https://orcid.org/0000-0001-7823-0349
Retraction
Following publication of [1], concerns were raised that this article contains a substantial degree of overlap with a previously published article [2] by the same author with regard to reported figures, tables, methodology, references cited, datasets, and overall conclusions.
The author stated that [1] is an extension of the study reported in [2], that additional data were included in [1], and that the focus of the two articles is different. The PLOS One Editors consider the extent of overlap is such that this article [1] constitutes a redundant publication.
The PLOS One Editors retract this article [1] because, per our editorial assessment, it does not meet PLOS One’s publication criteria #2 that requires the reported results have not previously been published elsewhere [3].
The author did not agree with the retraction.
Tables 1-4 and Figs 3 and 5 report material previously published in [2] under a restrictive license. This material is not offered under a CC BY license and is therefore excluded from this article’s [1] license. Please provide due attribution to the original publication when referring to this content.
17 Mar 2026: The PLOS One Editors (2026) Retraction: Drivers of domestic migration in Vietnam: The characteristics of the households and their heads, environmental factors and living conditions. PLOS ONE 21(3): e0345057. https://doi.org/10.1371/journal.pone.0345057 View retraction
Figures
Abstract
Vietnam has achieved significant economic growth, poverty reduction, and social transformation since its 1986 major economic reform. However, industrialization, a key pillar supporting this achievement, has resulted in massive domestic migration from certain parts of the country into industrialized provinces mainly located in the south of Vietnam, leading to various challenges for society. This study investigates the effects of the characteristics of Vietnamese households and their heads, environmental factors, and living conditions of the households in the regions where they have decided to leave behind for a migration decision on domestic migration. Our study also compares the migration trends in the past decade using the Vietnamese Household Living Standard Surveys (VHLSS) in 2010 and 2020, together with a logit model. We find that the average probability that a household and their members migrate is about 10 per cent in 2010, reduced to approximately 6.8 per cent in 2020. Our empirical findings also confirm that a migration decision is strongly associated with the characteristics of the households and their heads, particularly for the household size and the educational level of the household head. The average radiation and rainfall are also associated with a migration decision. Weather temperature, water sources, and electricity supply also play an essential role in Vietnamese households’ migration decisions. Policy implications have emerged based on these empirical findings that the Vietnamese government should consider.
Citation: Vo DH (2024) RETRACTED: Drivers of domestic migration in Vietnam: The characteristics of the households and their heads, environmental factors and living conditions. PLoS ONE 19(6): e0304321. https://doi.org/10.1371/journal.pone.0304321
Editor: Gouranga Lal Dasvarma, Flinders University, AUSTRALIA
Received: November 5, 2023; Accepted: May 9, 2024; Published: June 27, 2024
Copyright: © 2024 Duc Hong Vo. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying the results presented in the study are available from Vietnam's General Statistics Office at https://www.gso.gov.vn/en/data-and-statistics/2022/06/results-of-the-viet-nam-household-living-standards-survey-2020/. Alternatively, data can be provided from the corresponding author upon reasonable request.
Funding: This study is funded by the Ministry of Education and Training of Vietnam under Grant B2023-MBS-07. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
There is no universal definition or comprehensive migration theory due to the complexity and context-specific nature of migration experiences [1]. Over the years, social scientists have introduced various concepts of migration. In the early work of Eisenstadt [2], migration is defined as the relocation of an individual or a group of persons from one society to another. The focus is to abandon the social norms and practices of the individual’s previous residence and adopt a new set of social customs in the new or subsequent place of migration where they have chosen to reside. Lee [3] defines migration as a permanent or semi-permanent change of residence, regardless of the distance of the move or the reason behind it. The definition excludes continual movements of nomads and migratory workers and temporary moves. The United Nations [4] considers migrating more than one year from the original residence to be permanent, whereas a residence of one year or less is called temporary migration. For internal migration, most countries use the place of birth at the time of the census in the destination region to determine migration. Zelinsky [5] describes migration as a perceptible and simultaneous shift in both spatial and social location, implying a permanent or semi-permanent change of residence, and distinguishes migration from other forms of spatial mobility such as commuting, vacation travel, or student mobility. Mishra [6] defines migration as the immigration or emigration of a population from a defined region to another region for permanent or semi-permanent settlement. Bilsborrow [7] states that migration is a movement involving crossing a political or administrative boundary and a change in usual residence. This definition enables us to measure a migration flow using census data.
Nguyen at al. [8] consider that migration is generally viewed as a selective process influenced by multiple factors differentiating migrants from non-migrants. The definition of migration varies depending on the context and data sources. Migration can involve movement within a single country or between different countries and is often linked to better human capital, access to migration networks, and the potential for enhancing human development. This process can encompass individuals, families, or large groups [9, 10]. In the context of Vietnam, Coxhead et al. [11] define migrants as those aged 15–59 who move across provincial boundaries.
Migration decisions are driven by various factors that shape individuals’ or households’ decision-making processes regarding their choice of location. The economic literature on migration assumes that a decision is rational and aims to maximize expected gains. Todaro’s classical migration model [12] explains that migration decisions depend on the wage differentials between locations and the probability of finding a job. Other models, such as the human capital model developed by Mincer [13], explain migration decisions due to differences in the returns to human capital investment. The new economics of labour migration (NELM) developed by Stark & Bloom [14] suggests that migration is a joint household decision to manage and minimize household income and survival risk. Ravenstein’s [15] "laws of migration" also explain that migration is driven by labour force surpluses and shortages, with “push” factors at the origin and “pull” factors at the destination. These push and pull factors include insufficient job opportunities, political, social, or economic insecurity, and better living conditions or education opportunities. Personal characteristics, social networks, and vulnerability to poverty also shape migration decisions [16].
The theory of individual migration has emerged with human capital theory developed in the early work of Mincer [13] and Becker [17]. This theory assumes that wages in migrants’ potential origin and destination depend on the skills of individuals, affecting their productivity in both places [10, 16]. Torado’s [12] model suggests that an individual’s human capital characteristics can affect their wages and ability to find a job after migration. In addition, a person’s characteristics can influence the cost of a move. For example, Sjaastad [18] points out that age significantly influences the decision to migrate, and younger workers are more likely to migrate than older ones. The human capital perspective considers that those with the highest expected income gap between migrants and non-migrants and/or those with the lowest migration costs are the most likely to migrate. Overall, this theory suggests that those with the highest level of education will migrate, provided that schooling at the destination has a more positive impact on income than at the origin. However, this may not be true for internal migration to the agricultural sector or other low-skill jobs [16, 19].
Environmental hazards and climate change’s influence on migration have recently gained considerable attention. Previous studies argue that global environmental change will displace large populations and affect the cost of migration for individuals and households. Gray & Bilsborrow [20] argue that households respond to environmental factors in diverse ways, resulting in complex migratory responses, challenging the existing narratives about vulnerability to environmentally induced migration. Climate change may constrain migration, representing an exacerbating force concerning environmental hazards [21–23]. The argument has moved beyond linear environmental "push" theory toward greater context integration, highlighting that migration is often a household strategy to diversify risk [24].
Among the empirical studies regarding the relationship between environmental factors and migration, water-related factors are an environmental issue recognized in recent migration studies. For example, Rakib et al. [25] report high migration risk due to socioeconomic vulnerability, drinking water scarcity, health threats from salinity hazards, coastal communities’ poverty and low adaptive capacities. They argue that improving socioeconomic conditions, providing alternative potable water sources, and enhancing local awareness of coastal disasters and their associated consequences is important to address the mass migration problem caused by climate change. Meanwhile, Stoler et al. [26] argue that household water insecurity is a significant factor in shaping a migration decision in socio-environmental change, such as climate change. These scholars present evidence that water-related physical and mental health disruptions, livelihoods beyond agriculture, and social relationships can motivate households to migrate. A framework for linking climate change, household water insecurity, and environmental migration is considered and provides implications for anti-poverty and development initiatives and water interventions to mitigate forced climate migration.
The lack of working opportunities in rural areas has been one of the main factors that increase the likelihood of migration. VanWey [27] argues that owning a significant amount of land can lower migration costs and increase the likelihood of migration, as it is a form of wealth associated with better economic conditions. However, owning agricultural land can also provide job opportunities for rural residents, reducing migration. Bhandari & Ghimire [28] argue that the agricultural sector was experiencing an increase in productivity due to technological advances, which directly affected the migration patterns in developing countries because of the decreased labour force. In addition, the transition from subsistence farming to market-based commercial agriculture has led to a surplus of agricultural labour, leading to greater use of modern agricultural techniques, such as modern chemical fertilizers, which require less manual work than traditional farm activities.
Few attempts have examined internal migration in Vietnam, such as Coxhead et al. [11]. Nevertheless, the Vietnamese economy has enjoyed a miracle of economic growth in the past three decades after its major economic reform (“Doi Moi” in Vietnamese) in 1986. Urbanization plays an important role in achieving this miracle in economic growth, poverty reduction and social transformation. During the urbanization process, domestic migration has emerged as a key contributor to this fast-paced process in Vietnam. Millions of workers from various provinces in the central region of Vietnam have migrated to provinces located in the south of the country for opportunities for themselves and their families. Cities and provinces such as Ho Chi Minh City, Ha Noi, Binh Duong, Dong Nai, Ba Ria–Vung Tau & Long An have attracted millions of workers and immigrants from other parts of the country. As such, it is about time to revisit the important issues of domestic migration in Vietnam with a focus on the role of important factors identified from previous academic studies globally, including (i) the characteristics of the households and their heads, (ii) environmental factors; and (iii) living conditions.
Following this introduction, the remaining paper is structured as follows. Section 2 provides an overview of the data and the research methodology used in this study. We present the descriptive statistics in Section 3. Section 4 presents and discusses the empirical findings, followed by the policy implications in section 5 of the paper.
2. Data and methodology
2.1. Data
The Viet Nam Household Living Standards Survey (VHLSS) has been conducted every two years by the General Statistics Office (GSO) to examine living standards for policy-making and socio-economic development planning from 1993 until now. The VHLSS surveys are designed to collect data and information to monitor the living standards of different population groups in Viet Nam the implementation of the Comprehensive Poverty Reduction and Growth Strategy, and the evaluation of achievement of the Sustainable Development Goals (SDGs) and Vietnam’s socio-economic development goals (VHLSS, 2020). Data from these surveys include essential demographic characteristics, education, health and health care, labour–employment, income, consumption expenditure, durable goods, housing, electricity, water, sanitation facilities, participation in poverty alleviation programs, household businesses, and community general characteristics.
Fifty per cent of households participating in the sample are generally rotated from one VHLSS survey to the next. The VHLSS rotates the enumeration areas instead of rotating households. It means that the enumeration areas of half of the communes are retained while the other half is rotated from one survey to another. For example, a maximum of 50 per cent of households living in the 2018 VHLSS survey areas already participated in the VHLSS 2016. Among these, 50 per cent and 25 per cent of the households participated in the 2016 and 2014 VHLSS surveys. As such, it is only possible to form panel data for up to three VHLSS surveys with a sufficiently large amount of data. However, this does not appear to be the case for the 2020 VHLSS survey. As such, forming the panel for the 2020 VLHSS survey–the latest survey with data availability when this study is conducted- does not appear to be feasible.
This study uses two cross-sectional data sets from the Vietnamese Household Living Standard Survey (VHLSS) from 2010 to 2020. The survey sample consisted of 9,402 observations in 2010 and 9,383 observations in 2020. The sample consists of data for the households and their heads, as well as various data and information regarding environmental issues and living conditions of the households.
We also utilize two data sets from the General Statistics Office of Vietnam (GSO) in 2010 and 2020. The GSO is an agency directly under the Ministry of Planning and Investment (MPI) that serves as an adviser to the MPI Minister in state management for statistics, conducts statistical activities, and provides economic and social information to institutions and individuals domestically and internationally following the law. These data sets comprise details about monthly average temperatures, rainfall, and sunshine hours of 63 provinces, which are used as environmental-level indicators in our analysis.
“Migration" is used as our dependent variable in this study. This variable is classified into two different groups: "Migrate" and "Not migrate". The term "Migrate" is used to refer to migration in general. As such, a household member is considered a migrant regardless of types of migration, such as inter-provincial, inter-district, or intra-district movement. In addition, a household member is considered a migrant if he or she has been away from home for at least six months over the last ten years. This group of migrants is also classified into two categories: (i) migrants who continue to visit their origin households and (ii) migrants who have permanently departed from their origin households.
2.2. The research methodology
The logit model [29] is used in this study. Outputs from the linear regression model are assigned a probability value between 0 and 1 using the logistic functions. The model provides the probabilistic prediction regarding a group an input belongs to rather than just a simple binary classification. In this study, our model is written as follows:
Where P denotes the probability that a household has at least one migrant. Householdi denotes a vector of the characteristics of the households and their heads, EnvironmentalFactori denotes a vector of the environments, and LivingConditioni is a vector of living conditions in the region in which household i lives. f represents the logit function, which is written as follows:
Where βX denotes , yj is the binary variable of the household i, which takes the value of 1 if the household has at least one migrant and 0 otherwise.
The three groups of variables are used to examine their effect on migration in Vietnam. First, the characteristics of the households and heads include total household income, the total number of members of the family, and also the age, gender, and education level of the household heads. Second, environmental factors include the average radiation, rainfall, and temperature for the regions in which the households reside before making a migration decision. Third, living conditions include water sources, electricity bills, chemical fertilizers, and pesticides. Table 1 presents the list of variables used in this study and their descriptions.
3. Empirical results
3.1. The descriptive statistics
Descriptive statistics for these two surveys are presented in Table 2. The final dataset includes 8,693 observations for the VHLSS survey 2010 and 8,567 observations in 2020. Collected data are grouped into different categories: (i) the characteristics of the households and their heads, (ii) the environmental factors, and (iii) living conditions. The descriptive statistics indicate that the average migration in 2010 was 0.10, higher than in 2020 (0.07), indicating more migrants in 2010 than in 2020. In addition, the mean value of gender is higher in 2020 than in 2010, showing that more households were headed by women in 2020. However, the total number of households headed by males is still higher in both years. The oldest age of the household head in 2010 was 99 years old, while the mean age of the entire sample was slightly over 48 years old. A family comprising 15 people was the maximum number of persons for a family in 2010, whereas it was 12 for the 2020 VHLSS survey. The average annual income of the households was VND 90 million in 2010, which increased to VND 154 million in 2020.
In addition, drinking water is a nominal variable that takes a value from 1 to 4. The average temperature of provinces in 2010 was between 20.7 and 29.6 degrees Celsius, whereas a minimum of 20.8 degrees and a maximum of 28.9 degrees were recorded in 2020. In this analysis, we substitute the average temperature with five dummy variables representing five different temperature ranges from 20–30 Celsius to avoid the potential correlation between average temperature and radiation. In addition, the average monthly rainfall was 177.324 mm in 2010 and 162.224 mm in 2020. The maximum average monthly sunshine reached 1,864 hours in 2010 and only 399.3 hours in 2020.
Fig 1 presents the percentage of drinking water types in 2020 and 2010. In 2010, the most used type of water was well water, accounting for 45 per cent of the total water sources. However, in 2020, tap water became the top drinking water, standing at 50 per cent compared to 35 per cent of well water. Water from rain and other remaining sources also exhibited a decline, at 14 per cent in 2010 and 8 per cent in 2020. The same goes for spring water, at 15 per cent in 2010 and 7 per cent in 2020.
3.2. The effects of the characteristics of the households and heads, environmental factors and living conditions on domestic migration in Vietnam
We employ logit regressions to examine the effect of the characteristics of Vietnamese households and their heads, environmental factors, and living conditions on domestic migration in Vietnam using the VHLSS surveys from 2010 and 2020. Three regressions are used for each VHLSS survey. The first regression includes all households in the entire sample. The second and third models include households in rural and urban regions, respectively. Table 3 provides the empirical results of these regressions. The estimated coefficients show the odd ratios of having at least one migrant member in the family.
Our empirical results indicate that the characteristics of the households and the household heads are related to the likelihood that the household has a migrant. While the effect of household income on domestic migration is not very clear, the effect of household size on domestic migration is very significant across the entire sample and the sub-samples of households residing in urban and rural regions, as well as in both surveys in 2010 and 2020. The age of the household head has a non-linear relationship in all sub-samples in 2010 and 2020. The probability of having a migrant in the family will increase as the age of the household’s head rises. However, this likelihood diminishes after the "peak" age, estimated to be approximately 44 years in Vietnam. Higher educational attainment of the household’s heads supports a migration decision for them and other household members. For example, with an upper-secondary level of education, the probability of migration increases by 2.987 and 2.101 times compared to those without any educational degree in 2010 and 2020.
Furthermore, a migration decision is also affected by environmental factors. Our empirical results indicate that weather factors such as temperature, rainfall, and average sunshine hours impacted migration decisions in 2010 and 2020. Specifically, households living in rural and urban areas with radiation tend to migrate in 2020. However, the effect is not apparent in the 2010 VHLSS survey. In both surveys, the average rainfall is associated with an increased probability of domestic migration in both rural and urban regions in Vietnam. Interestingly, our findings confirm that households living in the highest temperature area of 28–30 degrees Celsius experienced a higher migration rate in 2010 than the average temperature of 24–26 degrees Celsius. This relationship is especially strong in rural areas.
Regarding the effect of living conditions on a migration decision, our results indicate that the use of well water and rainwater is not significantly correlated with a migration decision in 2010. However, using spring water will increase the likelihood of having a migrant member in the entire sample. This relationship is positive and significant in both 2010 and 2020. Notably, using rainwater and other water sources increases the probability of migration in the urban sample in 2020. In addition, the amount of money paid for electricity is also associated with an increased probability of migration in both the 2010 and 2020 surveys.
We now turn our attention to geographical factors. Our results confirm that households in the Southeast and the Mekong Delta rivers have a lower probability of migration in 2010 than those in the Red River Delta. In contrast, households living in the North Midlands and the mountains exhibited an increased probability of migration in 2010. The same conclusion applies to the central highlands. We also find that living in cities on the central coast increases the probability of migration in both 2010 and 2020.
3.3. Predicted probability of migration
Table 4 presents the predicted probability regarding the model’s average probability of a migration decision for the entire sample. Specifically, the average probability that a household member migrates is about 10 per cent in 2010 and 6.8 per cent in 2020. As presented in Figs 2 and 3, when examining households with migrant members predicted by the model, we find that, in general, households living in the urban regions had higher residences in 2010 and lower in 2020 compared to those living in the rural regions. Figs 4 and 5 show the relationship between the persons identifying as migrants, the household head’s age, and the total number of members in the household. There is little difference between these two surveys in 2010 and 2020. Migration is more significant for young household heads and large families and vice versa.
Predicted probability of migration by income (2020).
4. Conclusions and policy implications
Vietnam has achieved great success in its economic reform, poverty reduction, and social transformation since its Economic Reform in 1986. Since then, with ups and downs, the economy has generally done well in achieving its ambition to be a middle-income country by 2030. However, regardless of its constant effort, various socio-economic issues have emerged as fundamental challenges for the economy, particularly income inequality and environmental degradation. Urbanization and industrialization of the national economy to support its economic growth and development have resulted in a massive scale of domestic migration from some regions of the country, such as the central coast provinces, to industrial provinces mainly located in the south. This massive scale of domestic migration has put significant pressure on the destinations regarding infrastructure for public transport, housing, educational opportunities, environmental degradation, and others. Even more fundamentally, the origin regions where households and their members have decided to leave behind may not be given the opportunity to grow and transform their local economies to be better ones for future generations. This vicious circle may never be ended if no public policies are considered and implemented.
On these observations and a lack of previous academic studies with a focus on Vietnam, this study examines fundamental determinants in driving domestic migration in Vietnam with an emphasis on (i) the characteristics of the households and their heads, (ii) the environmental factors such as the average rainfall, and (ii) living conditions of the households who decide to migrate to other destinations for economic opportunities and various other reasons. This study has utilized two surveys from the VHLSS in 2010 and 2020 to understand the changes within the decade. Key results are summarized below.
First, our empirical results confirm that the households’ characteristics and heads matter in a migration decision. We find that the size of a household is a more dominant factor leading to a migration decision by the households and their family members than the family income. The educational level of the household heads appears to be a leading factor for a migration decision, implying that these individuals are confident with the job opportunities available at the destinations for their migration.
Second, both the average rainfall and radiation have appeared to affect a migration decision for Vietnamese households and their members. This effect is particularly strong in 2020. These findings imply that households can see barriers to their achievements in life as the weather is out of their control. As such, moving to another region with a less adverse weather condition can support their effort to work.
Third, our empirical findings also confirm that the living conditions of the households can also be considered a factor associated with a migration decision. Water sources and electricity supply play a role in the decision to migrate to other destinations for Vietnamese households and their members. We find that households and their members in the central coast provinces appear to migrate significantly in comparison with other regions.
Various policy implications have emerged based on these empirical findings. Going back to the basis is the starting point the government and its agencies will need to consider and implement as soon as practical. Clean water supply, stable and affordable electricity, and basic infrastructure supporting local jobs, including farming activities, are fundamental considerations for the government. These focuses are doable with the view that disadvantaged provinces regarding economic and investment opportunities are given opportunities first for their local economic development. Instead of investing resources into well-developed regions and then redistributing the benefits to the disadvantaged provinces in a way the Vietnamese government have been currently doing for the past 20 years, disadvantaged provinces and regions should be given significant investment from the public resources to ensure that they can catch up with other wealthy regions in economic opportunities for their local economies and their people. Special consideration regarding tax concession for investments and attracting scholars to these areas may be appropriate. Doing so will limit the massive scale of domestic migration from these provinces, leading to benefits for all provinces, including those wealthy ones in the country.
References
- 1. De Haas H. Migration theory: Quo vadis? 2014.
- 2. Eisenstadt SN. Analysis of patterns of immigration and absorption of immigrants. Population studies. 1953;7(2):167–80.
- 3. Lee ES. A theory of migration. Demography. 1966;3:47–57.
- 4. Nations U. Department of Economic and Social Affairs. Population Division. 2015.
- 5. Zelinsky W. The hypothesis of the mobility transition. Geographical review. 1971:219–49.
- 6. Mishra S. Factors and Process of Migration in Developing Economy. Frontiers in Migration Analysis. 1981;227.
- 7.
Bilsborrow RE. Migration, urbanization, and development: new directions and issues: Springer Science & Business Media; 1998.
- 8. Thu Phuong N, Tran T, Thi Nguyet N, Oostendorp R. Determinants and Impacts of Migration in Vietnam. Development and Policies Research Center (DEPOCEN), Vietnam; 2008.
- 9. Sriskandarajah D. Migration and Development. World Economics. 2005; 6(2):141–6.
- 10. Stark O, & Taylor J. E. Migration Incentives, Migration Types: The Role of Relative Deprivation. The Economic Journal. 1991;101(408):1163–78.
- 11.
Coxhead I, Nguyen V. C., & Vu H. L. Internal Migration in Vietnam, 2002–2012. In Liu A. Y. C & Meng X. (Eds.), Rural-Urban Migration in Vietnam Springer International Publishing. 2019:(pp. 67–96).
- 12. Todaro MP. A Model of Labor Migration and Urban Unemployment in Less Developed Countries. The American Economic Review. 1969;, 59(1), 138–148. JSTOR.
- 13. Mincer J. Family migration decisions. Journal of political Economy. 1978;86(5):749–73.
- 14. Stark O, Bloom DE. The new economics of labor migration. The american Economic review. 1985;75(2):173–8.
- 15. Ravenstein EG, editor The laws of migration1885: Royal Statistical Society.
- 16. Mora J, Taylor JE. Determinants of migration, destination, and sector choice: Disentangling individual, household, and community effects. International migration, remittances, and the brain drain. 2006;21(4):21–52.
- 17.
Becker GS. Investment in human capital: effects on earnings. Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, Second Edition: NBER; 1975. p. 13–44.
- 18. Sjaastad LA. The costs and returns of human migration. Journal of political Economy. 1962;70(5, Part 2):80–93.
- 19. Taylor JE. Undocumented Mexico—US migration and the returns to households in rural Mexico. American journal of agricultural economics. 1987;69(3):626–38.
- 20. Gray C, Bilsborrow R. Environmental influences on human migration in rural Ecuador. Demography. 2013;50(4):1217–41. pmid:23319207
- 21. Cattaneo C, Beine M, Fröhlich CJ, Kniveton D, Martinez-Zarzoso I, Mastrorillo M, et al. Human migration in the era of climate change. Review of Environmental Economics and Policy. 2019.
- 22.
McCarthy JJ. Climate change 2001: impacts, adaptation, and vulnerability: contribution of Working Group II to the third assessment report of the Intergovernmental Panel on Climate Change: Cambridge University Press; 2001.
- 23. Paul BK, Rahman MK, Lu M, Crawford TW. Household migration and intentions for future migration in the climate change vulnerable lower Meghna estuary of coastal Bangladesh. Sustainability. 2022;14(8):4686.
- 24. Hunter LM, Luna JK, Norton RM. Environmental dimensions of migration. Annual Review of Sociology. 2015;41:377–97.
- 25. Rakib M, Sasaki J, Matsuda H, Fukunaga M. Severe salinity contamination in drinking water and associated human health hazards increase migration risk in the southwestern coastal part of Bangladesh. Journal of environmental management. 2019;240:238–48. pmid:30952044
- 26. Stoler J, Brewis A, Kangmennang J, Keough SB, Pearson AL, Rosinger AY, et al. Connecting the dots between climate change, household water insecurity, and migration. Current opinion in environmental sustainability. 2021;51:36–41.
- 27. VanWey LK. Land Ownership as a Determinant of International and Internal Migration in Mexico and Internal Migration in Thailand 1. International Migration Review. 2005;39(1):141–72.
- 28. Bhandari P, Ghimire D. Rural agricultural change and individual out‐migration. Rural Sociology. 2016;81(4):572–600. pmid:28082752
- 29. Berkson J. Application of the logistic function to bio-assay. Journal of the American statistical association. 1944;39(227):357–65.