Drawing Inferences Research Articles

When Composite Likelihood meets Stochastic Approximation

A composite likelihood is an inference function derived by multiplying a set of likelihood components. This approach provides a flexible framework for drawing inferences when the likelihood function of a statistical model is computationally intractable. While composite likelihood has computational advantages, it can still be demanding when dealing with numerous likelihood components and a large sample size. This paper tackles this challenge by employing an approximation of the conventional composite likelihood estimator based on a stochastic optimization procedure. This novel estimator is shown to be asymptotically normally distributed around the true parameter. In particular, based on the relative divergent rate of the sample size and the number of iterations of the optimization, the variance of the limiting distribution is shown to compound for two sources of uncertainty: the sampling variability of the data and the optimization noise, with the latter depending on the sampling distribution used to construct the stochastic gradients. The advantages of the proposed framework are illustrated through simulation studies on two working examples: an Ising model for binary data and a gamma frailty model for count data. Finally, a real-data application is presented, showing its effectiveness in a large-scale mental health survey.

Manner implicatures in large language models

In human speakers’ daily conversations, what we do not say matters. We not only compute the literal semantics but also go beyond and draw inferences from what we could have said but chose not to. How well is this pragmatic reasoning process represented in pre-trained large language models (LLM)? In this study, we attempt to address this question through the lens of manner implicature, a pragmatic inference triggered by a violation of the Grice manner maxim. Manner implicature is a central member of the class of context-sensitive phenomena. The current work investigates to what extent pre-trained LLMs are able to identify and tease apart different shades of meaning in manner implicature. We constructed three metrics to explain LLMs’ behavior, including LLMs-surprisals, embedding vectors’ similarities, and natural language prompting. Results showed no striking evidence that LLMs have explainable representations of meaning. First, the LLMs-surprisal findings suggest that some LLMs showed above chance accuracy in capturing different dimensions of meaning, and they were able to differentiate neutral relations from entailment or implications, but they did not show consistent and robust sensitivities to more nuanced comparisons, such as entailment versus implications and equivalence versus entailment. Second, the similarity findings suggest that the perceived advantage of contextual over static embeddings was minimal, and contextual LLMs did not notably outperform static GloVe embeddings. LLMs and GloVe showed no significant difference, though distinctions between entailment and implication were slightly more observable in LLMs. Third, the prompting findings suggest no further supportive evidence indicating LLM’s competence in fully representing different shades of meaning. Overall, our study suggests that current dominant pre-training paradigms do not seem to lead to significant competence in manner implicature within our models. Our investigation sheds light on the design of datasets and benchmark metrics driven by formal and distributional linguistic theories.

Disentangling decision errors from action execution in mouse-tracking studies: The case of effect-based action control.

Mouse-tracking is regarded as a powerful technique to investigate latent cognitive and emotional states. However, drawing inferences from this manifold data source carries the risk of several pitfalls, especially when using aggregated data rather than single-trial trajectories. Researchers might reach wrong conclusions because averages lump together two distinct contributions that speak towards fundamentally different mechanisms underlying between-condition differences: influences from online-processing during action execution and influences from incomplete decision processes. Here, we propose a simple method to assess these factors, thus allowing us to probe whether process-pure interpretations are appropriate. By applying this method to data from 12 published experiments on ideomotor action control, we show that the interpretation of previous results changes when dissociating online processing from decision and initiation errors. Researchers using mouse-tracking to investigate cognition and emotion are therefore well advised to conduct detailed trial-by-trial analyses, particularly when they test for direct leakage of ongoing processing into movement trajectories.

Epigenetics of Hypertensive Nephropathy.

Hypertensive nephropathy (HN) is a leading cause of chronic kidney disease (CKD) and end-stage renal disease (ESRD), contributing to significant morbidity, mortality, and rising healthcare costs. In this review article, we explore the role of epigenetic mechanisms in HN progression and their potential therapeutic implications. We begin by examining key epigenetic modifications-DNA methylation, histone modifications, and non-coding RNAs-observed in kidney disease. Next, we discuss the underlying pathophysiology of HN and highlight current in vitro and in vivo models used to study the condition. Finally, we compare various types of HN-induced renal injury and their associated epigenetic mechanisms with those observed in other kidney injury models, drawing inferences on potential epigenetic therapies for HN. The information gathered in this work indicate that epigenetic mechanisms can drive the progression of HN by regulating key molecular signaling pathways involved in renal damage and fibrosis. The limitations of Renin-Angiotensin-Aldosterone System (RAAS) inhibitors underscore the need for alternative treatments targeting epigenetic pathways. This review emphasizes the importance of further research into the epigenetic regulation of HN to develop more effective therapies and preventive strategies. Identifying novel epigenetic markers could provide new therapeutic opportunities for managing CKD and reducing the burden of ESRD.

Inference from biased polls

People often attempt to present a positive image by overstating virtuous behaviors when responding to unincentivized “polls.” We examine whether others account for this “socially desirable responding” (SDR) when drawing inferences from such unincentivized responses. In an experiment, we incentivize “predictors” to guess others' choice behaviors across actions with varying social desirability. Predictors observe random subsamples of either (i) incentivized choices or (ii) hypothetical claims. The hypothetical claims exhibit systematic SDR and predictors are reasonably skeptical of them. However, their skepticism is not tailored to the direction or magnitude of SDR. This under-correction occurs even though subjects' stated sentiment toward the actions can predict SDR.

Investigation of geochemically induced permeability alteration in geothermal reservoirs and its implications for sustainable geothermal energy production

Geothermal resources constitute a significant portion of the world's low-carbon, renewable energy potential, with about 75% classified as low-temperature. One such potential resource exists in Precambrian basement rocks underlying the Williston Basin in southern Saskatchewan, Canada, with a reservoir temperature of 120 °C. However, geochemically induced permeability alteration in these highly reactive low-temperature granitoid resources poses a significant risk to long-term heat production. To assess and potentially mitigate this risk, we conducted a geochemical and mineralogical study of both altered and unaltered samples. Our findings facilitated the parameterization of geochemical simulations of water-rock interactions to predict mineral volume changes and, by extension, draw inferences on porosity and permeability changes resulting from these interactions. The simulations indicate an increased mineral volume in both samples, yet geothermal alteration of the unaltered, and thus more reactive, rocks induced relative mineral volume changes about 30% greater than those in the altered rocks. The resulting absolute change in porosity is 0.5 vol% for the unaltered rocks and 0.35 vol% for the altered rocks. Utilizing an empirical porosity-permeability relationship, the computed change in permeability indicates that the unaltered basement rock experienced a greater change in total permeability than the altered basement rocks. Additional calculations demonstrate the sensitivity of the porosity-permeability equation to critical porosity and power exponent, forecasting various scenarios with permeability changes ranging from 1.0 × 10−13 to 1 × 10−20 m2. Consequently, we infer that altered, permeable zones of the examined Precambrian basement rocks are likely to offer favourable conditions for sustained, multi-decade heat production, and thus should be targeted over less altered zones to justify initial capital expenditures. Globally, geothermal heat extraction from these rocks remains an underexplored yet promising resource for generating reliable, low-carbon renewable energy, crucial in our efforts to decarbonize the global economy.

Disrupted Candidacy: A Longitudinal Examination of the Constrained Healthcare-Access Journeys of National Health Service Workers in Scotland Seeking Supports for Long COVID Illness.

Evidence examines how persons experiencing Long COVID (LC) struggle to secure healthcare for symptoms. However, few studies examine healthcare workers experiencing LC, nor the complex and multiple difficulties faced when seeking and receiving healthcare. This study is based on two phases of longitudinally conducted qualitative interviews, 6 months apart, with National Health Service (NHS) workers experiencing LC, from different occupational roles at NHS locales in Scotland (first interviews, n = 50; second interviews, n = 44). Multiple factors restricted healthcare access, including worries about pressuring the NHS and concerns over LC being legitimised. When healthcare was sought, workers struggled to secure support, referrals and treatment. The following reasons were included: (1) context: the restrictive pandemic healthcare context; (2) illness climate: low GP knowledge surrounding LC and how this could be treated, trends for ascribing symptoms to other causes and reluctance to diagnose LC; (3) sense-making of LC: healthcare availability linked to occupational role identity. To visualise and examine healthcare barriers, candidacy theory is applied, drawing inferences between healthcare context, illness climate, sense-making and identities. NHS workers' complex journeys represent Disrupted Candidacy, intersecting challenges across candidacy domains, restricting the seeking and receiving of LC healthcare. Findings provide insights into why NHS workers resisted and withdrew from healthcare-seeking, and the barriers they faced when attempting to secure LC support. This study presents a pathway for future LC illness research to use a modified candidacy theory framework. This research focuses on amplifying and learning from lived experiences, and the voices of NHS workers in Scotland experiencing LC. Interviews represent primary data for this study; thus, participants and their healthcare journeys are centred in this research and all aspects of production, reporting and output. Explicit discussions of stakeholder group involvement are highlighted in the methods section.

Implementation of Inclusive Education Programs in Reading Practice for Low-Grade Slow Learners in Elementary Schools

Education is a fundamental entitlement granted to children, aimed at preparing forthcoming generations. However, it's imperative to acknowledge that not all children grasp education at the same pace, often referred to as slow learners. Additionally, the lack of specialized mentorship in primary schools exacerbates this issue. Therefore, it's crucial for teachers to employ effective strategies to address these challenges. This study seeks to outline the utilization of an approach method in supplementary reading sessions tailored for slow learners. Employing a descriptive qualitative research design, data was gathered through observations, in-depth interviews, and document analysis. The research, structured as an intrinsic case study, involved two class teachers and slow learner students in grades one and two at SDN Wiropaten. Data analysis comprised data reduction, display, and inference drawing, with data validity ensured through source and technique triangulation, along with crosschecking. Findings indicate the effectiveness of the approach method in catering to the needs of slow learner students in lower grades. Several programs and outcomes stemming from this approach were identified. Grade one initiatives encompassed building rapport with slow learners, fostering trust, facilitating social interaction, and implementing follow-up strategies. Grade two initiatives included assigning independent tasks, engaging in random student participation (such as alphabet memorization), word games, and guided reading exercises.

Asymmetric Biotic Interactions Cannot Be Inferred Without Accounting for Priority Effects.

Understanding biotic interactions is a crucial goal in community ecology and species distribution modelling, and large strides have been made towards improving multivariate computational methods with the aim of quantifying biotic interactions and improving predictions of species occurrence. Yet, while considerable attention has been given to computational approaches and the interpretation of these quantitative tools, the importance of sampling design to reveal these biotic interactions has received little consideration. This study explores the influential role of priority effects, that is, the order of habitat colonisation, in shaping our ability to detect biotic interactions. Using a simple set of simulations, we demonstrate that commonly used cross-sectional co-occurrence data alone cannot be used to make reliable inferences on asymmetric biotic interactions, even if they perform well in predicting the occurrence of species. We then show how sampling designs that consider priority effects can recover the asymmetric effects that are lost when priority effects are ignored. Based on these findings, we urge for caution when drawing inferences on biotic interactions from cross-sectional binary co-occurrence data, and provide guidance on sampling designs that may provide the necessary data to tackle this longstanding challenge.

EDNA State and Medium Affect DNA Degradation Patterns in Seminatural Systems of Southern African Waterholes

ABSTRACTEnvironmental DNA (eDNA) has evolved into a valuable asset of the ecologists' toolkit, enabling time‐ and cost‐efficient biodiversity assessments in a wide variety of ecosystems. Since eDNA can be isolated from a broad range of environmental substrates, its persistence times in those media are of decisive importance for drawing inferences about species presence. For the first time, we characterize eDNA persistence in water and sediment samples of seminatural waterholes in a savanna system in South Africa to gain a better understanding of eDNA decay in these waterbodies. Using mesocosm experiments, we tracked eDNA decay in two different DNA states (extracellular and membrane bound), during two different seasons (wet and dry), and from two different substrates (surface water and sediment). Extracellular DNA degraded rapidly in a first‐order exponential decay fashion and membrane‐bound DNA exhibited a slower decline with more intricate patterns, involving initial reduction followed by a subsequent increase in measured DNA concentrations. The latter we attribute to cell disassociation and cell lysis at 24–48 h after introduction into the environment. Higher stochasticity of membrane‐bound DNA capture in the dry season highlights the need for higher sampling efforts in natural systems in which eDNA is presumably more patchily distributed. Additionally, we observed longer eDNA persistence in sediments than in water samples, presumably due to better protection from nucleases. We could not reveal any effects of environmental parameters on eDNA decay, emphasizing that further research is needed to better understand eDNA dynamics in those waterbodies in order to exploit their full potential for eDNA‐based bioassessments in those systems.

From runway to homepage: Synthesizing luxury brand strategies in digital scape

The increasing preference of consumers to purchase luxury online has caught the attention of luxury retailers over time. It has encouraged retailers to formulate strategies that boost their online revenue. Owing to the importance of the topic, this article presents a detailed and critical review of brand strategies that induce luxury consumption using digital platforms. Using the PRISMA approach and following the systematic review process, the authors have identified five major brand strategies that can help them strengthen their digital game. The article presents a conceptual framework that will help luxury retailers understand the current state of online brand strategies, draw inferences from them, and use them to achieve their future objectives. Academically, the article includes a synthesis of publication trends, methodologies, prominent journals, theories, constructs, and directions for future research using the TCM approach. This will help future researchers gain meaningful insights into the extant work and help them discover gaps to advance the online luxury literature. Further, the paper highlights the industrial implications of the identified brand strategies in the Asian market, as luxury brands have shown an increasing interest in this growing market.

The impact of policy and technology infrastructure on telehealth utilization.

The COVID-19 outbreak resulted in an unprecedented surge in telehealth utilization. However, the effects of policy enactment on utilization remain understudied in the literature. Our research examines the impact of policy mandates relating to federal incentive programs on telehealth utilization across states during the pandemic by considering technology infrastructure, such as broadband penetration and equipment possession. This study also considers the impact of political orientation and control factors, including income attributes, on telehealth utilization. Considering telehealth utilization in the Medicare program as a test case, this study employed the partial least square and structural equation model to analyze data from the Centers for Medicare & Medicaid Services, the American Community Survey, and the Microsoft Airband Project to draw inferences. Our study finds that telehealth legislation, broadband penetration, political orientation, and control factors have a direct significant impact on telehealth utilization, whereas incentive programs and equipment possession have an indirect impact through broadband penetration.

A non‐stationary factor copula model for non‐Gaussian spatial data

AbstractWe introduce a new copula model for non‐stationary replicated spatial data. It is based on the assumption that a common factor exists that controls the joint dependence of all the observations from the spatial process. As a result, our proposal can model tail dependence and tail asymmetry, unlike the Gaussian copula model. Moreover, we show that the new model can cover a full range of dependence between tail quadrant independence and tail dependence. Although the log‐likelihood of the model can be obtained in a simple form, we discuss its numerical computational issues and ways to approximate it for drawing inference. Using the estimated copula model, the spatial process can be interpolated at locations where it is not observed. We apply the proposed model to temperature data over the western part of Switzerland, and we compare its performance with that of its stationary version and with the Gaussian copula model.

Retail investors and ESG news: A discussion

Li, Watts, and Zhu (2024) provide evidence that retail investors trade in response to financially material ESG news. This evidence is consistent with retail investors trading in response to the financial implications of ESG-related information in much the same way that they trade in response to the financial implications of other information. The authors suggest that their evidence is inconsistent with retail investors making investment decisions based on their nonpecuniary preferences. I suggest that their research design is constructed to identify financially motivated trading and is not well suited to drawing inferences about retail investors’ nonpecuniary investment decisions.

Evaluating elements of artificial cover object design for terrestrial salamander monitoring

AbstractDeclines in amphibian populations have led to greater research and monitoring of species of interest, including terrestrial salamanders. Several methods have been used to study salamanders, including artificial cover objects (ACOs). Artificial cover objects are easy to install and use, limit destruction to habitats, are easily reproducible both across and within studies, and can reduce variability between observers and sites by not relying on observer search time or effort, or on natural cover. However, different ACO designs and spacing may yield different capture rates, and thus different estimates of population densities. We examined the influence of ACO wood type, size, shape, and spacing arrays on the number of individually marked red‐backed salamanders (Plethodon cinereus) captured under ACOs. White pine (Pinus strobus) ACOs yielded the most (average of 14 across 3 replicates) individuals, while white ash (Fraxinus americana) ACOs had the least (average of 7 across 3 replicates). Larger ACOs had similar capture rates compared to equal sampled areas of smaller ACOs, but a single smaller ACO tended to have fewer individual salamander captures than a single large ACO. Our results suggested that high‐density ACO spacing arrays may increase capture probability but reduce the area uniquely sampled by each ACO. In contrast, low‐density ACO arrays may produce population estimates more comparable to other arrays. Future studies should consider monitoring goals and available sampling effort when choosing a spacing array. As ACO study design impacts salamander capture rates, a clear understanding of the effects of each element of ACO design is important when drawing inferences or making comparisons among studies.

Developing Cogent Strategies for the Lexical Development of Non-Dominant Languages: Empowering Linguistic have-Nots and Maximizing Linguistic Performance

We begin this paper by emphasizing the importance of language policy and language planning processes, and investigate its history in different contexts. We also draw reference to our previously published paper on the empowerment of linguistic have nots, and attempt to present and propose classificatory mechanisms of languages from the aforesaid perspective. We also then review the concepts of pedagogy, verbal intelligence, linguistic enculturation, linguistic ethnography, the Sapir Whorf hypothesis, linguistic workarounds and econolexia, and review and study patterns of lexical development among individuals in different contexts, and in different parts of the world. We interface these observations with first and second language acquisition patterns and theories as well. We then end this paper by debating and proposing various strategies for the lexical development of languages, particularly strategies for the lexical development of less dominant and future-ready languages. We do this by studying different case studies and experiences drawn from real-time studies carried out and conducted in different parts of the world, particularly experiences from developing countries, and then drawing inferences and generalizations there from. We anticipate that this would constitute an extremely important component of our globalization of science movement by boosting linguistic and economic performance of people in different cultures and different parts of the world, particularly those who lag behind or are otherwise underrepresented in science and economic development.

“Zeitenwende” as coming of age? EU foreign & security policy through war & peace*

ABSTRACT The Russian invasion of Ukraine has brought war back to the European continent and led to considerable change in EU member states’ foreign policies. The consequential degree of EU foreign policy unity, as well as shifts in long-lasting national and collective security and defence taboos, has represented a significant departure from past practices. We use these processes of change as a starting point to set the scene for this special issue and to inform its main research question: in what manner, if at all, has the EU come of age as a foreign and security actor during Russia’s war on Ukraine? This introduction situates the main question of the special issue into the wider scholarly debates on actorness and the EU’s geopolitical ambitions. It conceptually develops the analogy of “coming of age” to examine a prospective maturation process of the EU as a foreign and security actor. In doing so, it not only interrogates what the EU as a mature foreign and security actor would look like, but it also develops the framework, identifies four maturation processes and reflects on necessary caveats for drawing inferences about the state of maturation of the EU as foreign and security actor.

Bayesian modal regression based on mixture distributions

Compared to mean regression and quantile regression, the literature on modal regression is very sparse. A unifying framework for Bayesian modal regression is proposed, based on a family of unimodal distributions indexed by the mode, along with other parameters that allow for flexible shapes and tail behaviors. Sufficient conditions for posterior propriety under an improper prior on the mode parameter are derived. Following prior elicitation, regression analysis of simulated data and datasets from several real-life applications are conducted. Besides drawing inference for covariate effects that are easy to interpret, prediction and model selection under the proposed Bayesian modal regression framework are also considered. Evidence from these analyses suggest that the proposed inference procedures are very robust to outliers, enabling one to discover interesting covariate effects missed by mean or median regression, and to construct much tighter prediction intervals than those from mean or median regression. Computer programs for implementing the proposed Bayesian modal regression are available at https://github.com/rh8liuqy/Bayesian_modal_regression.

Pathways to Early Science Literacy: Investigating the Different Role of Language and Reading Skills in Science Literacy Among Early Primary School Children

The purpose of this study is to examine the distinct roles of language and reading skills in science literacy among primary school students through a qualitative approach. Science literacy, defined as an understanding of scientific concepts and processes, and the capacity to apply them in everyday life, is a crucial competency for students in the modern era. This research employed a case study method, with data collection techniques comprising in-depth interviews, classroom observations and document analysis. The findings indicate that language skills play a pivotal role in assisting students in comprehending and communicating scientific concepts in an effective manner. The aforementioned skills encompass an individual's capacity to comprehend and employ scientific terminology, construct sentences with correct grammatical structures, and convey ideas in a clear and logical manner. In contrast, reading abilities are pivotal for students to obtain, comprehend and integrate information from diverse sources, including scientific texts. Well-developed reading abilities facilitate students to understand intricate scientific texts, construct arguments supported by evidence and draw inferences from available data. This research offers insights into the development of language and reading skills holistically, with the aim of improving science literacy among primary school students. It also emphasises the importance of an integrated and comprehensive teaching approach in science education.

English literacy learning loss during the COVID-19 pandemic in Indonesia

This study explores the impact of the COVID-19-induced school closures in Indonesia on the decline of secondary school students’ learning, particularly in their English literacy. While numerous studies have delved into the repercussions of school closures on literacy skills in European countries, Australia, and the USA, such investigations have been relatively limited in Asia, particularly within the Indonesian context. Employing a qualitative approach for impact assessment, this research gathered data from diverse sources, including online surveys, one-on-one interviews, and focus group discussions. The survey involved 140 English teachers from four Indonesian provinces, including Java, Sumatra, Riau Islands, and Bali provinces. Ten English teachers were selected to participate in the individual online interviews, followed by a focus group discussion with four teachers. Within an inductive approach and thematic analysis framework, data were analysed, and five primary literacy skills emerged from the analysis that show a decline during the period of school closures: having critical thinking while reading, drawing inferences from texts, summarising text content, having good attitudes toward writing, and composing longer paragraphs. These initial findings, although on a small scale, are anticipated to help educators, school administrators, and policymakers in the Indonesian education sector in devising more effective plans for addressing unforeseen challenges in the future, including the development of programs to help students recover from learning loss.