (SMD) for continuous outcome variables, and odds ratios for dichotomous outcome variables. We will report these treatment effects along with standard error or 95 per cent confidence intervals. Treatment effects will be calculated as the difference between, or the ratio of, treated and control observations in a consistent way, such that outcome measures are comparable across studies. We shall also explicitly state and clarify how the effect size measure is defined and interpreted both in terms of sign and magnitude, in order to ensure that the sign of the effect size consistently measures an improvement or deterioration in outcomes across studies.

Unit of analysis

If the unit of treatment assignment is different from the unit of analysis, then we will assess whether the study has taken into account the issue of clustering and adjusted the standard errors accordingly (e.g. whether cluster-robust standard errors are provided). For studies that pose a risk of unit of analysis error, the standard errors will be adjusted according the formula provided in (Waddington et al., 2012):

S E_{c o r r e c t e d} = S E_{u n c o r r e c t e d} * \sqrt{(1 + (m - 1) * I C C}

Where m is the number of observations per cluster and ICC is the intra-cluster correlation coefficient. If information on the cluster size is not reported, then we will estimate m by dividing the total number of participants (or number of participants in each analysis, if data are available) by the number of clusters. If the data for estimating the ICC are not available, then we shall approximate ICC based on studies that have the same or a similar subject, if such studies are found.

We will also consider the unit of analysis issue in our own calculation of effect sizes (SMD or Odds Ratio) and their standard errors. This will be either based on a cluster-adjusted measure of the effective sample size or we will inflate the standard error using the design effect formula presented in Waddington et a. (2012).

Data synthesis

Synthesis of the evidence from the included studies will be presented through narrative and statistical analysis of comparable effect sizes using meta-analysis. Meta-analysis is useful in synthesizing quantitative evidence as it considers statistical power of the estimated effect. However, since we are considering a number of interventions and outcomes, we will first categorize the studies based on their type. Secondly, the scope of the meta-analysis will also depend on the nature of study design. Calculation of standardized mean difference, or the risk ratios are appropriate for similar type of treatment effects, hence they can be widely used for studies that use randomized control trials. However, in case of quasi-experimental studies, the treatment effects may not be strictly comparable. For instance, studies that use a regression discontinuity design or an instrumental variable method typically estimate local average treatment effect (LATE), while those using propensity score matching would estimate the average treatment effect on treated (ATT). For instance, it is usually not possible to make a conversion between the local average treatment effect (LATE) and the intent to treat effect (ITT). Although, if information on compliance rates are available, then one can make comparable conversions between the intent to treat effect (ITT) and the average treatment effect on the treated (ATT). This will depend on whether these information are available in the included studies. Therefore, we will present findings pooled across study designs or treatment effect estimates where evidence does not suggest effects are statistically different. We shall conduct meta-analysis where it is possible to convert the treatment effects into comparable measures (Duvendack, Hombrados, Palmer-Jones, & Waddington, 2012). Specifically, we shall carry out meta-analysis if the effect sizes can be computed for comparison, and the outcome measures are sufficiently similar. We shall conduct subgroup or moderator analysis, and meta-regression where possible, to determine empirical differences for outcome groups. Initially we will conduct separate analyses of randomized control trials (RCT) and quasi-experimental studies due to the differences in treatment effects estimated in these studies. However, we can include them in the same meta-regression analysis where a dummy explanatory variable will be included to control for the differences in study types. We will analyse effects sizes from studies with active controls separately from those with zero or business-as-usual controls.

These results will be presented using conventional methods such as forest plots. We will use Stata software for this purpose. When meta-analysis is not possible, we shall use narrative synthesis where we shall also discuss about the sample size and magnitude of effects.

Assessment of heterogeneity

When meta-analysis is possible, we shall test for heterogeneity across studies using the I-squared statistic along with the Tau-squared statistic. If heterogeneity is present, then we shall investigate what factors explain it by conducting moderator analysis, including sub-group meta-analysis and meta-regression if possible. Otherwise we shall discuss potential factors behind the heterogeneity through a narrative method. The moderator variables to be considered are race, ethnicity, socio-economic status, region, and number of intervention components. We shall collect data on these variables depending on their availability from the included studies.

Sensitivity analysis

To check if the results are sensitive to quality of data and approaches to analysis, we shall report sub group analysis particularly based on study design, and would carry out a weighted ANOVA and meta-regression according to the overall risk of bias classification, and if possible, risk of bias status for each category. If these methods are not implementable, then we shall analyse studies separately based on their study design and use narrative to analyse the methodological factors that might moderate the effect size. This will also involve describing studies based on design and risk of bias categories.

We shall use funnel plots, qualitative assessment and sub-group analysis comparing published versus unpublished studies to assess potential publication bias.

Treatment of qualitative research

We do not plan to include qualitative research.

Footnotes

Appendices

Review authors

Roles and responsibilities

Please give a brief description of content and methodological expertise within the review team. It is recommended to have at least one person on the review team who has content expertise, at least one person who has methodological expertise and at least one person who has statistical expertise. It is also recommended to have one person with information retrieval expertise. Please note that this is the recommended optimal review team composition. •

Content: Prof. Vollmer will take the lead on the content of the focus on the impact of female economic empowerment on the next generation. The chapter will particularly focus on child health and overall development. Prof. Vollmer leads a research group in development economics with a focus on maternal and child health at the University of Göttingen and is Adjunct Assistant Professor of Global Health at the Harvard T.H. Chan School of Public Health. He is the lead PI of various randomized field experiments, particularly of an impact evaluation of a female self-help group program to empower women in the state of Bihar in India. His field experience as well as grasp on the literature surrounding child development will provide guidance on the conceptualization of the review question.

•

Systematic review methods: Sarah Khan and Soham Sahoo will provide guidance on questions of systematic review methodology. They both have previous experience in conducting systematic reviews and have knowledge on the statistical methods required to conduct meta-analyses. Their research work on gender inequality and its impact on education would form a good basis for developing a research strategy for the review.

•

Statistical analysis: Statistical analysis in the first chapter will be carried out jointly by Sarah Khan, Le Thi Ngoc Tu, Atika Pasha and Soham Sahoo. They have experience with data analysis and familiarity with statistical analysis in meta-analysis. Mr Waddington will serve as a statistical advisor. He has substantial experience in systematic reviews, meta-analyses and generalized evidence synthesis methods.

•

Information retrieval: Sarah Khan, Le Thi Ngoc Tu, Atika Pasha and Soham Sahoo and will jointly develop the information retrieval for the review, starting with the title and abstract screening. Data extraction will be conducted by research assistants, under the guidance of Le Thi Ngoc Tu and Atika Pasha, who will also ensure that the data entry will be accurate and as expansive as required for the meta analysis. All of them have experience in literature reviews and good understanding of statistical methods. They have access to University of Göttingen library that will be the main source of information retrieval.

Sources of support

Describe the source(s) of financial and other support for the proposed review.

The systematic review will be a part of the ‘GrOW’ (Growth and Economic Opportunities of Women) Program which has received funding from IDRC, the Hewlett Foundation, and DFID. The final review is due by the end of 2017, before we have to submit progress reports.

Moreover, the resources at the University of Göttingen will be the source of all data extraction and information retrieval.

Declarations of interest

Please declare any potential conflicts of interest. For example, have any of the authors been involved in the development of relevant interventions, primary research, or prior published reviews on the topic?

We all have participated in research that is related to this research question in some way, but if any publications from our own work are determined to be eligible for inclusion into the study, we will have an independent evaluator assess the quality of the study.

Preliminary timeframe

Approximate date for submission of the systematic review.

Please note this should be no longer than two years after protocol approval. If the review is not submitted by then, the review area may be opened up for other authors. •

Date you plan to submit a draft protocol: November 11, 2016

•

Date you plan to submit a draft review: December 11, 2017

Plans for updating the review

Reviews should include in the protocol specifications for how the review, once completed, will be updated. This should include, at a minimum, information on who will be responsible and the frequency with which updates can be expected.

The authors agree to update the review if sufficient new studies and adequate funding to undertake this venture become available.

AUTHOR DECLARATION

References

Afridi

Mukhopadhyay

Sahoo

(2016). Female labor force participation and child education in India: evidence from the National Rural Employment Guarantee Scheme. IZA Journal of Labor & Development, 5(1). https://doi.org/10.1186/s40175-016-0053-y

Almond

Currie

(2011). Killing Me Softly: The Fetal Origins Hypothesis. The Journal of Economic Perspectives?: A Journal of the American Economic Association, 25(3), 153–172. https://doi.org/10.1257/jep.25.3.153

Bandiera

Natraj

(2013). Does Gender Inequality Hinder Development and Economic Growth? Evidence and Policy Implications. World Bank Research Observer, 28(1), 2–21. https://doi.org/10.1093/wbro/lks012

Baum

C. L.

II (2003). Does Early Maternal Employment Harm Child Development? An Analysis of the Potential Benefits of Leave Taking. Journal of Labor Economics, 21(2), 409–448. https://doi.org/10.1086/345563

Baydar

Brooks-Gunn

(1991). Effects of maternal employment and child-care arrangements on preschoolers' cognitive and behavioral outcomes: Evidence from the Children of the National Longitudinal Survey of Youth. Developmental Psychology, 27(6), 932–945. https://doi.org/10.1037/0012-1649.27.6.932

Beaman

Chattopadhyay

Duflo

Pande

Topalova

(2009). Powerful Women: Does Exposure Reduce Bias? The Quarterly Journal of Economics, 124(4), 1497–1540. https://doi.org/10.1162/qjec.2009.124.4.1497

Beaman

Duflo

Pande

Topalova

(2012). Female Leadership Raises Aspirations and Educational Attainment for Girls: A Policy Experiment in India. Science, 335(6068), 582–586. https://doi.org/10.1126/science.1212382

Bhutta

Z. A.

Das

J. K.

Arjumand

Gaffey

M. F.

Walker

Horton

, … Black

R. E.

(2013). Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? Lancet, 382(9890), 452–477.

Blumberg

R. L.

(1988). Income under female versus male control: hypotheses from a theory of gender stratification and data from the Third World. Journal of Family Issues, 9(1), 51–84.

10.

Bold

V. D.

Mara Quisumbing

A. R.

Gillespie

(2013). Women's Empowerment and Nutrition: An Evidence Review (SSRN Scholarly Paper No. ID 2343160). Rochester, NY: Social Science Research Network. Retrieved from https://papers.ssrn.com/abstract=2343160

11.

Borenstein

(Ed.). (2009). Introduction to meta-analysis. Chichester, U.K: John Wiley & Sons.

12.

Borenstein

Hedges

L. V.

Higgins

J. P. T.

Rothstein

H. R.

(2013). Introduction to meta-analysis. Hoboken, N.J.: Wiley. Retrieved from http://rbdigital.oneclickdigital.com

13.

Branisa

Klasen

Ziegler

(2009). The Construction of the Social Institutions and Gender Index (SIGI) (Ibero America Institute for Econ. Research (IAI) Discussion Paper No. 184). Ibero-America Institute for Economic Research. Retrieved from http://econpapers.repec.org/paper/gotiaidps/184.htm

14.

Brody

De Hoop

Vojtkova

Warnock

Dunbar

M.S.

Murthy

Dworkin

S.L.

, 2016. Economic self-help group programmes for improving women's empowerment A systematic review (3ie Systematic Review No. 23). Intenational Initiative for Impact Evaluation (3ie), London.

15.

Brooks-Gunn

Duncan

G. J.

(1997). The effects of poverty on children. The Future of Children, 55–71.

16.

Burger

(2010). How does early childhood care and education affect cognitive development? An international review of the effects of early interventions for children from different social backgrounds. Early Childhood Research Quarterly, 25(2), 140–165. https://doi.org/10.1016/j.ecresq.2009.11.001

17.

Buvinic

Furst-Nichols

(2014). Promoting Women's Economic Empowerment: What Works? (SSRN Scholarly Paper No. ID 2519302). Rochester, NY: Social Science Research Network. Retrieved from http://papers.ssrn.com/abstract=2519302

18.

Buvinic

Furst-Nichols

Pryor

E. C.

(2013). A roadmap for promoting women's economic empowerment. UN Foundation and the ExxonMobil Foundation.

19.

Cameron

D. B.

Mishra

Brown

A. N.

(2016). The growth of impact evaluation for international development: how much have we learned? Journal of Development Effectiveness, 8(1), 1–21. https://doi.org/10.1080/19439342.2015.1034156

20.

Campbell

Conti

Heckman

J. J.

Moon

S. H.

Pinto

Pungello

Pan

(2014). Early Childhood Investments Substantially Boost Adult Health. Science, 343(6178), 1478–1485. https://doi.org/10.1126/science.1248429

21.

Chattopadhyay

Duflo

(2004). Women as Policy Makers: Evidence from a Randomized Policy Experiment in India. Econometrica, 72(5), 1409–1443. https://doi.org/10.1111/j.1468-0262.2004.00539.x

22.

Conti

Heckman

J. J.

Pinto

(2015). The Effects of Two Influential Early Childhood Interventions on Health and Healthy Behaviors (NBER Working Paper No. 21454). National Bureau of Economic Research, Inc. Retrieved from https://ideas.repec.org/p/nbr/nberwo/21454.html

23.

Cunha

Heckman

J. J.

(2009). The Economics and Psychology of Inequality and Human Development (NBER Working Paper No. 14695). National Bureau of Economic Research, Inc. Retrieved from https://ideas.repec.org/p/nbr/nberwo/14695.html

24.

Cunningham

Ruel

Ferguson

Uauy

(2015). Women's empowerment and child nutritional status in South Asia: a synthesis of the literature. Maternal and Child Nutrition, 11(1), 1–19.

25.

De La Rue

Polanin

Espelage

Pigott

(2014). School-Based Interventions to Reduce Dating and Sexual Violence: A Systematic Review. Campbell Systematic Reviews 2014:7. Campbell Collaboration. Retrieved from https://eric.ed.gov/?id=ED557967

26.

Deininger

Goyal

Nagarajan

(2013). Women's Inheritance Rights and Intergenerational Transmission of Resources in India. Journal of Human Resources, 48(1), 114–141. https://doi.org/10.3368/jhr.48.1.114

27.

Desai

Chase-Lansdale

P. L.

Michael

R. T.

(1989). Mother or Market? Effects of Maternal Employment on the Intellectual Ability of 4-Year-Old Children. Demography, 26(4), 545–561. https://doi.org/10.2307/2061257

28.

Doyle

Harmon

Heckman

J. J.

Logue

Moon

(2013). Measuring Investment in Human Capital Formation: An Experimental Analysis of Early Life Outcomes (NBER Working Paper No. 19316). National Bureau of Economic Research, Inc. Retrieved from https://ideas.repec.org/p/nbr/nberwo/19316.html

29.

Doyle

Harmon

C. P.

Heckman

J. J.

Tremblay

R. E.

(2009). Investing in early human development: Timing and economic efficiency. Economics & Human Biology, 7(1), 1–6. https://doi.org/10.1016/j.ehb.2009.01.002

30.

Duflo

(2003). Grandmothers and Granddaughters: Old-Age Pensions and Intrahousehold Allocation in South Africa. The World Bank Economic Review, 17(1), 1–25. https://doi.org/10.1093/wber/lhg013

31.

Duflo

(2005). Why Political Reservations? Journal of the European Economic Association, 3(2-3), 668–678. https://doi.org/10.1162/jeea.2005.3.2-3.668

32.

Duflo

(2012). Women Empowerment and Economic Development. Journal of Economic Literature, 50(4), 1051–1079. https://doi.org/10.1257/jel.50.4.1051

33.

Duvendack

Hombrados

J. G.

Palmer-Jones

Waddington

(2012). Assessing ‘what works’ in international development: meta-analysis for sophisticated dummies. Journal of Development Effectiveness, 4(3), 456–471.

34.

Eswaran

Malhotra

(2011). Domestic violence and women's autonomy in developing countries: theory and evidence: Domestic violence and women's autonomy. Canadian Journal of Economics/Revue Canadienne D'économique, 44(4), 1222–1263. https://doi.org/10.1111/j.1540-5982.2011.01673.x

35.

Fantuzzo

J. W.

Mohr

W. K.

(1999). Prevalence and Effects of Child Exposure to Domestic Violence. The Future of Children, 9(3), 21–32. https://doi.org/10.2307/1602779

36.

Gertler

Heckman

Pinto

Zanolini

Vermeersch

Walker

, … Grantham-McGregor

(2013). Labor Market Returns to Early Childhood Stimulation: a 20-year Followup to an Experimental Intervention in Jamaica (NBER Working Paper No. 19185). National Bureau of Economic Research, Inc. Retrieved from https://ideas.repec.org/p/nbr/nberwo/19185.html

37.

Ghuman

S. J.

(2003). Women's Autonomy and Child Survival: A Comparison of Muslims and Non-Muslims in Four Asian Countries. Demography, 40(3), 419–436. https://doi.org/10.1353/dem.2003.0021

38.

Haddad

Hoddinott

(1994). Women's income and boy-girl anthropometric status in the Cote d'Ivoire. World Development, 22(4), 543–553.

39.

Hammerstrøm

Wade

Jørgensen

A. M.

(2010). Searching for studies: a guide to information retrieval for Campbell Systematic Reviews. Campbell Systematic Reviews, Supplement, 1, 1–75.

40.

Heath

(2014). Women's Access to Labor Market Opportunities, Control of Household Resources, and Domestic Violence: Evidence from Bangladesh. World Development, 57, 32–46. https://doi.org/10.1016/j.worlddev.2013.10.028

41.

Hoddinott

Haddad

(1995). Does Female Income Share Influence Household Expenditures? Evidence from Côte D'ivoire. Oxford Bulletin of Economics and Statistics, 57(1), 77–96. https://doi.org/10.1111/j.1468-0084.1995.tb00028.x

42.

Ibanez

Khan

Minasyan

Sahoo

Balasubramanian

, 2017. The impacts of interventions for female economic empowerment at the community level on human development: a systematic review of the evidence in low- and middle-income countries. Protocol for a Systematic Review. Campbell Systematic Reviews. Available at: https://www.campbellcollaboration.org/library/women-economicempowerment-low-and-middle-income-countries.html

43.

James-Burdumy

(2005). The Effect of Maternal Labor Force Participation on Child Development. Journal of Labor Economics, 23(1), 177–211. https://doi.org/10.1086/425437

44.

Kabeer

(1999). Resources, Agency, Achievements: Reflections on the Measurement of Women's Empowerment. Development and Change, 30(3), 435–464. https://doi.org/10.1111/1467-7660.00125

45.

Kitzmann

K. M.

Gaylord

N. K.

Holt

A. R.

Kenny

E. D.

(2003). Child witnesses to domestic violence: A meta-analytic review. Journal of Consulting and Clinical Psychology, 71(2), 339–352. https://doi.org/10.1037/0022-006X.71.2.339

46.

Laverack

(2006). Improving Health Outcomes through Community Empowerment: A Review of the Literature. Journal of Health, Population and Nutrition, 24(1), 113–120.

47.

Lawry

Samii

Hall

Leopold

Hornby

Mtero

2014. The impact of land property rights interventions on investment and agricultural productivity in developing countries: a systematic review. Campbell Systematic Reviews 2014:1 DOI: 10.4073/csr.2014.1

48.

Luke

Munshi

(2011). Women as agents of change: Female income and mobility in India. Journal of Development Economics, 94(1), 1–17. https://doi.org/10.1016/j.jdeveco.2010.01.002

49.

Lundberg

Pollak

Wales

T. J.

(1997). Do Husbands and Wives Pool Their Resources? Evidence from the United Kingdom Child Benefit. Journal of Human Resources, 32(3), 463–480.

50.

Macdonald

Higgins

J. P.

Ramchandani

Valentine

J. C.

Bronger

L. P.

Klein

, … Taylor

(2012). Cognitive-behavioural interventions for children who have been sexually abused. In The Cochrane Collaboration (Ed.), Cochrane Database of Systematic Reviews. Chichester, UK: John Wiley & Sons, Ltd. Retrieved from http://doi.wiley.com/10.1002/14651858.CD001930.pub3

51.

Macmillan

Gartner

(1999). When She Brings Home the Bacon: Labor-Force Participation and the Risk of Spousal Violence against Women. Journal of Marriage and Family, 61(4), 947–958. https://doi.org/10.2307/354015

52.

Mader

(2013). Rise and Fall of Microfinance in India: The Andhra Pradesh Crisis in Perspective. Strategic Change, 22(1-2), 47–66. https://doi.org/10.1002/jsc.1921

53.

Maertens

Verhofstadt

(2012). Horticultural exports, female wage employment and primary school enrolment: Theory and evidence from a natural quasi-experiment in Senegal. 2012 Conference, August 18-24, 2012, Foz Do Iguacu, Brazil. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&db=edsrep&AN=edsrep.126856&site=eds-live

54.

Malhotra

Schuler

Boender

(2002). Measuring Women's Empowerment as a Variable in International Development. Presented at the Workshop on Poverty and Gender: New Perspectives. Retrieved from http://www.aed.org/LeadershipandDemocracy/upload/MeasuringWomen.pdf

55.

Menon

van derMeulen Rodgers

Nguyen

(2014). Women's Land Rights and Children's Human Capital in Vietnam. World Development, 54, 18–31. https://doi.org/10.1016/j.worlddev.2013.07.005

56.

Osmani

Sen

(2003). The hidden penalties of gender inequality: fetal origins of ill-health. Economics & Human Biology, 1(1), 105–121. https://doi.org/10.1016/S1570-677X(02)00006-0

57.

Osofsky

J. D.

(2003). Prevalence of Children's Exposure to Domestic Violence and Child Maltreatment: Implications for Prevention and Intervention. Clinical Child and Family Psychology Review, 6(3), 161–170. https://doi.org/10.1023/A:1024958332093

58.

Paul-Majumder

(1996). Health Impact of Women's Wage Employment: A Case Study of the Garment Industry of Bangladesh. The Bangladesh Development Studies, 24(1/2), 59–102.

59.

Pick

W. M.

Ross

M. H.

Dada

(2002). The reproductive and occupational health of women street vendors in Johannesburg, South Africa. Social Science & Medicine, 54(2), 193–204. https://doi.org/10.1016/S0277-9536(01)00288-X

60.

Pitt

M. M.

Khandker

S. R.

Chowdhury

O. H.

Millimet

D. L.

(2003). Credit programs for the poor and the health status of children in rural bangladesh*. International Economic Review, 44(1), 87–118. https://doi.org/10.1111/1468-2354.t01-1-00063

61.

Prost

Colbourn

Seward

Azad

Coomarasamy

Copas

, … Costello

(2013). Women's groups practising participatory learning and action to improve maternal and newborn health in low-resource settings: a systematic review and meta-analysis. The Lancet, 381(9879), 1736–1746. https://doi.org/10.1016/S0140-6736(13)60685-6

62.

Qian

(2008). Missing Women and the Price of Tea in China: The Effect of Sex-Specific Earnings on Sex Imbalance. The Quarterly Journal of Economics, 123(3), 1251–1285. https://doi.org/10.1162/qjec.2008.123.3.1251

63.

Quisumbing

A. R.

Maluccio

J. A.

(2003). Resources at Marriage and Intrahousehold Allocation: Evidence from Bangladesh, Ethiopia, Indonesia, and South Africa*. Oxford Bulletin of Economics and Statistics, 65(3), 283–327. https://doi.org/10.1111/1468-0084.t01-1-00052

64.

Rangel

M. A.

(2006). Alimony Rights and Intrahousehold Allocation of Resources: Evidence from Brazil. The Economic Journal, 116(513), 627–658.

65.

Ruel

M. T.

(2008). Addressing the underlying determinants of undernutrition: Examples of successful integration of nutrition in poverty-reduction and agriculture strategies. In SCN News (Vol. 36). Retrieved from https://www.unscn.org/web/archives_resources/html/resource_000001.html

66.

Ruhm

C. J.

(2004). Parental Employment and Child Cognitive Development. Journal of Human Resources, XXXIX(1), 155–192. https://doi.org/10.3368/jhr.XXXIX.1.155

67.

Ruhm

C. J.

(2008). Maternal employment and adolescent development. Labour Economics, 15(5), 958–983. https://doi.org/10.1016/j.labeco.2007.07.008

68.

Schicks

(2014). Over-Indebtedness in Microfinance - An Empirical Analysis of Related Factors on the Borrower Level. World Development, 54, 301–324. https://doi.org/10.1016/j.worlddev.2013.08.009

69.

Shroff

Griffiths

Adair

Suchindran

Bentley

(2009). Maternal autonomy is inversely related to child stunting in Andhra Pradesh, India. Maternal & Child Nutrition, 5(1), 64–74. https://doi.org/10.1111/j.1740-8709.2008.00161.x

70.

Smith

L. C.

Ramakrishnan

Ndiaye

Haddad

Martotell

(2003). The Importance of Women's Status for Child Nutrition in Developing Countries (Research Report No. 131). Intl Food Policy Res Inst.

71.

Taylor

(2011). ‘Freedom from Poverty is Not for Free’: Rural Development and the Microfinance Crisis in Andhra Pradesh, India. Journal of Agrarian Change, 11(4), 484–504. https://doi.org/10.1111/j.1471-0366.2011.00330.x

72.

UNDP (Ed.). (1995). Gender and human development. New York: Oxford Univ. Press.

73.

UNDP (Ed.). (1996). Economic growth and human development. New York: Oxford Univ. Press.

74.

United Nations . (2015). United Nations: Gender equality and women's empowerment. Retrieved 29 September 2016, from http://www.un.org/sustainabledevelopment/gender-equality/

75.

Vaessen

Rivas

Duvendack

Palmer Jones

Leeuw

Van Gils

Lukach

Holvoet

Bastiaensen

Hombrados

Waddington

. The Effects of Microcredit on Women's Control over Household Spending in Developing Countries: A Systematic Review and Meta-analysis Campbell Systematic Reviews 2014:8 DOI: 10.4073/csr.2014.8

76.

Waddington

White

Snilstveit

Hombrados

J. G.

Vojtkova

Davies

, … Tugwell

(2012). How to do a good systematic review of effects in international development: a tool kit. Journal of Development Effectiveness, 4(3), 359–387. https://doi.org/10.1080/19439342.2012.711765

77.

Waldfogel

Han

W.-J.

Brooks-Gunn

(2002). The Effects of Early Maternal Employment on Child Cognitive Development. Demography, 39(2), 369–392. https://doi.org/10.1353/dem.2002.0021

78.

World Bank . (2014). Voice and Agency: Empowering women and girls for shared prosperity. World Bank Group.

79.

World Economic Forum . (2015). The Global Gender Gap Index 2015. Retrieved 29 September 2016, from http://wef.ch/1Mkpfot

80.

Yoong

Rabinovich

Diepeveen

(2012). The impact of economic resource transfers to women versus men: a systematic review (Technical Report). London: EPPI - Centre, Social Science Research Unit, Institute of Education, University of London.