Unequal Probability Sampling Research Articles

The spatial distribution of intimate partner violence prevalence and its associated factors among women aged 15–49 years in Zambia: evidence from the 2018 Demographic and Health Survey

BackgroundIntimate partner violence (IPV) includes any physical, sexual, or emotional harm experienced in any intimate relationship that results in negative outcomes. Zambia is one of the countries with the highest prevalence of IPV amongst women in sub-Saharan Africa. This study aims to investigate and describe the prevalence, associated risk factors, and geo-spatial distribution of IPV amongst women aged 15–49 years from the 2018 Zambia Demographic and Health Survey (ZDHS).MethodsThis study is a secondary data analysis of the 2018 ZDHS, which used the women’s individual dataset to extract a representative sample of 9 503 women from the domestic violence module. Analyses were adjusted using survey weights to account for unequal sampling probabilities. The bivariate and multivariable logistic regression models were applied to determine the factors associated with IPV. Stata MP version 14 was used to perform all analyses and QGIS software was used to map the geospatial distribution of IPV across provinces.ResultsThe overall IPV prevalence amongst women aged 15–49 years in this study was 36.5% (95%CI: 34.9 to 38.2), with Muchinga province having the highest prevalence at 55.2% (95% CI 50.4 to 59.8) and North western with the lowest prevalence at 22.6% (95% CI 19.9 to 25.6). In the adjusted analyses, women who justified wife beating were at a higher odds of experiencing IPV compared to those who did not (aOR = 1.75; 95% CI: 1.37 to 2.25; p < 0.001). Women with husbands or partners who consume alcohol were at higher odds of experiencing IPV (aOR = 3.81; 95% CI: 3.21 to 4.53; p < 0.001). The study also found that women who reported witnessing parental violence from their father to mother had increased odds of experiencing IPV (aOR = 1.75, 95% CI: 1.45 to 2.12; p < 0.001).ConclusionThis study has shown that women who witnessed parental violence from father to mother, justified wife beating, or had partners who consumed alcohol, had increased odds of experiencing IPV in Zambia. There is need to tailor interventions that address the cessation of alcohol consumption, and the promotion of awareness and education on IPV and its associated harms, especially in hot spot provinces.

A Matched Design for Causal Inference With Survey Data: Evaluation of Medical Marijuana Legalization in Kentucky and Tennessee.

A concern surrounding marijuana legalization is that driving after marijuana use may become more prevalent. Survey data are valuable for estimating policy effects, however their observational nature and unequal sampling probabilities create challenges for causal inference. To estimate population-level effects using survey data, we propose a matched design and implement sensitivity analyses to quantify how robust conclusions are to unmeasured confounding. Both theoretical justification and simulation studies are presented. We found no support that marijuana legalization increased tolerant behaviors and attitudes toward driving after marijuana use, and these conclusions seem moderately robust to unmeasuredconfounding.

Association between maternal and child health care and neonatal death in Angola: a secondary analysis of Angola Demographic Health Survey 2015–16

BackgroundNeonatal mortality has decreased in Angola; however, it remains high. Quality maternal and child health (MCH) care is key to improving neonatal survival. In Angola, however, the association between neonatal mortality and MCH care has not yet been investigated. Therefore, this study aimed to identify the association between neonatal mortality and MCH services in Angola.MethodsWe used the Angola Demographic Health Survey 2015–16, which is the latest nationally representative dataset of Angola. The associations between early/late neonatal death and MCH care utilization were identified by a multivariable logistic regression analysis, adjusted by the sex of the child, type of residence, wealth index, caesarian section, size of the child at birth and delivery assisted by skilled birth attendant. The individual sample weight, sample strata for sampling errors/design, and cluster number were incorporated in a descriptive and logistic regression analysis to account for the unequal probability sampling in different strata.ResultsThe early and late neonatal mortality rates were 22 and 2 per 1000 live births, respectively. We found that having none, one, two or three tetanus injections before the last pregnancy compared to five or more injections, and no postnatal health check for children before discharge were associated with the occurrence of late neonatal death. During the first 2 days after childbirth, no observation of breastfeeding, no counseling on breastfeeding, no counseling on newborn danger signs, no measurement of child body temperature, and no examination of the cord as well as not checking child health before discharge were associated with early neonatal death.ConclusionsDoses of maternal tetanus vaccination and postnatal child health check before discharge were modifiable factors associated to late neonatal death. Further studies to improve MCH care coverage are needed.

Measurements and medication data are indispensable for population blood pressure monitoring

Abstract Background The National Study on Health of Older People in Germany “Gesundheit 65+” included blood pressure (BP) measurements at home by a study nurse and classification of all current medication. Methods Nationally representative examination survey of the population aged 65 years and older residing in Germany using two-stage population registry sampling from 128 sample points, with a standardized examination at home from June 2022 to April 2023. Three standardized blood pressure measurements at two-minute intervals were taken after 5 minutes rest with an oscillometric device (Mobil-O-Graph) using one of five cuffs according to upper arm circumference. The mean of the second and third measurement was used for analysis. Hypertension was defined as mean systolic BP (SBP) at or above 140 mmHg, or mean diastolic BP (DBP) at or above 90 mmHg, or the use of antihypertensives (antihypertensives, diuretics, beta-blockers, calcium-channel blockers, ACEI and ARB) in participants with self-reported hypertension diagnosis. Provisional weights account for unequal sampling and participation probabilities. Results The “Gesundheit 65+” examination part included 1,493 participants of whom 1,475 (98.8%) had successful BP measurements (763 men and 712 women, aged 66 to 101 years). Based on self-reported diagnoses alone, the prevalence of hypertension would be 58.3% [95% CI 53.2-63.3] in men and 59.5% [53.8-64.9] in women. However, overall hypertension prevalence was estimated at 70.4% [65.8-74.7] in men and 72.3% [66.9-77.1] in women based on the study definition which combines self-reported diagnoses, standardized BP measurements and medication. Conclusions The prevalence of hypertension is greatly underestimated in the absence of examination survey data with standardized measurements and recording of medication use. In addition, measurement and medication data are needed for monitoring the hypertension cascade, i.e. awareness, treatment and control of hypertension. Key messages • Hypertension prevalence is underestimated in the absence of health examination surveys with measurements and medication data. • The hypertension cascade, i.e. assessment of awareness, treatment and control of hypertension needs population surveys with measurements and medication assessment.

Pairwise likelihood estimation and limited-information goodness-of-fit test statistics for binary factor analysis models under complex survey sampling.

This paper discusses estimation and limited-information goodness-of-fit test statistics in factor models for binary data using pairwise likelihood estimation and sampling weights. The paper extends the applicability of pairwise likelihood estimation for factor models with binary data to accommodate complex sampling designs. Additionally, it introduces two key limited-information test statistics: the Pearson chi-squared test and the Wald test. To enhance computational efficiency, the paper introduces modifications to both test statistics. The performance of the estimation and the proposed test statistics under simple random sampling and unequal probability sampling is evaluated using simulated data.

The Bahadur representations of quantile estimators in general unequal probability sampling

In this study, we establish the Bahadur representations of quantile estimators based on the Hájek and the Horvitz–Thompson estimators for the population distribution function in unequal probability sampling designs under general conditions. We also present the asymptotic normality of the above quantile estimators. The simulation studies using real data confirm the validity of the asymptotic normality.

Prevalence and predictors of HIV testing among young men in Papua New Guinea: A cross-sectional analysis of a nationally representative sample.

HIV testing is an important component of HIV prevention and serves as a gateway to other HIV-related services. However, the uptake remains suboptimal among young people, particularly in highly prevalent settings such as Papua New Guinea (PNG). This study aimed to assess the prevalence and determine the predictors of HIV testing uptake among young men aged 15-24 years in PNG. The 2016-2018 PNG Demographic and Health Survey (DHS) data was used. A total of 1,275 young men aged 15-24 years were included in the final analysis. Descriptive, bivariate, and multivariable logistic regression analyses were performed to determine independent predictors of HIV testing. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were reported. All analyses were adjusted using survey weights to account for unequal sampling probabilities. The overall prevalence of HIV testing among young men was 17.1% (95% CI: 15-19). Of those who were tested for HIV, about one-third (32.9%) had experienced a sexual debut at age <15 years, and 33.9% inconsistently used condoms during sex. In multivariable analysis, men aged 20-24 years (AOR 1.18, 95% CI: 1.00-2.31), who owned mobile phones (AOR 1.43, 95% CI: 1.00-2.55), who were aware that consistent condom use during sex can reduce HIV risk (AOR 2.18, 95% CI: 1.18-4.04), who had paid for sex (AOR 1.75, 95% CI: 1.01-5.83), and who had two or more sexual partners (AOR 1.37, 95% CI: 1.01-3.14) had increased odds of HIV testing. However, decreased odds of HIV testing were found among men who were never married (AOR 0.51, 95% CI: 0.29-0.88), lived in rural areas (AOR 0.54, 95% CI: 0.32-0.92), and consistently used condoms during sex (AOR 0.59, 95% CI: 0.34-1.01). The findings show that HIV testing is low among young men in PNG. To increase HIV testing uptake among young men, it is crucial to implement comprehensive youth-friendly HIV/STI education and tailored sensitization programs and enable more accessible and affordable HIV testing services. Also, outreach and community-based testing programs for young men in rural and prioritized areas requiring urgent prevention interventions are feasible options in PNG.

Review of how the generalized regression estimators contribute to estimating the financial and economic data with missing observations under unequal probability sampling

PurposeKnowing financial and economic information beforehand benefits in planning and developing policies for every country especially for a developing country like Thailand and for other Asian countries. Unfortunately, missing data or non-response plays an essential role in many areas of studies including finance and economics. Eradication of missing data in a proper way before further analysis can gain remarkable outcomes and can be effective for planning policies. This review on the generalized regression estimators for population total can be applied to financial, economic and other data when missing data are present.Design/methodology/approachThe generalized regression estimators for estimating population total, including the variance estimators under unequal probability sampling without replacement with missing data are explored under the reverse framework. Applications to financial and economic data in Thailand are also reviewed.FindingsThe review of literatures related to the proposed estimator shows the best performance, giving smaller variances in all scenarios.Originality/valueThe generalized regression estimators can assist in estimating financial and economic data that contain missing values with different missing mechanisms and can be used in other applications which help gain more superior estimators.

Detection of outliers in survey–weighted linear regression

ABSTRACT Regression diagnostics help identify influential data points in a model. Detecting outliers in complex survey design data involving stratification, clustering, and unequal probability sampling is difficult due to the presence of masking, where one outlier makes it hard to detect others. The masking factor for survey–weighted linear regression is developed and applied to analyzing the Household Consumer Expenditure dataset of 68th round of the National Sample Survey Organization survey of India. Regression parameters are calculated before and after detection and removal of outliers. The standard error of regression parameters for survey-weighted least squares models is reduced by 2% for the intercept, 5% for variable “meat” (X 5), 4% for “served processed food” (X 9), and 4% for “packaged processed food” (X 10). Inference alters the significance of regression coefficients of the variable “served processed food” (X 9) leading to the emergence of significance. There is no change in inference for other variables.

NATIONAL EXAMINATION SURVEY ON BLOOD PRESSURE OF OLDER AND VERY OLD PEOPLE IN GERMANY 2022-2023: WHY SELF-REPORTS, MEASUREMENTS AND MEDICATION DATA ARE NEEDED

Objective: The National Study on Health of Older People in Germany “Gesundheit 65+” aimed to provide up-to-date information on the health of older adults in Germany including blood pressure (BP) measurements at home by a study nurse. Design and method: Nationally representative examination survey of the population aged 65 years and older residing in Germany and with sufficient German language skills, using two-stage population registry sampling from 128 sample points from June 2022 to April 2023. Participation of physically or cognitively impaired persons was possible via proxy interviews and consent from legal representatives. Recruitment and data collection used a mixed-mode design. Three standardized blood pressure measurements at two-minute intervals were taken after 5 minutes rest with an oscillometric device (Mobil-O-Graph) using one of five cuffs according to upper arm circumference. The mean of the second and third measurement was used for analysis. Hypertension was defined as mean systolic BP (SBP) at or above140 mm Hg, or mean diastolic BP (DBP) at or above 90 mm Hg, or the use of antihypertensives (ATC-codes C02 antihypertensives, C03 diuretics, C07 beta-blockers, C08 calcium-channel blockers or C09 ACEI and ARB) in participants with self-reported hypertension diagnosis. Provisional weights account for unequal sampling and participation probabilities. Results: The “Gesundheit 65+” examination survey included 1,493 participants and 1,475 (98.8%) had successful BP measurements (763 men and 712 women aged 65 to 100 years). External quality control showed very good standardization of BP measurements in the home environment. Central BP data were available from 1,356 participants. Based on self-reported diagnoses alone the prevalence of hypertension would be 58.3% [95% CI 53.2-63.3] in men and 59.5% [53.8-64.9] in women. However, overall hypertension prevalence was estimated at 70.4% [65.8-74.7] in men and 72.3% [66.9-77.1] in women based on the study definition which combines self-reported physician diagnoses, standardized BP measurements and medication. Conclusions: Our study shows very good feasibility, acceptability and data completeness of standardized BP measurements using home visits. The prevalence of hypertension would be greatly underestimated in the absence of examination survey data with standardized measurements and detailed recording of medication.

Exploring the Influence of Village Social Capital and Rural Development on Farmers’ Entrepreneurial Decision-Making: Unveiling the Path to Local Entrepreneurship

This paper aims to comprehensively define village social capital, assess its multifaceted dimensions in relation to farmers’ entrepreneurial decision-making, and investigate the mediating role of rural development. The research sample consists of 14,133 farmers selected through a multi-stage proportional-to-size unequal probability sampling method. Data collection was carried out using questionnaires as part of the “Chinese Thousand Village Survey” project. Empirical analysis was conducted using binary logit and basic linear regression models, supplemented by probit models and other robustness testing methods. The findings indicate that village social capital exerts a significant and favorable influence on farmers’ entrepreneurial decision-making. Controlling for other variables, a unit increment in farmers’ village social network, village social participation, and village social trust is associated with a respective increase in farmers’ odds of entrepreneurial decision-making by 3.96, 5.42, and 16.28%. Village social trust emerges as the most influential factor. Additionally, rural harmony and rural economy partially mediate the relationship between village social capital and farmers’ entrepreneurial decision-making. This study contributes to a deeper understanding of the complex interplay between village social capital and farmers’ entrepreneurial decision-making, offering valuable guidance for farmers looking to leverage village social capital in their entrepreneurial endeavors. It lays a theoretical foundation for relevant entities to strengthen village social capital and cultivate a supportive environment for rural entrepreneurship. To our knowledge, this paper is among the earlier studies that investigate farmers’ entrepreneurial decision-making. Besides, the relationship between village social capital and rural development is rarely studied in the literature as two determinants of farmers’ entrepreneurial decision-making.

New Ratio Estimators for Population Mean under Unequal Probability Sampling Without Replacement in the Presence of Missing Data: A Case Study on Fine Particulate Matter in Bangkok, Thailand

Missing data are frequently present in datasets and give rise to a myriad of issues that significantly affect data utilization. The missing data needs to be handled before data can be efficiently estimated and applied. New ratio estimators for population mean were proposed for use when data are missing completely at random and for a more flexible situation where missing data are missing at random in the study variable under unequal probability sampling without replacement. Furthermore, the variance estimators of the proposed ratio estimators were investigated under a reverse framework. We show theoretically that the proposed estimators were approximately unbiased estimators. The proposed estimators were utilized in simulation studies and were applied to the study of fine particulate matter data in Suan Luang District, Bangkok, Thailand in order to see how the proposed estimators performed. The results from the application to fine particulate matter showed that the ratio estimators and their variance estimators worked better than existing estimators, producing less estimated variances. Therefore, they could be applied to estimate the average fine particulate matter even when missing values appeared.

Ship encounter scenario generation for collision avoidance algorithm testing based on AIS data

Collision avoidance algorithms play a crucial role in assisting onboard officers and ensuring navigation safety. To ensure the effectiveness of these algorithms, comprehensive testing with a wide range of scenarios is necessary before widespread deployment. However, there are currently few systematic methods available for generating such testing scenarios. In this paper, a general framework is proposed to generate testing scenarios based on Automatic Identification System (AIS) data, consisting of three modules: encounter scenario extraction, scenario importance evaluation, and testing scenario sampling. Firstly, ship encounter scenarios are extracted from the AIS data according to spatial-temporal proximity relationships between ships. Subsequently, a two-factor model is developed to evaluate the importance of extracted scenarios, considering both exposure frequency and collision risk. Finally, the unequal probability sampling method, which takes into account the importance of each scenario, is utilized to construct the testing scenario set. The proposed framework is validated through a systematic experiment using real AIS data. The results demonstrate that the framework effectively extracts diverse encounter scenarios from AIS data and rationally evaluates their importance. On this basis, an appropriate scenario set can be generated for collision avoidance algorithm testing. The generated scenario set ensures comprehensive coverage while prioritizing high-importance scenarios.

Integrating spatial and ecological information into comprehensive biodiversity monitoring on agricultural land

Biodiversity loss on agricultural land is a major concern. Comprehensive monitoring is needed to quantify the ongoing changes and assess the effectiveness of agri-environmental measures. However, current approaches to monitoring biodiversity on agricultural land are limited in their ability to capture the complex pattern of species and habitats. Using a real-world example of plant and habitat monitoring on Swiss agricultural land, we show how meaningful and efficient sampling can be achieved at the relevant scales. The multi-stage sampling design of this approach uses unequal probability sampling in combination with intermediate small-scale habitat sampling to ensure broad representation of regions, landscape types, and plant species. To achieve broad coverage of temporary agri-environmental measures, the baseline survey on permanent plots is complemented by dynamic sampling of these specific areas. Sampling efficiency and practicality are ensured at all stages of sampling through modern sampling techniques, such as unequal probability sampling with fixed sample size, self-weighting, spatial spreading, balancing on additional information, and stratified balancing. In this way, the samples are well distributed across ecological and geographic space. Despite the high complexity of the sampling design, simple estimators are provided. The effects of stratified balancing and clustering of samples are demonstrated in Monte Carlo simulations using modelled habitat data. A power analysis based on actual survey data is also presented. Overall, the study could serve as a useful example for improving future biodiversity monitoring networks on agricultural land at multiple scales.

An Investigation into the Attitude Toward Special Education Curriculum and Teaching of Special Education Teachers: The Data from China’s Guangdong Province

To further understand the curriculum and teaching attitude of teachers in special education schools in Guangdong Province, this study adopted a stratified whole-group unequal probability sampling method and selected 768 teachers from special education schools in Guangdong Province’s 21 cities for a questionnaire survey. The data on special education teachers’ attitudes towards special education curriculum and teaching were selected. Descriptive statistics, one-sample t-test, independent sample t-test, and one-way analysis of variance (ANOVA) were used to analyze the data and compared with the data of special education teachers’ attitudes towards special education curriculum and teaching in the National Survey of Special Education Teacher Professional Development (NSETPDS). The results indicated that the cognitive response of special education teachers in Guangdong Province to the special education curriculum and teaching attitude is positive (M=1.81, SD=0.51), significantly higher than that of the national special education teachers. The emotional response and general attitude were positive but significantly lower than that of the national special education teachers. The attitude of male special education teachers toward the special education curriculum and teaching was significantly higher than that of female special education teachers. Special education teachers with more than 20 years of teaching experience had a more positive attitude towards special education curriculum and teaching than those with 1-5 years of teaching experience. Based on the analysis of the research results, some suggestions are put forward to provide a basis for the professional development of teachers in Guangdong Province.

A survey on public welfare cognition of college students in 2023

This paper describes the situation of public welfare volunteerism, and conducted a survey in Zhejiang Province. According to the existing data, a survey method was designed, and a comparative study was carried out in the form of multi-stage sampling, simple random sampling and unequal probability sampling. In addition, a questionnaire was set up according to the existing data, and we also conducted a study on sample quality control. For example, the unqualified samples are discarded, and the final conclusion is that most people have a strong purpose to participate in public welfare, while the appeal and credibility of public welfare itself are almost ignored.

Internet Media and Universal Trust: An Analysis of the Impact on the Psychology of Social Interaction among Chinese Netizens.

Objectives: With the rapid advancement of the "Internet +" era, network media have an increasingly greater influence on human production and lifestyle. However, the network security situation has become more serious, which has affected the trust of netizens in society. In this study, we explore the interaction between online media and universal trust, and further analyze the social psychological impact on Chinese netizens. Methods: We collected data from the "China General Social Survey" (CGSS) in 2017, using a 4-stage unequal probability sampling. We set 3 hypotheses and used a binary logistic regression model for data analysis. Results: We found that the universal trust of respondents who use online media as their main source of information is lower than that of traditional media, while social interaction and class status have a significant weakening effect on the negative impact of online media and its universal trust. Conclusions: These results contribute to further expansion of the theoretical and practical implications. The use of data ensures robust research conclusions; we propose factors that increase Chinese netizens' trust in online media.

Determinants of comprehensive knowledge on mother-to-child transmission of HIV and its prevention among childbearing women in Rwanda: insights from the 2020 Rwandan Demographic and Health Survey

BackgroundMaternal knowledge on mother-to-child transmission (MTCT) and its prevention has been identified to enhance maternal testing and adherence to antiretroviral therapy (ART) regimen. Examining prevalence and associated factors on MTCT and its prevention among women provides empirical evidence for design and implementation of health strategies aimed at increasing MTCT knowledge and its elimination. This study therefore examined women’s comprehensive knowledge and associated factors on MTCT and its prevention among childbearing women in Rwanda.MethodsAnalysis was conducted on a weighted sample of 14,634 women from the 2020 Rwanda Demographic and Health Survey (RDHS). Dataset cleaning and missing value analysis was conducted. Chi square, bivariate and multivariable regression was then conducted in complex samples in SPSS. Alpha level set at p < 0.05 and at 95% Confidence Interval (95% CI). All analysis were adjusted for unequal probability sampling using survey weights. Bivariate and multivariable results were reported with crude and adjusted odds ratios.ResultsThe mean age was 29.2 years, SD-9.1. Prevalence of HIV testing and comprehensive knowledge on MTCT and its prevention among women in Rwanda was 79.6% and 65.1% respectively. Findings from this study showed that married women have higher odds (aOR = 1.18, 95% CI = 1.04–1.35) of comprehensive knowledge on MTCT and its prevention compared to those unmarried. Women who were living in southern (aOR = 1.23, 95%CI = 1.02–1.48) and eastern (aOR = 1.37, 95% CI = 1.13–1.66) parts of Rwanda were more likely to have adequate knowledge on MTCT of HIV and its prevention than those in Kigali. Also, women who received post-test counselling (aOR = 1.26, 95% CI = 1.01–2.11) have increased knowledge on MTCT than those who did not. Women with access to radio (aOR = 1.18, 95% CI = 1.06–1.32) and television (aOR = 1.25, 95% CI = 1.07–1.45) at least once a week were more likely to have adequate knowledge on MTCT and its prevention compared to those who do not in Rwanda.ConclusionThere is inadequate knowledge on MTCT and its elimination among women of reproductive age in Rwanda. Strategies to enhance knowledge on MTCT and its prevention among childbearing women should be adopted through rigorous educational sensitization campaigns using local media such as radio and television. Health services that focus on prevention of MTCT must emphasize post-test counselling.

Prevalence and factors associated with condom use among women aged 15-49 years in Rwanda using a survey logistic regression model: evidence from the 2019/20 Rwanda Demographic and Health Survey.

although Rwanda's HIV prevalence has declined, many people are still acquiring or living with it. Among other methods of HIV prevention, condoms are a safe and reliable method in addition to preventing pregnancy and other sexually transmitted infections, especially when used properly. This study aimed to determine the prevalence and determinants of condom use during last sexual intercourse among reproductive-aged women in Rwanda. using data from the cross-sectional, nationally representative Rwanda Demographic and Health Survey (RDHS) conducted in 2019/2020, we carried out secondary data analysis. A multivariable logistic regression model was applied to determine the factors associated with condom use. All analyses from the model were adjusted for unequal sampling probabilities using survey weights. results showed a 10.8% prevalence of condom use. The odds of condom use during last sex were significantly lower for women who lived with a man (adjusted odds ratio [aOR]= 0.10, CI=0.08: 0.13) and those from the Southern region (aOR=0.69, CI= 0.52 to 0.92) but were significantly higher for those with primary education (aOR=1.38, CI= 1.00: 1.88). Also, the rich were more significantly associated with condom use compared to the poor (aOR=1.53, 95% CI= 1.20: 1.93). Those who had three or more sexual partners had higher odds of condom use than those with one partner (aOR=3.12, CI= 2.50: 3.89). based on the results, health promotion interventions aimed at raising awareness on HIV prevention should, therefore, target the groups that were found to have a high risk of not using condoms.

New estimators for estimating population total: an application to water demand in Thailand under unequal probability sampling without replacement for missing data.

Water shortage could play an imperative role in the future due to an influx of water demand when compared to water supplies. Inadequate water could damage human life and other aspects related to living. This serious issue can be prevented by estimating the demand for water to bridge the small gap between demand and supplies for water. Some water consumption data recorded daily may be missing and could affect the estimated value of water demand. In this article, new ratio estimators for estimating population total are proposed under unequal probability sampling without replacement when data are missing. Two situations are considered: known or unknown mean of an auxiliary variable and missing data are missing at random for both study and auxiliary variables. The variance and associated estimators of the proposed estimators are investigated under a reverse framework. The proposed estimators are applied to data from simulation studies and empirical data on water demand in Thailand which contain some missing values, to assess the efficacies of the estimators.