Midday meals for schoolchildren in India: More good than harm

On July 16th at least 23 children in the Indian state of Bihar died after eating a midday meal that was provided for free by their school. Nearly as many are in critical condition in a local hospital. Tests have revealed that adulterated cooking oil, perhaps containing pesticides, is likely to blame. A government inquiry has determined that the principal of the school, who is in hiding, must be held responsible for the bad ingredients or unsafe methods used in preparing these meals.

This event is horrific, without a doubt. Yet its damage could be even worse, if it raises too many doubts about the value of a largely successful programme. The midday-meal scheme, which began on a small scale decades earlier, received the support of India’s Supreme Court in 2001. Since then most Indian states have adopted it, offering free meals to children in state-run or state-assisted schools. More than 120m children, including many who would otherwise go hungry, receive these meals every school day.

According to a recent analysis by Farzana Afridi of Syracuse University and the Delhi School of Economics, at a cost of three cents per child per school day, the scheme “reduced the daily protein deficiency of a primary-school student by 100%, the calorie deficiency by almost 30% and the daily iron deficiency by nearly 10%.” Ms Afridi also found that, after controlling for all other factors, the meals scheme has boosted the school attendance of girls by 12%. Abhijeet Singh of Oxford University found that, in some parts of India where children were born during a drought, the health of those who had been brought into the meals scheme before the age of six was compensated for earlier nutritional deficits.

What the disaster in Bihar has done, at the very least, is to highlight some of the pitfalls of the scheme. As with any programme of this size in a country rife with corruption, the meals scheme is riddled with problems. The corruptible state is not alone in funding the programme; it is joined by private companies and NGOs. Corruption exists not just among state entities but among the supporting agencies too, as was demonstrated in 2006 when a Delhi NGO was caught dipping into rice that was meant for midday meals. In the states of Bihar and Uttar Pradesh, where the levels of malnutrition are among the highest in the country, it was found that only three-fourths of the food meant for children reached them. Food is often stolen by the administrators’ faking their students’ attendance. Beyond that, reports of adulteration—not only with shoddy or unsafe foodstuffs, but including finding worms, lizards and snakes—are common.

Next month, the Indian government will be voting on a food security bill which aims to provide food to 60% of the entire population, by means of a public distribution system. This one school’s tragedy comes at an especially crucial moment, when officials ought to be forced to inspect the leaky pipeline of distribution. At the same time it will be important to bear in mind: This scheme has done a lot more good than harm.

First published on economist.com.

Image credit: GlobalPartnership for Education

To kill, cheetahs use agility and acceleration not top speed

Researchers have used gadget-laden collars to record cheetahs’ movements in the wild. They found that cheetahs succeed not because it is the fastest animal on land, but because of its incredible acceleration and unmatched turning speeds.

Most of what we know about cheetahs in the wild is based on direct observation, or through videos from remote cameras. This limits our understanding of cheetahs to open habitats and daytime. Alan Wilson at the University of London’s Royal Veterinary College wanted to study cheetahs better.

Over the past ten years, Wilson and his team have been perfecting devices to study the locomotion of animals. For cheetahs, they assembled a collar that carries a GPS to record location data, an accelerometer to measure speed, a gyroscope to understand angular motion, and a magnetometer to make location data more accurate, which it does by measuring tiny changes in Earth’s magnetic field. The data were transmitted back to the researchers in real time through radio.

“The key development,” Wilson said, “was to pack all that in a low-power device”. The collar relies only on solar cells for recharging, but carries a battery in case of failure.

After tracking 367 runs by five cheetahs in the wild, Wilson found many surprising results.

First, the top speed of most cheetah hunts is on average half the “record speed”. That record speed is 102 km per hour, and was noted in 1965 (though not published until 1997), by a veterinary surgeon in Kenya.

The average length of a cheetah’s hunt was about 180 meters. Instead, on average, cheetahs covered about six kilometers every day. With only two hunts made every three days, high speed runs make for only a tiny fraction of a cheetah’s daily routine.

Second, he found that cheetahs can successfully hunt in all terrains, not just open fields. The run data were overlaid on Google Earth to visualise the landscape the cheetahs were operating in. This showed that only 20% of chases in open fields were successful, compared to 31% in dense cover. Wilson thinks that dense cover, such as trees, might give cheetahs vantage points that open fields cannot.

Third, cheetahs can decelerate faster than they can accelerate, much as sports cars with powerful engines need beefed-up brakes. While both these processes require different sets of muscles and depend on different conditions, the rates of acceleration and deceleration beat those of any other land-dwelling animal. Based on the recorded data, Wilson calculates that the muscle power output of cheetahs is about four times that of Usain Bolt, three times that of polo horses, and nearly double that of greyhounds.

The top speed of a cheetah hunt had no correlation to the successful outcome of the hunt. Instead, Wilson found that success depended more on how fast the cheetah could slow down, rather than on how fast it could speed up. It is this last phase of a hunt that was critical for success, where the cheetah slows down. When these two observations are put together, Wilson thinks that it seems cheetahs don’t abandon hunts early to save energy or reduce risk of injury.

Finally, cheetahs are not built to be able to turn at their highest speed. In an artificial setting, which astronauts and fighter pilots are put into for training, the force felt by a cheetah trying to turn around at top speed could knock it unconscious. Instead they use their ability to slow down and their ridged footpads and claws to grip the ground well enough to turn quickly.

The results of Wilson’s work are published in the journal Nature today. Craig McGowan at the University of Idaho, an expert in understanding animal locomotion who was not involved in this, was impressed by Wilson’s work. “This research has been able to collect a huge amount of data from animals behaving naturally in their environment. No other dataset of this kind exists,” he said.

Roger Kram at the University of Colorado, Boulder, another biomechanics expert who was not involved in the study, said, “The technology used is absolutely fantastic. Most people studying biomechanics of running do so in labs. I’d like to see this technology applied to prey, such as impala and Thomson’s gazelle.”

Wilson is keen to see the technology used widely. “My aim is not to commercialise this. We’ve revealed all the technology and methods in our paper,” he said. His team has already started using it on lions and wild dogs.The Conversation

First published on The Conversation.

Image credit: photosbyflick

This story made it to the front page of Reddit and Digg, receiving over 130,000 views in two days.

Social entrepreneurs in India: Water for all

Nearly three-fourths of all diseases caused in India are due to water contaminants. Despite that, one in eight Indians still lacks access to clean drinking water. The poor now realise that paying for clean water can save much more in health-care costs later. It was this market that Sarvajal, a social enterprise in India, wanted to cater to.

Founded in 2008, Sarvajal—which in Sanskrit means “water for all”—now sells clean drinking water to more than 70,000 people in rural India. In bigger villages, it employs local people to man filtration plants and sell water. In small villages it installs solar-powered water dispensing machines (pictured) that use prepaid (or pay-as-you-go) smart cards that can be topped up just like a mobile phone. The machines send data to a central server via SMS, which helps Sarvajal ensure regular supply of clean water.

Sarvajal started with some help from the Piramal Foundation, a charity. And it is not alone: Water Health International was launched with an investment from the Acumen Fund and the Naandi Foundation’s not-for-profit company was backed by a charity with the same name. What sets Sarvajal apart is that it has stayed away from government subsidies while still keeping the price of water low. It sells 10 litres of water for four pence (or six cents), just as much or lower than its competitors.

“Subsidies are not a long-term solution,” says Anand Shah, Savajal’s founder, who grew up in America and moved to India to become a social entrepreneur. It took a healthy bit of tinkering to lower the price of installation and maintenance for its water supply infrastructure. It costs on average $2,500 to install a filtration plant, which is about half the expense of similar projects. Sarvajal claims to recover those costs within three years.

Setting up its project was not easy. Savajal needed to deal with things that few businesses in rich countries have to worry about: lack of proper roads in villages, irregularity of power supply, unreliability of water sources and devising a system of money transfer. Having reached a respectable size, Mr Shah is hopeful that scaling up his business further will be less challenging.

Apart from villages, Sarvajal’s other obvious market is the urban poor. Nearly 100m people live in very densely populated slums in India’s cities. They are more willing to pay a higher price for water than villagers who have a much smaller disposable income. But Mr Shah says that “water barons”, sellers of bottled-water, have been trying to block Sarvajal’s entry into cities. After many months of efforts, this time not without help from the government, Sarvajal will soon be launching its first filtration plant in Delhi.

First published on economist.com.

Image credit: Sarvajal