Calculation Of Standard Errors Research Articles

A mixture distribution for modelling bivariate ordinal data

Ordinal responses often arise from surveys which require respondents to rate items on a Likert scale. Since most surveys contain more than one question, the data collected are multivariate in nature, and the associations between different survey items are usually of considerable interest. In this paper, we focus on a mixture distribution, called the combination of uniform and binomial (CUB), under which each response is assumed to originate from either the respondent’s uncertainty or the actual feeling towards the survey item. We extend the CUB model to the bivariate case for modelling two correlated ordinal data without using copula-based approaches. The proposed model allows the associations between the unobserved uncertainty and feeling components of the variables to be estimated, a distinctive feature compared to previous attempts. This article describes the underlying logic and deals with both theoretical and practical aspects of the proposed model. In particular, we will show that the model is identifiable under a wide range of conditions. Practical inferential aspects such as parameter estimation, standard error calculations and hypothesis tests will be discussed through simulations and a real case study.

Quantile Regression for Longitudinal Functional Data with Application to Feed Intake of Lactating Sows

AbstractThis article focuses on the study of lactating sows, where the main interest is the influence of temperature, measured throughout the day, on the lower quantiles of the daily feed intake. We outline a model framework and estimation methodology for quantile regression in scenarios with longitudinal data and functional covariates. The quantile regression model uses a time-varying regression coefficient function to quantify the association between covariates and the quantile level of interest, and it includes subject-specific intercepts to incorporate within-subject dependence. Estimation relies on spline representations of the unknown coefficient functions and can be carried out with existing software. We introduce bootstrap procedures for bias adjustment and computation of standard errors. Analysis of the lactation data indicates, among others, that the influence of temperature increases during the lactation period.Supplementary materials accompanying this paper appear on-line.

Meta-analysis of Physicochemical Characteristics in Beekeeping, Wild, and Stingless Bee Honey

The two most extensively beekeeping honeybee species were Apis mellifera and A. cerana. Other species that produced honey, albeit with infrequent cultivation, include A. dorsata, A. florea, and some stingless bees, including Trigona and Melipona. Different types of honeybees were known to affect the quality of honey. Hence, this investigation aimed to conduct a meta-analysis to examine the similarities in honey quality between honey sourced from beekeeping and wild honeybees. Data analysis was performed using the OpenMEE software, facilitating the calculation of effect size and standard error. The effect size and common error data were subsequently organized into separate columns within a CSV file. This CSV file was then imported into the JASP 0.16.2 software to conduct heterogeneity and Egger tests to detect potential publication bias. The findings indicated significant disparities in the quality of honey produced by beekeeping, wild, and stingless bee honey, as determined by various parameters, including pH value, moisture, total sugar, acidity, HMF (hydroxymethylfurfural), and diastase enzyme levels (p < 0.05). According to the results of the meta-analysis, honey from beekeeping exhibited superior quality to that of wild and stingless bees. However, the average values of all parameters still adhered to the established honey quality standards set by the Standar Nasional Indonesia and the International Honey Standard (IHS).
 
 Keywords: honey, honeybees, meta-analysis, physicochemical quality

Participation in organised sport to improve and prevent adverse developmental trajectories of at-risk youth: A systematic review.

Healthy after-school activities such as participation in organised sport have been shown to serve as important resources for reducing school failure and other problem/high-risk behaviour. It remains to be established to what extent organised sport participation has positive impacts on young people in unstable life circumstances. What are the effects of organised sport on risk behaviour, personal, emotional and social skills of young people, who either have experienced or are at-risk of experiencing an adverse outcome? The database searches were carried out in March 2023 and other sources were searched in May 2023. We searched to identify both published and unpublished literature. The intervention was participation in leisure time organised sport. Young people between 6 and 18 years of age, who either have experienced or are at-risk of experiencing an adverse outcome were eligible. Primary outcomes were problem/high-risk behaviour and a secondary outcomes social and emotional outcomes. Studies that used a control group were eligible for. Studies that utilised qualitative approaches were not. The number of potentially relevant studies was 43,716. Thirteen studies met the inclusion criteria. Only seven studies could be used in the data synthesis. Five studies were judged to have a critical risk of bias and were excluded from the meta-analysis. One study did not report data that enabled the calculation of effect sizes and standard errors. Meta-analyses were conducted on each conceptual outcome separately. All analyses were inverse variance weighted using random effects statistical models. Two studies were from Canada, one from Australia, and the remaining from the USA. The timespan of the interventions was 23 years, from 1995 to 2018. The median number of participants analysed was 316, and the median number of controls was 452. A number of primary outcomes were reported but each in a single study only. Concerning secondary outcomes, two studies reported the effect on overall psychosocial adjustment at post-intervention. The standardised mean difference was 0.70 (95% CI 0.28-1.11). There was a small amount of heterogeneity. Three studies reported on depressive symptoms at 0-3 years follow-up. The standardised mean difference was 0.02 (95% CI -0.01 to 0.06). There was no heterogeneity between the three studies. In addition, a number of other secondary outcomes were reported each in a single study only. There were too few studies included in the meta-analyses in order for us to draw any conclusion. The dominance of Northern America clearly limiting the generalisability of the findings. The majority of the studies were not considered to be of overall high quality and the process of excluding studies with critical risk of bias from the meta-analysis applied in this review left us with only 7 of a total of 13 possible studies to synthesise. Further, because too few studies reported results on the same type of outcome, at most three studies could be combined in a particular meta-analysis and no meta-analysis could be performed on any of the primary outcomes.

Reproducibility of peak moment for isometric and isokinetic knee extension exercise

BackgroundAccurate measurements of muscular performance are important for diagnostics, for example during rehabilitation after traumatic injuries but also in competitive sports. For these purposes, dynamometric devices are widely used and considered the gold standard for muscle strength testing. However, few previous studies have tested the reproducibility of peak moment (PM) at velocities close to the maximum device capability, and in general, reproducibility results cannot be transferred to other devices or test protocols. The purpose of this study was to evaluate the reproducibility of PM for different isometric and isokinetic knee extension exercises using the IsoMed 2000.MethodsThirty subjects volunteered in three repeated test sessions, including isometric knee extension (100° and 140° knee angle) and isokinetic knee extension (30°/s and 400°/s). Statistical analysis for comparison of sessions two and three included paired sample t-test, calculation of intraclass correlation coefficient (ICC) and standard error of measurement (SEM). Additionally, Bland Altman statistics and corresponding plots were created.ResultsA significant difference between sessions in PM was found for isometric knee extension in one leg (140° left). Reproducibility was high for all conditions with ICC ranging from 0.964 to 0.988 and SEM in the range of 7.6 to 10.5 Nm. Bland Altman statistics revealed a bias between − 7.3 and 0.7 Nm.ConclusionsReproducibility of PM using the IsoMed 2000 was good after an initial familiarization trial with high values of relative reproducibility. Absolute reproducibility can be interpreted as appropriate for most common practical applications.

ESG and dividend distribution decisions: evidence of moderation by shareholder activism

Purpose Environment, social and governance (ESG) practices and shareholder activism are making significant strides in the decision-making policies and processes for all firms. This study aims to assess the impact of ESG on the dividend payout decisions of firms in India. In addition, it also aims to determine how shareholder activism influences the impact of ESG on dividend distribution decisions. Design/methodology/approach The authors gather relevant data from 78 non-financial listed Indian firms from 2016 to 2020. This study undertakes longitudinal data analysis, with fixed effects and calculation of robust standard errors. In addition, the slope test is used to examine the effects of the interaction between ESG and shareholder activism. Findings It is found in the study that not only does ESG positively impact the dividends but also shareholder activism positively impacts the dividend distribution decisions. Surprisingly, the authors see a significant but negative interaction impact of shareholder activism on the positive association of ESG with dividend distribution decisions. In other words, ESG impacts dividend distribution decisions differently at levels of shareholder activism. When shareholder activism is low, ESG positively influences dividend distribution decisions. However, when shareholder activism is high, ESG negatively influences dividend distribution decisions. Practical implications This result has significant implications for all the stakeholders, including shareholders. A shareholder expecting a dividend could decide correctly through the current study’s findings. In cases of high shareholder activism, investors may skip picking a stock if investors expect high ESG to influence the dividend distribution decisions favourably. On the contrary, investors may choose a stock if shareholder activism is low and all else remains the same. Originality/value Literature has some evidence of the influence of shareholder activism and ESG (in silos) on the dividend distribution decisions in the firms. This study attempts to contribute by bringing forth the interaction effects of shareholder activism and ESG on dividend distribution decisions.

COVID-19 incidence in women of reproductive age: a population-based study in Reggio Emilia, northern Italy

BackgroundDespite being at higher risk of severe disease and pregnancy complications, evidence on susceptibility to SARS-CoV-2 infection in pregnancy is still limited. The aim of the study is to compare the likelihood of undergoing a SARS-CoV-2 test and testing positive for COVID-19 in pregnancy and puerperium with that of the general female population of reproductive age.MethodsThis is a retrospective population-based cohort study including 117,606 women of reproductive age (March 2020-September 2021) with 6608 (5.6%) women having ≥ 1 pregnancy.Women were linked to the pregnancy registry to be classified as “non-pregnant”, “pregnant”, and “puerperium”; then, according to the national case-based integrated COVID-19 surveillance system, all women undergoing a SARS-CoV-2 test during the study period were identified. The Incidence Rate Ratio was calculated to compare the likelihood of being tested for SARS-CoV-2 in pregnant, puerperium and non-pregnant women among all women included. The likelihood of having a COVID-19 diagnosis was calculated using two comparators (not-pregnant women and the person-time before/after pregnancy) by means of Cox proportional hazards models, adjusting for age and with the cluster option to control standard error calculation in repeated pregnancies. Only first infection and swabs before the first one positive were included.ResultsThe probability of being tested for SARS-CoV-2 was 4.9 (95% CI: 4.8–5.1) and 3.6 times higher (95%CI: 3.4–3.9) in pregnancy (including spontaneous miscarriages) and in the puerperium, respectively.The Hazard Ratio (HR) of covid-19 diagnosis during pregnancy vs. non-pregnancy was 1.17 (95% CI 1.03–1.33) with similar results when comparing the risk during pregnancy with that of the same women outside pregnancy (puerperium excluded), with an HR of 1.13 (95% CI 0.96–1.33); the excess decreased when excluding the test performed at admission for delivery (HR 1.08 (95%CI 0.90–1.30). In the puerperium, the HR was 0.62 (95% CI 0.41–0.92) comparing women with ≥ 1childbirth with all other women and excluding the first two weeks of puerperium.ConclusionsWomen during pregnancy showed a small increase in the risk of infection, compatible with a higher likelihood of being tested. A lower probability of infection during the puerperium was observed during the entire pandemic period, suggesting likely protective behaviors which were effective in reducing their probability of infection.

Determine the Occurrence of Latent Myofascial Trigger Points Through Binary Logistic Regression Model in Sportsmen

This study aimed to ascertain the prevalence of latent myofascial trigger points (L-MTrPs) among male athletes. Ninety participants were randomly included in the study. The presence of L-MTrPs was identified based on Simon’s criteria, leading to the categorization of participants into two groups: L-MTrPs (n=45) and non-myofascial trigger points (N-MTrPs) (n=45). Pain pressure thresholds (PPT) of muscles were quantified using a pressure algometer, while force production was measured with HUMAC NORM isokinetic. The range of motion (ROM) for knee flexors and extensors was determined using Kinovea software. These parameters were treated as independent variables, whereas L-MTrPs and N-TrPs were considered dependent variables. Binary Logistic Regression (Enter Method) was employed for data analysis. Model fitness was evaluated through standard error computation, Wald’s χ2, and odds ratios with a 95% confidence interval. Model adequacy was assessed via the likelihood ratio, Cox & Snell (R2), Nagelkerke (R2), and Hosmer and Lemeshow tests. The results effectively predicted L-MTrPs and N-MTrPs, revealing that force production, PPT, and ROM were the most significant predictive variables for L-MTrPs. In conclusion, reduced force production, lower PPT, and restricted ROM were indicative of L-MTrPs. Consequently, regular evaluation of muscular strength, PPT, and ROM are recommended for athletes to prevent the development of L-MTrPs.

The Bayesian simulation study (BASIS) framework for simulation studies in statistical and methodological research.

Statistical simulation studies are becoming increasingly popular to demonstrate the performance or superiority of new computational procedures and algorithms. Despite this status quo, previous surveys of the literature have shown that the reporting of statistical simulation studies often lacks relevant information and structure. The latter applies in particular to Bayesian simulation studies, and in this paper the Bayesian simulation study framework (BASIS) is presented as a step towards improving the situation. The BASIS framework provides a structured skeleton for planning, coding, executing, analyzing, and reporting Bayesian simulation studies in biometrical research and computational statistics. It encompasses various features of previous proposals and recommendations in the methodological literature and aims to promote neutral comparison studies in statistical research. Computational aspects covered in the BASIS include algorithmic choices, Markov-chain-Monte-Carlo convergence diagnostics, sensitivity analyses, and Monte Carlo standard error calculations for Bayesian simulation studies. Although the BASIS framework focuses primarily on methodological research, it also provides useful guidance for researchers who rely on the results of Bayesian simulation studies or analyses, as current state-of-the-art guidelines for Bayesian analyses are incorporated into theBASIS.

Flexible modeling of multiple nonlinear longitudinal trajectories with censored and non-ignorable missing outcomes.

Multivariate nonlinear mixed-effects models (MNLMMs) have become a promising tool for analyzing multi-outcome longitudinal data following nonlinear trajectory patterns. However, such a classical analysis can be challenging due to censorship induced by detection limits of the quantification assay or non-response occurring when participants missed scheduled visits intermittently or discontinued participation. This article proposes an extension of the MNLMM approach, called the MNLMM-CM, by taking the censored and non-ignorable missing responses into account simultaneously. The non-ignorable missingness is described by the selection-modeling factorization to tackle the missing not at random mechanism. A Monte Carlo expectation conditional maximization algorithm coupled with the first-order Taylor approximation is developed for parameter estimation. The techniques for the calculation of standard errors of fixed effects, estimation of unobservable random effects, imputation of censored and missing responses and prediction of future values are also provided. The proposed methodology is motivated and illustrated by the analysis of a clinical HIV/AIDS dataset with censored RNA viral loads and the presence of missing CD4 and CD8 cell counts. The superiority of our method on the provision of more adequate estimation is validated by a simulation study.

A Markov random field model with cumulative logistic functions for spatially dependent ordinal data

This paper presents a class of regression models with cumulative logistic functions that are chiefly designed to analyse spatially dependent ordinal data. In contrast to previous works, the proposed model requires neither the sites to be regularly spaced nor the assumption of an underlying continuous variable. It belongs to a more general class of Markov random field models, and can be considered an extension of the ordinal regression model with the proportional odds link function. Our proposed model allows practitioners to interpret the model parameters using odds ratios. Apart from the theoretical developments, this work also highlights the practical aspects of model fitting, including parameterisation, selection of neighbourhood, and calculation of standard errors. Simulation studies with regularly and irregularly spaced sites were conducted. Modelling strategies including pseudo-likelihood methods were found to be useful in both settings. The proposed model and the non-spatial counterpart were applied to the daily air quality index measured in the United Kingdom. The results indicate the presence of spatial effects and the incorporation of spatial effects led to better model performance in terms of various goodness-of-fit measures.

The mid-Holocene sea-level change in the Arabian Gulf

The mid-Holocene sea-level highstand is a well-known phenomenon in sea-level science, yet the knowledge on the highstand’s spatial and temporal distribution remains incomplete. Here we study the southwest coast of the Arabian-Persian Gulf where a mid-Holocene sea-level highstand and subsequent sea-level fall may have occurred due to the Earth crustal response to meltwater load. Sea-level indicators were established using standard facies analysis and error calculations, then constrained through glacio-isostatic adjustment (GIA) modelling and though procedures based on Gaussian Process and exponential decay analysis. This work allowed to identify the highstand at 1.6 ± 0.4 m occurring 6.7–6.0 ka, in excellent agreement with GIA model results. The subsequent shoreline migration followed the geophysical constraint by prograding in line with the sea-level fall until around 3 ka. Then, the strength of the external control weakened and internal processes, in particular sediment binding through microbial activity, started controlling the geometry of the accommodation space.

Finite-sample performance of the robust variance estimator in the presence of missing data

Theoretically, the maximum likelihood estimator has the sandwich-type asymptotic variance-covariance matrix under model misspecification. Its empirical estimator, that is called the robust variance estimator, is consistent. Thus, the estimator is asymptotically valid even under model misspecification. In practice, the robust variance estimator is used for computation of standard errors in longitudinal data analysis. Recently, Golden et al. (2019 Econometrics, 7, 1-27) showed that the maximum likelihood estimator retains a sandwich-type asymptotic variance-covariance matrix in the presence of missing data even when the missing-data mechanism is missing not at random. Although they revealed the asymptotic validity of the robust variance estimator in the simultaneous presence of both model misspecification and missing data, its finite-sample performance did not be investigated. In this article, we evaluated the finite-sample performance via simulation studies and clarify its small-sample problems. In addition, we illustrated the robust variance estimator using longitudinal CD4 count data from a randomized double-blind study.

Эффективность применения природной минеральной воды «Увинская» в комплексной эрадикации Helicobacter pylori у больных хроническим гастритом на фоне сахарного диабета 2 типа

The purpose — to assess the effectiveness of therapeutic use of «Uvinskaya» natural mineral water for Helicobacter Pylori eradication in patients with chronic gastritis and type 2 diabetes mellitus. Material and methods. The study included 120 patients with type 2 diabetes mellitus, who were diagnosed with chronic gastritis associated with Helicobacter Pylori. The average age of patients was 61.6 ± 4.13 years, the duration of diabetes — from 2 to 15 years (on average – 9.9 ± 2.5 years), gender composition — 21 men and 99 women. The patients were divided into two groups by simple randomization. In the first group (n = 60), the first line eradication therapy was carried out according to Maastricht V recommendations, in the second group (n = 60) underwent eradication therapy combined with «Uvinskaya» mineral water. The diagnosis of «chronic gastritis» was histologically verified in all patients. The eradication control was carried out 4 weeks after therapy. Changes in the severity of gastroenterological symptoms in patients were determined by the Gastrointestinal Symptom Rating Scale questionnaire. Statistical processing was carried out with the calculation of average values and standard errors of the average. The differences were considered reliable at p ≤ 0.05. Results. The study shows the positive effect of the «Uvinskaya» mineral water inclusion into the Helicobacter Pylori eradication treatment complex in patients with type 2 diabetes mellitus: the condition of patients improved, inflammatory changes in the stomach mucosa were reduced and the efficiency of eradication therapy increased. Conclusion. The amplification of the positive therapeutic effects of drug treatment and drinking balneotherapy provides a significant increase in the efficiency of eradication, improving the clinical course and forecast of chronic gastritis associated with Helicobacter Pylori in patients with type 2 diabetes mellitus. It enables to recommend the intake of «Uvinskaya» mineral water as part of eradication therapy in patients with type 2 diabetes mellitus.

Effect of Aloe vera (Barbadensis Mill) on Letrozole Induced Polycystic Ovarian Syndrome in Swiss Albino Mice.

ABSTRACTBackground:Polycystic ovarian syndrome (PCOS) is the predominant endocrine abnormality in premenopausal women characterised by hyperandrogenism, ovulatory dysfunction, metabolic disturbances such as, insulin resistance and obesity.Aim:The key objective of this study was to evaluate the efficacy of Aloe vera as a treatment drug to identify the factors among female Swiss albino mice (PCOS) models displaying ovarian and metabolic abnormalities and serum sex steroids that are associated with insulin sensitivity.Study Setting and Design:The laboratory breed 4 months adult virgin female Swiss albino mice showing regular estrous cycle and weighing about 20–30 g were employed in present study under CPCSEA guidelines with ethical clearance from the institutional ethical committee. These mice were categorized into two groups with 10 mice in each as control and test animals.Materials and Methods:The mice which were categorised into two groups with 10 mice in each. Carboxy methyl cellulose control and letrozole were induced for 21 days followed by 30 days treatment with distilled water for control, Diane as standard drug and Aloe vera gel formulation. Body weight, estrous cycle, biochemical estimations, hormone assay and histopathology were the parameters studied.Statistical Analysis:IBM SPSS version 20.0 was used for the mean and standard error calculations of biochemical and antioxidants estimation.Results:During the induction period, we observed a significant increase in the body weight, decrease in the number of estrous cycle, and the concomitant increase in diestrous. The biochemical estimations showed changes in the regular activity; hormonal imbalance was observed in hormone assay and cyst formation was also observed in the histopathological study. After the treatment of Aloe vera and Diane, all the parameters showed considerable good results.Conclusion:The findings of the current study provide a baseline data for designing further investigations on the therapeutic benefits of Aloe vera as an adjunct therapy in the management of PCOS.

Estimation in multivariate linear mixed models for longitudinal data with multiple outputs: Application to PBCseq data analysis

In many biomedical studies or clinical trials, we have data with more than one response variable on the same subject repeatedly measured over time. In analyzing such data, we adopt a multivariate linear mixed-effects longitudinal model. On the other hand, in longitudinal data, we often find features that do not impact modeling the response variable and are eliminated from the study. In this paper, we consider the problem of simultaneous variable selection and estimation in a multivariate t linear mixed-effects model (MtLMM) for analyzing longitudinally measured multioutcome data. This work's motivation comes from a cohort study of patients with primary biliary cirrhosis. The interest is eliminating insignificant variables using the smoothly clipped and absolute deviation penalty function in the MtLMM. The proposed penalized model offers robustness and flexibility to accommodate fat tails. An expectation conditional maximization algorithm is employed for the computation of maximum likelihood estimates of parameters. The calculation of standard errors is affected by an information-based method. The methodology is illustrated by analyzing Mayo Clinic Primary Biliary Cirrhosis sequential (PBCseq) data and a simulation study. We found drugs and sex can be eliminated from the PBCseq analysis, and over time the diseaseprogresses.

Large-Scale Ideal Point Estimation

AbstractRecent advances in the study of voting behavior and the study of legislatures have relied on ideal point estimation for measuring the preferences of political actors, and increasingly, these applications have involved very large data matrices. This has proved challenging for the widely available approaches. Limitations of existing methods include excessive computation time and excessive memory requirements on large datasets, the inability to efficiently deal with sparse data matrices, inefficient computation of standard errors, and ineffective methods for generating starting values. I develop an approach for estimating multidimensional ideal points in large-scale applications, which overcomes these limitations. I demonstrate my approach by applying it to a number of challenging problems. The methods I develop are implemented in an r package (ipe).

Clustered into control: Heterogeneous causal impacts of water infrastructure failure

AbstractWe estimate economic impacts from decaying water infrastructure in the United States. Using water main breaks in Washington, DC, and a yearlong panel of hourly traffic speeds, we estimate causal effects of water main failures on traffic congestion. We use k‐means clustering to create clusters of streets that are similar to each other: treated observations are compared to other units in the cluster. We identify heterogeneous treatment effects algorithmically while retaining straightforward standard error calculations. We find strong evidence of heterogeneous treatment effects across clusters but small welfare impacts of water main breaks on traffic patterns overall.

Cross cultural adaptation and psychometric properties of the Bengali version of the Scale of Oral Health Outcomes for 5-year-old children (SOHO-5)

BackgroundThe oral health related quality of life (OHRQoL) of children in Bangladesh has not yet been measured, as there is no validated OHRQoL measure for that population. The aim of this study was to cross-culturally adapt the child self-report and parental proxy report versions of the Scale of Oral Health Outcomes for 5-year-old children (SOHO-5) into Bengali and test their psychometric properties: face validity, construct validity (convergent and discriminant validity) and reliability (internal consistency and test–retest reliability), among 5–9-year-old children and their parents in Bangladesh and assess associations between dental caries/sepsis and OHRQoL in this population.MethodsThe forward–backward translated Bengali SOHO-5 was piloted among 272 children and their parents to test its face validity. The questionnaire was administered to 788 children and their parents to evaluate its psychometric properties. Internal consistency of Bengali SOHO-5 was assessed using Cronbach’s alpha, and test–retest reliability was assessed using Kappa. Convergent and discriminant validity were assessed through nonparametric tests. The calculation of effect sizes and standard error of measurement facilitated the assessment of minimally important difference (MID) for SOHO-5. The associations of reporting an oral impact with caries and sepsis were assessed via logistic regression models.ResultsBoth child self-report and parental proxy report questionnaires showed good face validity. Cronbach’s alpha scores were 0.79 and 0.87 for child and parental questionnaire, respectively. A weighted Kappa score of 0.85 demonstrated test–retest reliability of child questionnaire. SOHO-5 scores were significantly associated with subjective oral health outcomes and discriminated clearly between different caries severity and sepsis groups. These differences were considerably higher than the MID. After adjusting for child’s age, sex, setting, maternal education and family income, the odds of reporting an oral impact were 2.25 (95% CI 1.98–2.56) and 4.44 (95% CI 3.14–6.28) times higher for each additional tooth with caries and sepsis, respectively.ConclusionThis study provided strong evidence supporting the validity and reliability of both versions of Bengali SOHO-5 as OHRQoL measures. Dental caries and sepsis were associated with poor OHRQoL in this population. The Bengali SOHO-5 is expected to be a useful outcome measure for research and clinical purposes in Bengali speaking child populations.

Robust Post-Matching Inference

Nearest-neighbor matching is a popular nonparametric tool to create balance between treatment and control groups in observational studies. As a preprocessing step before regression, matching reduces the dependence on parametric modeling assumptions. In current empirical practice, however, the matching step is often ignored in the calculation of standard errors and confidence intervals. In this article, we show that ignoring the matching step results in asymptotically valid standard errors if matching is done without replacement and the regression model is correctly specified relative to the population regression function of the outcome variable on the treatment variable and all the covariates used for matching. However, standard errors that ignore the matching step are not valid if matching is conducted with replacement or, more crucially, if the second step regression model is misspecified in the sense indicated above. Moreover, correct specification of the regression model is not required for consistent estimation of treatment effects with matched data. We show that two easily implementable alternatives produce approximations to the distribution of the post-matching estimator that are robust to misspecification. A simulation study and an empirical example demonstrate the empirical relevance of our results. Supplementary materials for this article are available online.