Learning from Pakistan

Schools in Pakistan, like those in much of the developing world, have few of the extracurricular activities many in the developed world take for granted.

Of the five and a half hours that a typical Pakistani schoolchild typically spends in school, more than three quarters are spent in formal classes. Less than 5 percent involves supervised sports, music and art, or activities such as drama and debates. The majority of schools do not even offer these activities. The issue is an important one. Families make conscious decisions about, “which child to ‘push’ through school and which child to exert less effort on,” researchers found. “The fact that schools are boring means that those children whose parents do not push them to attend school will likely not attend.”

Rote Learning

While developing their testing tools at one small private school, the LEAPS team could not understand how the children could struggle with a simple reading comprehension, while they had earlier done quite well in a much more advanced English reading comprehension passage in their last internal examination.

The puzzle was solved when they found out that the passage on the internal test was taken verbatim from the textbook used in the class, which the children had practiced and mostly memorized. At an “elite” school, children were shown the picture of a parallelogram and a rectangle and asked, in a “free response” format, to describe the shapes’ similarities, and differences. The question drew a complete blank and the children told testers they had never been exposed to that type of question.

“Testing children using template questions not only leads to official exams overstating children’s subject mastery, it also results in them forgetting the important skill of decoding instructions,” the authors write.

A Quarter Mile

The distance from home to school is a vastly important factor for Pakistani schoolchildren. Half the children surveyed live within 200 meters of the closest eligible school, and the vast majority live within 500 meters, no more than a 10-minute walk.

That proximity matters most for girls. As the distance to the closest eligible school increases, the drop in enrollment among girls is three to four times as great as that for boys: 11 to 16 percent for every 500 meters, compared with 4 percent for boys.

In analyzing what they called girls’ “distance penalty,” researchers found that neither wealth (and thus the ability to pay for transport or a chaperone) nor a girl’s age (looking at differences between pre- and post-pubescent girls) had any effect. The only thing that reduced the “distance penalty” was the presence of adult educated females in the household.

The importance of girls’ education is not to be underestimated. The researchers found that when a village had a girls’ secondary school, its chances of having a private school increased by 300 percent, because many of those educated girls went on to become private school teachers.

In the world of development economics, where Kennedy School Public Policy Professor Asim Ijaz Khwaja does much of his work, education is seen, along with health, as one of those great human capital investments that can help lift a country from poverty.

Despite much of the troubling news that has come out of Khwaja’s native Pakistan in the past decade, he and his co-authors — Tahir Andrabi, of Pomona College, and Jishnu Das, of the World Bank’s research group — looked more carefully and saw a development few others were noticing. Not the story that was being exaggerated across much of the Western media: the rise of madrassas, Islamic schools that were being blamed for increased fundamentalism and militancy throughout the Muslim world. Those schools serve no more than 3 percent of schoolchildren in the country.

Rather, they saw an increase in the number and spread of small private schools, many of them in rural villages. In a country receiving billions of dollars in international aid, it was a development worth studying. And with the recent decision by the U.S. government to pour in another $7.5 billion in nonmilitary aid over the next five years, it may prove to be invaluable information.

The research project they conducted, with support from the World Bank and especially lead economist Tara Vishwanath, has provided a unique view into the changing educational environment of a country; a dramatically detailed study of students, their families, and teachers in a developing country; and insight into possible development policies for a country that is in dire need of them.

The project was ambitious. The team began to design a project looking at thousands of subjects and following them over several years.

“We realized we could focus on a few select topics, or we could step back and see the entire educational universe,” Khwaja says.

They launched their project in the province of Punjab, Pakistan’s most populous, home to a little more than half the country’s 164 million people, and where about half the rural population live in villages with private schools.

The group’s academic credentials and its consistent interest in the country’s development, expressed in previous contacts with local economists and statisticians, helped it gain the trust and cooperation of the government. With that, government schools were on board, and private schools were surprisingly quick to follow.

The study, known as the LEAPS Project (Learning and Educational Achievements in Punjab Schools) became a huge undertaking: all 812 government and private schools in 112 villages, 12,000 schoolchildren, 2,000 households, and 5,000 teachers.

From 2003 to 2007, an in-country team administered carefully designed tests independently to students so as to have comparable data on their knowledge and progress. Schools, households, teachers, and school principals were surveyed. Information was collected on school infrastructure and costs, teachers’ socioeconomic status, geographic location, and much more. The team even collected information on families with school-age children who were not enrolled in schools.

The project found an education system in flux. From 2001 to 2005, student enrollment in Pakistan went up 10 percentage points. Yet, learning across the board was poor — most children at the end of third grade couldn’t answer simple questions. And learning remained stagnant or even declined as enrollment increased. During that same time, the number of private schools increased from 32,000 to 47,000. These were not urban schools catering to the elite, but rather affordable, rural “mom and pop” schools run by local village women and offering modern progressive education. By 2005, one in three Pakistani children in primary school was enrolled in a private school.

The researchers found that private schools outperformed government schools: On average, students in government schools needed as much as two and a half years to catch up to their private school counterparts. As growth in private schools exploded during the course of the research, Khwaja says, he and his colleagues began to realize that the future they had envisaged was becoming the present for many villages.

With that change came debates over the future of education in Pakistan, and the data the LEAPS project had been collecting and analyzing played an important role. The new realities of the Pakistani education system meant more choices for parents when pondering their children’s education; they also raised questions of whether and how to regulate this vast new network of schools and of teacher certification and qualification, and a focus on learning rather than just enrollment. Khwaja and his colleagues were uniquely able to inform those debates.

Their household surveys had given them a window into how parents made decisions on schooling. They found that parents thought carefully about which children should receive the largest educational investments and reduced the money they spent on educating the ones they saw as “less intelligent.” And they found that parents were relatively aware, regardless of whether they were literate or illiterate, of how different schools performed and how hardworking, or prone to absenteeism, the teachers were.

On the contentious issue of increasing teachers’ minimum qualifications — requiring a bachelor’s degree as opposed to just a secondary education — the data showed that although certification mattered, the difference between private and public school settings mattered much more. LEAPS data were able to correlate child test scores with teacher qualifications. They showed a gap of less than 2 percent between children taught by a teacher with secondary education and children taught by one with post-secondary education working in a public school. But children taught by a more educated teacher working in a private school scored 19 percent higher.

Raising the minimum qualification would mean that villages would have to import outside teachers rather than the secondary school-trained women who made up an important part of the teaching force in private schools, most likely raising absenteeism and forcing schools to either shut down or raise fees.

The project also contributes to a debate on school regulation. It found by studying the location of private schools, that they did not behave as monopolies, because they were typically clustered tightly in the villages, and their prices were kept low by competition. (They charged an average of 60 rupees a month — less than a day’s unskilled wage — and 18 percent of the poorest third of the population sent their children to private schools where they existed.)

But the role of government schools was very important in ensuring access to education, suggesting that the two should be thought of as complementary. Whereas private schools clustered in the centers of villages, or in wealthier neighborhoods, public schools were scattered more equitably.

Khwaja and his colleagues are integrating the policy and research focus of the LEAPS project. Together with Andrabi, Das, Vishwanath, and Tristan Zajonc of Harvard University, he is in the process of turning their insights into a book, which they hope will make their work more accessible — and, with billions of aid dollars about to be distributed, more applicable. At the same time, they have brought standard economic tools to bear on important topics such as the benefits of maternal education, the process through which children learn, and the impact of providing information.

“Pakistan is a fascinating and challenging environment for this work,” Khwaja says. “The country’s educational system has come into sharp focus around the world. However, the real insights from the LEAPS project come from the opportunities and challenges that small scale private schools — perhaps best thought of as ‘microeducation’ — have opened up for parents and children. Given that this phenomenon is spreading at a speed that no one anticipated, the results from this study are relevant to our understanding of rural education for a large number of countries. Remaining sensitive to the particular context of Pakistan yet realizing the implications of our findings for other countries is a fine balancing act. But that’s the nice thing about being at a policy school. Where we can also say ‘this is a valuable project because it becomes valuable policy — not only for Pakistan, but also for a larger developing world.’”