In real testing, examinees may manifest different types of test‐taking behaviours. this paper we focus on two that appear to be among the more frequently occurring behaviours – solution behaviour and rapid guessing behaviour. Rapid usually happens in high‐stakes tests when there is insufficient time, low‐stakes lack effort. These qualitatively behaviours, if ignored, will lead violation local independence assumption and, as a result, yield biased item/person parameter estimation. We propose...

The recent surge of interests in cognitive assessment has led to developments novel statistical models for diagnostic classification. Central many such is the well-known Q-matrix, which specifies item-attribute relationships. This article proposes a data-driven approach identification Q-matrix and estimation related model parameters. A key ingredient flexible T-matrix that relates response patterns. flexibility allows construction natural criterion function as well computationally amenable...

Diagnostic classification models have recently gained prominence in educational assessment, psychiatric evaluation, and many other disciplines. Central to the model specification is so-called Q-matrix that provides a qualitative of item-attribute relationship. In this paper, we develop theories on identifiability for under DINA DINO models. We further propose an estimation procedure through regularized maximum likelihood. The applicability not limited or it can be applied essentially all...

This article focuses on a family of restricted latent structure models with wide applications in psychological and educational assessment, where the model parameters are via matrix to reflect prespecified assumptions attributes. Such is often provided by experts assumed be correct upon construction, yet it may subjective misspecified. Recognizing this problem, researchers have been developing methods estimate from data. However, fundamental issue identifiability has not addressed until now....

Statistical latent class models are widely used in social and psychological researches, yet it is often difficult to establish the identifiability of model parameters. In this paper, we consider issue a family restricted models, where restriction structures needed reflect pre-specified assumptions on related assessment. We results strict sense specify which types structure would give The not only guarantee validity many popularly but also provide guideline for experimental design, current...

Several two-sample tests for high-dimensional data have been proposed recently, but they are powerful only against certain limited alternative hypotheses. In practice, since the true hypothesis is unknown, it unclear how to choose a test. We propose an adaptive test that maintains high power across wide range of situations, and study its asymptotic properties. Its finite sample performance compared with existing tests. apply other detect possible associations between bipolar disease large...

Latent class models have wide applications in social and biological sciences. In many applications, prespecified restrictions are imposed on the parameter space of latent models, through a design matrix, to reflect practitioners' assumptions about how observed responses depend subjects' traits. Though widely used various fields, such restricted suffer from nonidentifiability due their discreteness nature complex structure restrictions. This work addresses fundamental identifiability issue by...

The modern web-based technology greatly popularizes computer-administered testing, also known as online testing. When these tests are administered continuously within a certain “testing window,” many items likely to be exposed and compromised, posing type of test security concern. In addition, if the testing time is limited, another recognized aberrant behavior rapid guessing, which refers quickly answering an item without processing its meaning. Both cheating guessing result in extremely...

Many high-dimensional hypothesis tests aim to globally examine marginal or low-dimensional features of a joint distribution, such as testing mean vectors, covariance matrices and regression coefficients. This paper constructs family U-statistics unbiased estimators the

Recurrent event analyses have found a wide range of applications in biomedicine, public health, and engineering, among others, where study subjects may experience sequence interest during follow-up. The R package reReg offers comprehensive collection practical easy-to-use tools for regression analysis recurrent events, possibly with the presence an informative terminal event. framework is general scalechange model which encompasses popular Cox-type model, accelerated rate mean as special...

Cognitive assessment is a growing area in psychological and educational measurement, where tests are given to assess mastery/deficiency of attributes or skills. A key issue the correct identification associated with items test. In this paper, we set up mathematical framework under which theoretical properties may be discussed. We establish sufficient conditions ensure that required by each item learnable from data.

It has been reported that red blood cell width (RDW) is a marker associated with the presence and adverse outcomes of various diseases. However, no data are available on correlation RDW presence, stage, grade in patients renal carcinoma (RCC) yet. By retrospectively analyzing clinical laboratory at baseline histologically confirmed RCC cases controls, present study demonstrated values were significantly higher than those value was independently RCC. Besides, revealed positive association...

Differential item functioning (DIF) analysis refers to procedures that evaluate whether an item's characteristic differs for different groups of persons after controlling overall differences in performance. DIF is routinely evaluated as a screening step ensure items behave the same across groups. Currently, majority studies focus predominately on unidimensional IRT models, although multidimensional (MIRT) models provide powerful tool enriching information gained modern assessment. In this...

Hypergraph data, which capture multi-way interactions among entities, are becoming increasingly prevalent in the big data eta. Generating new hyperlinks from an observed, usually high-dimensional hypergraph is important yet challenging task with diverse applications, such as electronic health record analysis and biological research. This fraught several challenges. The discrete nature of renders many existing generative models inapplicable. Additionally, powerful machine learning-based often...

This study extends the loss function-based parameter estimation method for diagnostic classification models proposed by C. Ma, de la Torre, et al. (2023, Psychometrika) to consider prior knowledge and uncertainty of sampling. To this end, we integrate with generalized Bayesian method. We establish consistency attribute mastery patterns The is compared in a simulation found be superior previous nonparametric method—a special case Moreover, applied real data parametric methods. Finally,...

This study extends the loss function-based parameter estimation method for diagnostic classification models proposed by C. Ma, de la Torre, et al. (2023, Psychometrika) to consider prior knowledge and uncertainty of sampling. To this end, we integrate with generalized Bayesian method. We establish consistency attribute mastery patterns The is compared in a simulation found be superior previous nonparametric method—a special case Moreover, applied real data parametric methods. Finally,...

Abstract Factor analysis is a widely used statistical tool in many scientific disciplines, such as psychology, economics, and sociology. As observations linked by networks become increasingly common, incorporating network structures into factor remains an open problem. In this paper, we focus on high-dimensional involving network-connected observations, propose generalized model with latent factors that account for both the structure dependence among variables. These can be shared variables...