Introduction

The growing interest to find a measurable metric that can indicate the quality of life of a country’s residents has swamped the emerging economies across the world. National and international organizations, governments, academia, think tanks, investors and development partners are either searching for reliable indicators or building their own. Accumulating data at regular frequency, finding the right metric that corroborates with the socio-politico-economic condition of a country and conducting traditional surveys and censuses incurs massive economic investment. This is particularly challenging for the emerging economies.

However, massive computational power to harness non-traditional data from sensors, signals, and satellite imagery is also showing their potential for continuous monitoring of the lives and environments of the localities.  The alternatives are showing promise due to their high frequency, spread and cost effectiveness. Moreover, use of new technologies for data accumulation, analysis and visualization are getting popular. With the advent of such data centric initiatives various governmental and non-governmental entities have popularized initiatives such as open data platforms.  The world wants to monitor the socio-politico-economic health of the counties.

bdqoL web portal is one such endeavor dedicated to Bangladesh. Here we envision to visualize various  traditional and non-traditional indicators that can help us monitor the quality of lives of our residents. Data and Design Lab hopes that this platform will grow as we acquire newer data and learn novel visual methods with the help of participation, comments,  criticisms and suggestions of the platform users. The endeavor is a continuous process, a work in progress. At this nascent stage we only managed to visualize few indicators that were generated or accumulated from raw data from national and international sources. Along with presenting various time series, correlational, map-based visualizations of traditional data we also present ranks of districts based on their most recent standing on specific indicators.  As indicated earlier as the traditional data were collected by different organizations they vary in frequency, time, methodology and coverage. We tried our best to visualize the most recent data and the process will be continuously updated.  

One of the challenges of building composite indicators to capture the progress of living standard lies in choosing appropriate indicators and then assigning appropriate weighs to the indicators. We wanted to give the researchers and enthusiasts a chance to decide on their own weight, create their own rankings and clusters, and finally visualize them.  

Bangladesh turns 50 on 26th of March 2021. However, in absence of recent data we have to rely upon works conducted few years back. Hence the recent uplift of the country is largely unrepresented or underrepresented in most of data centric socio-economic research. Research conducted around 2016 shows that despite progress in attaining several development goals and strong political commitments, we are still facing substantial challenge in dealing with raising regional inequity (NIPORT-measure evaluation-RDM, 2016). Literature also indicates that regional disparity within the country has now been a well-established fact (Khondker and Mehzab, 2016). A vast number of studies have tried to investigate the nature and cause of this inequality and lag. Studies came up with various finding like- an east-west division exists where the eastern districts have shown more progress (GED 2008; Narayan et al. 2007 and WB 2008;  Deb et al. 2008); less integrated regions such as Rajshahi, Khulna, Barisal were lagging behind (Shilpi, 2008; Zohir, 2011); lagging areas even exist in developed districts (Zohir, 2011); West was lagging behind in income and consumption expenditure, but in terms of human capital, west was equally good or to some extent better than the east (Sen et al., 2014); Rajshahi division was lagging behind in terms of density of poor people but Barishal division had the most severe poverty  (Khondker & Wadud, 2010) and so on. This shows the importance to visualize how individual localities are faring in comparison with others and pinpointing the regions that have similarity based on certain characteristics.

bdqol.dndlab.org project aims to give a chance to its users to look at Bangladesh at district level. We wanted to users to have a chance to see how different districts are faring in comparison with others. We have conducted both rank and cluster analysis at the district level based on some selected indicators to reflect the overall socio-economic and livelihood condition of the people living in those administrative units. We have considered five basic categories of indicators- living standard, health status, education, infrastructure, and environment. Description and rationale for selecting the categories and indicators are provided in the data catalogue page. Furthermore, we have constructed a composite indicator to incorporate all the dimension simultaneously. In constructing this composite indicator, we have mainly followed the same methodology applied in constructing the conventional composite indicators like, wealth index, MPI or HDI. However, the difference is we allow the user to maneuver the selection of indicators and as well as the weight of each indicator. Such option allows to analyze the district ranking and cluster construction from different developmental aspect and objective as different development programs may have different set of priorities. Once the user selects the indicators and the weights of each indication, the system will automatically generate a cluster map of the similar districts (based on k-means method) and will show the ranking of each individual district. So, rather than using a fixed weight or fixed set of indicators, the user has the freedom to set the priority. Follow this link to the page on ranking and clustering based on the composite indicator.

This district level ranking and cluster analysis is crucial as over the years the different regions of the county have shown varying progress in different development indicators as a consequence the development goals of local areas also vary significantly. Therefore, an updated knowledge on the state of regional inequality is an essential requirement for the development planners and economists for priority setting, program monitoring and evaluation purpose. Besides, in Bangladesh, information on development indicators comes from multiple sources, e.g., HIES, DHS MICS, BMSS, SVRS and so on. Often it becomes technically challenging and time consuming for users to incorporate all the required data in their analysis. In many cases the data are not widely available at the district level. Therefore, this common (data) platform gives the users a comparative advantage.  Overall, this project will facilitate the use of district-level data for local-level planning and monitoring of various socio-economic and development indicators.

bdqol.dndlab.org is a work in progress of data and design lab and we believe it will remain to be so for the coming years. We hope that this platform will usher a new era of how we gather, value and view data and help policymakers, researchers, development workers, and enthusiasts to aid their respective endeavors.

Acknowledgement: Visualizations are done using leaflet.js and c3.js.

Disclaimer: The data visualized in this website is collected as is from the sources cited. The Data and Design lab shall not be liable for any errors and inconsistencies in the data taken from these sources.

Introduction

The growing interest to find a measurable metric that can indicate the quality of life of a country’s residents has swamped the emerging economies across the world. National and international organizations, governments, academia, think tanks, investors and development partners are either searching for reliable indicators or building their own. Accumulating data at regular frequency, finding the right metric that corroborates with the socio-politico-economic condition of a country and conducting traditional surveys and censuses incurs massive economic investment. This is particularly challenging for the emerging economies.

However, massive computational power to harness non-traditional data from sensors, signals, and satellite imagery is also showing their potential for continuous monitoring of the lives and environments of the localities.  The alternatives are showing promise due to their high frequency, spread and cost effectiveness. Moreover, use of new technologies for data accumulation, analysis and visualization are getting popular. With the advent of such data centric initiatives various governmental and non-governmental entities have popularized initiatives such as open data platforms.  The world wants to monitor the socio-politico-economic health of the counties.

bdqoL web portal is one such endeavor dedicated to Bangladesh. Here we envision to visualize various  traditional and non-traditional indicators that can help us monitor the quality of lives of our residents. Data and Design Lab hopes that this platform will grow as we acquire newer data and learn novel visual methods with the help of participation, comments,  criticisms and suggestions of the platform users. The endeavor is a continuous process, a work in progress. At this nascent stage we only managed to visualize few indicators that were generated or accumulated from raw data from national and international sources. Along with presenting various time series, correlational, map-based visualizations of traditional data we also present ranks of districts based on their most recent standing on specific indicators.  As indicated earlier as the traditional data were collected by different organizations they vary in frequency, time, methodology and coverage. We tried our best to visualize the most recent data and the process will be continuously updated.  

One of the challenges of building composite indicators to capture the progress of living standard lies in choosing appropriate indicators and then assigning appropriate weighs to the indicators. We wanted to give the researchers and enthusiasts a chance to decide on their own weight, create their own rankings and clusters, and finally visualize them.  

Bangladesh turns 50 on 26th of March 2021. However, in absence of recent data we have to rely upon works conducted few years back. Hence the recent uplift of the country is largely unrepresented or underrepresented in most of data centric socio-economic research. Research conducted around 2016 shows that despite progress in attaining several development goals and strong political commitments, we are still facing substantial challenge in dealing with raising regional inequity (NIPORT-measure evaluation-RDM, 2016). Literature also indicates that regional disparity within the country has now been a well-established fact (Khondker and Mehzab, 2016). A vast number of studies have tried to investigate the nature and cause of this inequality and lag. Studies came up with various finding like- an east-west division exists where the eastern districts have shown more progress (GED 2008; Narayan et al. 2007 and WB 2008;  Deb et al. 2008); less integrated regions such as Rajshahi, Khulna, Barisal were lagging behind (Shilpi, 2008; Zohir, 2011); lagging areas even exist in developed districts (Zohir, 2011); West was lagging behind in income and consumption expenditure, but in terms of human capital, west was equally good or to some extent better than the east (Sen et al., 2014); Rajshahi division was lagging behind in terms of density of poor people but Barishal division had the most severe poverty  (Khondker & Wadud, 2010) and so on. This shows the importance to visualize how individual localities are faring in comparison with others and pinpointing the regions that have similarity based on certain characteristics.

bdqol.dndlab.org project aims to give a chance to its users to look at Bangladesh at district level. We wanted to users to have a chance to see how different districts are faring in comparison with others. We have conducted both rank and cluster analysis at the district level based on some selected indicators to reflect the overall socio-economic and livelihood condition of the people living in those administrative units. We have considered five basic categories of indicators- living standard, health status, education, infrastructure, and environment. Description and rationale for selecting the categories and indicators are provided in the data catalogue page. Furthermore, we have constructed a composite indicator to incorporate all the dimension simultaneously. In constructing this composite indicator, we have mainly followed the same methodology applied in constructing the conventional composite indicators like, wealth index, MPI or HDI. However, the difference is we allow the user to maneuver the selection of indicators and as well as the weight of each indicator. Such option allows to analyze the district ranking and cluster construction from different developmental aspect and objective as different development programs may have different set of priorities. Once the user selects the indicators and the weights of each indication, the system will automatically generate a cluster map of the similar districts (based on k-means method) and will show the ranking of each individual district. So, rather than using a fixed weight or fixed set of indicators, the user has the freedom to set the priority. Follow this link to the page on ranking and clustering based on the composite indicator.

This district level ranking and cluster analysis is crucial as over the years the different regions of the county have shown varying progress in different development indicators as a consequence the development goals of local areas also vary significantly. Therefore, an updated knowledge on the state of regional inequality is an essential requirement for the development planners and economists for priority setting, program monitoring and evaluation purpose. Besides, in Bangladesh, information on development indicators comes from multiple sources, e.g., HIES, DHS MICS, BMSS, SVRS and so on. Often it becomes technically challenging and time consuming for users to incorporate all the required data in their analysis. In many cases the data are not widely available at the district level. Therefore, this common (data) platform gives the users a comparative advantage.  Overall, this project will facilitate the use of district-level data for local-level planning and monitoring of various socio-economic and development indicators.

bdqol.dndlab.org is a work in progress of data and design lab and we believe it will remain to be so for the coming years. We hope that this platform will usher a new era of how we gather, value and view data and help policymakers, researchers, development workers, and enthusiasts to aid their respective endeavors.

Acknowledgement: Visualizations are done using leaflet.js and c3.js.

Disclaimer: The data visualized in this website is collected as is from the sources cited. The Data and Design lab shall not be liable for any errors and inconsistencies in the data taken from these sources.