This study examines the Programme for International Student Assessment (PISA), one of the most influential tools of global education reform discourses in the 21st century. The study focuses on the governing role of statistical data in the discourse constructed by the international comparative assessment, referring to the global educational governance of the OECD. Disturbingly, the systematic collection and distribution of data does not merely quantify student achievement. Rather, students and participating countries are also qualified and classified. Here, (...) OECD’s PISA statistics act as a governing technology enabling the comparisons; through PISA’s statistics, the OECD has largely fostered the test-driven education reform trends worldwide. In this study, statistical data are examined as political instruments visualizing intangible concepts, constructing definitions of good practice and improvement, and fabricating the individuals whose knowledge and skills are normalized. In conclusion, it is necessary to rethink the new sociopolitical instrument – PISA’s statistics – not as a master rationale to reform the educational system of each country, but as a mere reference point based on the system’s distinctiveness. (shrink)

Statistical tests of the primality of some numbers look similar to statistical tests of many nonmathematical, clearly empirical propositions. Yet interpretations of probability prima facie appear to preclude the possibility of statistical tests of mathematical propositions. For example, it is hard to understand how the statement that n is prime could have a frequentist probability other than 0 or 1. On the other hand, subjectivist approaches appear to be saddled with ‘coherence’ constraints on rational probabilities that require (...) rational agents to assign extremal probabilities to logical and mathematical propositions. In the light of these problems, many philosophers have come to think that there must be some way to generalize a Bayesian statistical account. In this article I propose that a classical frequentist approach should be reconsidered. I conclude that we can give a conditional justification of statistical testing of at least some mathematical hypotheses: if statistical tests provide us with reasons to believe or bet on empirical hypotheses in the standard situations, then they also provide us with reasons to believe or bet on mathematical hypotheses in the structurally similar mathematical cases. (shrink)

Peter Spirtes. Building Causal Graphs from Statistical Data in the Presence of Latent Variables.

This article is a longitudinal analysis of the relation between gendered labour divisions and new data processing technologies at the Dutch Central Bureau of Statistics. Following social-constructivist and evolutionary economic approaches, the authors hold that the relation between technology and work organization is a two-way process. This means that technology does not only affect the relations between men and women at work, but that these relations also influence technological choices. The proportional numbers of men and women on the labour (...) market and changing conceptions of which work is deemed appropriate for women have been important factors in the strongly varying extent to which women participated in office work. They have also been important factors in determining the course of technological development. Shortages of women on the labour market have even determined search heuristics for new technological solutions in specific directions. (shrink)

Statistical significance is almost universally equated with the attribution to some population of nonchance influences as the source of structure in the data. But statistical significance can be divorced from both parameter estimation and probability as, instead, a statement about the atypicality or lack of exchangeability over some distinction of the data relative to some set. From this perspective, the criticisms of significance tests evaporate.

Data mining is on the interface of Computer Science and Statistics, utilizing advances in both disciplines to make progress in extracting information from large databases. It is an emerging field that has attracted much attention in a very short period of time. This article highlights some statistical themes and lessons that are directly relevant to data mining and attempts to identify opportunities where close cooperation between the statistical and computational communities might reasonably provide synergy for further (...) progress in data analysis. (shrink)

Statistical methods used by healthcare entities enable the collection of various information about the structure and characteristics of treated patients. They are an important source of knowledge, and form a database that plays an important role in entity management theory. In the presented study, we analysed the hospital stays of patients treated in all hospital wards of the 3rd City Hospital in Łodź during 2012. The following, in particular, were taken into account: admittance procedure, discharge procedure, age and sex (...) of hospitalised persons. Patients in over 55% of cases were admitted using the sud- den admittance procedure. At the same time, over 3/4 of the stays ended with a referral for further treatment in ambulatory conditions, and death occurred in approx. 5% of hospitalisations. By comparing the discharge procedures, the percentage of deaths in the Anaesthesiology and Intensive Care Wards can be seen clearly. Internal wards are next in turn. The biggest differences in the length of hospitalisation between the studied institution and the NFZ data concern the E77, A49, A48, A87, A33, D18, E16, E61 and G37 groups. (shrink)

Combining information from two or possibly several blocks of data is gaining increased attention and importance in several areas of science and industry. Typical examples can be found in chemistry, spectroscopy, metabolomics, genomics, systems biology and sensory science. Many methods and procedures have been proposed and used in practice. The area goes under different names: data integration, data fusion, multiblock analyses, multiset analyses and a few more. This book is an attempt to give an up-to-date treatment of (...) the most used and important methods within an important branch of the area; namely methods based on so-called components or latent variables. These methods have already obtained an enormous attention in for instance chemometrics, bioinformatics, machine learning and sensometrics and have proved to be important both for prediction and interpretation. The book is primarily a description of methodologies, but most of the methods will be illustrated by examples from the above-mentioned areas. The book is written such that both users of the methods as well as method developers will hopefully find sections of interest. In the end of the book there is a description of a software package developed particularly for the book. This package is freely available in R and covers many of the methods discussed. To distinguish the different type of methods from each other, the book is divided into five parts. Part I is introduction and preliminary concepts. Part II is the core of the book containing the main unsupervised and supervised methods. Part III deals with more complex structures and, finally, Part IV discusses alternative unsupervised and supervised methods. The book ends with Part V discussing the available software. Our recommendation for reading the book are as follows. A minimum read of the book would involve chapters 1, 2, 3, 5 and 7. Chapters 4, 6 and 8 are more specialized and chapters 9 and 10 contain methods we think are more advanced or less obvious to use. We feel privileged to have so many friendly colleagues who were willing to spend their time on helping us to improve the book by reading separate chapters. We would like to express our thanks to: Rasmus Bro, Margriet Hendriks, Ulf Indahl, Henk Kiers, Ingrid MaÌ⁽ge, Federico Marini, AÌ⁽smund Rinnan, Rosaria Romano, Lars Erik Solberg, Marieke Timmerman, Oliver Tomic, Johan Westerhuis and Barry Wise. Of course, the correctness of the final text is fully our responsibility! (shrink)

This book examines several epistemological regimes in studies of numerical data over the last four centuries. It distinguishes these regimes and mobilises questions present in the philosophy of science, sociology and historical works throughout the 20th century. Attention is given to the skills of scholars and their methods, their assumptions, and the socio-historical conditions that made calculations and their interpretations possible. In doing so, questions posed as early as Émile Durkheim’s and Ernst Cassirer’s ones are revisited and the concept (...) of symbolic form is put to the test in this particular survey, conducted over long period of time. Although distinct from a methodological and epistemological point of view, today these regimes may be found together in the toolbox of statisticians and those who comment on their conclusions. As such, the book is addressed to social scientists and historians and all those who are interested in numerical productions. (shrink)

Statistical distributions have great applicability for modeling data in almost every applied sector. Among the available classical distributions, the inverse Weibull distribution has received considerable attention. In the practice of distribution theory, numerous methods have been studied and suggested/introduced to increase the flexibility level of the traditional probability distributions. In this paper, we implement different distribution methods to obtain five new different versions of the inverse Weibull model. The new modifications of the inverse Weibull model are called the (...) logarithm transformed-inverse Weibull, a flexible reduced logarithmic-inverse Weibull, the weighted TX-inverse Weibull, a new generalized-inverse Weibull, and the alpha power transformed extended-inverse Weibull distributions. To illustrate the flexibility and applicability of the new modifications of the inverse Weibull model, a biomedical data set is analyzed. The data set consists of 108 observations and represents the mortality rate of the COVID-19-infected patients. The practical application shows that the new generalized-inverse Weibull is the best modification of the inverse Weibull distribution. (shrink)

The quality of the products coming from different lines of production requires some tests called comparative life tests. For lines having the same facility, the lifetime of the product is distributed by Burr XII, the lifetime distribution, and units are tested under type-I generalized hybrid censoring scheme. The observed censoring data are used under maximum likelihood and the Bayes method to estimate the model parameters. The theoretical results are discussed and assessed through data analysis and Monte Carlo simulation (...) study. Finally, we reported some brief comments obtained from numerical computation. (shrink)

The rise of Big Data changes the context in which organisations producing official statistics operate. Big Data provides opportunities, but in order to make optimal use of Big Data, a number of challenges have to be addressed. This stimulates increased collaboration between National Statistical Institutes, Big Data holders, businesses and universities. In time, this may lead to a shift in the role of statistical institutes in the provision of high-quality and impartial statistical information (...) to society. In this paper, the changes in context, the opportunities, the challenges and the way to collaborate are addressed. The collaboration between the various stakeholders will involve each partner building on and contributing different strengths. For national statistical offices, traditional strengths include, on the one hand, the ability to collect data and combine data sources with statistical products and, on the other hand, their focus on quality, transparency and sound methodology. In the Big Data era of competing and multiplying data sources, they continue to have a unique knowledge of official statistical production methods. And their impartiality and respect for privacy as enshrined in law uniquely position them as a trusted third party. Based on this, they may advise on the quality and validity of information of various sources. By thus positioning themselves, they will be able to play their role as key information providers in a changing society. (shrink)

There is an increasing interest in expanding the one-parameter Lindley distribution to two-parameter, three-parameter, and five-parameter. The univariate one-parameter Lindley distribution is still one of the most applicable distributions in data analysis especially in lifetime data. Modeling dependent random quantities required bivariate parametric probability distributions. This study presents a new bivariate three-parameter probability distribution called bivariate modified Lindley distribution. The one-parameter modified Lindley distribution is used as a base line to construct the new model. Its statistical properties (...) including cumulative function, density function, marginals, moments, conditional distributions, and copula are discussed. Simulation is constructed to declare theoretical properties, to show the flexibility of the new model and to investigate the goodness of fit. Two sets of real data, financial data and UEFA Champion’s League data, are used to show the applicability of the proposed model for different types of data. (shrink)

People made forecasts from graphically presented time series. Series were sinusoids overlaid by a zero or positive linear trend and a zero, low, moderate, or high level of noise. Forecasting performance was affected by both these variables. However, it did not correlate with ability to identify the trend and correlated significantly with ability to detect the sinusoidal pattern only when series were noise-free. A second experiment showed that the effect of data noise was not influenced by the number of (...) forecasts that people made from a series. These findings are consistent with the view that data noise does not affect the way that forecasts are made but that it impairs them for two reasons. First, it renders the anchor-and-adjust heuristics underlying them less effective. Second, it causes people to add more noise to their judgements in their attempts to make them representative of the data. (shrink)

The diversity of state confidentiality laws governing public health data presents a significant challenge for public health initiatives. This challenge is further complicated by the array of confidentially laws that are relevant within a state as disclosure and usage standards vary depending upon data holder, type, and source. These laws often have not been updated to address modern confidentiality risks such as unlawful data linkage or breach, leaving many public health organizations without clear guidance in the contentious (...) area of individual privacy. To address these challenges, public health organizations have increasingly turned to the science of de-identification, but whether de-identification adequately meets the many and varied state confidentiality legal requirements remains an unanswered question. (shrink)

The exponential growth of social data both in volume and complexity has increasingly exposed many of the shortcomings of the conventional frequentist approach to statistics. The scientific community has called for careful usage of the approach and its inference. Meanwhile, the alternative method, Bayesian statistics, still faces considerable barriers toward a more widespread application. The bayesvl R package is an open program, designed for implementing Bayesian modeling and analysis using the Stan language’s no-U-turn (NUTS) sampler. The package combines the (...) ability to construct Bayesian network models using directed acyclic graphs (DAGs), the Markov chain Monte Carlo (MCMC) simulation technique, and the graphic capability of the ggplot2 package. As a result, it can improve the user experience and intuitive understanding when constructing and analyzing Bayesian network models. A case example is offered to illustrate the usefulness of the package for Big Data analytics and cognitive computing. (shrink)

Comparing datasets, that is, sets of numbers in context, is a critical skill in higher order cognition. Although much is known about how people compare single numbers, little is known about how number sets are represented and compared. We investigated how subjects compared datasets that varied in their statistical properties, including ratio of means, coefficient of variation, and number of observations, by measuring eye fixations, accuracy, and confidence when assessing differences between number sets. Results indicated that participants implicitly create (...) and compare approximate summary values that include information about mean and variance, with no evidence of explicit calculation. Accuracy and confidence increased, while the number of fixations decreased as sets became more distinct , demonstrating that the statistical properties of datasets were highly related to comparisons. The discussion includes a model proposing how reasoners summarize and compare datasets within the architecture for approximate number representation. (shrink)

In European countries (excluding the UK and Ireland), official statistics do not use racial or ethnic categories, but instead rely on proxies to collect data about discrimination. In the German microcensus, the proxy category adopted is ‘migration background’ (Migrationshintergrund): an individual has a ‘migration background’ when one or more of their parents does not have German citizenship by birth. We apply a coupled ethical-epistemic analysis to the ‘migration background’ category to illuminate how the epistemic issues contribute to ethical ones. (...) Our central claim is that these ethical-epistemic issues with the ‘migration background’ category are best analysed in terms of Charles Mills's ‘white ignorance’. Appealing to Annette Martín's structural account of white ignorance, we highlight the cyclical reinforcement of ignorance and racial injustice in the use of the ‘migration background’ variable. Colourblind eliminativism about race perpetuates the use of ‘migration background’, which sustains and intensifies racial injustices. (shrink)

DNA sequencers, Twitter, MRIs, Facebook, particle accelerators, Google Books, radio telescopes, Tumblr: what do these things have in common? According to the evangelists of “data science,” all of these are instruments for observing reality at unprecedentedly large scales and fine granularities. This perspective ignores the social reality of these very different technological systems, ignoring how they are made, how they work, and what they mean in favor of an exclusive focus on what they generate: Big Data. But no (...) data, big or small, can be interpreted without an understanding of the process that generated them. Statistical data science is applicable to systems that have been designed as scientific instruments, but is likely to lead to confusion when applied to systems that have not. In those cases, a historical inquiry is preferable. (shrink)

Evidence, as viewed through the lens of statistical significance, is not always as it appears! In the investigation of clinical research findings arising from statistical analyses, a fundamental initial step for the emerging fraud detective is to retrieve the source data for cross-examination with the study data. Recognizing that source data are not always forthcoming and that, realistically speaking, the investigator may be uninitiated in fraud detection and investigation, this paper will highlight some key methodological (...) procedures for providing a sounder evidence base for withdrawing from a study on grounds of integrity. The promotion of patient safety is paramount. However, there is a broader rationale for disseminating these ideas. This includes empowering researchers to optimize their personal integrity, make informed choices regarding membership of future research collaborations and successfully voice their concerns to journal editors, particularly where a conflict of interests can render such dialogues particularly difficult. Recommendations will be supported by topical case studies and practical steps involving data exploration, testing of baseline data and application of Benford’s Law. While this paper has a clinical focus, the advice provided is transferrable to a wide range of multidisciplinary research settings outside of Medicine. (shrink)

Madole & Harden argue that the Mendelian reshuffling of genes and genomes is analogous to randomised controlled trials. We are not convinced by their arguments. First, their recipe for meeting the demands on randomised experiments is inherently inconsistent. Second, disequilibrium across chromosomes conflicts with their assumption of statistical independence. Third, the genome-wide association study (GWAS) method has many pitfalls, including low repeatability.

In empirical modeling, an important desiderata for deeming theoretical entities and processes as real is that they can be reproducible in a statistical sense. Current day crises regarding replicability in science intertwines with the question of how statistical methods link data to statistical and substantive theories and models. Different answers to this question have important methodological consequences for inference, which are intertwined with a contrast between the ontological commitments of the two types of models. The key (...) to untangling them is the realization that behind every substantive model there is a statistical model that pertains exclusively to the probabilistic assumptions imposed on the data. It is not that the methodology determines whether to be a realist about entities and processes in a substantive field. It is rather that the substantive and statistical models refer to different entities and processes, and therefore call for different criteria of adequacy. (shrink)

Ethical conduct is an essential component in research, especially in medical research. Statistical methods for design and analysis are powerful research tools if used properly. Abuse of these principles and methods are just as unethical as other laboratory or clinical misconduct. Inadequate research design can produce worthless results and thus wastes effort and valuable resources. For clinical research, patient resources are wasted. Inappropriate analysis of data can also produce misleading results and conclusions. For clinical research, inferior therapy might (...) be given to patients as a consequence. These ethical concerns can have implications for and affect the individuals responsible for the statistical design and analysis. Examples are provided which illustrate some possible abuses and inappropriate pressures. (shrink)

There is little empirical evidence showing a direct link between a capacity for statistical learning (SL) and proficiency with natural language. Moreover, discussion of the role of SL in language acquisition has seldom focused on literacy development. Our study addressed these issues by investigating the relationship between SL and reading ability in typically developing children and healthy adults. We tested SL using visually presented stimuli within a triplet learning paradigm and examined reading ability by administering the Wide Range Achievement (...) Test (WRAT-4; Wilkinson & Robertson, 2006). A total of 38 typically developing children (mean age of 9;5 years, range 6;4–12;5) and 37 healthy adults (mean age of 21 years, range 18–34) were assessed. In children, SL was significantly related to reading ability. Importantly, this relationship was independent of grade and also age. The adult data, too, revealed that SL was significantly related to reading ability. A regression analysis of the combined child and adult data revealed that SL accounted for a unique amount of variance in reading ability, after age and attention had been taken into consideration. For the first time, this study provides empirical evidence that a capacity for more effective SL is related to higher reading ability in the general population. (shrink)

The paper reviews statistical models for money, wealth, and income distributions developed in the econophysics literature since the late 1990s. By analogy with the Boltzmann-Gibbs distribution of energy in physics, it is shown that the probability distribution of money is exponential for certain classes of models with interacting economic agents. Alternative scenarios are also reviewed. Data analysis of the empirical distributions of wealth and income reveals a two-class distribution. The majority of the population belongs to the lower class, (...) characterized by the exponential (“thermal”) distribution, whereas a small fraction of the population in the upper class is characterized by the power-law (“superthermal”) distribution. The lower part is very stable, stationary in time, whereas the upper part is highly dynamical and out of equilibrium. (shrink)

This paper shows how to define probability distributions over linguistically realistic syntactic structures in a way that permits us to define language learning and language comprehension as statistical problems. We demonstrate our approach using lexical‐functional grammar (LFG), but our approach generalizes to virtually any linguistic theory. Our probabilistic models are maximum entropy models. In this paper we concentrate on statistical inference procedures for learning the parameters that define these probability distributions. We point out some of the practical problems (...) that make straightforward ways of estimating these distributions infeasible, and develop a “pseudo‐likelihood” estimation procedure that overcomes some of these problems. This method raises interesting questions concerning the nature of the data available to a language learner and the modularity of language learning and processing. (shrink)

Multivariate pattern analysis can reveal new information from neuroimaging data to illuminate human cognition and its disturbances. Here, we develop a methodological approach, based on multivariate statistical/machine learning and time series analysis, to discern cognitive processing stages from functional magnetic resonance imaging (fMRI) blood oxygenation level dependent (BOLD) time series. We apply this method to data recorded from a group of healthy adults whilst performing a virtual reality version of the delayed win-shift radial arm maze (RAM) task. (...) This task has been frequently used to study working memory and decision making in rodents. Using linear classifiers and multivariate test statistics in conjunction with time series bootstraps, we show that different cognitive stages of the task, as defined by the experimenter, namely, the encoding/retrieval, choice, reward and delay stages, can be statistically discriminated from the BOLD time series in brain areas relevant for decision making and working memory. Discrimination of these task stages was significantly reduced during poor behavioral performance in dorsolateral prefrontal cortex (DLPFC), but not in the primary visual cortex (V1). Experimenter-defined dissection of time series into class labels based on task structure was confirmed by an unsupervised, bottom-up approach based on Hidden Markov Models. Furthermore, we show that different groupings of recorded time points into cognitive event classes can be used to test hypotheses about the specific cognitive role of a given brain region during task execution. We found that whilst the DLPFC strongly differentiated between task stages associated with different memory loads, but not between different visual-spatial aspects, the reverse was true for V1. Our methodology illustrates how different aspects of cognitive information processing during one and the same task can be separated and attributed to specific brain regions based on information contained in multivariate patterns of voxel activity. (shrink)

In typical statistical learning studies, researchers define sequences in terms of the probability of the next item in the sequence given the current item (or items), and they show that high probability sequences are treated as more familiar than low probability sequences. Existing accounts of these phenomena all assume that participants represent statistical regularities more or less as they are defined by the experimenters—as sequential probabilities of symbols in a string. Here we offer an alternative, or possibly supplementary, (...) hypothesis. Specifically, rather than identifying or labeling individual stimuli discretely in order to predict the next item in a sequence, we need only assume that the participant is able to represent the stimuli as evincing particular similarity relations to one another, with sequences represented as trajectories through this similarity space. We present experiments in which this hypothesis makes sharply different predictions from hypotheses based on the assumption that sequences are learned over discrete, labeled stimuli. We also present a series of simulation models that encode stimuli as positions in a continuous two‐dimensional space, and predict the next location from the current location. Although no model captures all of the data presented here, the results of three critical experiments are more consistent with the view that participants represent trajectories through similarity space rather than sequences of discrete labels under particular conditions. (shrink)

Inductive logic generalizes the idea of logical entailment and provides standards for the evaluation of non-conclusive arguments. A main application of inductive logic is the generalization of observational data to theoretical models. In the empirical sciences, the mathematical theory of statistics addresses the same problem. This paper argues that there is no separable purely logical aspect of statistical inference in a variety of complex problems. Instead, statistical practice is often motivated by decision-theoretic considerations and resembles empirical science.

Statistical inference is often justified by long-run properties of the sampling distributions, such as the repeated sampling rationale. These are frequentist justifications of statistical inference. I argue, in line with existing philosophical literature, but against a widespread image in empirical science, that these justifications are flawed. Then I propose a novel interpretation of probability in statistics, the artefactual interpretation. I believe that this interpretation is able to bridge the gap between statistical probability calculations and rational decisions on (...) the basis of observed data. The artefactual interpretation is able to justify statistical inference without making any assumptions about probability in the material world. (shrink)

Gignac and Zajenkowski (2020) find that “the degree to which people mispredicted their objectively measured intelligence was equal across the whole spectrum of objectively measured intelligence”. This Comment shows that Gignac and Zajenkowski’s (2020) finding of homoscedasticity is likely the result of a recoding choice by the experimenters and does not in fact indicate that the Dunning-Kruger Effect is a mere statistical artifact. Specifically, Gignac and Zajenkowski (2020) recoded test subjects’ responses to a question regarding self-assessed comparative IQ onto (...) a linear IQ scale when a normal IQ scale would likely have been more appropriate. More generally, researchers studying self-assessed intelligence should be aware of potential measurement problems that may arise when transforming an ordinal scale onto an interval scale. (shrink)