Survey Statistics Research Articles

Ninety and not out-Understanding our oldest old.

This paper draws attention to the rapidly growing number of Australian nonagenarians, presents previously unpublished information on this sub-group of older people and explores the implications for future patterns of service delivery, planning and policy. Statistical analyses of Census data and other Australian Bureau of Statistics Surveys using Table Builder Pro, combined with analysis of de-identified Australian Institute of Health and Welfare (AIHW) unit record data on aged care and AIHW GRIM (mortality) data. Male nonagenarians almost doubled from 2006 to 2016, while their female counterparts grew by 55%. This cohort is the first to reap cumulative advantage from the dramatic reduction in death rates from 1970. Their demographic circumstances reveal both changes and continuities compared to the previous cohort. Men and women aged 90 and over use a substantial proportion of aged care services, and their characteristics and circumstances are highly relevant to planning current and future aged care services.

Individual Factors Associated with Women’s Use of Mental Health Services during Incarceration: Targeting Programmatic Resources

ABSTRACT This study analyzed data from the Bureau of Justice Statistics Survey of Inmates in State Correctional Facilities to identify factors associated with the use of mental health services among incarcerated women (N = 2571). One quarter of the sample (25.3%) reported accessing mental health counseling during their incarceration, while 30.8% reported using psychotropic medications. Survivors of childhood sexual victimization were more likely to use psychotropic medication (OR = 1.45, p = .016) compared to women who had not experienced this form of violence. Women convicted of violent offenses were more likely to use both psychotropic medication (OR = 2.13, p < .001) and counseling services (OR = 1.57, p = .002). Correctional practitioners should considertargeting resources to address women’s experiences as both victims and perpetrators of violence.

Model-Based Screening for Robust Estimation in the Presence of Deviations from Linearity in Small Domain Models

Abstract Small domain estimation models, like the Fay-Herriot (FH), often assume a normally distributed latent process centered on a linear mean function. The linearity assumption may be violated for domains that express idiosyncratic phenomena not captured by the predictors. The direct sample estimate for such domain will be viewed as an outlier by FH when, in fact, it reflects an underlying true value. The model interpretation is also confounded by the variances of direct sample estimates because, while typically treated as fixed and known, they are estimates and thus contain noise. In this article, we construct a joint model for the direct estimates and their variances with nonparametric mixtures of normal distributions with the goal to improve robustness in estimation quality for these idiosyncratic domains. We devise a model-based screening tool to nominate domains where the model may not accurately account for deviations from the linearity assumption. We replace the modeled values for nominated domains with the direct estimate which we show robustify our models. The US Bureau of Labor Statistics’ Current Employment Statistics (CES) survey publishes monthly employment estimates for domains defined by industry and geography. Model estimation is performed for smaller domains to improve the reliability of the direct estimator. We compare fit performances for our candidate models under data constructed to be similar to the CES and conduct a simulation study to assess the robustness of our candidate models in the presence of deviations from linearity. We apply our model-based screening method and quantify its ability to improve the quality of published estimates.

Establishing Transport Statistics: Results From the Medevac Transport Statistics Survey

ObjectiveThe purpose of this article was to report the results of a national survey of medical transport programs to establish national estimates of critical care transports and use those results combined with other data sources to generate annual transport volume estimates. MethodsAn online survey was administered to collect transport statistics from medical transport programs registered in the Association of Air Medical Services Atlas and Database of Air Medical services in 2015. ResultsRoughly 20% of all registered programs participated. An estimated 640,000 critical care transports are conducted annually; an additional breakdown by mode of transfer is presented. ConclusionLow participation rates preclude establishing precise critical care transport statistics. Future participation is encouraged to enable more accurate data reporting to establish resources that can support research and policy initiatives.

Statistical knowledge and training in second language acquisition

Abstract The present study investigated second language acquisition (SLA) doctoral students’ statistical training and knowledge of statistics. One hundred and twenty SLA doctoral students in North America took a comprehensive statistics survey, and 16 of them participated in follow-up interviews. The results showed that doctoral students were well trained in basic descriptive statistics, while their training in inferential statistics, particularly advanced statistics, was limited. When looking at their statistical knowledge, the results indicated that SLA doctoral students were good at understanding descriptive and inferential statistics, but they found it hard to interpret statistical analyses related to inferential statistics that are commonly encountered in SLA research. Several suggestions directed toward improving statistical literacy in SLA were provided.

Future projection of cancer patients with cardiovascular disease in Japan by the year 2039: a pilot study

BackgroundThe number of cancer patients in Japan is estimated to rise to 3.5 million by 2025. The disease burden may be further complicated by comorbidities caused by cardiovascular disease (CVD). Predicting the number of cancer patients with CVD can help anticipate future resource needs.MethodsWe used statistics derived from the Niigata Cancer Center CVD Study (2015) as well as population estimates from the National Cancer Center’s Cancer Registry and Statistics survey of 2017 for convenience. We simply multiplied the projected number of cancer patients through the year 2039 by the CVD prevalence in 2015, with patients classified by sex, age, and cancer type to estimate the number of cancer patients with CVD.ResultsThe total number of Japanese cancer patients with CVD was 253,000 in 2015 and is predicted to increase rapidly by 30,000 in 2020 and peak at 313,000 in 2030–2034. Men will dominate the CVD population at 2.5-fold the number of women. The growth rate of the population with both cancer and CVD will be greater than that of the cancer-only population (1.23 vs 1.18, P < 0.001), and will comprise notably high proportions of patients with prostatic, breast, and uterine cancers (1.80, 1.57, and 1.66, P < 0.001, respectively).ConclusionFuture cancer patients will be older and more likely to have CVD. Although men will continue to dominate this population, the increase in the number of women will be pronounced. Cancer care providers should be trained to recognize CVD and provide any necessary interventions concurrently with cancer therapy.

Unintentional injury mortality among children under age five in urban and rural areas in the Sichuan province of west China, 2009\u20132017

This study analysed trends in the unintentional injuries specific mortality rates among children under age five (UI-specific U5MRs) in urban and rural areas in the Sichuan province of western China. Data were obtained from the National Health Statistics Survey System. The Cochran-Armitage trend test was used to analyse the trends in UI-specific U5MRs and the proportion of unintentional injury deaths to total deaths. The Poisson regression model was used to compare the UI-specific U5MRs between rural and urban areas. The overall UI-specific U5MRs decreased from 3.8 to 1.7 per 1,000 live births from 2009 to 2017, with an average annual decline in the rates of 8.78% and 10.05% in urban and rural areas, respectively. The UI risk of death in rural areas was approximately 1.95 times that in urban areas (95% CI: 1.73–2.18; p < 0.01). A total of 49.9% of all the children in the study did not receive any treatment before death caused by UI. The UI-specific U5MRs significantly declined in Sichuan province from 2009 to 2017, but large disparities in UI-specific U5MRs in urban and rural areas still exist. Reducing the U5MRs due to UI should be a major public health concern in western China.

Evaluation of individuals’ satisfaction with health care services in Turkey

ObjectivesTo demonstrate whether features related to the institution and service provider affect individuals’ level of satisfaction regarding the public health care services they receive. MethodsThis study used hierarchical regression analysis and data from the Life Satisfaction Statistics survey implemented by the Turkish Statistical Institute, with 9,397 people in 2015. Level of satisfaction with public health care services was the dependent variable. Having any problems with securing an appointment for examination and analyses, cleanliness/hygiene, attitudes of physicians and nurses/caregivers towards patients, costs of examination and analyses, prices of medications, queuing for examination and/or analyses, co-payments for examination, and if they think the number of physicians and medical personnel is adequate were the independent variables. ResultsSecuring appointments, costs of examinations and analyses, inadequate number of physicians and medical personnel, prices of medications, queuing, cleanliness/hygiene, attitudes of physicians and nurses/caregivers, and co-payments were important determinants of individuals’ level of satisfaction with health care services. ConclusionsThis study revealed that despite the development in the health care sector in Turkey, individuals continue to experience problems with various areas related to health care services, and accordingly, further improvements are needed.

Life Course Perspective on Economic Shocks and Income Inequality Through Age‐Period‐Cohort Analysis: Evidence From Finland

Utilizing age‐period‐cohort analysis, this paper examines the development of income distribution across periodic economic fluctuations in relation to cohorts and age groups. The empirical analysis is based on the Finnish Income Distribution Statistics and Household Expenditure Surveys covering the period of 1966–2015. The findings suggest that the period and cohort effects can be identified as the main effects on relative income, while the age effects have no meaningful impact when the control variables are taken into account. This result reveals a connection between the effects of economic shocks and cohort placement on labor market entry. Additionally, absolute income analysis suggests that economic shocks create stagnation points in income development, which are especially detrimental to cohorts who are transitioning into labor markets. Additionally, middle‐income attainment has not changed due to periodic shocks but rather is related to inter‐cohort inequalities and relative income differences, where the baby boomer generation is a clear winner.

The Prevalence of Female Representation in Supervisory and Management Positions in American Law Enforcement: An Examination of Organizational Correlates

Increasing gender diversity in American policing has long been a focus of reform efforts since the 1960s. Although the proportion of women in the profession improved initially and research has focused on the gender breakdown of police organizations in general, less scholarly and empirical attention has been directed to female representation in positions of power (i.e., supervisory and management roles). Using data from the 2013 Law Enforcement Management and Administrative Statistics survey, the current study examined the prevalence as well as the organizational correlates of female representation in supervisory (e.g., sergeants), mid-level management (e.g., lieutenants) and chief executive roles in departments across the United States. The findings indicate a number of factors associated with female representation in supervisory and mid-level management positions, including the size and type of agencies, geographic region of the country, and potential indicators of professionalism. Policy implications and directions for future research are discussed.

Evaluating Factors for Predicting the Life Dissatisfaction of South Korean Elderly using Soft Margin Support Vector Machine based on Communication Frequency, Social Network Health Behavior and Depression

Since health and the quality of life are caused not by a single factor but by the interaction of multiple factors, it is necessary to develop a model that can predict the quality of life using multiple risk factors rather than to identify individual risk factors. This study aimed to develop a model predicting the quality of life based on C-SVM using big data and provide baseline data for a successful old age. This study selected 2,420 elderly (1,110 men, 1,310 women) who were 65 years or older and completed the Seoul Statistics Survey. The quality of life satisfaction, a binary outcome variable (satisfied or dissatisfied), was evaluated based on a self-report questionnaire. This study performed a Gauss function among the SVM algorithms. To verify the predictive power of the developed model, this study compared the Gauss function with the linear algorithm, polynomial algorithm, and sigmoid algorithm. Additionally, C-SVM and Nu-SVM were applied to four kernel algorithm types to create eight types, and prediction accuracies of the eight SVM types were estimated and compared. Among 2,420 subjects, 483 elderly (19.9%) were not satisfied with their current lives. The final prediction accuracy of this SVM using 625 support vectors was 92.63%. The results showed that the difference between C-SVM and Nu-SVM was negligible in the models for predicting the satisfaction of life in old age while the Gaussian kernel had the highest accuracy and the sigmoid kernel had the lowest accuracy. Based on the prediction model of this study, it is required to manage local communities systematically to enhance the quality of life in old age.

Economic Savings From Changing Anesthetic Agent Purchasing Must Include Costs Associated With Expected Changes in Case Times Known From Meta-analyses of Randomized Clinical Trials

To the Editor Zuegge et al1 reported that their department’s “tracking of anesthetic volatile agent purchasing show[ed] a >100% reduction (from 252 to 113 bottles/month) in desflurane usage and a 14% increase (from 262 to 305 bottles/month) in sevoflurane usage over this period, while isoflurane usage remained steady. This change in purchasing trend [over the period of 5 years] led to a savings estimate of $25,000 per month.” Purchasing reductions mathematically cannot exceed 100%. Were the intended values ≅55.2% for desflurane (ie, [252–113]/252) and ≅16.4% for sevoflurane (ie, [305–263]/262)? Anesthesia & Analgesia editorial policy requires that brief reports follow the relevant Enhancing the QUAlity and Trans parency Of health Research guideline, which for economic end points is the Consolidated Health Economic Evaluation Reporting Standard. Such requirements are designed to help ensure that research methodologies are valid. The authors’ analysis of “savings” failed to consider the following items, listed in sequence of the Consolidated Health Economic Evaluation Reporting Standard statement: cost savings perspective (eg, societal), resource use associated with the intervention, valuation of each resource changed, listing of cost model assumptions, incremental secondary costs, and estimation of outcomes of interest expected to differ. Valid economic analyses of anesthetic drugs must include effects on personnel and workflow costs, including nonoperative time. There are slower emergence times (end surgery to extubation) and turnover times with sevoflurane compared to desflurane.2 Differences in times for tracheal extubation can be quantified accurately using either prospective observational studies or randomized clinical trials.3 Time reductions in proportional scales are relatively homogeneous among studies, making meta-analyses useful and their findings easily applied to cost modeling.2 The values of the cost reductions are highly heterogeneous among operating rooms of the same hospital and among hospitals (eg, based on workday duration and personnel compensation).4 These economic relationships are well understood.2,4 When these factors are ignored, such as in the authors’ study, findings do not produce generalizable economic science. For example, suppose that 85% of the authors’ 2648 general anesthesia cases per month in fiscal year 2015 were performed principally in operating rooms with >8 hours of cases, averaging 3 cases per operating room per day.1,2,4 Suppose the mean time to extubations was 9.2 minutes.2 Then, the authors likely added a mean of 3.3 minutes per case because of the shift from desflurane to sevoflurane, including both extra emergence time in the operating room and delay in starting next cases, where 3.3 = 25% × 9.2 minutes + 0.5 minutes × 2. Using a cost of $3.44 per minute of operating room time, based on May 2016 Bureau of Labor Statistics Occupational Employment Statistics survey (Accessed: Thursday, December 14, 2017), this extra time may have increased hospital costs by $25,550 per month. Compared with the reported $25,000 reduction in purchasing costs,1 no substantive savings likely accrued. Note that if the hospitals’ public accounting reports were used, because costs per minute for operating room time are generally much larger than $3.44, the hospital’s net costs would have increased. Regardless of the details of our example, the authors overestimated the savings from their vaporizer labeling and education initiative. Franklin Dexter, MD, PhD, FASADepartment of Anesthesia, University of IowaIowa City, Iowa[email protected] Richard H. Epstein, MDDepartment of Anesthesiology, Perioperative Medicine and Pain ManagementUniversity of MiamiCoral Gables, FloridaDepartment of AnesthesiologySidney Kimmel Medical CollegeThomas Jefferson UniversityPhiladelphia, Pennsylvania

Familial hypercholesterolemia prevalence in an admixed racial society: Sex and race matter. The ELSA-Brasil

Background and aimsFamilial hypercholesterolemia (FH) is a genetic disorder associated with high cardiovascular burden of disease. FH prevalence may vary widely across populations and data in race/ethnically diverse and admixed populations is scarce. ELSA-Brasil epidemiology may be widely generalizable in this regard, and we calculated the ELSA-Brasil FH prevalence and its variation according to age, sex and race/ethnicity. MethodsIn 14,460 individuals aged from 35 to 75 years from the ELSA-Brasil cohort baseline, we classified FH according to the Dutch Lipid Clinic Network criteria score ≥6 (probable and definite FH). LDL-C levels were adjusted for statin use. We calculated the overall ELSA-Brasil FH prevalence and the weighted prevalence for age, sex and race/ethnic categories. We extrapolated those frequencies to the Brazilian population weighting for age-sex-race/ethnicity according to the 2015 Statistics and Geography Brazilian Institute survey. ResultsThe overall FH prevalence per 1000 individuals in ELSA-Brasil was 3.8 (2.9, 4.9) or 1 in 263. The age/sex/race-ethnicity-weighted FH prevalences were: male, 3.0 (1.7, 4.4) or 1 in 333; female, 4.1 (3.0, 5.2) or 1 in 244 (p<0.001). White race prevalence was 2.4 (1.9, 3.0) or 1 in 417; Brown, 4.9 (4.0, 5.9) or 1 in 204; and Black 6.4 (41.1, 8.7) or 1 in 156 (p<0.001). The weighted extrapolation for the Brazilian population derived similar magnitude frequencies. ConclusionsFH affects 1 in 263 in ELSA-Brasil and affects disproportionally more Brown (1 in 204), and Black (1 in 156), than White (1 in 417). Weighted extrapolation for the Brazilian population derived similar magnitude frequencies.

Automated selection of post‐strata using a model‐assisted regression tree estimator

AbstractDespite having desirable properties, model‐assisted estimators are rarely used in anything but their simplest form to produce official statistics. This is due to the fact that the more complicated models are often ill suited to the available auxiliary data. Under a model‐assisted framework, we propose a regression tree estimator for a finite‐population total. Regression tree models are adept at handling the type of auxiliary data usually available in the sampling frame and provide a model that is easy to explain and justify. The estimator can be viewed as a post‐stratification estimator where the post‐strata are automatically selected by the recursive partitioning algorithm of the regression tree. We establish consistency of the regression tree estimator and a variance estimator, along with asymptotic normality of the regression tree estimator. We compare the performance of our estimator to other survey estimators using the United States Bureau of Labor Statistics Occupational Employment Statistics Survey data.

Sustainability of Italian families' food practices: Mediterranean diet adherence combined with organic and local food consumption

The present research intends to contribute to the literature on sustainable food consumption by investigating the present-day dietary patterns of Italian households to assess the degree of adherence to the Mediterranean diet together with the level of organic and local food consumption. The analysis is based on data from the Italian National Institute of Statistics survey “Aspects of Daily Life” for the year 2014; this is a Multipurpose Survey conducted in Italy with 44,984 individuals across 18,864 households. A Heckman two-step probit model was implemented to estimate equations to assess the impact of Italian household composition and adherence to the Mediterranean diet upon organic and local food consumption. The study's findings show that both those households with children and those with a higher degree of adherence to the Mediterranean diet have higher probabilities of buying both organic and local products, while increasing household size reduces the likelihood of buying local products. Furthermore, good economic status and a higher level of education increase the probability of consuming organic food. Some of the differences highlighted are of low amplitude; however, these results offer novel and original insights for Italy and support the need to for both policy makers and marketers to implement a wide range of initiatives aimed at educating consumers about, and facilitating access to sustainable food products.

A Dynamic Inhomogeneous Latent State Model for Measuring Material Deprivation

SummaryMaterial deprivation can be used to assess poverty in a society. The status of poverty is not directly observable, but it can be measured with error for instance through a list of deprivation items. Given two unobservable classes, corresponding to poor and not poor, we develop a time inhomogeneous latent Markov model which enables us to classify households according to their current and intertemporal poverty status, and to identify transitions between classes that may occur year by year. Households are grouped by estimating their posterior probability of belonging to the latent status of poverty. We then estimate an optimal weighting scheme, associated with the list of items, to obtain an optimal deprivation score. Our score is arguably better at predicting the poverty status than simple item counting (equal weighting). We use the longitudinal component of the European Union statistics Survey on Income and Living Conditions for evaluating poverty patterns over the period 2010–2013 in Greece, Italy and the UK.

EVALUATION OF THEORETICAL AND EMPIRICAL CHARACTERISTICS OF THE COMMUNICATION, LANGUAGE, AND STATISTICS SURVEY (CLASS)

The interaction between language and the learning of statistical concepts has been receiving increased attention. The Communication, Language, And Statistics Survey (CLASS) was developed in response to the need to focus on dynamics of language in light of the culturally and linguistically diverse environments of introductory statistics classrooms. This manuscript presents a cross-cultural evaluation of the characteristics of the CLASS III (third generation of the instrument) and provides a user-friendly cross-culturally valid version of the CLASS. Mixed methods are employed to investigate further characteristics of the CLASS III and provide a scale (CLASS IV) that may be utilized in diverse settings. These research results have implications for instructors, professional developers, and researchers to improve instruction with culturally and linguistically diverse student populations. First published May 2018 at Statistics Education Research Journal Archives

The unmet support needs of carers of older Australians: prevalence and mental health.

Utilizing new data from the 2015 Australian Bureau of Statistics (ABS) Survey of Disability, Ageing and Carers, we calculated the prevalence of carers experiencing specific and multiple unmet needs for support, using single and multiple item measures. Logistic regression models were fitted to examine the association between unmet needs and psychological distress (using the Kessler psychological distress scale), once demographic and health factors were controlled for. In 2015, 35% of carers of older Australians cited at least one unmet need for support. Among this group, almost two-thirds cited multiple unmet support needs (64.7%). The most prevalent types of unmet needs included financial (18%), physical (13%), and emotional support (12%), as well as additional respite care and support to improve carer health (12%). After controlling for demographic and health characteristics of the carer, having any unmet need for support increased the odds of psychological distress by twofold (OR = 2.20, 95% CI = 1.65, 2.94). With each successive unmet need for support, the odds of psychological distress increased 1.37 times (OR = 1.36, 95% CI = 1.22, 1.54). Those who had received assistance with care, but required further support were 1.95 times more likely (OR = 1.95, 95% CI = 1.17, 3.24) to be in distress and those who had not received care assistance were about 2.4 times more likely (OR = 2.38 95% OR = 1.56, 3.62) to be in distress relative to those with no unmet need. Addressing unmet support needs of carers is important, not only for the planning of services for carers in an aging population, but also because of the association between unmet support needs and carers mental health.

Poverty and Income Inequality in Nigeria: An Illustration of Lorenz Curve from NBS Survey

This study adopted the Nigerian Bureau of Statistics survey to examine poverty and income inequality in Nigeria. The objective was to examine the rate of poverty and income distribution in Nigeria using the Lorenz curve and Gain coefficient. Food poverty line, absolute poverty line, subjective poverty measure and dollar per day poverty line were used to measure poverty while Gani coefficient was used to measure income inequality. Findings reveal that significant proportions of Nigerian population are living below the poverty line adopted in this study. The study also found that there is wide gap between the rich and the poor in Nigeria. The study recommend implementable polices to reduce poverty and reduce income inequality in Nigeria.

RELATIONSHIP OF FACTOR-FACTOR WITH EARLY PREGNANCY IN WORKING AREA PUBLIC HEALTH CENTER GUGUK PANJANG

&lt;p&gt;&lt;em&gt;According to the Bureau of Statistics survey in 2012 in Indonesia was 48 / 1,000 pregnancies at the age of 15-19, while in 2007 was 8.5% of pregnancies at age &amp;lt;20 years. Public health center Guguk Panjang 20% of women with risk pregnancies with child-bearing age &amp;lt;20 years. This study aims to determine the factors associated with Pregnancy Early in Puskesmas Guguk Panjang . &lt;/em&gt;&lt;/p&gt;&lt;p&gt;&lt;em&gt;This type of research is descriptive analytic with cross sectional approach. The population in the study were all pregnant women in Public health center Guguk Panjang. Retrieval using random sampling techniques to obtain 100 respondents. Data collected by using a questionnaire. Data analysis included univariate and bivariate analysis using chi-square test. &lt;/em&gt;&lt;/p&gt;&lt;p&gt;&lt;em&gt;Results showed 60% of respondents are highly educated, 57% of respondents with high economic status, 52% of respondents with negative behavior, 62% of respondents with a social and cultural good, 75% of respondents affected by the mass media, and 52% of respondents underwent teen pregnancy. The results of the bivariate analysis showed that there is a relationship of education (p = 0.003, OR = 3.857), the economy (p = 0.018, OR = 2.893), behavior (p = 0.003, OR = 3.821), sociocultural (p = 0.01, OR = 3.261) and the media (p = 0.011, OR = 4.030) with a teen pregnancy. &lt;/em&gt;&lt;/p&gt;&lt;em&gt;It can be concluded that education, economic status, behavioral, social, cultural, and media associated with teen pregnancy. For the expected role of the various parties to raise public awareness, especially young women about the risks of pregnancy at an early age so that young women are not affected by various factors that may increase the risk of teen pregnancy and teen daughter was able to behave avoid the incidence of pregnancy early age.&lt;/em&gt;