Traditional Statistical Tests Research Articles

Improving Admission Temperature in Infants ≥34 Weeks' Gestation: A Quality Improvement Initiative.

NICU admission for hypothermia is a problem worldwide, with associated morbidity, mortality, and financial costs. Many interventions have been studied for smaller infants, but there has been little focus on infants born ≥34 weeks' gestational age (GA), though most deliveries occur at this gestation. Our primary aim was to improve the proportion of infants ≥34 weeks' GA with normal NICU admission temperature. Secondary outcomes included improvement of the proportion of normal first temperature in all infants ≥34 weeks' GA, independent of NICU admission, and improvement of predefined outcome measures. We completed a root cause analysis, using fishbone and process mapping to determine what factors were contributing to hypothermia. A series of changes were trialed using plan-do-study-act cycles to develop a standard operating procedure, covering both vaginal and cesarean section births. Outcome measures were analyzed using a P-chart as well as traditional statistical tests. We successfully increased the proportion of infants ≥34 weeks' GA with normothermia on NICU admission from 62% to 80% without increasing hyperthermia. In addition, the interventions improved the proportion of delivery room normothermia in all infants born ≥34 weeks' GA and were associated with a decreased need for intravenous therapy for hypoglycemia and the incidence of metabolic acidosis. This quality improvement initiative was successful at improving our institution's rates of normal infant temperature. The methodology used can be applied to other similar centers to improve this common problem.

The Effect of Employee Performance on Telecommunication Industry in Bangladesh

The study delves into the intricate relationship between employee training and performance outcomes within the telecommunications sector, particularly focusing on Bangladesh. It underscores the crucial role of employee training in driving innovation, maintaining competitiveness, and enhancing organizational performance in an ever-evolving industry landscape. The study defines training effectiveness as the extent to which training programs contribute to improving employee performance across various dimensions such as technical skills, customer service, sales, compliance, and innovation. Using a mixed-methods approach, including survey data and qualitative insights from industry professionals, the study aims to uncover the mechanisms through which training impacts employee performance. It emphasizes the need for robust analysis methods beyond traditional statistical tests like the F-test to capture the complexity of these relationships. Firstly, it provides an overview of the research background, significance, and methodology, setting the stage for subsequent chapters. It identifies the research problem, formulates research questions, and outlines the study's structure, emphasizing the importance of a comprehensive literature review to inform theoretical perspectives and hypothesis formulation. Moreover, the study explores the theoretical basis underlying employee training and performance, defining core concepts such as training effectiveness, performance, satisfaction, and contribution. It discusses theories like Training Value Management Theory, Innovation Theories of Training, and Core Competitiveness Theories of Training, which inform the understanding of how training influences organizational outcomes. Then, it delves into the mechanisms and hypotheses formulation for empirical testing. The research methodology, including sample selection, data collection, and model construction, is discussed, ensuring methodological rigor and ethical considerations. It also presents the data analysis and findings, incorporating both quantitative regression analysis and qualitative insights from interviews. It discusses the effectiveness of training initiatives, challenges faced, and implications for practice, along with limitations and future research directions. Lastly, it offers a comprehensive discussion of the findings, highlighting theoretical and practical implications for organizations in the telecommunications sector. It emphasizes the role of management in optimizing training programs and fostering a culture of continuous learning. Overall, the study sheds light on the critical role of training in shaping employee performance and organizational success in the telecommunications industry of Bangladesh. It offers valuable insights for practitioners, policymakers, and scholars, contributing to the understanding of training effectiveness and its implications in a rapidly evolving digital age.

Parent-adolescent discrepancies in educational expectations, relationship quality, and study engagement: a multi-informant study using response surface analysis.

Whether parental educational expectations for adolescents serve as a source of motivation or stress depends on the extent to which adolescents hold expectations for themselves. Previous research on the discrepancies between parental and adolescent educational expectations and their impact on learning engagement has been limited by traditional statistical tests, and lacking an examination of the internal mediating mechanism of parent-child relational quality from both parental and adolescent perspectives. This cross-sectional study, utilizing a multi-informant design, examined the association between discrepancies in parents' and adolescents' reports of expectations, and adolescents' study engagement, as well as the mediating role of parent-child relational qualities perceived by both parties. The sample for this study consisted of 455 adolescents and their parents from 10 classes in a junior high school in Wuhan, Hubei Province, China. The adolescents had an average age of 12.8 years, and 51.6% of them were boys. Both parents and adolescents reported on their expectations and perceived relational quality, while adolescents also filled out questionnaires assessing their learning engagement. Data were analyzed using polynomial regressions with response surface analysis. The results revealed that when adolescents reported high expectations, regardless of whether their parents reported high or low expectations, adolescents reported satisfied relationships and high learning engagement. In contrast, parents reported satisfied relationships when both parties reported high expectations, or when parents reported higher expectations than adolescents. Lastly, the association between discrepancies in expectations and learning engagement was significantly mediated by adolescent-reported relationships but not parent-reported ones. These findings highlight the importance of considering multiple perspectives when studying the association between expectations and adolescent study engagement. This research advances our comprehension of the dynamics between parent-adolescent educational expectation discrepancies and adolescent learning engagement, offering insights for more nuanced and effective parenting strategies tailored to foster optimal educational outcomes.

Hormonal intrauterine devices and heat exchange during exercise.

Synthetic progestins in oral contraceptives are thought to blunt heat dissipation by reducing skin blood flow and sweating. However, whether progestin-releasing intrauterine devices (IUDs) modulate heat loss during exercise-heat stress is unknown. We used direct calorimetry to measure whole-body total (dry + evaporative) heat loss in young, physically active women (mean (SD); aged 24(4)years, 39.3(5.3) ml/kg/min) with (IUD; n=19) and without (Control; n=17) IUDs in the follicular and luteal phases of the menstrual cycle during light- and moderate-intensity exercise at fixed rates of heat production (∼175 and ∼275W/m2 ) in 30°C, ∼21% relative humidity. Between-group and -phase differences were evaluated using traditional hypothesis testing and statistical equivalence testing within pre-determined bounds (±11W/m2 ; difference required to elicit a ±0.3°C difference in core temperature over 1h) in each exercise bout. Whole-body total heat loss was statistically equivalent between groups within ±11Wm-2 (IUD-Control [90% CIs]; Light: -2 [-8, 5] W/m2 , P=0.007; Moderate: 0 [-6, 6] W/m2 , P=0.002), as were dry and evaporative heat loss (P ≤ 0.023), except for evaporative heat loss during moderate-intensity exercise (equivalence: P=0.063, difference: P=0.647). Whole-body total and evaporative heat loss were not different between phases (P≥0.267), but dry heat loss was 3 [95% CIs: 1, 5] W/m2 greater in the luteal phase (P≤0.022). Despite this, all whole-body heat loss outcomes were equivalent between phases (P≤0.003). These findings expand our understanding of the factors that modulate heat exchange in women and provide valuable mechanistic insight of the role of endogenous and exogenous female sex hormones in thermoregulation. KEY POINTS: Progestin released by hormonal intrauterine devices (IUDs) may negatively impact heat dissipation during exercise by blunting skin blood flow and sweating. However, the influence of IUDs on thermoregulation has not previously been assessed. We used direct calorimetry to show that IUD users and non-users display statistically equivalent whole-body dry and evaporative heat loss, body heat storage and oesophageal temperature during moderate- and high-intensity exercise in a warm, dry environment, indicating that IUDs do not appear to compromise exercise thermoregulation. However, within IUD users and non-users, dry heat loss was increased and body heat storage and oesophageal temperature were reduced in the luteal compared to the follicular phase of the menstrual cycle, though these effects were small and unlikely to be practically meaningful. Together, these findings expand our understanding of the factors that modulate heat exchange in women and have important practical implications for the design of future studies of exercise thermoregulation.

Quality by Design (QbD) and Design of Experiments (DOE) as a Strategy for Tuning Lipid Nanoparticle Formulations for RNA Delivery.

The successful development of nonviral delivery systems for nucleic acids has been reported extensively over the past years. Increasingly employed to improve the delivery efficiency and therapeutic efficacy of RNA are lipid nanoparticles (LNPs). Many of the various critical formulation parameters can affect the quality attributes and effectiveness of these nano-formulations. Therefore, the systematic drug development approach (QbD) and multivariate design and statistical analysis (DOE) can be very helpful and recommended for the optimization of the composition and production of RNA-LNPs. This review addresses the concepts and applications of QbD and/or DOE for the development of lipid nanoparticles for the delivery of different types of RNA, reporting examples published in the ten recent years presenting the latest trends and regulatory requirements as well as the modern mathematical and statistical design methods. As the topic explored in this review is a novel approach, the full QbD has been described in only a few papers, and a few refer only to some aspects of QbD. In contrast, the DOE approach has been used in most of the optimization works. Different approaches and innovations in DOE have been observed. Traditional statistical tests and modeling (ANOVA, regression analysis) are slowly being replaced by artificial intelligence and machine learning methods.

Identification of monotonically expressed long non-coding RNA signatures for breast cancer using variational autoencoders.

As breast cancer is a multistage progression disease resulting from a genetic sequence of mutations, understanding the genes whose expression values increase or decrease monotonically across pathologic stages can provide insightful clues about how breast cancer initiates and advances. Utilizing variational autoencoder (VAE) networks in conjunction with traditional statistical testing, we successfully ascertain long non-coding RNAs (lncRNAs) that exhibit monotonically differential expression values in breast cancer. Subsequently, we validate that the identified lncRNAs really present monotonically changed patterns. The proposed procedure identified 248 monotonically decreasing expressed and 115 increasing expressed lncRNAs. They correspond to a total of 65 and 33 genes respectively, which possess unique known gene symbols. Some of them are associated with breast cancer, as suggested by previous studies. Furthermore, enriched pathways by the target mRNAs of these identified lncRNAs include the Wnt signaling pathway, human papillomavirus (HPV) infection, and Rap 1 signaling pathway, which have been shown to play crucial roles in the initiation and development of breast cancer. Additionally, we trained a VAE model using the entire dataset. To assess the effectiveness of the identified lncRNAs, a microarray dataset was employed as the test set. The results obtained from this evaluation were deemed satisfactory. In conclusion, further experimental validation of these lncRNAs with a large-sized study is warranted, and the proposed procedure is highly recommended.

Correlation in Dental Studies: Why It Matters and What to Do About It.

In dental research, it is particularly common for studies to collect data that are fundamentally correlated. Some common dental situations in which correlation arises include patients being observed across multiple teeth and/or across multiple time points, such as before and after treatment, or groups of patients being clustered (ie, familial units). For a number of traditional statistical tests and modeling techniques, the assumption of independence between observations is imperative in order to receive valid results and make accurate conclusions. This article describes how ignoring inherent correlations in data can lead to erroneous results when using traditional methods as well as the types of modeling techniques that are available to handle correlated data. Furthermore, two simulation studies are performed to further illustrate and prove the advantages of adequately handling correlated data in statistical analyses. Int J Oral Maxillofac Implants 2023;38:417-421. doi: 10.11607/jomi.10285.

Mortality prediction in patients with hyperglycaemic crisis using explainable machine learning: a prospective, multicentre study based on tertiary hospitals

BackgroundExperiencing a hyperglycaemic crisis is associated with a short- and long-term increased risk of mortality. We aimed to develop an explainable machine learning model for predicting 3-year mortality and providing individualized risk factor assessment of patients with hyperglycaemic crisis after admission.MethodsBased on five representative machine learning algorithms, we trained prediction models on data from patients with hyperglycaemic crisis admitted to two tertiary hospitals between 2016 and 2020. The models were internally validated by tenfold cross-validation and externally validated using previously unseen data from two other tertiary hospitals. A SHapley Additive exPlanations algorithm was used to interpret the predictions of the best performing model, and the relative importance of the features in the model was compared with the traditional statistical test results.ResultsA total of 337 patients with hyperglycaemic crisis were enrolled in the study, 3-year mortality was 13.6% (46 patients). 257 patients were used to train the models, and 80 patients were used for model validation. The Light Gradient Boosting Machine model performed best across testing cohorts (area under the ROC curve 0.89 [95% CI 0.77–0.97]). Advanced age, higher blood glucose and blood urea nitrogen were the three most important predictors for increased mortality.ConclusionThe developed explainable model can provide estimates of the mortality and visual contribution of the features to the prediction for an individual patient with hyperglycaemic crisis. Advanced age, metabolic disorders, and impaired renal and cardiac function were important factors that predicted non-survival.Trial Registration Number: ChiCTR1800015981, 2018/05/04.

Beyond Statistical Significance: A Holistic View of What Makes a Research Finding "Important"

Students often believe that statistical significance is the only determinant of whether a quantitative result is “important.” In this paper, I review traditional null hypothesis statistical testing to identify what questions inferential statistics can and cannot answer, including statistical significance, effect size and direction, causality, generalizability, and changeability of the independent variable. I illustrate these issues with examples from an empirical study of the association between how much time teenagers spent playing video games and time spent reading. I describe how study design and context determine each of those aspects of “importance,” and close by summarizing how to provide a holistic view of importance when writing about a quantitative analysis. I also include exercises to guide students through applying these concepts to articles in newspapers and scholarly journals.

Visualization According to Statisticians: An Interview Study on the Role of Visualization for Inferential Statistics.

Statisticians are not only one of the earliest professional adopters of data visualization, but also some of its most prolific users. Understanding how these professionals utilize visual representations in their analytic process may shed light on best practices for visual sensemaking. We present results from an interview study involving 18 professional statisticians (19.7 years average in the profession) on three aspects: (1) their use of visualization in their daily analytic work; (2) their mental models of inferential statistical processes; and (3) their design recommendations for how to best represent statistical inferences. Interview sessions consisted of discussing inferential statistics, eliciting participant sketches of suitable visual designs, and finally, a design intervention with our proposed visual designs. We analyzed interview transcripts using thematic analysis and open coding, deriving thematic codes on statistical mindset, analytic process, and analytic toolkit. The key findings for each aspect are as follows: (1) statisticians make extensive use of visualization during all phases of their work (and not just when reporting results); (2) their mental models of inferential methods tend to be mostly visually based; and (3) many statisticians abhor dichotomous thinking. The latter suggests that a multi-faceted visual display of inferential statistics that includes a visual indicator of analytically important effect sizes may help to balance the attributed epistemic power of traditional statistical testing with an awareness of the uncertainty of sensemaking.

User adoption and value of e-government services (Citizen-centric empirical study from Hungary)

AbstractThis paper looks at the adoption of e-government technologies from a citizen-centric, value-based point-of-view. We analyse e-government technology adoption and value creation on a large, representative Hungarian sample, using the data of the Good State Public Administration Opinion Survey. The paper examines the near total spectrum of the Hungarian government-to-citizen administration service areas: 11 e-government services, with a special focus on personal income tax administration and the use of government issued documents. The technology acceptance model and an e-government-specific adaptation of the DeLone – McLean information system (IS) success model are used as the theoretic base. Factor analysis, traditional association metrics and statistical tests are used for the analysis. Results confirm the relevance of the technology adoption factors suggested by the mainstream IS literature, while citizen-level value creation – in the form of cost or time saved, satisfaction level raised – was less demonstrable. Increasing citizens' internet trust or improving facilitating infrastructural conditions, as well as a significant value proposition in terms of time savings and ease of use would help increasing e-government service adoption levels and value creation potential.

Analysis of Injury Patterns in Men's Football between the English League and the Spanish League.

Background: Injuries in professional football lead to reduced team performance and large financial losses. The aim of this study was to analyse injuries in the two best team competitions in the world (LaLiga and Premier League), establishing similarities and differences, as well as determining injury causation patterns. Methods: A total of 277 on-field injuries requiring a substitution were analysed (142 in the Spanish league and 135 in the English league). The analysis was performed using traditional statistical tests (frequency analysis, chi-square test) with SPSS 25 and a T-Patterns sequence analysis with THEME 5.0. Results: In the Spanish league, there were a similar number of injuries in the first part of the season as in the second part of the season, while in the English league, they are more frequent in the first part of the season. In the Spanish league, injuries are more frequent in the first half of the match, while in the English league, they are more frequent in the second half. The type of player most frequently injured was the defender. Most of the injuries occurred without the presence of an opponent. The accumulated minutes during the season affect injuries of the muscular type. Conclusions: The most common type of injury in both leagues was a strain, followed by a sprain and a contusion. Although common injury patterns can be established between the two leagues, there are notable differences. One of the factors is determined by the English league calendar (many matches at Christmas). In Spain, there were more muscular injuries that were not caused by the opponent, while in England there were more tackling injuries. Age is a risk factor in the Spanish league. In this league, there is a moderately significant relationship between the number of injuries and the points won.

Comparison of the Sensitivity of Sterility Tests Based on the Analysis of 2x2 Contingency Tables

The work is devoted to statistical methods for determining the comparative sensitivity of sterility analyzers with a binary scale: 1/0, +/-. The paper considers both traditional statistical test methods namely the direct binomial test, the McNemar test, and the new, highly sensitive Yefimov Method, and Pearson-Yefimov statistical test. First, the comparative effectiveness of statistical methods for assessing the sensitivity of testers is demonstrated on a specially developed model system (the Test Bench), and then on real experimental data. The data were obtained during the validation of a new tester and a new method for analyzing samples for sterility. The new method is compared with the old one, the certified method. The sensitivity of both new and old methods (testers) was already determined earlier by a direct non-statistical method, which allowed us to compare the results of the two approaches. The factors influencing the efficiency of statistical tests are revealed and described.

A Surgeon's Guide to Machine Learning.

Machine learning (ML) represents a collection of advanced data modeling techniques beyond the traditional statistical models and tests with which most clinicians are familiar. While a subset of artificial intelligence, ML is far from the science fiction impression frequently associated with AI. At its most basic, ML is about pattern finding, sometimes with complex algorithms. The advanced mathematical modeling of ML is seeing expanding use throughout healthcare and increasingly in the day-to-day practice of surgeons. As with any new technique or technology, a basic understanding of principles, applications, and limitations are essential for appropriate implementation. This primer is intended to provide the surgical reader an accelerated introduction to applied ML and considerations in potential research applications or the review of publications, including ML techniques.

Dealing with trend uncertainty in empirical estimates of European rainfall climate for insurance risk management

AbstractThe insurance industry uses mathematical models to estimate the risks due to future natural catastrophes. For climate‐related risks, historical climate data are a key ingredient used in making the models. Historical data for temperature and sea level often show clear and readily quantified climate change driven trends, and these trends would typically be accounted for when building risk models by adjusting earlier values to render the earlier data relevant to the future climate. For other climate variables, such as rainfall in many parts of the world, the questions of whether there are climate change driven trends in the historical data, and how to quantify them if there are, are less simple to answer. We investigate these questions in the context of European rainfall with a specific focus on how to deal with the uncertainty around trend estimates. We compare 10 empirical methodologies that one might use to model and predict trends, including traditional statistical testing and alternatives to statistical testing based on standard methods from model selection and model averaging. We emphasize prediction and risk assessment, rather than detection of trends, as our goal. Viewed in terms of this goal, the methods we consider each have qualitative and quantitative advantages and disadvantages. Understanding these advantages and disadvantages can help risk modellers make a choice as to which method to use, and based on the results we present, we believe that in many common situations model averaging methods, as opposed to statistical testing or model selection, are the most appropriate.

Comparison study of differential abundance testing methods using two large Parkinson disease gut microbiome datasets derived from 16S amplicon sequencing

BackgroundTesting for differential abundance of microbes in disease is a common practice in microbiome studies. Numerous differential abundance (DA) testing methods exist and range from traditional statistical tests to methods designed for microbiome data. Comparison studies of DA testing methods have been performed, but none performed on microbiome datasets collected for the study of real, complex disease. Due to this, DA testing was performed here using various DA methods in two large, uniformly collected gut microbiome datasets on Parkinson disease (PD), and their results compared.ResultsOverall, 78–92% of taxa tested were detected as differentially abundant by at least one method, while 5–22% were called differentially abundant by the majority of methods (depending on dataset and filtering of taxonomic data prior to testing). Concordances between method results ranged from 1 to 100%. Average concordance for datasets 1 and 2 were 24% and 28% respectively, and 27% for replicated DA signatures. Concordances increased when removing rarer taxa before testing, increasing average concordances by 2–32%. Certain methods consistently resulted in higher concordances (e.g. ANCOM-BC, LEfSe), while others consistently resulted in lower (e.g. edgeR, fitZIG). Hierarchical clustering revealed three groups of DA signatures that were (1) replicated by the majority of methods on average and included taxa previously associated with PD, (2) replicated by a subset of methods and included taxa largely enriched in PD, and (3) replicated by few to one method(s).ConclusionsDifferential abundance tests yielded varied concordances, and amounts of detected DA signatures. Some methods were more concordant than others on both filtered and unfiltered data, therefore, if consistency with other study methodology is a key goal, one might choose among these methods. Even still, using one method on one dataset may find true associations, but may also detect false positives. To help lower false positives, one might analyze data with two or more DA methods to gauge concordance, and use a built-in replication dataset. This study will hopefully serve to complement previously reported DA method comparison studies by implementing and coalescing a large number of both previously and yet to be compared methods on two real gut microbiome datasets.

Finding associations in a heterogeneous setting: statistical test for aberration enrichment

Most two-group statistical tests find broad patterns such as overall shifts in mean, median, or variance. These tests may not have enough power to detect effects in a small subset of samples, e.g., a drug that works well only on a few patients. We developed a novel statistical test targeting such effects relevant for clinical trials, biomarker discovery, feature selection, etc. We focused on finding meaningful associations in complex genetic diseases in gene expression, miRNA expression, and DNA methylation. Our test outperforms traditional statistical tests in simulated and experimental data and detects potentially disease-relevant genes with heterogeneous effects.

With topological data analysis, predicting stock market crashes

We are investigating the evolution of four big US stock market indexes' regular returns after the 2000 technology crash and the 2007-2009 financial crisis. Our approach is based on topological data processing (TDA). To identify and measure topological phenomena occurring in multidimensional time series, we use persistence homology. We obtain time-dependent point cloud data sets using a sliding window, which we connect a topological space for. Our research indicates that a new method of econometric analysis is offered by TDA, which complements the traditional statistical tests. The tool may be used to predict early warning signs of market declines that are inevitable.

HypoML: Visual Analysis for Hypothesis-based Evaluation of Machine Learning Models.

In this paper, we present a visual analytics tool for enabling hypothesis-based evaluation of machine learning (ML) models. We describe a novel ML-testing framework that combines the traditional statistical hypothesis testing (commonly used in empirical research) with logical reasoning about the conclusions of multiple hypotheses. The framework defines a controlled configuration for testing a number of hypotheses as to whether and how some extra information about a "concept" or "feature" may benefit or hinder an ML model. Because reasoning multiple hypotheses is not always straightforward, we provide HypoML as a visual analysis tool, with which, the multi-thread testing results are first transformed to analytical results using statistical and logical inferences, and then to a visual representation for rapid observation of the conclusions and the logical flow between the testing results and hypotheses. We have applied HypoML to a number of hypothesized concepts, demonstrating the intuitive and explainable nature of the visual analysis.

Macroeconomics and the value premium

The value premium as measured by the “high minus low” returns from the Fama and French database has been negative or statistically indistinguishable from zero for the past decade. The value premium is highly variable and non-normally distributed, making traditional statistical tests like t tests or regressions difficult to implement or misleading. The authors hypothesize that the Great Moderation—a period of generally lower variability in economic growth and inflation—has not been great for the value premium. To demonstrate this, the authors use a variety of nonparametric techniques and panel analysis to show that the value premium was generally challenged by the Great Moderation. This is consistent with the idea that as a risk premium, if risk is lower, the premium should also be lower.