Theories of statistical testing may be seen as attempts to provide systematic means for evaluating scientific conjectures on the basis of incomplete or inaccurate observational data. The Neyman-Pearson Theory of Testing (NPT) has purported to provide an objective means for testing statistical hypotheses corresponding to scientific claims. Despite their widespread use in science, methods of NPT have themselves been accused of failing to be objective; and the purported objectivity of scientific claims based upon NPT has been called into (...) question. The purpose of this paper is first to clarify this question by examining the conceptions of (I) the function served by NPT in science, and (II) the requirements of an objective theory of statistics upon which attacks on NPT's objectivity are based. Our grounds for rejecting these conceptions suggest altered conceptions of (I) and (II) that might avoid such attacks. Second, we propose a reformulation of NPT, denoted by NPT*, based on these altered conceptions, and argue that it provides an objective theory of statistics. The crux of our argument is that by being able to objectively control error frequencies NPT* is able to objectively evaluate what has or has not been learned from the result of a statistical test. (shrink)

Several of Krueger & Funder's (K&F's) suggestions may promote more balanced social cognition research, but reconsidered null hypothesis statistical testing (NHST) is not one of them. Although NHST has primarily supported negative conclusions, this is simply because most conclusions have been negative. NHST can support positive, negative, and even balanced conclusions. Better NHST practices would benefit psychology, but would not alter the balance between positive and negative approaches.

Let an unbiased coin be used to form an ω-sequence S of independent tosses. Let N be the positive integers. The finite initial segment of length n ∈ N is denoted by Sn (thus, S1 holds exactly the first toss). For n ∈ N , let Hn be the proportion of heads that show up in Sn.

While orthodox (Neyman-Pearson) statistical tests enjoy widespread use in science, the philosophical controversy over their appropriateness for obtaining scientific knowledge remains unresolved. I shall suggest an explanation and a resolution of this controversy. The source of the controversy, I argue, is that orthodox tests are typically interpreted as rules for making optimal decisions as to how to behave--where optimality is measured by the frequency of errors the test would commit in a long series of trials. Most philosophers of (...) statistics, however, view the task of statistical methods as providing appropriate measures of the evidential-strength that data affords hypotheses. Since tests appropriate for the behavioral-decision task fail to provide measures of evidential-strength, philosophers of statistics claim the use of orthodox tests in science is misleading and unjustified. What critics of orthodox tests overlook, I argue, is that the primary function of statistical tests in science is neither to decide how to behave nor to assign measures of evidential strength to hypotheses. Rather, tests provide a tool for using incomplete data to learn about the process that generated it. This they do, I show, by providing a standard for distinguishing differences (between observed and hypothesized results) due to accidental or trivial errors from those due to systematic or substantively important discrepancies. I propose a reinterpretation of a commonly used orthodox test to make this learning model of tests explicit. (shrink)

Gillies introduced a propensity interpretation of probability which is linked to experience by a falsifying rule for probability statements. The present paper argues that general statistical tests should qualify as falsification rules. The ‘goodness-of-fit paradox’ is introduced: the confirmation of a probability model by a test refutes the model's validity. An example is given in which an independence test introduces dependence. Several possibilities to interpret the paradox and to deal with it are discussed. It is concluded that (...) the propensity interpretation properly reflects statistical practice, but it is not as objective as some adherents claim. (shrink)

Classical statistical inferences have been criticised for various reasons. To assess the soundness of such criticisms is a very important task because they are widely used in everyday scientific research. This is one of the reasons why the philosophy of statistics is an exciting field of study. In this paper, I focus on two such criticisms. The first one claims that the use of the p-value violates the principle of total evidence. It is a thesis that has been defended (...) by Elliott Sober and Bengt Autzen. The second one says that the result of classical tests does not only depend on the data but on the sampling plan of the experimenter also. The underlying criticism of course is that the sampling plan is not part of the evidence and that classical tests therefore violate PTE. The intentions of the experimenter should not affect the result of an inference. My aim is to show that both criticisms are unsound. Doing so, I hope to clarify the concept of p-value and the nature of the evidence in classical statistical tests. The point of my paper is to show that the identification of the evidence on which those criticisms rest is inadequate. (shrink)

One approach to validating simulation models is to formally compare model outputs with independent data. We consider such model validation from the point of view of Frequentist statistics. A range of estimates and tests of goodness of fit have been advanced. We review these approaches, and demonstrate that some of the tests suffer from difficulties in interpretation because they rely on the null hypothesisHypothesis that the model is similar to the observationsObservations. This reliance creates two unpleasant possibilities, namely, a model (...) could be spuriously validated when dataData are too few, or inappropriately rejected when data are too many. Finally, these tests do not allow a principled declaration of what a reasonable level of difference would be considering the purposes to which the model will be put. We consider equivalence tests, and demonstrate that they do not suffer from the previously identified shortcomings. We provide two case studies to illustrate the claims of the chapter. (shrink)

The origins of testing scientific models with statistical techniques go back to 18th century mathematics. However, the modern theory of statistical testing was primarily developed through the work of Sir R.A. Fisher, Jerzy Neyman, and Egon Pearson in the inter-war period. Some of Fisher's papers on testing were published in economics journals (Fisher, 1923, 1935) and exerted a notable influence on the discipline. The development of econometrics and the rise of quantitative economic models in the mid-20th century made (...) statistical significance testing a commonplace, albeit controversial tool within economics. -/- In the debate about significance testing, methodological controversies intertwine with epistemological issues and sociological developments. Our aim in this chapter is to expound these connections and to show how the use of, and the debate about, significance testing in economics differs from other social sciences, such as psychology. (shrink)

In a recent note Roger Carlson presented a rather negative appraisal of my treatment of the logic of Fisherian significance testing in [10]. The main issue between us involves Carlson's thesis that, within the limits set by Fisher, standard significance tests are valuable tools of data analysis as they stand, i.e., without modification of the structure of the reasoning they employ. Call this the adequacy thesis. In my paper I argued that the pattern of reasoning employed by tests of significance (...) needs to be justified in spite of its unquestioned acceptance by most researchers, The best justification offered to date—by R. A. Fisher—is seriously defective. Therefore the adequacy thesis is not justified. I proposed some alterations of the pattern of reasoning employed by tests of significance and argued that they guarantee that tests are adequate epistemological tools. (shrink)

Chow sets his version of statistical significance testing in an impoverished context of “theory corroboration” that explicitly excludes well-posed theories admitting of strong support by precise empirical evidence. He demonstrates no scientific usefulness for the problematic procedure he recommends instead. The important role played by significance testing in today's behavioral and brain sciences is wholly inconsistent with the rhetoric he would enforce.

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.

In educational measurement, exploring the method of generating multiple high-quality parallel tests has become a research hotspot. One purpose of this research is to construct parallel forms item by item according to a seed test, using two proposed item selection heuristic methods [minimum parameters–information–distance method and minimum information–parameters–distance method ]. Moreover, previous research addressing test assembly issues has been limited mainly to situations in which the information curve of the item pool or seed test has a normal (...) or skewed distribution. However, in practice, the distributions of information curves for tests are diverse. These include multimodal distributions, the most common type of which is the bimodal distribution. Therefore, another main aim of this article is to extend the information curves of unimodal distributions to bimodal distributions. Thus, this study adopts simulation research to compare the results of two item, response, theory -based item matching methods using different information curve distributions for item pools or seed tests. The results show that the MPID and MIPD methods yield rather good performance in terms of both two statistical targets when the information curve has a unimodal distribution, and two new methods yield better performance than two existing methods in terms of test information functions target when the information curve has a bimodal distribution. (shrink)

Chow's endorsement of a limited role for null hypothesis significance testing is a needed corrective of research malpractice, but his decision to place this procedure in a hypothetico-deductive framework of Popperian cast is unwise. Various failures of this version of the hypothetico-deductive method have negative implications for Chow's treatment of significance testing, meta-analysis, and theory evaluation.

The growing availability of computer power and statistical software has greatly increased the ease with which practitioners apply statistical methods, but this has not been accompanied by attention to checking the assumptions on which these methods are based. At the same time, disagreements about inferences based on statistical research frequently revolve around whether the assumptions are actually met in the studies available, e.g., in psychology, ecology, biology, risk assessment. Philosophical scrutiny can help disentangle 'practical' problems of model (...) validation, and conversely, a methodology of statistical model validation can shed light on a number of issues of interest to philosophers of science. (shrink)

Unfortunately, reading Chow's work is likely to leave the reader more confused than enlightened. My preferred solutions to the “controversy” about null- hypothesis testing are: (1) recognize that we really want to test the hypothesis that an effect is “small,” not null, and (2) use Bayesian methods, which are much more in keeping with the way humans naturally think than are classical statistical methods.

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)

To estimate causal relationships, time series econometricians must be aware of spurious correlation, a problem first mentioned by Yule (1926). To deal with this problem, one can work either with differenced series or multivariate models: VAR (VEC or VECM) models. These models usually include at least one cointegration relation. Although the Bayesian literature on VAR/VEC is quite advanced, Bauwens et al. (1999) highlighted that “the topic of selecting the cointegrating rank has not yet given very useful and convincing results”. The (...) present article applies the Full Bayesian Significance Test (FBST), especially designed to deal with sharp hypotheses, to cointegration rank selection tests in VECM time series models. It shows the FBST implementation using both simulated and available (in the literature) data sets. As illustration, standard non informative priors are used. (shrink)

Chow's book provides a thorough analysis of the confusing array of issues surrounding conventional tests of statistical significance. This book should be required reading for behavioral and social scientists. Chow concludes that the null-hypothesis significance-testing procedure (NHSTP) plays a limited, but necessary, role in the experimental sciences. Another possibility is that – owing in part to its metaphorical underpinnings and convoluted logic – the NHSTP is declining in importance in those few sciences in which it ever played a role.

This replication study aims to investigate a potential bias toward addition in the German language, building upon previous findings of Winter and colleagues who identified a similar bias in English. Our results confirm a bias in word frequencies and binomial expressions, aligning with these previous findings. However, the analysis of distributional semantics based on word vectors did not yield consistent results for German. Furthermore, our study emphasizes the crucial role of selecting appropriate translational equivalents, highlighting the significance of considering language‐specific (...) factors when testing for such biases for languages other than English. (shrink)

Male and female participants were instructed to produce an altered response pattern on an Implicit Association Test measure of gender identity by slowing performance in trials requiring the same response to stimuli designating own gender and self. Participants’ faking success was found to be predictable by a measure of slowing relative to unfaked performances. This combined task slowing (CTS) indicator was then applied in reanalyses of three experiments from other laboratories, two involving instructed faking and one involving possibly motivated (...) faking. Across all studies involving instructed faking, CTS correctly classified 75% of intentionally faking participants. Using the CTS index to adjust faked Implicit Association Test scores increased the correlation of CTS-adjusted measures with known group membership, relative to unadjusted (i.e., faked) measures. (shrink)

This article focuses on the role of statistical concepts in both experiment and theory in various scientific disciplines, especially physics, including astronomy, and psychology. In Sect. 1 the concept of uncertainty in astronomy is analyzed from Ptolemy to Laplace and Gauss. In Sect. 2 theoretical uses of probability and statistics in science are surveyed. Attention is focused on the historically important example of radioactive decay. In Sect. 3 the use of statistics in biology and the social sciences is examined, (...) with detailed consideration of various Chi-square statistical tests. Such tests are essential for proper evaluation of many different kinds of scientific hypotheses. (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 examines the claim that significance testing violates the Principle of Total Evidence. I argue that p-values violate PTE for two-sided tests but satisfy PTE for one-sided tests invoking a sufficient test statistic independent of the preferred theory of evidence. While the focus of the paper is to evaluate a particular claim about the relationship of significance testing and PTE, I clarify the reading of this methodological principle along the way.

The computations involved in statistical learning have long been debated. Here, we build on work suggesting that a basic memory process, chunking, may account for the processing of statistical regularities into larger units. Drawing on methods from the memory literature, we developed a novel paradigm to test statistical learning by leveraging a robust phenomenon observed in serial recall tasks: that short‐term memory is fundamentally shaped by long‐term distributional learning. In the statistically induced chunking recall (SICR) task, (...) participants are exposed to an artificial language, using a standard statistical learning exposure phase. Afterward, they recall strings of syllables that either follow the statistics of the artificial language or comprise the same syllables presented in a random order. We hypothesized that if individuals had chunked the artificial language into word‐like units, then the statistically structured items would be more accurately recalled relative to the random controls. Our results demonstrate that SICR effectively captures learning in both the auditory and visual modalities, with participants displaying significantly improved recall of the statistically structured items, and even recall specific trigram chunks from the input. SICR also exhibits greater test–retest reliability in the auditory modality and sensitivity to individual differences in both modalities than the standard two‐alternative forced‐choice task. These results thereby provide key empirical support to the chunking account of statistical learning and contribute a valuable new tool to the literature. (shrink)

Of the many proposals for inferring genetic regulatory structure from microarray measurements of mRNA transcript hybridization, several aim to estimate regulatory structure from the associations of gene expression levels measured in repeated samples. The repeated samples may be from a single experimental condition, or from several distinct experimental conditions; they may be “equilibrium” measurements or time series; the associations may be estimated by correlation coefficients or by conditional frequencies (for discretized measurements) or by some other statistic. This paper describes two (...) elementary statistical difficulties for all such procedures, no matter whether based on Bayesian updating, conditional independence testing, or other machine learning procedures such as simulated annealing or neural net pruning. One difficulty obtains if large numbers of cells are aggregated in a measurement of expression levels from a common population of cells; the other obtains if small numbers of cells are aggregated or if samples are separately aggregated over different populations of cells. (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)

Despite the widespread use of key concepts of the Neyman–Pearson (N–P) statistical paradigm—type I and II errors, significance levels, power, confidence levels—they have been the subject of philosophical controversy and debate for over 60 years. Both current and long-standing problems of N–P tests stem from unclarity and confusion, even among N–P adherents, as to how a test's (pre-data) error probabilities are to be used for (post-data) inductive inference as opposed to inductive behavior. We argue that the relevance of (...) error probabilities is to ensure that only statistical hypotheses that have passed severe or probative tests are inferred from the data. The severity criterion supplies a meta-statistical principle for evaluating proposed statistical inferences, avoiding classic fallacies from tests that are overly sensitive, as well as those not sensitive enough to particular errors and discrepancies. Introduction and overview 1.1 Behavioristic and inferential rationales for Neyman–Pearson (N–P) tests 1.2 Severity rationale: induction as severe testing 1.3 Severity as a meta-statistical concept: three required restrictions on the N–P paradigm Error statistical tests from the severity perspective 2.1 N–P test T(): type I, II error probabilities and power 2.2 Specifying test T() using p-values Neyman's post-data use of power 3.1 Neyman: does failure to reject H warrant confirming H? Severe testing as a basic concept for an adequate post-data inference 4.1 The severity interpretation of acceptance (SIA) for test T() 4.2 The fallacy of acceptance (i.e., an insignificant difference): Ms Rosy 4.3 Severity and power Fallacy of rejection: statistical vs. substantive significance 5.1 Taking a rejection of H0 as evidence for a substantive claim or theory 5.2 A statistically significant difference from H0 may fail to indicate a substantively important magnitude 5.3 Principle for the severity interpretation of a rejection (SIR) 5.4 Comparing significant results with different sample sizes in T(): large n problem 5.5 General testing rules for T(), using the severe testing concept The severe testing concept and confidence intervals 6.1 Dualities between one and two-sided intervals and tests 6.2 Avoiding shortcomings of confidence intervals Beyond the N–P paradigm: pure significance, and misspecification tests Concluding comments: have we shown severity to be a basic concept in a N–P philosophy of induction? (shrink)