Principal investigator: Adriano Valladão Pires Ribeiro
Article title: HOW DOES MOTHER'S SCHOOLING AFFECT FAMILY HEALTH, NUTRITION, MEDICAL CARE USAGE, AND HOUSEHOLD SANITATION?
Article authors: Jere R. Behrman and Barbara L. Wolfe
Location of the intervention: Nicaragua
Sample size: 1699 Families
Main theme: Health Insurance
Type of Intervention: The role of maternal education in family health.
Primary variable of interest: Mother's education level
Evaluation method: LISREL Function
Problem Politics
The mother's education level within a family plays a significant role in increasing the family's overall health needs, such as food, access to medical care, and sanitation. However, a problem is that many factors related to health are not observable, making it difficult to establish a direct link between these factors and the level of education. Therefore, exploring the effects of maternal education on family health requires addressing this issue appropriately.
Evaluation Context
Starting from the idea that the mother's level of education is linked to important factors such as health and nutrition, the World Bank, for example, suggested increasing women's education as a means of improving the health conditions of the population in developing countries. Following this, a study conducted in Nicaragua on the impact of maternal education and other unobserved factors on health conditions is presented. The data comes from a national socioeconomic survey of women aged 15 to 45 conducted in 1977 and 1978. Information from mothers in 1699 families with health indicators for children under 5 years of age was used.
Details da Intervention
Considering a model, parents, given their family income, would choose to consume health-related items for themselves and their children, as well as goods and services unrelated to health. The health of each family member would be determined by nutrient intake, water and sanitation conditions, use of medical care, income, education and age of the mother, in addition to other characteristics of the mother and the district where they live. The characteristics of the district where they live include, among other things, the availability of and information about health services, while those of the mother include her knowledge and habits regarding health and her past health status.
With the exception of consumption of goods and services unrelated to health, income, and the mother's age and education level, all other variables mentioned above are not directly observed. In the process of determining the relationship between the mother's education level and the family's consumption of health items, the lack of or measurement error of an important variable can end up placing more importance on education level than it would have if the other factors were present. Thus, results obtained without addressing or mitigating this problem may end up providing a distorted view of reality. idp.edu.br
Exclusive
The first step is to circumvent the problem of unobserved health variables. Therefore, various related and available research data were used to infer about them, allowing us to say that they are imperfectly observed. The relationship between the data used and the variables is described below.
Child Health: Standardized measurements of weight, height, and biceps circumference in children.
Mother's Health: Four health condition indices based on information from mothers, such as the number of days too sick to work or perform similar activities in the last 6 months and having had, at some point in their lives, parasitic diseases, preventable diseases (examples: diphtheria and tetanus) or diseases with therapeutic treatment (example: typhoid and high blood pressure).
Nutrition: Two standardized measures of the family's nutrient intake (calories and protein) in the week prior to the survey and qualitative information on whether or not they had a refrigerator (an important indicator given Nicaragua's tropical climate).
Use of Medical Assistance: Deviation from the standard in the number of injections the children received, in which trimester of the last pregnancy the mother had her first prenatal medical exam, and a qualitative variable of social security participation, since such participation covers subsidized medical assistance.
Water and Sanitation Conditions: Three indicators related to the absence and condition of the bathroom in the house.
Location Features: Index based on the district's total population, population density, hospital beds per 1000 inhabitants, and the proportion of the population that is literate.
Mother's characteristics: Index based on the mother's education level, whether she was brought to an urban area, whether her mother was present during adolescence, whether her father was present during adolescence, and the number of siblings.
After constructing the measurements of the latent variables using this information, one can examine the effects of both the mother's education level and the district characteristics and other maternal characteristics on the first 5 (child health, mother's health, nutrition, use of medical care, and water and sanitation conditions). First, the effect is estimated using only the mother's education level without considering her other characteristics and the district characteristics. Second, the location characteristics are added to the system. Finally, the effect is estimated by also adding the mother's other characteristics.
Results
The first procedure only indicates the impact of the mother's education level on health items. The effect is positive and significant for the child's health, nutrition, use of medical care, and water and sanitation conditions; that is, only her own health is not as impacted by her education level. Furthermore, there is a positive indirect effect on the child's health via nutrition and sanitation. The second procedure also considers the characteristics of the location. Although these characteristics have a positive effect on all health conditions, the impact of the mother's education level remains quite similar. In other words, the inclusion of these characteristics increases the explanatory power regarding health items, but without affecting the impact of education. Finally, the other characteristics of the mother in the system, when added to the system, have a positive impact on all health items. Unlike the second procedure, here the mother's education level loses all relevance, and it would not be possible to say that this is a factor that has a positive impact on health and nutrition.
Lessons Politics
When dealing with relationships involving imperfectly observed factors, great caution must be exercised when drawing conclusions and recommendations. The study presented reveals that the apparent importance of maternal education on family health and nutrition may be overestimated. By adding other maternal characteristics to the analysis, her education loses its positive effect on health in Nicaragua. If such a result holds true for other developing countries, recommendations like those of the World Bank may be misleading.
Reference
Behrman, Jere R.; Wolfe, Barbara L.. “How does mother's schooling affect family health, nutrition, medical care usage, and household sanitation?”, Journal of Econometrics, Volume 36, Issues 1–2. 1987.
