Bulletin of Geography. Socio-economic Series, No. 51 (2021): 109–122 http://doi.org/10.2478/bog-2021-0009 BULLETIN OF GEOGRAPHY. SOCIO–ECONOMIC SERIES journal homepages: https://content.sciendo.com/view/journals/bog/bog-overview.xml ISSN 1732–4254 quarterly http://apcz.umk.pl/czasopisma/index.php/BGSS/index Multivariate classification of provinces of Vietnam according to the level of sustainable development Van Canh Truong University of Da Nang, University of Science and Education, 459 Ton Duc Thang – Da Nang – Vietnam, tel: (+48) 729 451 495, e-mail: [email protected], https://orcid.org/0000-0001-6024-6692 How to cite: Truong, V.C. (2021). Multivariate classification of provinces of Vietnam according to the level of sustainable development. Bulle- tin of Geography. Socio-economic Series, 51(51): 109-122. DOI: http://doi.org/10.2478/bog-2021-0009 Abstract. The research aims to classify the level of sustainability of 63 provinc- Article details: es in Vietnam upon 24 indicators reflecting three main dimensions of sustainable Received: 27 March 2020 development by using multivariate classification method for the year 2014-2016. Revised: 29 September 2020 First of all, the principal component analysis (PCA) was applied to group quanti- Accepted: 20 October 2020 tative variables that reflect important aspects of each component of sustainable de- velopment of localities in Vietnam into a number of limited dimensions (factors). The results of PCA illustrate 8 principal components in which 3 main components of economic and social pillar, and 2 main components for environmental pillar. After that, the second method was applied by using the hierarchical methods of cluster analysis for the set of 8 principal components conducted by PCA. With the candidate solution divided all of the provinces into 6 clusters, the analysis shows that the higher level of sustainable development belongs to South East, Red River Delta, and North Central, in which 4 provinces of South East (Ho Chi Minh City, Key words: Binh Duong, Dong Nai, and Ba Ria – Vung Tau) form a group with the highest principal component analysis, level of sustainability. The regions such as South Central Coastal, Highland area, cluster analysis, and Mekong River Delta have the medium level; the North midland and moun- level of sustainability, tain areas are regions at the low level of sustainability. indicators Contents: 1. Introduction . 110 2. Indicators and data source............................................................... 110 3. Results and discussion .................................................................. 111 3.1 Principal component analysis (PCA) for sustainable development ........................ 111 3.2. Classification of level of sustainability by using cluster analysis .......................... 116 © 2021 (Van Canh Truong) This is an open access article licensed under the Creative Commons Attribution-NonCommercial-NoDerivs License (http://creativecommons.org/licenses/by-nc-nd/4.0/). 110 Van Canh Truong / Bulletin of Geography. Socio-economic Series / 51 (2021): 109–122 4. Conclusion ............................................................................ 121 References ............................................................................... 121 1. Introduction Indicators, as variables, are indispensable tool to make the concept of sustainable development op- erational. They help us to construct a comprehen- One of the most striking characteristics of sustain- sive picture of state of a complex system, to translate able development is that it means so many different the concepts and objectives of sustainable devel- things to so many different people and organiza- opment into practical terms (Hardi et al., 1997; tions. Sustainable development involves not just one Bossel, 1999; Li-Yin Shen, 2011). Based on 24 indi- but three main complex interacting systems (eco- cators which cover a wide range of three main di- nomic, social and environmental subsystem). In mensions of sustainable development, the research each subsystem, it also contains many things and aims to classify the level of sustainability of 63 prov- a great deal of relationships. Sustainable economy inces in Vietnam by using multivariate classifica- means diversity, efficiency, creativity and its bal- tion method. The multivariate classification aims to anced relations with society and environment. Sus- bring diverse features into certain groups, with sim- tainable society is revealed by justice, diversity, ilar characteristics, expressed in certain sets of cri- security and its balanced relations with economy teria. First of all, the principal component analysis and environment. Finally, sustainable environment (PCA) was utilized to group quantitative variables means closeness of ecosystem, biodiversity, troph- that reflect important aspects of each component ic structure, resource renewal and its balanced re- of sustainable development of localities in Vietnam lations with economy and society (Cohen et al., into a number of limited dimensions (factors, prin- 1998). Adam Szirmai (2015), Professor of Develop- cipal components). After that, the second method ment Economics at Maastricht University consid- was applied by using the hierarchical methods of ers, that “sustainable development’ is rather fuzzy cluster analysis for the set of principal components concept which is sometimes broadened to include conducted by PCA. The analysis use the complete a variety of desirable goals such as equitable devel- raw existing data set provided for 24 indicators for opment, social inclusion or poverty reduction. In the years 2014 – 2016 of 63 administrative units of addition to these three major aspects, many people Vietnam. also mention other aspects of sustainable develop- ment such as governance, culture, spirit, ethnicity, law ... and require calculation and balance them in 2. Indicators and data source planning strategies and policies for socio-econom- ic development for each specific country and local- ity. Other proposed aspect of spatial sustainability Primary and secondary data from government agen- which provides for a balance at different territorial cies and academic institutes have been collected. levels and results into local sustainable communi- Primary data are comprised of information gath- ties. This emprical research only take into account ered directly by technological monitoring, includ- the three main components of sustainability (eco- ing satellite-derived estimates of air quality. Data nomic, social, environmental) which are accepted for annual mean concentration of Particulate Matter widely. 2.5 (µg/m3) by province was synthesized from the We desire a tool that includes all relevant dimen- satellite data of air quality for Environmental Per- sions of sustainability in the evaluation process to formance Index of Yale University of United States assess their progress, level toward sustainable de- and retrieved from remote sensing data provided velopment and understand the interlinkages within by the Department of Physics and Atmospheric Sci- each component and between components of sus- ence, Dalhousie University in Canada. Secondary tainability (Kaivo-Oja et al., 2014; Fredericks, 2014). data include official statistical data formally report- Van Canh Truong / Bulletin of Geography. Socio-economic Series / 51 (2021): 109–122 111 ed by General Statistics Office (GSO) of Vietnamese same dimension, rather than with variables of oth- government. The set of 24 sustainable development er dimensions. It is then possible to interpret each indicators has been constructed. It covers a wide dimension according to the meaning of the load- range of issues with 8 indicators for each compo- ings of principal component assigned to variables. nent of sustainability, as following: The loadings (correlation coefficients) mean how • Economic component (8 indicators): GDP per many percent of the variance of the i-th variable capita (PPP current USD), GDP density (million is explained by the j-th principal component. The USD PPP per km2), Proportion of employment reduction of the number of variables allows to bet- in agriculture (%), Incremental capital-output ra- ter identify the characteristics of the research ob- tio (ICOR), Unemployment rate (% labor force), ject (in this study is the sustainable development Percentage of trained employed workers (%), of provinces and cities). Principal components de- Competitiveness Index, and Budget surplus as scribe the significant predictors of sustainability in percentage of GDP (%). each subsystem and factor scores were used to as- • Social component (8 indicators): Adult litera- sess the level of sustainability taking into consider- cy rate (%), Proportion of household own per- ation the positive or negative effect of these factors. manent house (%), Poverty rate (%), Gini index, One of the rules used in the analysis assumes that Female labor force participation rate (% male), cumulatively retained principal components should Prevalence of underweight children, weight for explain at least 60-70% of total variance. age (% of children under 5), Average life expec- The first step for PCA of each component is test- tancy at birth (year), and Proportion of death ing the correlation matrix. The research used KMO due to traffic accident (per 100.000 people). and Bartlett’s Test to test the assumption that varia- • Environmental component (8 indicators): For- bles are not correlated with each other in the whole est cover (% total land area), Agricultural