Household-level Characteristics Research Articles

A Multilevel Analysis of Farmers’ Adoption of Water-saving Irrigation Technology: Evidence From the North China Plain

Agricultural water scarcity poses a significant threat to food security and sustainable development. Farmers’ adoption of agricultural water-saving irrigation technologies (WSIT) is essential for addressing this challenge. However, decisions to adopt such technologies are shaped by both individual and structural factors across multiple levels. This study investigated the multilevel determinants of WSIT adoption in the North China Plain (NCP) through a field survey conducted from late 2012 to early 2013. We collected retrospective data from 818 households spanning 2010 to 2012. Initially, we employed a binomial logistic regression model to identify significant household- and village-level factors influencing WSIT adoption. Recognizing the nested nature of adoption decisions within individual and village contexts, and the importance of multilevel interactions, we further utilized a multilevel mixed-effects logistic model. Our findings indicated that household-level factors, such as farm size, access to irrigation resources, education, input costs, and village leadership, were the primary drivers of WSIT adoption. Approximately 22.7% of the variation in adoption was attributable to differences between villages. At the village level, larger village size decreased the likelihood of adoption, while the presence of Water User Associations (WUAs) had a positive influence. Mechanism analysis suggested that household-level characteristics affected farmers’ adoption behavior through possible channels such as technical training, farmers’ perception of climate change, and the availability of alternative technologies. These findings highlight the importance of considering multiple contextual factors in studying farmers’ adoption behavior and provide important policy implications for decision-makers. JEL classification: D1; Q16; Q25

Factors Affecting Formal Micro-Credits in the Wetland Regions of Bangladesh: A Discriminant Analysis

Micro-credit has been playing a vital role in developing wetland people’s socio-economic status by facilitating their income-generating activities such as seasonal fish farming, raising ducks and cattle, and floating vegetable gardens. Wetland people’s livelihoods are largely dependent on borrowing micro-credits, but access to formal credits is somewhat intricate for them due to procedural complexity. Therefore, this study aims to explore the potential discriminating factors of formal credits in the wetland region of Bangladesh. This study used data from 1607 micro-credit receiver households that were collected by a research project sponsored by the GARE Program, Ministry of Education, Government of Bangladesh. The discriminant analysis was performed to explore the factors discriminating between formal and informal credit-receiving groups. This study considered three types of factors such as borrower’s individual-level, household-level, and micro-credit related characteristics as the potential factors distinguishing between formal and informal credits. The findings show that micro-credit-related factors such as the number of installments, duration of the loan, and rate of interest are more crucial for affecting formal credit in the wetland peoples of Bangladesh. Most vulnerable groups in the wetland region still struggle to obtain formal credit because of obstacles such as a lower number of installments, lower duration of loans, and high interest rates. The government policy should remove these obstacles to ensure the people's easy access to formal loans, which would ultimately improve their socio-economic conditions. International Journal of Statistical Sciences, Vol. 24(2), November, 2024, pp 85-96

Predictive modelling of linear growth faltering among pediatric patients with Diarrhea in Rural Western Kenya: an explainable machine learning approach

IntroductionStunting affects one-fifth of children globally with diarrhea accounting for an estimated 13.5% of stunting. Identifying risk factors for its precursor, linear growth faltering (LGF), is critical to designing interventions. Moreover, developing new predictive models for LGF using more recent data offers opportunity to enhance model accuracy, interpretability and capture new insights. We employed machine learning (ML) to derive and validate a predictive model for LGF among children enrolled with diarrhea in the Vaccine Impact on Diarrhea in Africa (VIDA) study and the Enterics for Global Heath (EFGH) ― Shigella study in rural western Kenya.MethodsWe used 7 diverse ML algorithms to retrospectively build prognostic models for the prediction of LGF (≥ 0.5 decrease in height/length for age z-score [HAZ]) among children 6–35 months. We used de-identified data from the VIDA study (n = 1,106) combined with synthetic data (n = 8,894) in model development, which entailed split-sampling and K-fold cross-validation with over-sampling technique, and data from EFGH-Shigella study (n = 655) for temporal validation. Potential predictors (n = 65) included demographic, household-level characteristics, illness history, anthropometric and clinical data were identified using boruta feature selection with an explanatory model analysis used to enhance interpretability.ResultsThe prevalence of LGF in the development and temporal validation cohorts was 187 (16.9%) and 147 (22.4%), respectively. Feature selection identified the following 6 variables used in model development, ranked by importance: age (16.6%), temperature (6.0%), respiratory rate (4.1%), SAM (3.4%), rotavirus vaccination (3.3%), and skin turgor (2.1%). While all models showed good prediction capability, the gradient boosting model achieved the best performance (area under the curve % [95% Confidence Interval]: 83.5 [81.6–85.4] and 65.6 [60.8–70.4]) on the development and temporal validation datasets, respectively.ConclusionOur findings accentuate the enduring relevance of established predictors of LGF whilst demonstrating the practical utility of ML algorithms for rapid identification of at-risk children.

Factors Associated With Transmission Across Three Waves of SARS-CoV-2 in a Prospective Community-Based Study of Households With School-Aged Children-Dane County, Wisconsin, 2020-2022.

Household transmission of SARS-CoV-2 is a driver of the ongoing COVID-19 pandemic. Understanding factors that contribute to secondary infection risks (SIRs) can define changing trends and inform public health policies. The ORegon CHild Absenteeism due to Respiratory Disease Study (ORCHARDS) prospectively monitors respiratory viruses within the Oregon School District (OSD) in southcentral Wisconsin. Households with students who had ≥ 2 respiratory symptoms were eligible and opted to participate in ORCHARDS. Between October 28, 2020, and May 16, 2022, all household members provided self-collected nasal specimens on days 0, 7, and 14 for SARS-CoV-2 detection using real-time reverse-transcription-polymerase chain reaction. We used logistic regression to investigate individual- and household-level characteristics associated with SARS-CoV-2 transmission. Overall, 127 households comprising 572 individuals (48% female; 52% male; 0.4% nonbinary; 77% ≥ 18 years) had at least one detection of SARS-CoV-2. The overall SIR was 47% and decreased over time (pre-Delta = 72% [95% CI: 58%-83%]; Delta = 51% [40%-63%]; and Omicron = 41% [36%-47%]). Odds of household transmission were 63% lower during the Omicron period compared with the pre-Delta period (OR = 0.36 [95% CI: 0.13-0.94] p = 0.037). Greater household density (members/bedroom) was significantly associated with household transmission during the Omicron period (OR = 6.8, [2.19-21.37] p = 0.001). Index case age, illness severity, and individual symptoms were not significantly associated with odds of household transmission. Greater household density was associated with a higher risk of SARS-CoV-2 transmission, but the risk declined over time with subsequent variants. Interplay between variants, prior infection, and individual/household factors may identify modifiable factors (e.g., behavior and vaccination) to reduce future transmission risk.

Gender and resources' role in parents' time with children in South Korea and the United States

AbstractObjectiveParental time with children varies across cultures but has mostly been studied in a European context. In this paper we estimate whether the association of work and gender with parental allocation of time is different in South Korea and the United States.BackgroundWe suggest that household level characteristics are associated with the time parents spend with their children, but that these associations are more gendered in South Korea than in the United States.MethodWe used data from the 2003–2018 American Time Use Survey (n = 38,562) and 2004–2014 Korean Time Use Survey (n = 12,828). Using ordinary least squares, we estimated the association of the intersection of gender and work in predicting parental time with children.ResultsGender is an important factor in time allocation in South Korea, where education was positively associated and working hours were negatively associated with mothers', but not fathers', time spent with their children. In the United States, resources and work have a dominant role, as both parents tend to spend more time with their children if they have higher education and income and work fewer hours.ConclusionIn South Korea, a gendered regime is associated with the allocation of parental time only; mothers are influenced by their spouses' time allocation and resources. The United States follows a trade‐off regime, where resources are associated with time spent with children, whereas gender plays a smaller role in parental time allocation.ImplicationsStudying parental time allocation should account for the intertwined relationship of culture, resources, and gender.

Agricultural Land Rental Arrangements in Nepal: Incidence, Forms, Determinants, and the Current Legal Provisions

Given the fragmented and small size of land holdings in Nepal, the viability of crop production is a pressing concern. Rental or tenancy arrangements, which equalise marginal returns from factors of production across households, offer a solution to this viability crisis. In this context, our study is unique as it not only examines the incidence and forms of tenancy, but also identifies the determinants of the land leasing decision of rural households, and discusses their implications for agrarian development in Nepal. Drawing on primary survey data from 350 households in six districts of plains and hills, we provide evidence of the widespread prevalence of informal tenancy in Nepal. Despite locational variations explained by the location-specific cropping patterns, sharecropping and fixed-rent are found to be the primary forms of tenancy contracts. A Tobit regression analysis identifies certain household level, farm level, and location-specific characteristics as important factors influencing land leasing decisions. Our analysis suggests that the land supply in the land lease market may increase in the future, and therefore, it is pertinent to examine the legal provisions that govern the use of such lands. Hence, we also examine the legal provisions concerning tenancy relations.

Child, family and household characteristics associated with physical activity in Samoan children aged 3-8 years: A cross-sectional study.

Physical activity is a key component of many obesity prevention strategies. The aim of this analysis was to identify child, family, and household characteristics associated with parent-reported physical activity in Samoan children aged 3-8 years. Children (n = 445; 51.2% female, mean age 5.4 years) were part of an ongoing, mixed-longitudinal study of child growth, development, and wellbeing (the Ola Tuputupua'e cohort). Bivariate analyses and multivariate generalized linear regressions were conducted to investigate the relationship of child, family, and household characteristics with physical activity level, measured using the Netherlands Physical Activity Questionnaire (NPAQ). Children were classified as being 'highly active' if they had NPAQ scores in the 75th percentile or above. Among the n = 111 children classified as 'highly active', n = 67 (60.4%) were boys. After adjusting for child, family, and household-level characteristics, hours of child sleep per night was the only variable significantly associated with odds of being highly active. Compared to children who slept less than 9 hours at night, those who slept 10-10.99 hours (OR: 5.97, 95% CI: 2.14-18.13) and 11+ hours (OR: 25.75, 95% CI: 8.14-90.12) had higher odds of being 'highly active'. Future research should examine the mechanisms driving the relationship between nighttime sleep and physical activity among Samoan children. Intervening on sleep duration and quality may improve physical activity and, in turn, obesity risk in this setting.

Water, sanitation, and hygiene insecurity and disease prevention behaviors during the COVID-19 pandemic in low-income neighborhoods of Beira, Mozambique.

Inadequate water, sanitation, and hygiene (WASH) are important drivers of the global burden of disease, and their impact is exacerbated during outbreaks. Directives to practice handwashing and physical distancing may be impractical for people that have limited access to WASH resources. In this study, which took place during the acute phase of the COVID-19 pandemic, we explore the relationship between control measures for global health crises and water, sanitation, and hygiene insecurity, with implications for other infectious diseases and future health emergencies. We investigated the relationship between WASH-related factors and disease prevention behaviors (handwashing, physical distancing, and masking), and the role of demographic characteristics and risk perceptions in influencing these relationships in low-income neighborhoods of Beira, Mozambique during the COVID-19 pandemic. We utilized data collected from 1,762 randomly selected households during a population-based survey. We fit multivariable logistic regression models to examine the associations between various WASH factors and disease prevention outcomes of interest, adjusting for individual- and household-level demographic characteristics and risk perceptions. Over 98% of people had access to improved drinking water and over 80% of people had access to improved sanitation facilities. There was a high level of reported adherence to handwashing (95.5%) and physical distancing (91.7%) practices during the COVID-19 pandemic. There was a lower odds of reported handwashing [aOR = 0.89; 95% CI: 0.81, 0.98] and physical distancing [0.85 (0.80, 0.92)] among respondents who had higher levels of water insecurity. Respondents that had a water source in their dwelling had a higher odds of reporting of physical distancing [2.03 (1.22, 3.41)] compared to people that had to leave their household to access water. There was a higher odds of reported handwashing and physical distancing among respondents who had their own sanitation facility, compared to a shared one [handwashing: 2.77 (1.35, 5.82); distancing: 1.61 (0.95, 2.73)], and those that had a sanitation facility inside their compound compared to outside their compound [handwashing: 2.11 (0.75, 5.71); distancing: 1.50 (0.65, 3.36)]. Respondents with a basic handwashing station, compared to no facility or a limited facility, had a higher odds of reported handwashing [4.45 (2.37, 8.65)], and those that had a connected handwashing station, compared to an unconnected handwashing station, had a higher odds of reporting handwashing and physical distancing [handwashing: 2.13 (0.68, 8.54); distancing: 1.77 (0.77, 4.53)]. Despite a high level of knowledge about the risks posed by COVID-19 and understanding of the benefits of handwashing and physical distancing, limitations in access to water, sanitation, and hygiene infrastructure acted as a barrier to people practicing disease prevention behaviors during the COVID-19 pandemic.

Natural history of shedding and household transmission of severe acute respiratory syndrome coronavirus 2 using intensive high-resolution sampling.

The COVID-19 pandemic has led to 775 million documented cases and over 7 million deaths worldwide as of March 2024 and is an ongoing health crisis. To limit viral spread within households and in the community, public health officials have recommended self-isolation, self-quarantine of exposed household contacts, and mask use. Yet, risk of household transmission (HHT) may be underestimated due to low frequency of sampling, and risk factors for HHT are not well understood. To estimate the secondary attack rate of SARS-CoV-2 within households and to define the risk factors for new infections in household members who are in close contact with the index case. In this prospective cohort study, from March 2020-December 2021 we enrolled 60 households with index cases who tested positive for SARS-CoV-2. All household contacts and index cases were tested daily for SARS-CoV-2 via reverse transcription polymerase chain reaction (RT-PCR) using self-collected anterior nares specimens. Households were followed until all study participants in the household tested negative for SARS-CoV-2 for seven consecutive days. We collected sex, age, race/ethnicity, comorbidities, and relationship to index case for secondary contacts, household level characteristics including primary income, household density, and square feet per person on property. We compared the sociodemographic variables between COVID-19 positive and negative household members and between households where secondary transmission did and did not occur. Daily anterior nares swabs were tested for SARS-CoV-2 using RT-PCR, in order to assess duration of nasal shedding of SARS-CoV-2, as well as risk of transmission to secondary household contacts. Of the 163 participants in this study, 84 (51.5%) were women; median age (IQR) was 36.0 (17.0-54.0) years of age; 78 (47.8%) were white and 48 (29.5%) were Hispanic/LatinX. Of the fifty households with household contacts, at least one secondary case occurred in twenty-six households (52.0%) and forty-five household contacts (43.7%) were infected. Secondary attack rate was lowest among children of index cases (6/23, 26.1%). Modified Poisson regression identified that the risk of transmission to household contacts increases significantly with age (Risk ratio for each increase in years of age = 1.01, 95% CI = 1.00-1.02). Mixed effects regression models identified that participants with chronic diseases, such as asthma, diabetes, cancer, or cardiac disease, had higher Cts at baseline when compared to participants without chronic diseases (6.62, 95% CI: 1.46-11.77, p = 0.02) and show a slower rate of increase in Ct over time (-0.43, 95% CI: -0.77 to -0.09, p = 0.02). This study suggests that HHT represents a key source of community-based infection of SARS-CoV-2. Allocation of resources for contact investigations and prevention interventions should focus on the individuals at highest risk of infection in households, especially those with higher density homes.

Analysis on adoption of sorghum innovation technologies in the tehuledere District, Eastern part of the Amhara National Regional State, Ethiopia

For Ethiopia's diverse agro-ecological zones, numerous open-pollinated and hybrid sorghum varieties have been released because of national and regional sorghum improvement research projects. Despite this fact, there is a persistent concern that technological advances for sorghum productivity enhancement have not been realized to the extent it could. In Ethiopia, sorghum production is constrained by biological, social, and abiotic constraints. The objective of the study was to analyze the adoption of sorghum technologies (Improved Seeds, Ferilizer, Irrigation) in the Eastern part of the Amhara region. The research used both descriptive research and explanatory research methodologies. Systematic random sampling method was used to select to select the 297 heads of household from the list of female-headed households and male-headed in each kebele. Data was collected through a structured questionnaire administered to sampled farmers. The survey covered a detailed community, household, and plot-level information. Households' demographic, institution and socioeconomic features were also collected. The data was collected using computer-assisted personal interview (Kobo Toolbox); then data were exported to and analyzed using SPSS and STATA. Descriptive test statistics and econometric methods were applied to analyze the data collected from smallholder household heads. Descriptive analyses were used to describe and analyze the household-level characteristics, including demographic, socioeconomic and institutional characteristics of the household. The factors explaining adoption of technologies (improved seeds, irrigation, fertilizer) and the intensity in sorghum production were analyzed using Logit, and Tobit models. The results indicated that demographic, socio-economic, and institutional factors have significant effect in the adoption of Sorghum Innovation Technology, and intensity in sorghum production among the farmer households in the study area. The analysis of Sorghum improvement project in the Eastern Amhara region, hence, underscores the need for holistic approaches that could consider the socio economic, environmental and technological dimensions of Agricultural development.

Gender identity, labor market outcomes, and socioeconomic status: Evidence from Chile

This paper provides the first evidence of differential labor market and socioeconomic outcomes for non-cisgender individuals in Latin America using nationally representative data. Very little is known about gender diverse populations in the developing world. Recent population-based data from the 2017 Chilean National Socioeconomic Characterization Survey allows for the identification of over 1800 non-cisgender adults (i.e., individuals whose reported gender identity differs from their sex assigned at birth.) I examine labor market outcomes and socioeconomic status in the context of gender identity, while controlling for relevant individual- and household-level characteristics. I find that non-cisgender individuals who were assigned male at birth are less likely to be employed, and more likely to be living in a household that is multidimensionally poor when compared to otherwise similar cisgender men. Non-cisgender individuals who were assigned female at birth are less likely to be employed, work fewer labor hours, report lower labor incomes, and are more likely to work in the informal sector relative to otherwise similar cisgender men.

Do car drivers respond differently to fuel price changes? Evidence from German household data

AbstractFor cutting down greenhouse gas emissions in road transport, core economic measures are relying on the response to increasing fuel prices—i.e., the fuel price elasticity. A focal point of the effectiveness of these measures is the heterogeneity in fuel price elasticities of different vehicle users. These effects were often neglected in the transport-related literature and were only incorporated recently. The results show, however, sometimes contradicting conclusions on influencing parameters such as income, region-type or household size. In this paper, we used a pooled OLS model estimated on a German refuelling diary data set and analysed the impact of various household level characteristics on fuel price elasticities through an analysis of interaction terms and their marginal effects. This analysis provides a cornerstone in this discussion on fuel price elasticities. We found out that the overall results contrast the existing literature by identifying heterogeneity in fuel price elasticities among German households for different socio-economic and regional characteristics. The results are highly relevant for policy modellers and for introducing effective policy measures for mitigating greenhouse gas emissions in road transport.

THE PREDICTORS OF QUALITY OF LIFE OF CORONARY HEART DISEASE PATIENTS: A STUDY IN TÜRKİYE

Aim: The aim of this study was to examine the predictors of the quality of life of patients who have coronary heart disease in Turkey. 
 Methods: The sample consisted of 796 participants with coronary heart disease selected from a nationally representative 2019 Turkey Health Survey. The effects of individual-level, individual-level healthy behaviors, household-level, and regional-level characteristics on quality of life were analyzed by using nested regression in STATA. 
 Results: The majority of respondents were male (62.3%), relatively more aged (41.0%), married (70.5%), and primary school educated (50.6%). More than 18 percent of respondents (144 people) said that they were unable to pay for the needed medical care. The variables lessening the quality of life for patients with coronary heart disease were found to be as female gender, increasing age, decreasing education level, living with a person with bad health status in the same house, inability to pay for medical care and living in statistical region-8 of Türkiye. 
 Conclusion: The results of this study contribute to the debate concerning the expected and unexpected relationships between QoL and its predictors among respondents with patients who have coronary heart disease. The study suggests that appropriate local and national policies should be developed to improve quality of life of coronary heart disease patients.

Perceived neighborhood disorder, social cohesion, and depressive symptoms in spousal caregivers

Objectives Prior research into the factors linked to mental health of caregivers of older adults have largely focused on individual- or household-level characteristics, but neighborhood supports and stressors may also matter for caregiver mental health. The current study fills this knowledge gap by examining the association of neighborhood social cohesion and disorder and depressive symptoms among spousal caregivers. Method We used data from the 2006 to 2016 waves of the Health and Retirement Study, which include 2,322 spousal caregivers. Negative binomial regression models were estimated to examine the association of perceived neighborhood social cohesion and disorder with depressive symptoms. Results A higher level of perceived neighborhood social cohesion was associated with fewer depressive symptoms (b = –0.06, 95% CI: −0.10, −0.02). On the other hand, greater perceived neighborhood disorder was associated with more symptoms (b = 0.04, 95% CI: 0.01, 0.08). The association of perceived social cohesion with depressive symptoms remained even after controlling for perceived disorder, but neighborhood disorder was no longer associated with depressive symptoms after accounting for reported neighborhood social cohesion. Conclusions This study suggests neighborhood supports and stressors matter for caregiver well-being. Neighborhood-based social support may be particularly important for caregivers as they navigate the challenges caregiving for an aging spouse can bring. Future studies should determine if enhancing positive characteristics of the neighborhood promotes well-being of spousal caregivers.

The prevalence and predictors of household food insecurity among adolescents in Canada

ObjectivesHousehold food insecurity is almost four times more prevalent among adolescents than among older adults in Canada, and it adversely affects their health. Our objective was to describe the sociodemographic and geographic patterning of household food insecurity among adolescents.MethodsOur analytic sample comprised all 12–17-year-old respondents to the 2017–2018 Canadian Community Health Survey with complete data on household food insecurity (n = 8416). We used bivariate and multivariable logistic regression models to identify respondent- and household-level sociodemographic characteristics associated with household food insecurity.ResultsThe prevalence of household food insecurity among adolescents was 20.7%. The adjusted odds of food insecurity were significantly elevated among adolescents who identified as Black or Indigenous (aOR 1.80), those living with a single parent (aOR 1.60), those living with a greater number of children ≤ 5 years (aOR 1.45) or 12–17 years (aOR 1.25), those in rented accommodation (aOR 1.98), those in households with only secondary school education (aOR 1.38), and those in households reliant on social assistance (aOR 2.03). Higher before-tax income was protective (aOR 0.99). In comparison with Ontario, the adjusted odds of food insecurity among adolescents were higher in Nunavut (aOR 6.77), Northwest Territories (aOR 2.11), and Alberta (aOR 1.48), and lower in Manitoba (aOR 0.66).ConclusionThe markedly higher odds of exposure to household food insecurity among adolescents who are Black or Indigenous and those living in households characterized by markers of social and economic disadvantage highlight the need for more effective policy interventions to protect vulnerable families from this hardship.

Characteristics and patterns of residential energy consumption for space cooling in China: Evidence from appliance-level data

The adoption of air conditioners (ACs) in households is rapidly growing in response to rising global temperatures, thereby posing significant sustainability challenges. It is vital to identify the characteristics and patterns of household demand for AC-induced electricity consumption. However, owing to huge costs and privacy concerns, appliance-level data sets are rare. Therefore, we conducted the Chinese Residential Energy Consumption Survey (CRECS) 2021, collected high-frequency operation data from 1140 active ACs from 673 households in 40 southern Chinese cities, and depicted the overall residential space-cooling electricity consumption profile. We found that, first, urban households in southern China showed a highly consistent pattern in AC time use. The peak usage in both living rooms and bedrooms occurred at noon (11 a.m.–3 p.m.) and night (9 p.m.–1 a.m.), when electricity consumption was 29%–38% higher than at other times of the day. Second, AC usage patterns showed great heterogeneity in terms of city-level and household-level characteristics. Households in large cities typically owned less energy-efficient equipment and consumed more AC-related electricity than did those in mega- and small-and-medium cities. Middle-income households, smaller room sizes, less-educated heads of households, and households with older adults were associated with lower AC-related electricity use.

Unexploded ordnance contamination and household livelihood choice in rural Vietnam

Vietnam is a country that found itself at the center of the Indochina wars and was subjected to the most intense aerial bombing in history. However, little research has been done on the effect of unexploded ordnance (UXO) contamination on household livelihoods in rural Vietnam. In this paper, we investigate the contaminating effect of unexploded ordnance on households. Livelihood choices are classified by cluster analysis techniques, and unexploded ordnance contamination is measured at the district level by the proportion of land at risk from unexploded bombs and mines. We examine the effect of UXO contamination on livelihood choices using a multinomial logit model, controlling for various important household and regional level characteristics. It was found that households in districts with greater contamination were less likely to adopt a formal wage-earning livelihood, characterized by higher income and less poverty, than they were to engage in an agricultural livelihood. This suggests that the Indochina wars have had a long-running effect, reducing the likelihood of non-farm diversification, which in turn diminishes economic well-being among rural households in Vietnam.

Determinants of clean cooking fuels utilization in three peri-urban communities in Ghana: key lessons to inform national policy

Use of polluting fuels has negative effect on human health and the environment. Use of clean cooking fuels (liquefied petroleum gas (LPG) can address this burden but in Ghana adoption is low with accessibility, convenience and safety concerns being potential barriers to LPG use. In order to address these issues, the Ghanaian government launched a new LPG distribution model through cylinder recirculation in 2019 on a pilot basis As part of the CLEAN-Air (Africa) project, we collaborated with National Petroleum Authority implementing the CRM to conduct face-to-face surveys of randomly selected households on household cooking fuels preferences and use in three peri-urban CRM pilot communities in Ghana. To analyse determinants of adoption and sustained use of LPG for cooking, individual, household and wider community characteristics were entered as covariates into logistic regression models as potential predictors. A total of 5,008 respondents completed the questionnaires across the three communities (Obuasi; 1,987 (40.0%), Kwaebibirem; 1,731 (34.9%) and Hohoe; 1,248 (25.1%)). A total of 48.6% (Obuasi (50.1%); Kwaebibirem (44.4%); Hohoe (52.2%)) households reported currently using LPG as the main cooking fuel. Being head of the household, increased household size, home ownership, an indoor cooking area, increased educational status and an increased number of LPG burners were associated with household’s choice of using LPG as a primary fuel. Households using an LPG stove with three or more-burners, were 68% [OR:1.68 (95%CI: 1.32, 2.14), 001] more likely to use LPG as a primary fuel than those using a single burner on top of the cylinder. Given the importance of the CRM, understanding household level characteristics for adoption are important. Better education and awareness raising of benefits will be important in scaling adoption of LPG as a clean fuel for cooking at the household level. Keywords: clean cooking fuels, LPG, Cylinder Recirculation Model, primary fuel use

Relatives, neighbors, or friends: Information exchanges among irrigators on new on-farm water management tools

CONTEXTOn-farm water application in Ethiopia, as in much of Sub-Saharan Africa, is dominated by furrow irrigation, which resulted in inefficient water uses and related economic and environmental problems. A recent project introduced two on-farm water management tools, called wetting front detectors and Chameleon sensors, to some farmers in Koga irrigation scheme and facilitated for other farmers in the quaternary canal, who did not receive the technology, to learn from farmers who installed the tools on their plots. OBJECTIVESThe objective of this study is to investigate the role of different social ties on information exchanges among farmers when some farmers have the signal on how long to irrigate a field during an irrigation event from on-farm water management tools. The study explored the relative importance of being neighbors, friends, spatial proximity of farms, and project induced pairings. METHODSThe study used a household survey data from all members of quaternary canals in the project that were in the technology, information, and control groups, as well as detailed network modules on how farmers with plots in the quaternary canal are associated with each other. A fixed effects econometric approach is used to control for time invariant household level and quaternary canal characteristics, while teasing out how the different social ties affect the information flow. RESULTS AND CONCLUSIONSThe results show that being in purposeful friendships as indicated by knowing each other's decision on the use of agricultural inputs and its outcome, as well as being spatially proximate as indicated by having farm plots next to each other or usually passing by each other's plots play a significant role in determining whether information-recipient farmers received information from the technology-recipient farmers as expected. Being relatives or neighbors played a minor role to facilitate information exchanges on how long to irrigate. In addition, ad-hoc pairs of farmers between technology-recipient and information-recipient created through the project within the quaternary canal did not play a significant role above and beyond the existing social ties of friendships and spatial proximity. SIGNIFICANCEThe findings have implications for effective ways of targeting in future scale up of such technologies as it informs that the roll out of such type of technologies and the extension services around it can better help technology diffusion and learning if they use friends and spatial proximates as anchors of information. That is, at times of over-subscription to such on-farm water management tools, information about the technology and the recommended duration of one irrigation turn can diffuse faster if the limited number of tools are distributed in such a way that friends and spatial proximates have access to a tool, rather than distributing the tools based on being neighbors or relatives.

Understanding the Geography of Victimization: A Spatial Analysis of Intimate Partner Violence in India.

Most studies on intimate partner violence (IPV) and its drivers have focused on individual-and household-level characteristics of the victim. Recent studies have acknowledged that it is a community-level phenomenon, using spatial analytical methods to analyze community-level determinants of IPV and its geographic dimensions. Such studies provide mixed evidence on the impact of different factors and need to be supplemented by similar studies-particularly in South Asian countries where IPV is common. The present study examines district-level variations in the incidence of various forms of IPV and identifies its determinants in India, a fast-growing South Asian country with poor gender indicators. The study combines data from the National Family Health Survey, District Level Household Survey, and the decadal Census. It applies spatial analytical methods such as the Global Moran's I, Getis-ord statistic, and Multivariate Local Geary to determine the nature of the spatial distribution of different categories of IPV. Spatial regression models are used to identify the community-level predictors of each category of IPV. The study finds non-random overlapping spatial clusters in the eastern part of India. The study also finds that neighborhoods characterized by low empowerment levels, and with a high child sex ratio, road connectivity, and proportion of socially marginalized groups are more likely to exhibit high levels of all types of IPV-although the impact of these determinants varies across districts. Furthermore, spill-overs in the incidence of IPV between neighboring districts are also observed. The study concludes by recommending the use of localized policies, rather than broad national or state policies, in reducing IPV.