How a mother's literacy skills, even at a basic level, translate into healthier kids.
It's an idea that's been around for a long time: The more formal schooling a mother gets, the better off her children's health will be. Educated women get prenatal care, boil their water, and take sick kids to the doctors. Illness decreases. Survival increases.
Study after study in public health, particularly in the developing world, has shown this to be absolutely true.
And yet, surprisingly, for many years, no one ever quite figured out why or how this happened. There were theories, of course. Education empowers women, who still serve as the primary caretakers for children. It liberates them from traditional bonds like family influence and old wives' tales. It improves status and access. However, few researchers actually gave much credit to the learning itself, assuming many schools in developing countries were of such low quality that they couldn't possibly be teaching useful skills to women.
Even Professor Emeritus Robert LeVine and his wife, Sarah, an anthropologist and former researcher at the Ed School, couldn't quite put their fingers on the school-health connection when they started looking at women's schooling, fertility, and child mortality in the early 1980s with the Ed School's Project on Maternal Schooling. As the LeVines and three Ed School alumni, Beatrice Schnell-Anzola, Ed.M.'86, Ed.D.'01, Meredith Rowe, Ed.M.'99, Ed.D.'03, and Emily Dexter, Ed.M.'90, Ed.D.'00, ask in their new book, Literacy and Mothering: How Women's Schooling Changes the Lives of the World's Children, "What is it about schooling that affects child survival, fertility, and the behavioral development of children?"
For nearly a decade, their guesses about that link between going to school as a young girl — even for only a few years — and behavior later as a mother that positively affects children's health focused primarily on attitude and behavior. Educated women feel more assertive, for example. But early on in the project, LeVine says, "We didn't have an account of what happens in the classroom that could explain later health and fertility outcomes."
And then, while planning to partially replicate a previous research project done in Cuernavaca, Mexico, an Ed School student, Patricia Velasco, C.A.S.'84, Ed.D.'89, suggested they look into literacy as a possible explanation — an aspect of schooling "strangely neglected," LeVine says. Velasco was studying under Ed School Professors Catherine Snow and Jeanne Chall, experts on language and reading.
At first, LeVine says he "pooh-poohed" the suggestion. As he writes in the book, "We were frankly skeptical that literacy or any other cognitive outcome of schooling could account for the impact of schooling on the maternal behavior that led to diminished fertility and mortality. We were inclined to believe" — as had earlier researchers focused on attitude and behavior — "that schooling empowered girls, influencing their aspirations for themselves and their children, their ability to assert themselves, and their sense of self-worth, regardless of what they learned in school." In part this was because many schools in less-developed countries employed poorly trained teachers and lacked basic equipment. It was reasonable to assume, LeVine says, that only limited literacy skills could be learned in those environments, and that retaining those skills many years later when these girls became mothers was uncertain.
Still, he decided to add a literacy test that Velasco helped create to the new Mexico project, mostly, he says, to prove "once and for all" that literacy didn't have an effect.
What they found from their initial work was surprising.
"The results came out exactly the opposite from what I expected," LeVine says. "The more schooling, the more literacy skills." But even just a couple of years in the classroom as young girls made a difference in what was used later, as mothers, to make informed health choices for their children. "The data was absolutely unequivocal. We thought: Now we finally have the intervening variable."
From that point on, maternal literacy became the project's main focus.
"It's not a model I'd recommend," LeVine jokes about the project's path, "but it worked for us."
And it worked through four studies in four different countries: After Mexico, they interviewed mothers in Zambia, Venezuela, and Nepal. The areas were chosen, in part, for their diversity, and also because there were Ed School doctoral students who had lived or worked in each of these countries, like Kathleen Stuebing, Ed.M.'89, Ed.D.'94, who taught for many years in Zambia, and Nepal native Arun Joshi, Ed.M.'86, Ed.D.'98, who showed up at LeVine's Cambridge office one day, urging him to include Nepal in the study. Schnell-Anzola, from Venezuela, became the invaluable point person in her country. LeVine says these students were critical to the success of the project, especially when doing fieldwork, which Sarah primarily directed, supervised, and coordinated.
"You need to find local informants in each community," LeVine says. The informants helped him and the team cast a wide enough net to find women in each country with varied amounts of schooling.
"As anthropologists, we needed to identify a village or urban neighborhood where there was a mixture of mothers," he says, noting that his team went door-to-door asking questions and looking for mothers of children three years old or younger. "If you're only talking to women who all have six years of education, that's no good."
It was particularly challenging in Nepal, where, at the time of the study, the mean number of years of schooling for women 25 years or older was less than one year. They often found entire villages where women had no education at all. Still, LeVine says they managed to find some diversity: Joshi found a village outside of Kathmandu that had women varying in the number of years they had gone to school, and Sarah found an urban neighborhood with similar variety.
Once the right mix was found, each mother was interviewed by the researchers, covering a range of topics like socioeconomic and educational backgrounds; the schooling of their parents, husbands, and siblings; reproductive and health behavior; knowledge of child development; and attitudes toward their children.
Language and reading skills were assessed, primarily using tests created by Snow and Chall as well as existing science textbooks and passages developed in collaboration with local pediatricians. This included reading comprehension, with women reading silently and out loud to researchers to determine if the women were only decoding words — breaking up words into understandable parts — or actually reading, and how much was understood. Mothers were also asked to identify nouns common in each culture by answering the question, "What is a … ?" Responses were scored based on how simple or complex the definition was. Answering that a dog is an animal that barks, for example, would have scored higher than simply: A dog is a dog.
Another test involved each mother recounting a health crisis involving one of her children — a simulation of how she would tell the story in a clinic. Her narrative was analyzed and scored on how well the researcher could follow what she was saying.
In Venezuela and Nepal, researchers also looked at functional literacy — the ability of the mothers to do everyday tasks such as tell time and read food labels or prescriptions.
All of the tests were tailored to each population, using native languages and common objects. This was a time-consuming, but critical, part of the fieldwork, LeVine told the Harvard Gazette in 1997, just prior to starting the study in Nepal, in urban and rural communities.
"The patterns of communication are different in each culture and setting," he said, noting that students from each country, like Clara Sunderland Correa, Ed.M.'83, and Medardo Tapia Uribe, C.A.S.'82, Ed.D.'89, both natives of Mexico, helped make this possible.
"In that sense, this is totally a project of the school," LeVine says. "Our methods were taken entirely from Snow's and Chall's work and all of the students were Catherine's and Jeanne's students."
In each country, women were also assessed for their comprehension of public health messages. For example, mothers were played taped recordings of health messages that had been broadcast regularly on local radio stations. Except for those with no or very little education, they were also asked to read health-related information in publications and recount everything they could remember. In Mexico, there were radio messages about breastfeeding and playing with infants. In Nepal, one radio broadcast emphasized using oral hydration salts when a child has diarrhea. In Venezuela, mothers read campaigns about AIDS, family planning, and cancer.
Across the board, in every country studied, what they found was that women retain the literacy skills they learned as young girls in school — even when they attended lowquality schools only for a couple of years — and then use those skills later, as mothers, to understand and connect with public health information disseminated through the media and through healthcare workers. Their trust of this information, as well as respect for experts like doctors and nurses, also increases.
"This clearly shows that schooling actually affects a woman and she does something with that education," LeVine says.
As Sarah told the Gazette in 1997, "Schooling allows women to learn hygiene more quickly because schools introduce them to a different way of learning. At home, they learn primarily by watching and imitating, by apprenticeship, not by following verbal instructions," she said. "In school, however, students are given instructions on how to accomplish tasks — in a language very different from home conversation — which later facilitates the process of assimilating information."
Learning to trust experts is a big part of this leap for mothers, says LeVine.
"A pregnant woman is told by her doctor to get three injections of tetanus toxoid," he says, referring to a vaccine used to prevent neonatal tetanus, a massive killer of newborn children, especially in developing countries. "She gets her first injection. A doctor or nurse says you need to come back to complete the series. A woman with less education, she just won't come back. She doesn't understand the importance."
But for the sake of her children, she needs to. As Michael Cole, a professor of psychology and communication at the University of California – San Diego, wrote in the introduction to Literacy and Mothering, which recently won the 2013 Eleanor Maccoby Award from the developmental psychology section of the American Psychological Association, "Whether or not one has been to school for some length of time, raising children requires adult decisions about how best to feed and clothe the child and how to protect them from disease and injury. Adults must learn to whom they should turn for help when normal caretaking measures do not suffice."
Skeptics, LeVine says, question whether a mother who has only been in school for a year or two could possibly have learned something like the germ theory of disease.
"The answer is no," he says. "What she understands is that when a doctor — the expert — says do this, you do it. You don't have to have an intellectual understanding of the disease. I think of myself as a cancer survivor who had no more than a superficial knowledge of what was wrong with me and what was being done to combat it, but I followed the experts to a cure."
When asked about the assumption that more education actually makes people question authority, not follow it blindly, LeVine explains that modern institutions such as Western schooling make individuals more likely to question traditional authority — an older member of a family, for example. However, they also teach people to follow the orders of authorized experts, such as schoolteachers or doctors.
People "learn to learn," he says. "Questioning the authority of those in white coats is not what these women learned. Quite the opposite. And they're right. They need not know microbiology or other biomedical processes in order to be a good patient or even to provide sufficiently sanitary conditions for child health. They need the guidance of experts, and that's what bureaucratic institutions are designed to provide."
Mother as Teacher
In addition to learning in the classroom how to be a pupil, which improves child health and survival, LeVine and his team found that mothers also learn how to be teachers at home, which improves school outcomes for their children. Although one could argue that all mothers teach their children, regardless of whether they went to school or for how long, in Literacy and Mothering, LeVine argues that teaching in school involves adults talking to children, using language for instruction — a skill girls eventually adopt when they become mothers with their own children.
In Mexico, for example, mothers with more formal schooling responded more frequently when their 10- and 15-montholds babbled or looked at them by talking and looking back. Follow-up visits when the children were 30 months showed that toddlers whose mothers talked to them more at 15 months scored higher on a language development test.
Asked if he ever considered replicating the study in the United States, LeVine says they had hopes of studying the large Latino population in eastern Massachusetts, but the funding never came through.
"I'd still love to see that happen," he says.
He'd also love to see an overall expansion of the work that he and his team started, particularly in the fields of public health and demography. Despite working on this research for more than three decades now, he still considers this a pilot study.
"By now, I had hoped that other groups like USAID would take this research to a larger level," he says. "My hope is that they adopt our methods and test so many hypotheses that we were unable to test with our small samples. I want to see this replicated with 5,000 women."
From a public policy standpoint, LeVine says their work is important because it shows — despite what many in education don't want to hear — that even low-quality schools and attending for only a few years can have a positive impact, especially in developing countries. One study found that an average of only one to three years of women's schooling reduces early childhood mortality by 10 percent.
Rowe, now an assistant professor of human development at the University of Maryland, says this highlights the fact that obtaining literacy skills is a process that starts with oral language skills.
"That is, girls who may not have been in school long enough to achieve high levels of reading comprehension were still exposed to the academic language used in schools," she says. "Having some experience with this type of discourse — for example, following instructions — helps them to better navigate other bureaucratic settings in the future, such as healthcare settings. Policymakers should encourage girls to attend school, even if it is only for a few years, because every year makes a difference."
Eventually, improving the quality of what women are learning, especially literacy skills, will make an even bigger difference.
"We might see even more dramatic, positive effects," says Rowe, "across generations."