Socioeconomic Datasets Research Articles

Gridded global dataset of industrial water use predicted using the Random Forest

Spatially distributed industrial water use (IWU) data are essential for effective region-specific water resource management. Such data are often scarce in underdeveloped and developing countries. We propose a random forest regression model to predict IWU at a spatial resolution of 0.5° by combining socioeconomic, climatic, and geographical datasets. These datasets included nighttime light (NL), global power plants, country-wise IWU, elevation data (DEM), gross domestic product (GDP), road density (RD), cropland (CRP), wetland (WLND), population (POP), precipitation (PCP), temperature (TEMP), wet days (WET) per year, and potential evapotranspiration (PET). The results show that RD, CRP, POP, GDP, DEM, and TEMP were the most influential variables. We assessed the accuracy of the global IWU map using published and observed datasets from various sources for the major industrialized countries such as the USA and China from 2000 to 2015. The predicted global map shows a reasonable distribution of grid-wise values for highly industrialized countries and data-scarce regions. Thus, fine-resolution maps can support local planning and decision-making for large basins worldwide.

Individual level risk factors associated with episodes of heat as recorded within primary care records: a time-stratified case-cross over study

Abstract Background A key recommended action outlined in Heat-Health Action Plans is to identify individuals at risk and deploy targeted interventions, however this is not routinely undertaken. A major contributing factor is a lack of understanding of individual-level risk factors to support an evidence-based approach to targeted prevention. Aim To use electronic health data to identify individual level characteristics associated with increased risk of mortality during heatwaves in England. Methods A time-stratified case-crossover analysis was undertaken of individual-level clinical and socio-environmental (SE) risk factors. Conditional logistic regression was used to characterise associations between temperature and the risk of death on hot days, with analysis stratified by potential risk modifying factors. Results Heat mortality risk was modified by a large range of chronic conditions, with cardio-respiratory, mental health and cognitive function conditions, diabetes and Parkinson’s, all increasing risk. Prescribed medications also increased odds of death during heatwaves. Heat mortality risk was modified by a range of SE determinants of health investigated, with ORs increasing with age; OR differences by sex, ethnicity and living arrangement; an increasing trend in ORs by alcohol intake and body mass index, excluding the obese-3 group. Results suggest the least deprived have the lowest OR and the most deprived have the highest. Conclusions This study has important implications for patient medication management during heat events, incorporating heat-risk considerations into other health policies. The results also provide important evidence on the role of disadvantage in driving the inequitable distribution of climate impacts, and the need for better socio-economic data sets linked to health records. The findings highlight the importance of incorporating an assessment of individual SE circumstances when prioritising patients at highest risk during heat events.

GRACE-derived groundwater variability and its resilience in north India: impact of climatic and socioeconomic factors

ABSTRACT Groundwater availability has become a global challenge due to climatic droughts and human interventions. North India has emerged as one of the most groundwater-depleted regions. Our study delves into the linkage between groundwater variability, climatic and anthropogenic factors for the last two decades using the Gravity Recovery and Climate Experiment (GRACE) and GRACE-Follow On (GRACE-FO) observations, groundwater levels, climatic and socioeconomic datasets. Our novel spatiotemporal analysis of groundwater storage (GWS) anomaly reveals five episodes of notable changes in north India affected by rainfall variability and groundwater abstraction. The standardized precipitation index (SPI) reveals that the prolonged drought and groundwater pumping during 2015 to 2017 significantly affected GWS. Our results suggest that sufficient rainfall contributed to the replenishment of GWS in some zones during 2018 to 2023, likely due to government recharge programmes. Our findings elucidate the current scenario and underscore a pressing necessity to accelerate groundwater management strategies, ensuring its resilience in future periods of intense drought.

Suitability mapping for alternative crops: A consistent, high-resolution approach for the United States

Alternative crops are a key aspect of many nature-based solutions in agriculture, and there is a need for more consistent geographic information on biophysical suitability to aid in both farm-level planning and larger-scale analyses. Here we describe an approach for generating consistent, replicable, high-resolution suitability maps for any alternative crop species across the USA. The method employs a criteria-based approach to map potential species performance using a simple suitability index. Criteria for suitable and ideal ranges of values for influential biophysical variables are created using data collected from published sources and are reviewed by experts. Publicly available soil and climate map data are used with the criteria to map suitability for each variable which are integrated into an overall suitability index map. Maps of unsuitable locations are combined to produce a limiting variable map showing which category of environmental variables is most limiting. We demonstrate the application of this approach for two alternative perennial crops, apple (Malus spp.) and Chinese chestnut (Castanea mollissima) and validate its accuracy using known farm locations. Maps of apple and chestnut in four distant study regions show varying patterns of suitability and potential performance depending on the climate and geophysical characteristics of the region. The maps can be used to identify areas of high suitability and compare across regions or between species and be combined with socio-economic and environmental datasets for further analysis. Compared to other approaches, ours can be applied to multiple species with a range of pre-existing knowledge in a consistent way, allowing for reliable mapping and subsequent research and planning for alternative crops and nature-based solutions in agriculture.

Prior emergency medical services utilization is a risk factor for in-hospital death among patients with substance misuse: a retrospective cohort study

BackgroundSubstance misuse poses a significant public health challenge, characterized by premature morbidity and mortality, and heightened healthcare utilization. While studies have demonstrated that previous hospitalizations and emergency department visits are associated with increased mortality in patients with substance misuse, it is unknown whether prior utilization of emergency medical service (EMS) is similarly associated with poor outcomes among this population. The objective of this study is to determine the association between EMS utilization in the 30 days before a hospitalization or emergency department visit and in-hospital outcomes among patients with substance misuse.MethodsWe conducted a retrospective analysis of adult emergency department visits and hospitalizations (referred to as a hospital encounter) between 2017 and 2021 within the Substance Misuse Data Commons, which maintains electronic health records from substance misuse patients seen at two University of Wisconsin hospitals, linked with state agency, claims, and socioeconomic datasets. Using regression models, we examined the association between EMS use and the outcomes of in-hospital death, hospital length of stay, intensive care unit (ICU) admission, and critical illness events, defined by invasive mechanical ventilation or vasoactive drug administration. Models were adjusted for age, comorbidities, initial severity of illness, substance misuse type, and socioeconomic status.ResultsAmong 19,402 encounters, individuals with substance misuse who had at least one EMS incident within 30 days of a hospital encounter experienced a higher likelihood of in-hospital mortality (OR 1.52, 95% CI [1.05 – 2.14]) compared to those without prior EMS use, after adjusting for confounders. Using EMS in the 30 days prior to an encounter was associated with a small increase in hospital length of stay but was not associated with ICU admission or critical illness events.ConclusionsIndividuals with substance misuse who have used EMS in the month preceding a hospital encounter are at an increased risk of in-hospital mortality. Enhanced monitoring of EMS users in this population could improve overall patient outcomes.

Spatiotemporal characteristics and driving factors of chemical oxygen demand emissions in China’s wastewater: An analysis based on spatial autocorrelation and geodetector

The advancement of industrialization and urbanization has exacerbated global water pollution. Chemical Oxygen Demand (COD) serves as a crucial metric for assessing organic contamination in aquatic environments. Consequently, the spatiotemporal characteristics and driving factors of COD emissions have emerged as a significant area of current research interest. This study employs spatial autocorrelation and geodetector methods to rigorously analyze the spatial–temporal patterns and principal drivers of COD emissions in China’s wastewater, drawing on eleven geographic and socio-economic datasets spanning from 2011 to 2019. This study indicates that COD emissions in China’s wastewater have consistently decreased annually between 2011 and 2019, with a marked reduction observed post-2016. Regarding spatial distribution, COD emissions in the eastern region significantly surpass those in the western region, highlighting pronounced regional disparities. The analysis of the global Moran’s Index indicates that the COD emissions across China’s provincial regions demonstrate spatial clustering of similar values, with this clustering intensifying annually. Furthermore, the local Moran’s Index identifies the precise locations of these spatial clusters. Geodetector analysis identified that the consumption of nitrogen fertilizers and the consumption of chemical fertilizers had substantial impacts on COD emissions, whereas the proportion of urban population at year-end and the number of granted inventions exerted lesser effects. Furthermore, the interaction between two factors notably enhances the explanatory power regarding COD emissions. This research underpins the formulation of efficacious pollution control and environmental protection strategies, contributing vital insights for the stewardship of water quality.

Exploring Drivers of Wildfires in Spain

Wildfires play a dual role in ecosystems by providing ecological benefits while posing catastrophic events; they also inflict non-catastrophic damage and yield long-term effects on biodiversity, soil quality, and air quality, among other factors, including public health. This study analysed the key determinants of wildland fires in Spain using openly available spatial data from 2008 to 2021, including fire perimeters, bioclimatic variables, topography, and socioeconomic datasets, at a resolution of 1 km2. Our methodology combined principal component analysis (PCA), linear regression analysis, and one-way analysis of variance (ANOVA). Our findings show that scrub/herbaceous vegetation (average 63 ± 1.45% SE) and forests (average 19 ± 0.76% SE) have been highly susceptible to wildfires. The population density exhibited a robust positive correlation with wildfire frequency (R2 = 0.88, p < 0.0001). Although the study provides insights into some fire-related climatic drivers over Spain, it includes only temperature and precipitation-based variables and does not explicitly consider fuel dynamics. Therefore, a more advanced methodology should be applied in the future to understand the local specifics of regional wildfire dynamics. Our study identified that scrub/herbaceous areas and forests near densely populated regions should be prioritised for wildfire management in Spain, particularly under changing climate conditions.

Transit-oriented development and effective supply on the tenant-occupied multifamily housing market in Washington, DC

ABSTRACT A range of socioeconomic factors, including income distribution, productivity and land use policies, are thought to influence the development of contemporary multifamily housing markets in the United States. Focusing on Washington, DC, as a case study, we use data from the American Community Survey to assemble socioeconomic panel datasets for the years 2010–2019. Through spatial regression analysis, we explore which socioeconomic measures and land use policies are associated with the effective supply of tenant-occupied multifamily housing. We interpret our findings considering issues such as displacement, lessor habit formation, demographic shifts and the uneven growth of transit-oriented housing in DC's current housing landscape.

Enhancing vulnerability assessment through spatially explicit modeling of mountain social-ecological systems exposed to multiple environmental hazards

The evaluation of the vulnerability of coupled socio-ecological systems is critical for addressing and preventing the adverse impacts of various environmental hazards and devising strategies for climate change adaptation. The initial step in vulnerability assessment involves exposure assessment, which entails quantifying and mapping the risks posed by multiple environmental hazards, thereby offering valuable insights for the implementation of vulnerability assessment methodologies. Consequently, this study sought to model the exposure of coupled social-ecological systems in mountainous regions to various environmental hazards. By a set of socio-economic, climatic, geospatial, hydrological, and demographic data, as well as satellite imagery, and examining 11 hazards, including droughts, pests, dust storms, winds, extreme temperatures, evapotranspiration, landslides, floods, wildfires, and social vulnerability, this research employed machine learning (ML) techniques and the fuzzy analytical hierarchy process (FAHP). Expert opinions were utilized to guide hazard weighting and calculate the exposure index (EI). Through the precise spatial mapping of EI variations across the socio-ecological systems in mountainous areas, this investigation provides insights into vulnerability to multiple environmental hazards, thereby laying the groundwork for future endeavors in supporting national-level vulnerability assessments aimed at fostering sustainable environments. The findings reveal that social vulnerability and pests receive the highest weighting, while floods and landslides are ranked lower. All hazards demonstrate significant correlations with the EI, with droughts exhibiting the strongest correlation (r > 0.81). Spatial analysis indicates a north-south gradient in forest exposure, with southern regions showing higher exposure hotspots (EI 29.08) compared to northern areas (EI 10.60). Validation based on Area Under Curve (AUC) and Consistency Rate (CR) in FAHP demonstrates robustness, with AUC values exceeding 0.78 and CR values below 0.1. Considering the anticipated intensification of hazards, management strategies should prioritize reducing social vulnerability, restore degraded areas using drought-resistant species, combat pests, and mitigate desertification. By integrating multidisciplinary data and expert opinions, this research contributes to informed decision-making regarding sustainable forest management and climate resilience in mountain ecosystems.

Spatio-temporal variation and dynamic risk assessment of drought and flood disaster (DFD) in China

It is important to understand the variations of Drought-Flood-Disaster (DFD) from the past to future for disaster management decision-making under the backdrop of global warming. This study explored the spatio-temporal variations of DFD risks in China from the past (2000–2020) to future (2020–2100) based on the natural and socio-economic datasets. The main findings include: (1) The spatial pattern of disaster vulnerability decreased from 2000 to 2100. The overall vulnerability distribution showed a spatial pattern of high east and low west, and its eastern distribution presented high north and low south. (2) The drought hazard danger degree showed a trend of decreasing first and then increasing with obvious spatial heterogeneity. The flood hazard danger degree had a decreasing spatial distribution from southeast to northwest, with an overall gradually increasing trend. (3) The DFD risks showed a spatial pattern of high east and low west, and were more closely related to disaster vulnerability. Overall, the flood and drought risks had good consistency at the highest level (VI), mainly distributed in the North China Plain, Northeast Plain, eastern Sichuan, and southern Guangdong. This level means that the most severe damage to life and property will occur in these regions. (4) The DFD showed the direction of “north(east)-south(west)”, and the main axis of the ellipse moved north eventually. The DFD hot spots were mainly distributed in North China and the middle and lower reaches of the Yangtze River. The DFD cold spots were mainly distributed in the southwest, Qinghai-Tibet region, and Gansu province.

Flood susceptibility mapping using fuzzy analytical hierarchy process for Cedar Rapids, Iowa

ABSTRACT Floods affect over 2.2 billion people worldwide, and their frequency is increasing at an alarming rate compared to other disasters. Presidential disaster declarations have issued increasingly almost every year in Iowa for the past 30 years, indicating that the state is on the rise of flood risk. A multi-disciplinary approach is required, in which the underlying hydrologic processes that cause floods are closely linked with watershed-level socio-economic functions using effective collaboration tools to ensure community participation in the co-production of mitigation plans while paying attention to socio-environmental justice principles. Considering the existing limitations and needs, we conducted a flood risk assessment by utilizing geophysical and socio-economic datasets for a case study in Cedar Rapids, Iowa. Flood risk outputs are generated based on three main risk groups: geophysical-based flood risk, socioeconomic risk, and combined flood risk. An extensive literature review is conducted to determine the pairwise comparison matrices for the parameters that are used in the analytical hierarchy process (AHP) and fuzzy AHP methods. Our results indicate that high- and very-high-risk flood susceptibility zones are primarily located in central urban areas with lower elevations, regardless of the method type (AHP or FAHP). According to overall results, a large area of Cedar Rapids consists of a medium risk level according to the flood risk map combined with the fuzzy AHP method. The results show that high and very high-risk areas are 16% of the studied region, medium, low and very low-risk areas correspond to 84%. Besides, nearly 40% of the population lives in high to very high flood risk zones.

The Spatial Patterns of the Crime Rate in London and Its Socio-Economic Influence Factors

This paper analyses the spatial trends and patterns of the crime rates in London and explores how socio-economic characteristics affect crime rates with consideration of the geographic context across London. The 2015 London Crime Statistics and Socio-economic Characteristics datasets were used. First, we investigated the spatial patterns of crime rates through exploratory spatial analysis at the ward level. In addition, both the ordinary least square (OLS) model and geographically weighted regression (GWR) model, which allow the effects of factors to vary in spatial scales, were adopted and compared to explore the potential spatially varying effect across London. The results showed that there exists obvious spatial clustering for the crime rate in central London. Both global and local Moran’s I values indicated the spatial dependence of crime at the ward level. The GWR model performed better in explaining crime rates than the OLS model. Only two factors, namely, the percentage of children aged from 0 to 15 and employment rates, had significant spatial variability in London. The influences of the percentage of children aged 0 to 15 on crime rates are constantly negative over a spatial scale; however, employment rates positively affect crime rates in the north-western areas near the centre of London. Therefore, this paper focuses more on the spatial perspective, which fills the gap in traditional crime analysis, especially on the spatially varying influence of socio-economic status.

Generating synthetic personal health data using conditional generative adversarial networks combining with differential privacy

A large amount of personal health data that is highly valuable to the scientific community is still not accessible or requires a lengthy request process due to privacy concerns and legal restrictions. As a solution, synthetic data has been studied and proposed to be a promising alternative to this issue. However, generating realistic and privacy-preserving synthetic personal health data retains challenges such as simulating the characteristics of the patients’ data that are in the minority classes, capturing the relations among variables in imbalanced data and transferring them to the synthetic data, and preserving individual patients’ privacy. In this paper, we propose a differentially private conditional Generative Adversarial Network model (DP-CGANS) consisting of data transformation, sampling, conditioning, and network training to generate realistic and privacy-preserving personal data. Our model distinguishes categorical and continuous variables and transforms them into latent space separately for better training performance. We tackle the unique challenges of generating synthetic patient data due to the special data characteristics of personal health data. For example, patients with a certain disease are typically the minority in the dataset and the relations among variables are crucial to be observed. Our model is structured with a conditional vector as an additional input to present the minority class in the imbalanced data and maximally capture the dependency between variables. Moreover, we inject statistical noise into the gradients in the networking training process of DP-CGANS to provide a differential privacy guarantee. We extensively evaluate our model with state-of-the-art generative models on personal socio-economic datasets and real-world personal health datasets in terms of statistical similarity, machine learning performance, and privacy measurement. We demonstrate that our model outperforms other comparable models, especially in capturing the dependence between variables. Finally, we present the balance between data utility and privacy in synthetic data generation considering the different data structures and characteristics of real-world personal health data such as imbalanced classes, abnormal distributions, and data sparsity.

A hybrid machine learning-mathematical programming optimization approach for municipal solid waste management during the pandemic.

The online version contains supplementary material available at 10.1007/s10668-023-03354-2.

Analysis on absolute conflict and relative conflict of land use in Xining metropolitan area under different scenarios in 2030 by PLUS and PFCI

As the highest level of urbanization in the Qinghai-Tibet Plateau, the Xining metropolitan area has unique development conditions and its own characteristics. Based on land use dataset, socio-economic dataset, natural conditions dataset, and restricted development area dataset, we forecast the 6 land use types in 7 districts and 6 counties of Xining metropolitan area under different scenarios in 2030 using the Patch-generating Land Use Simulation (PLUS) model. Then we summarize the characteristics of the land use in 2010, 2020 and different scenarios in 2030. Subsequently, the absolute land use conflict and relative land use conflict of Xining metropolitan area in 2010, 2020 and different scenarios in 2030 are measured, via Spatial Comprehensive Conflict Index (SCCI) and the Pythagorean Fuzzy Conflict Information (PFCI) system. Results show that: Expansion of built-up land in the core and the peripheral in 2030 is comparable, but cultivated land in the peripheral shows a more obvious decrease trend. There is a high proportion of cultivated land and built-up land with strong absolute conflicts. The absolute conflict in the peripheral area is more serious. In addition, the peripheral area is the main source of relative conflict, and has advantages of development. Finally, we discuss the logics of land use conflict and the relationship between absolute land use conflict and relative land use conflict. Ultimately, development strategies for Xining metropolitan area are given. This paper aims to provide reference suggestions to alleviate the conflict of land use and to promote the sustainable development in Xining metropolitan area.

A Nuanced Analysis on Livelihood Resilience of Vietnamese Upland Households: An Intersectional Lens of Ethnicity and Gender

“How to gauge the resilience of a household’s livelihood?” and “who is resilience for?”; as the world’s volatility increases, especially with unprecedented changes in climate, interest in these questions continues to rise. While many previous conventional attempts to measure resilience at the household level have largely employed the “objective” top-down framework relying mostly on a macro observed socioeconomic dataset, this present work seeks to estimate household resilience through an alternative bottom-up method, called the “subjective” resilience approach. With specific reference to the context of two ethnic Pa Cô and Tà Ôi minorities living in upland areas of Central Vietnam, this study aims to (1) measure household resilience to climate change by scoring five livelihood capitals, financial, human, social, physical, and natural, by applying the Household Livelihood Resilience (HLR) framework; and (2) provide an intersectional lens of ethnicity and gender in relation to the household’s livelihood resilience. To achieve these goals, in addition to two focus group discussions and eleven in-depth interviews, an empirical survey of 236 households was conducted between September to December 2021. Our findings indicated that there are differences in observed livelihood resilience among the ethnic minority groups and, within that, between gender (sex, status) factors. Ethnic women and poor households have a lower resilience than other groups in society due to their lower human capital and limited access to public and financial resources, which are to some degree linked to the difference in the gender division of labor and the cultural norms of patriarch traditions. This study highlights the importance of considering gender and poverty in resilience-building efforts and offers insights for future programs in multi-ethnic developing countries such as Vietnam.

Geospatial analysis of patients' social determinants of health for health systems science and disparity research.

Geospatial analysis of patients' social determinants of health for health systems science and disparity research.

Convolutional neural networks in estimating the spatial distribution of electric vehicles to support electricity grid planning

From the perspective of electricity distribution networks and the energy system, the increasing numbers of electric vehicles are among the most topical and challenging problems. The paper investigates a novel approach of a convolutional neural network-based modeling method for estimating the spatial distribution of electric vehicles. The proposed model extracts features from multilayer socioeconomic input raster data that are sequenced in strides and outputs a spatial estimation of EV distribution. Spatial forecasting or area forecasting is at the core of the distribution system operators’ planning and development process as it provides a solid foundation for stochastic load modeling and load development analysis. Present models mostly focus on stochastic load modeling, lacking the spatial forecasting aspect of EV distribution. The proposed model aims to enhance EV load modeling by providing a more accurate spatial approach to the models. The study uses large actual socioeconomic and vehicle registration data sets to tackle the modeling challenge. In comparison with previous studies on similar topics, the present study benefits from more samples resulting from an increase in the adoption of electric vehicles. The proposed model architecture performs adequately in predicting a spatial electric vehicle distribution; the CNN model reached a weighted average precision score of 0.91. The proposed methodology greatly enhances stochastic EV load modeling by providing a good spatial forecast of the initial EV locations, and the results can be further aggregated to support the electricity distribution system planning process. An energy-, material-, and cost-efficient electricity distribution system is the backbone of the modern energy system.

Evaluating the climatic and socio-economic influences on the agricultural drought vulnerability in Jharkhand.

Environmental hazards like drought lead to degrading food production and adversely impact the agro-economy. This study investigates the contributions of different climatic and socio-economic variables to agricultural drought in Jharkhand. The three primary criteria, i.e., exposure (E), sensitivity (S), and adaptive capacity (AC), responsible for agricultural drought vulnerability, were examined to identify the drought-prone areas. Long-term (1958-2020) gridded climatic datasets obtained from the Terra-climate global dataset, MODIS vegetation index dataset (MOD13Q1) for the years 2001-2020, different soil parameters obtained from the ISRIC global soil database and state agricultural portal of Jharkhand, and different socio-economic datasets obtained from census data (2011) provided by Govt. of India, were utilized for this study. Analytic Hierarchy Process (AHP) was used to estimate the weighted contribution of the indicator variables falling under each criterion (E, S, and AC), and three criteria index maps were generated. These separate maps were further integrated to generate the final vulnerability index map. Finally, the study area was categorized into different zones based on the drought vulnerability index value ranging from 0 to 1, according to the severity of the drought. It was observed that about 4.05%, 28.12%, and 37.07% of the total geographical area is very highly, highly, and moderately vulnerable to agricultural drought, respectively. Amongst the three primary criteria, exposure showed a significant positive correlation (R = 0.61), and sensitivity showed a strong positive correlation (R = 0.55) with vulnerability. The adaptive capacity was negatively correlated (R = -0.75) with the vulnerability. However, putting equal weights to the variables to calculate the vulnerability, the exposure and sensitivity indicators showed a significant positive correlation with the vulnerability, with an R-value of 0.82 and 0.79, respectively. In contrast, the adaptive capacity showed a negative correlation with the vulnerability with R = -0.75.

Three Common Machine Learning Algorithms Neither Enhance Prediction Accuracy Nor Reduce Spatial Autocorrelation in Residuals: An Analysis of Twenty‐five Socioeconomic Data Sets

Machine learning (ML) is being applied in an increasing volume of geographical research. However, the aspects of spatial autocorrelation (SAC) in the residuals produced by ML models have been understudied compared to the benefit of ML, namely, reduction of prediction errors. In this study, we examined the relationship between predictive accuracy and the reduction in the residual SAC for 597 variables from 25 geographical socio‐economic data sets using spatial and nonspatial cross‐validation of three ML algorithms such as random forests, support vector machine, and artificial neural network (ANN) to provide an extensive empirical diagnosis—but not a definitive theory—of the relationship between SAC and ML. Our results highlighted that the ML algorithms with tuned hyperparameters yielded marginal predictive accuracy gains and the minimal decreases in residual SAC. ANN revealed lower accuracy and higher reduction in the residual SAC than others. This implies ML algorithms in geographical research in socio‐economic domains would not always result in higher prediction accuracy. We suggest that ML in geographical research should be cautiously employed when the main objective is related to the residual SAC. We also showed that spatial cross‐validation neither improves predictive accuracy substantially nor reduce the residual SAC effectively.