Annotated Bibliography

Ziyue Feng

Word Count: 2,307

Li, Yuheng et al. “Realizing Targeted Poverty Alleviation In China”. China Agricultural Economic Review, vol 8, no. 3, 2016, pp. 443-454. Emerald, `doi:10.1108/caer-11-2015-0157.

This article attempts to assess and evaluate how the targeted poverty alleviation policy by the Chinese government performs in terms of effectiveness and people’s expectations. Researchers decided to focus their investigation on the people’s voices, implementation challenges and implications for policymaking and improvement. Poverty eradication by China and the Chinese government is under the worldwide spotlight due to the promise of ending absolute poverty by 2020, which is this year and ten years ahead of the deadline set by the United Nations. Back to 2015, there were still 7017 million population under the government’s official extreme poverty line of 2300 RMB ($362.5) per capita annual net income and becoming the targeted population of the poverty alleviation strategy. A new 5-year poverty alleviation plan has been conducted, highlighting the importance of accurate poverty identification, appropriate projects arrangement and accurate effect. Though the government believes it is promising to ensure projected assistance to reach needed villages and households, researchers chose to critically evaluate the implementation based on a nationwide survey, one of the most traditional means in data collection. This national survey was carried out in August 2015 and supported by the State Council. It managed to cover 2075 households in 22 impoverished counties of 13 provinces across China. Not only examination of the policy implementation, but also “voices” like willingness, preference or advice from the local bureaus, enterprises and households had been collected, which I think well fits the idea of development as freedom by Amartya Sen. The assessment of poverty eradication should be done in terms of people’s internal thoughts that whether their (material) freedoms or living standards are enhanced. Ultimately, a total amount of 2139 questionnaires were made and collected. The survey intended to find the linkage among the causes of poverty, the practical needs of people in poverty and their intending usage of expected funds aid. Though this report does not conclude with a clear relationship between these factors, by visualizing the data, some trends and signs are indicated and easy to understand. For instance, for people whose poverty were caused by lack of education, they were much more interested in occupational training and characteristic industry development; while if it was diseases that led to their poverty, people then had a focusing need on social security. The article also evaluated the limitation and challenges of poverty assessment in this case of China. Researchers considered that a sole standard for poverty identification as an income standard of RMB 2300 was a poor mechanism which can malfunction due to demo-geographic factors including no official statistics of income level in rural villages and different purchasing power of 2300 RMB in different areas. This is an important reflection on the judgement of poverty. I think it is this kind of views and opinions that actually stimulate the development of poverty assessment and its methodology.

Ma, Libang et al. “Rural Poverty Identification And Comprehensive Poverty Assessment Based On Quality-Of-Life: The Case Of Gansu Province (China)”. Sustainability, vol 11, no. 17, 2019, p. 4547. MDPI AG, doi:10.3390/su11174547.

Wang, Wen et al. “Poverty Assessment Using DMSP/OLS Night-Time Light Satellite Imagery At A Provincial Scale In China”. Advances In Space Research, vol 49, no. 8, 2012, pp. 1253-1264. Elsevier BV, doi:10.1016/j.asr.2012.01.025.

This paper introduces a new assessing method for poverty level by using DMSP/OLS night-time light satellite imagery. It first discussed traditional studies on regional poverty evaluations, which are based on statistics collected by the economic census, and its deficiencies including a huge amount of economic costs, long period to update existing data and lack of spatial information for special demand. By comparing using satellite imagery with traditional methods, this paper points out that satellite remote sensing has an advantage to provide efficient and accurate spatial data because of its high temporal resolution and extensive spatial coverage. Satellite imagery has been conducted in many socio-economic fields including population, energy consumption, urban sprawl and greenhouse gas emission; and it should be also well used in studies on poverty issues, typically in China with vast areas and multiple characteristics. Researchers combine the 3-year DMSP/OLS night-time light data with other socio-economic indicators to establish DMSP/OLS night-time average light indices at a provisional scale in China and analyze the relationship between them and an integrated poverty index, in order to prove the representativeness of the satellite data on poverty level. DMSP/OLS night-time data are annual night-time cloud-free image composites of lights of the globe collected by the DMSO/OLS sensors on a low-earth orbiting satellite. By using this satellite technology, statistics like average light index (ALI) can be conducted from total regional luminance of night-time light and the number of total pixels in satellite imagery. Researchers aim to find out the positive and accurate correlation between intensiveness of night-time light and the evaluated poverty level in a region, and the latter one will be represented by integrated poverty index (IPI). They used 17 socio-economic variables to extract this IPI that can become a multidimensional community-level poverty indicator. The chosen variables not only include economic indicators such as GDP per capita, but also contain statistics like life expectancy, illiteracy ratio and living space per capita. This again well explained Amartya Sen’s point of view that the degree of freedom itself is just as important as economic indicators. Thirty-one provinces and municipalities in mainland China have been selected to carry out this study, and the 3-year socio-economic statistical data (from 2007 to 2009) for the selected provinces and municipalities is obtained from the National Bureau of Statistics of China. By analyzing the relationship of ALI and IPI, researchers manage to find out a good correlation between them with a coefficient of determination of 0.854 and in this case, a conclusion that DMSP/OLS night-time light data is a valid data source for estimating regional poverty issues. In addition, this technological application is still at a preliminary stage so there will be many other probabilities in its application. For instance, some scholars have tested the combination of the LandScan population data and the night-time data. Though there are many limitations to it, it is a great signal for trying new technological application which will finally lead to progress and development.

Steele JE et al. 2017, “Mapping poverty using mobile phone and satellite data.” J. R. Soc. Interface 14: 20160690. http://dx.doi.org/10.1098/rsif.2016.0690

This article intends to demonstrate the application of mobile phone data and satellite data for poverty assessment and examine its effectiveness. Scholars believe that traditional approaches to measuring and targeting poverty heavily rely on census data, and they are inappropriate to be conducted in most low- and middle-income countries due to lack of availability or update. Therefore, this study attempts to find alternative measures available and accessible for low- and middle-income countries which can provide novel insight into the spatial distribution of poverty. Researchers decide to use aggregate data from mobile operators and geospatial data for the evaluation of three poverty assessing models: DHS WI, PPI and income-based measures. In brief, these models tend to show poverty level from asset, consumption and income level. Researchers choose remote sensing and geographic information system data (RS data) and mobile operator call detail records (CDR data) as their data sources. In this study, they intend to evaluate the predictive power and ability to measure poverty in integrating both data sources. The logic behind using these data to assess the poverty level is that they can capture distinct and complimentary correlates of human living conditions and behaviour. For instance, RS data obtain information like distance to roads and cities, which can reflect access to market and information to a certain degree; and similarly, monthly credit consumption on mobile phones and the proportion of people in an area using mobile phones indicate household access to financial resources. These two sources of data are complementary in terms of not only the information they can reflect, but also geographical area each of them can cover in this study. For CDR indicators, the spatial resolution is determined by the coverage of physical cell towers, which is larger in rural areas and fine-scaled in urban areas. By contrast, RS data can be relatively coarse in urban areas. Researchers use overlapping sources of RS, CDR and traditional survey-based data from Bangladesh to provide the first systematic evaluation of the extent to which different sources of input area can accurately estimate the three different measures of poverty mentioned above, which were obtained in advance from Bangladesh officials. First, researchers will evaluate the correlations of RS and CDR data to the three measures. In addition, prediction mapping is then to be conducted at unsampled locations across the population based on the previous results. Through the experiment and research, they find models employing a combination of CDR and RS data generally provide an advantage over models based on either data source alone, which verifies the predicted complementarity, though RS-only and CDR-only also make a good result on their own. Moreover, an external finding makes scholars think about the correlations between socio-economic measures and night-time light intensity, access to roads and cities, the entropy of contacts and mobility features. This article draws the conclusion that methods that exploit information from, and correlations between, many different data sources will provide the greatest benefit in understanding the distribution of human living conditions.

Sedda, Luigi et al. “Poverty, Health And Satellite-Derived Vegetation Indices: Their Inter-Spatial Relationship In West Africa”. International Health, vol 7, no. 2, 2015, pp. 99-106. Oxford University Press (OUP), doi:10.1093/inthealth/ihv005.

This report aims to reveal the relationship existed between poverty, health and the environment. Researchers decide to conduct their study in West Africa because in 2012 while only 3.7% of the population in Europe and Asia was under an international poverty line of US$1.25 a day, that proportion increased to half in sub-Saharan Africa (SSA) area. Despite improvements in socio-economic conditions, poverty increases are still reported for some areas of SSA. They believe it is inequality, unequal redistribution and disparities in access to resources like food and health care make a negative impact on the most vulnerable groups, such as women and children, promoting poverty, poor health and disadvantages. Moreover, in SSA, 75% of the population lives in rural areas with agriculture and livestock as their main source of livelihood and therefore are often threat due to drought. Researchers have firmly believed the existence of correlations between poverty, health and the environment in SSA before this experiment. Nonetheless, the calculation of quantitative measures to represent poverty is a complex issue, and they think the spatial representation of various poverty indices would be increasingly required to achieve greater understandings of poverty and causes of it. So they carry out this analysis to test whether incorporating environmental information into spatial models for prediction of the spatial distribution of multidimensional poverty index (MPI) increases predictive accuracy. Researchers use the following equation to determine MPI: MPI=H*A, where H is called headcount ratio, meaning the percentage of people who are poor, and A measures the degree and share of dimensions in which poor people are deprived (intensity of poverty). And for the environmental variables, this study uses indicators that commonly related with poverty and health, which are day- and night-time land surface temperature, normalized difference vegetation index and elevation measured via a digital elevation model obtained from the MODIS sensor of NASA’s satellites. Researchers employed the methods that belong to the field of geostatistics, which is based on the fitting of explicit spatial and Spatio-temporal correlation functions to parameterize a Random Function model for subsequent use in spatial prediction. This will be the spatial interconnections between poverty indices and the environment in this study. By this research, they have demonstrated the spatial interconnections between poverty and vegetation in West Africa by finding a greatest correlation between normalized difference vegetation index and A (intensity of poverty). Because the NDVI is independent of the process of calculating A, the probability of being process-driven for this correlation results are excluded. Moreover, researchers consider that such conclusion should be extended to drylands in general (40% of the world’s land surface) instead of the only studied area (West Africa) and similar statistical procedures can be applied to other environmental components such as land use or population intensity to improve poverty mapping system.