The literature on the definition of women's empowerment and its association with children's nutrition outcomes is diverse and often contentious. The combined 2016 Ethiopia Demographic and Health Survey (EDHS) and 2019 Ethiopia Mini Demographic and Health Survey (EMDHS) data were used to develop and test the effect of women's empowerment on children's nutrition status in Ethiopia. In our analysis, based on the intrinsic agency and instrumental agency dimensions of women empowerment, we found a significant number of empowered women compared to the disempowered ones. This result is supported by a study by (58), where they found that women in East Africa may hold much more autonomy in, for example, economic decision-making than their South Asian and Middle Eastern counterparts. However, considering the information presented below, it is possible to question the validity of these findings using scientific inquiry.
In this study, multivariable probit regression analysis was used to examine whether women's empowerment contributed to reducing wasting and stunting among children aged 0 to 59 months. In the study, we found that improvements in women's empowerment indices were not associated with a reduction in stunting or wasting, unlike previous studies conducted in the country, for instance, a study by (53) and (49). This result is also in contrast to results obtained from Burkina Faso (36), Bangladesh (59), Pakistan (28) and East Africa (58).
The result is, however, consistent with a study conducted in Mozambique, where they found no clear-cut conclusion concerning the possible impact of women's empowerment on the nutritional status of children (60) and Ghana, where they found no statistically significant association between the composite empowerment score measuring women's empowerment in agriculture and its decomposed components and the latent health status of children (61). A related study in India (62) and Nepal (63) found no association between a woman's ability to control her own health care and child stunting. Furthermore, (36) reported that the hypothesis that women's empowerment will improve child nutrition has limited empirical support. Based on these findings, we argue that the association between women's empowerment and child nutrition outcomes may be affected by additional underlying factors not captured by the analyses.
These findings might provide the impression that women's empowerment is not a pathway to improving child stunting and wasting in Ethiopia. However, it is advisable to take these results with some level of caution for some significant reasons. First of all, it is possible that the current lack of evidence is likely not due to the absence of an underlying relationship between women's empowerment and child nutrition but rather to limitations in survey design and data collection techniques. For instance, the survey might be subject to a typical study problem often caused by respondents' self-reports, namely social desirability bias (64, 65, 66). Social desirability bias is the tendency of survey respondents to answer questions in a way that conforms to social norms and expectations, rather than providing honest answers, resulting in inaccurate measurements of women's empowerment (67, 68, 69, 70). Nederhof's definition of social desirability bias is widely cited, and it is considered the most common reason why a survey may fail to reveal actual correlations (71). The bias arises when respondents deceive themselves or explicitly seek to mislead others by indicating that their opinion conforms to the socially expected one. The presence of desirability bias in survey responses can be challenging to detect, but some variables are more susceptible than others (71). For instance, questions about specific facts, such as educational attainment, are less likely to be affected by the bias because respondents can provide certificates as proof. However, questions about personal experiences, such as domestic violence, are more likely to be influenced by the bias, especially if there is no official proof of the facts (72). Social desirability bias can be present in all responses to a survey, and it can make a survey completely unrealistic if its non-systematic effect cannot be eliminated (73).
Furthermore, prevailing gender norms within a particular cultural context can have an impact on women's empowerment. Social customs in Ethiopia encourage people to conform to organizational norms in all circumstances, and the political environment is also very conducive to encourage people to conform to organizational norms in all circumstances (74, 75). For instance, according to (76), when Ethiopian women were asked how often they gave birth at home, almost all respondents answered that they never did. However, collected data indicate that most mothers give birth at home. These authors explained this phenomenon by pointing out that there was a lack of care in asking these questions. Similar ideas were expressed by (74) and (75). In addition, the survey results contradict the commonly held belief that women in developing countries, including Ethiopia, have limited independence in making decisions on their own or household affairs (77). The other factor might be the way the sample respondents were recruited. These inconsistencies might also likely be caused by methods/conceptualizations of women's empowerment, as well as the specific domains researched. In earlier studies, women's empowerment was defined and operationalized differently, which might have contributed to its significance.
However, other variables such as maternal BMI (56, 58, 60), gender of the child (78), child's perceived weight at birth (56, 79), region (56, 80), wealth index (31, 56, 58, 60) were found to have a significant counteracting effect on stunting and wasting. Maternal Body Mass Index (BMI) is an important signal of the quality of how mothers feed their children (30, 58, 81). Mothers with a healthy BMI usually provide much healthier food to their children as well.
The religion of the respondent, the working status of the husband, and the type of cooking fuel (56) were found to have a significant mitigating effect on stunting. The place of residence (56) was found also to decrease the level of wasting. Norms and factors linked with labour market opportunities guide sex preference towards male children (82). Giving birth to the preferred sex will likely give women more access to resources and decision-making (49). In our study, female children still showed less stunting and wasting than male children (83, 84). These differences could be explained by biological factors, such as differences in immune and endocrine systems between boys and girls, as well as social factors, such as gender dynamics and preferential feeding practices (83). The more educated the husband was the better nutritional status his child has (85). In our case, we have such evidence of stunting, despite no significant difference in wasting. However, the maternal working status had an insignificant decreasing effect on child malnutrition. Because most of women work in low-paying jobs, it is sometimes possible that women who stay at home take better care of their children.
Longer childbirth interval had a significant mitigating effect on child stunting, despite having an inconclusive effect on child wasting. Child age had a significant increasing effect on child stunting and a significant decreasing effect on child wasting. The age of the child is positively associated with stunting, perhaps suggesting that risk of stunting increases with age (86), while the risk of wasting decreases with age (21). This can easily be explained by the fact that the growth of the child's length starts to occur mainly after infancy so in children who are stunted, this becomes apparent mainly after the age of one. The paper (87) directly demonstrates that stunting most often occurs between the ages of 2-3 years.
Undernutrition in children has a spatial dimension that is strongly influenced by the degree of urbanization, as measured by nutrition indices such as stunting and wasting (88). It appeared that rural Ethiopian children are less wasted than their urban counterparts, despite no significant difference in stunting. The coefficients of the region dummies are interpreted using the omitted Tigray region dummy. Apart from Amhara, Benishangul Gumuz, and Dire Dawa City Administration, children in all other regions are more stunted than children in the Tigray region. However, suppose the data were collected after 2019/20, the results of this study may have been reversed as Ethiopia has experienced widespread internal displacement and a devastating civil war (especially in the Amhara, Afar, Tigray, and Oromia regions). Furthermore, the study shows that children in Somalia and Southern Nations, Nationalities, and Peoples Region (SNNPR) are not only stunted but also wasted. This makes sense given that Somalia region lies within the severely food-insecure and dry eastern part of Ethiopia. As a region, SNNPR is also characterized by a majority of its inhabitants who live far away from minimum standard infrastructures, which might result in poor infant feeding practices (food choice).
In the study, it was found that there was no statistically significant association between any of the women's empowerment indices and the adequacy of women’s child nutrition. This lack of association may be due to the difficulty of establishing the direct effect of women's empowerment because personal statements may be biased. However, the study further explored variables associated with women’s empowerment and promoting better child nutrition and found that women's educational attainment had a significant impact on child nutrition status. It is worth noting that the beneficial effect of women's educational attainment on their children's health is less prone to desirability bias and therefore highly relevant information. Because none of the three latent empowerment variables include mothers' educational attainment, there can only be an indirect relationship between our empowerment variables and mothers' educational attainment.
The literature provides numerous examples of how women's literacy and education can greatly improve the health of their children. For example, in Sub-Saharan countries, improved educational level is a very important factor in women's conscious behavior (32), and in Malawi, Tanzania, Zimbabwe, and Mozambique, there is a positive impact of women's education on their decision-making ability in the household (89). Similarly, studies conducted in Uganda (90) and Nigeria (91) have shown a positive relationship between women's educational attainment and effective care for their children. Similar thoughts can be found in (51) and (52). It is known that knowledge empowers women, enabling them to change both their own lives and their children's lives for the better.
Thus, while the direct effect of women's empowerment on child nutrition adequacy may be difficult to establish, indirect data, like women’s educational attainment suggests that women's empowerment plays an important role in their children's health status. Many studies have shown that women with good circumstances and proper education feed their children much healthier than their less empowered counterparts.