Technological advances enable us to have economic advantages and helps nations combat some of the worst economic evils including poverty. While combating is an important aspect, identification of poverty is of paramount importance as without identifying how poor a country and its people are, there is no way of devising steps to combat the evil.
Traditional methods of surveys are time consuming and many a times ineffective because far flung remote areas of a country are difficult to reach and hence a complete and accurate picture of poverty in a country can’t be drawn. This is where a research by Stanford scientists will help in a great deal as it enables use of satellites to predict poverty.
The technology developed by scientists involves combining satellite data with sophisticated machine learning to estimate household consumption and income from space. This eventually helps determine poverty levels of a country paving way for effective policies including targeted interventions.
Some of the poorest countries in the world are located in Africa and this is the continent that provides a particularly striking example of limited insights into economic wellbeing. World Bank has revealed in one of its report based on from 2000 to 2010, that out of the 59 African countries 39 conducted less than two surveys substantial enough to result in poverty measures. Surveys are costly, infrequent, and cannot always reach countries or regions within countries, for instance, due to armed conflict.
Recent studies show that satellite data capturing nightlights can be used to predict wealth in a given area; however, nightlight data alone is not effective at differentiating between regions at the bottom end of the income distribution, where satellite images appear uniformly dark. To circumvent this problem researchers at Stanford turned their attention to daylight imagery, which offers higher resolution and can capture features such as paved roads and metal roofs, markers that can help distinguish poor and ultra-poor regions. The researchers then developed a sophisticated learning algorithm that categorizes these features.
Several different validation methods reveal a high level of accuracy in their approach. The new model outperforms nightlight models by 81 per cent in predicting poverty in regions under the poverty line, the researchers say, and by 99 per cent in areas that are two times below the poverty line. Importantly, the new method uses publicly available daytime satellite data, can be repeated more frequently than surveys, and is inexpensive to use. Furthermore, initial evidence suggests that a model “trained” in one country can be used in another.