Uncovering Substantive Patterns in Student Responses in International Large-Scale Assessments—Comparing a Latent Class to a Manifest DIF Approach

In this study, we contrast results from two differential item functioning (DIF) approaches (manifest and latent class) by the number of items and sources of items identified as DIF using data from an international reading assessment. The latter approach yielded three latent classes, presenting evidence of heterogeneity in examinee response patterns. It also yielded more DIF items with larger effect sizes and more consistent item response patterns by substantive aspects (e.g., reading comprehension processes and cognitive complexity of items). Based on our findings, we suggest empirically evaluating the homogeneity assumption in international assessments because international populations cannot be assumed to have homogeneous item response patterns. Otherwise, differences in response patterns within these populations may be under-detected when conducting manifest DIF analyses. Detecting differences in item responses across international examinee populations has implications on the generalizability and meaningfulness of DIF findings as they apply to heterogeneous examinee subgroups.     

Is Work Engagement Exhausting? The Longitudinal Relationship Between Work Engagement and Exhaustion Using Latent Growth Modeling

The relationship between exhaustion and work engagement has received considerable attention during the past decades. Although the theoretical proposition exists that work engagement may increase exhaustion over time, previous research has been mixed. Drawing on the transactional stress model and applying latent growth modeling, we aim to provide a more comprehensive picture of the work engagement–exhaustion relationship over time. In two longitudinal studies, with four measurement points each, we found consistent evidence that a higher initial work engagement related to increased exhaustion over time. Consistent with our hypotheses, a higher initial work engagement also related to less initial exhaustion, and increases in work engagement related to decreases in exhaustion over time. However, contrary to our expectations, a higher initial exhaustion related to elevated work engagement over time. In conclusion, our findings suggest that engaged employees are less exhausted but face a higher risk of exhaustion over time. At the same time, exhausted employees are less engaged, but they have the potential to become more so over time. The theoretical and practical implications of these findings will be discussed in this paper.     

A Latent Class Multidimensional Scaling Model for Two-Way One-Mode Continuous Rating Dissimilarity Data

In this paper, we propose a cluster-MDS model for two-way one-mode continuous rating dissimilarity data. The model aims at partitioning the objects into classes and simultaneously representing the cluster centers in a low-dimensional space. Under the normal distribution assumption, a latent class model is developed in terms of the set of dissimilarities in a maximum likelihood framework. In each iteration, the probability that a dissimilarity belongs to each of the blocks conforming to a partition of the original dissimilarity matrix, and the rest of parameters, are estimated in a simulated annealing based algorithm. A model selection strategy is used to test the number of latent classes and the dimensionality of the problem. Both simulated and classical dissimilarity data are analyzed to illustrate the model.     

Variable Neighborhood Search Heuristics for Selecting a Subset of Variables in Principal Component Analysis

The selection of a subset of variables from a pool of candidates is an important problem in several areas of multivariate statistics. Within the context of principal component analysis (PCA), a number of authors have argued that subset selection is crucial for identifying those variables that are required for correct interpretation of the components. In this paper, we adapt the variable neighborhood search (VNS) paradigm to develop two heuristics for variable selection in PCA. The performances of these heuristics were compared to those obtained by a branch-and-bound algorithm, as well as forward stepwise, backward stepwise, and tabu search heuristics. In the first experiment, which considered candidate pools of 18 to 30 variables, the VNS heuristics matched the optimal subset obtained by the branch-and-bound algorithm more frequently than their competitors. In the second experiment, which considered candidate pools of 54 to 90 variables, the VNS heuristics provided better solutions than their competitors for a large percentage of the test problems. An application to a real-world data set is provided to demonstrate the importance of variable selection in the context of PCA.     

Impact of Immigration on Economic Growth in Canada and in its Smaller Provinces

This paper evaluates the potential impact of education levels of immigrants and Canadian-born on economic growth in Canada and its smaller provinces by using data for the period 2006–2013. We specify a production function in which levels of educational attainments of immigrants and Canadian-born workers are entered separately. Feasible generalized least square (FGLS) method is applied to estimate the production function separately for all immigrants, and also for established immigrants (those who have been in Canada for 10 years or longer). The results show that all educational levels of immigrants have positive and statistically significant effects on economic growth. A similar conclusion applies to Canadian-born workers, although the impacts of their university degree holders is lower than that of immigrant university degree holders. Both immigrant and Canadian-born workers have smaller effects on economic growth in smaller provinces, which have attracted larger numbers of immigrants in recent years. The results also show that the economic growth effects are similar for all and established immigrants. Although these results are consistent with previous findings on discounting of immigrants’ educational credentials, more data are needed to strengthen their validity. We also suggest that the higher economic growth impact of immigrant university degree holders than that of Canadian-born is indicative of greater social returns to higher education resulting from increased diversity of population which in turn, as some previous studies suggest, can result in increased technological innovation, new ideas, and production of a wide variety of goods and services.     

A Two-Tier Full-Information Item Factor Analysis Model with Applications

Motivated by Gibbons et al.’s (Appl. Psychol. Meas. 31:4–19, 2007) full-information maximum marginal likelihood item bifactor analysis for polytomous data, and Rijmen, Vansteelandt, and De Boeck’s (Psychometrika 73:167–182, 2008) work on constructing computationally efficient estimation algorithms for latent variable models, a two-tier item factor analysis model is developed in this research. The modeling framework subsumes standard multidimensional IRT models, bifactor IRT models, and testlet response theory models as special cases. Features of the model lead to a reduction in the dimensionality of the latent variable space, and consequently significant computational savings. An EM algorithm for full-information maximum marginal likelihood estimation is developed. Simulations and real data demonstrations confirm the accuracy and efficiency of the proposed methods. Three real data sets from a large-scale educational assessment, a longitudinal public health survey, and a scale development study measuring patient reported quality of life outcomes are analyzed as illustrations of the model’s broad range of applicability.     

Examining Trajectories of Early Adolescents’ Life Satisfaction in South Korea Using a Growth Mixture Model

Various aspects of early adolescence can have a key impact on adult life satisfaction. This study applied a growth mixture model to examine the trajectories of early adolescents’ life satisfaction and test the effects of individual (sex, subjective achievement, and self-esteem) and social factors (school adjustment and peer attachment) thereon. For this purpose, a growth mixture model was used on the fifth- (grade five), sixth- (grade six), and seventh- (grade seven) year panel data of 1917 grade one students from the Korean Children and Youth Panel Survey (2010–2016, National Youth Policy Institute), excluding missing samples. Participants were 990 boys (51.6%) and 927 girls (48.4%). The results indicated that the trajectories of life satisfaction could be categorized into three groups: “High Level-decreasing” (57.19%), “Moderate Level-stable” (37.62%), and “Low Level-increasing” (5.19%). Predictors for each class exhibited that female students and individuals with higher levels of self-esteem and school adjustment were more likely to be in the “Low Level-increasing” and “Moderate Level-stable” groups compared to male students and those reporting lower levels of self-esteem and school adjustment. Moreover, students with higher peer attachment were more likely to be in the “Low Level-increasing” group compared to the “High Level-decreasing” group. Educational implications to achieve positive life satisfaction by determining the influence of the predictors for each trajectory are outlined.

     

Measuring Adolescents’ Affective Civic Competence: Validity and Cross-Group Equivalence of a Second-Order Hierarchical Latent Construct

Applied Research in Quality of Life - In the last few decades, educational literature and policy makers have increasingly supported research into affective competencies—attitudes, emotions,...     

A Note on Comparing the Estimates of Models for Cluster-Correlated or Longitudinal Data with Binary or Ordinal Outcomes

When using linear models for cluster-correlated or longitudinal data, a common modeling practice is to begin by fitting a relatively simple model and then to increase the model complexity in steps. New predictors might be added to the model, or a more complex covariance structure might be specified for the observations. When fitting models for binary or ordered-categorical outcomes, however, comparisons between such models are impeded by the implicit rescaling of the model estimates that takes place with the inclusion of new predictors and/or random effects. This paper presents an approach for putting the estimates on a common scale to facilitate relative comparisons between models fit to binary or ordinal outcomes. The approach is developed for both population-average and unit-specific models.     

A Bayesian Semiparametric Item Response Model with Dirichlet Process Priors

In Item Response Theory (IRT), item characteristic curves (ICCs) are illustrated through logistic models or normal ogive models, and the probability that examinees give the correct answer is usually a monotonically increasing function of their ability parameters. However, since only limited patterns of shapes can be obtained from logistic models or normal ogive models, there is a possibility that the model applied does not fit the data. As a result, the existing method can be rejected because it cannot deal with various item response patterns. To overcome these problems, we propose a new semiparametric IRT model using a Dirichlet process mixture logistic distribution. Our method does not rely on assumptions but only requires that the ICCs be a monotonically nondecreasing function; that is, our method can deal with more types of item response patterns than the existing methods, such as the one-parameter normal ogive models or the two- or three-parameter logistic models.     

Sufficiency and Conditional Estimation of Person Parameters in the Polytomous Rasch Model

Rasch models are characterised by sufficient statistics for all parameters. In the Rasch unidimensional model for two ordered categories, the parameterisation of the person and item is symmetrical and it is readily established that the total scores of a person and item are sufficient statistics for their respective parameters. In contrast, in the unidimensional polytomous Rasch model for more than two ordered categories, the parameterisation is not symmetrical. Specifically, each item has a vector of item parameters, one for each category, and each person only one person parameter. In addition, different items can have different numbers of categories and, therefore, different numbers of parameters. The sufficient statistic for the parameters of an item is itself a vector. In estimating the person parameters in presently available software, these sufficient statistics are not used to condition out the item parameters. This paper derives a conditional, pairwise, pseudo-likelihood and constructs estimates of the parameters of any number of persons which are independent of all item parameters and of the maximum scores of all items. It also shows that these estimates are consistent. Although Rasch’s original work began with equating tests using test scores, and not with items of a test, the polytomous Rasch model has not been applied in this way. Operationally, this is because the current approaches, in which item parameters are estimated first, cannot handle test data where there may be many scores with zero frequencies. A small simulation study shows that, when using the estimation equations derived in this paper, such a property of the data is no impediment to the application of the model at the level of tests. This opens up the possibility of using the polytomous Rasch model directly in equating test scores.     

Varieties of Religious and Pornographic Experience: Latent Growth in Adolescents’ Religiosity and Pornography Use

The relationship between adolescent religiosity and pornography use has been longitudinally tested only in the United States. Given the social relevance of hypothesized mechanisms underlying the association, this study offers a three-wave longitudinal assessment of parallel latent growth in the two constructs carried out in a Southern European country. Using responses of 1,041 Croatian adolescents from the capital city (M_age = 16.14 years, SD = .45; 64.6% of female students) and latent growth curve modeling approach, we explored links between individual trajectories of change in religiosity and pornography use over a period of 24 months. In the observed period, religiosity decreased and pornography use increased among both male and female adolescents, but their dynamics were independent of each other—pointing to other (unmeasured) processes responsible for both adolescents’ sexualization and secularization. Important to note, the findings also pointed to an important role of age at first exposure to pornography for its frequency of use in middle to late adolescence.     

Who Speaks First? Analyzing Response Waves in a Large-Scale Campus Climate Survey

ABSTRACT

There has been a recent surge in the number of campus climate surveys conducted across the United States; however, little research has examined response rates. Using data from 5,524 students who participated in a Web-based survey at a large university in a South Atlantic part of the United States, four waves of data corresponding with participation rates after reminder e-mails were analyzed. These waves are first compared across a number of demographic variables. Next, the data on rates of stalking, intimate partner violence, sexual assault, and stranger sexual harassment for each wave are examined. The results show few significant differences in the characteristics of the students participating in each wave, and the victimization rates remained relatively constant across the 4 weeks of data collection.     

Measuring central executive functioning: what's in a reading span?

Although variations of Daneman and Carpenter's (1980) Reading Span Test (RST) have seen increasing use in both cognitive and neuropsychological research, the specific mental operations involved in performing it remain unclear. We tested 80 undergraduates to examine the extent to which speed of processing, manipulation capacity, and susceptibility to interference contributed to RST performance. The results suggest that, rather than unitary central executive or processing speed functions underlying RST performance, at least two factors, manipulation capacity and susceptibility to interference, underlie the task. Further study of RST operations may lead to a better understanding of the nature of the central executive itself.     

Reviewing Reduction in a Preferential Model‐Theoretic Context

In this article, we redefine classical notions of theory reduction in such a way that model‐theoretic preferential semantics becomes part of a realist depiction of this aspect of science. We offer a model‐theoretic reconstruction of science in which theory succession or reduction is often better—or at a finer level of analysis—interpreted as the result of model succession or reduction. This analysis leads to ‘defeasible reduction’, defined as follows: The conjunction of the assumptions of a reducing theory T with the definitions translating the vocabulary of a reduced theory T′ to the vocabulary of T, defeasibly entails the assumptions of reduced T′. This relation of defeasible reduction offers, in the context of additional knowledge becoming available, articulation of a more flexible kind of reduction in theory development than in the classical case. Also, defeasible reduction is shown to solve the problems of entailment that classical homogeneous reduction encounters. Reduction in the defeasible sense is a practical device for studying the processes of science, since it is about highlighting different aspects of the same theory at different times of application, rather than about naive dreams concerning a metaphysical unity of science.     

Acting Nasty in the Face of Failure? Longitudinal Observations of “Hard-to-Manage” Children Playing a Rigged Competitive Game with a Friend

Peer problems are almost universal among children with disruptive behavior disorders, and have been linked to social information processing deficits that lead to heightened threat responses (K. A. Dodge, 1980). This 2-year longitudinal study uses direct observations to examine the real-life significance of this finding. Forty hard-to-manage children and 40 typically developing control children were filmed at ages 5 and 7 playing a rigged competitive game in which they experienced a clear threat of losing. Group differences in negative behavior (hard-to-manage > controls) were stable over time and independent of verbal ability. Predictors of individual differences in negative behavior were also examined. Previous studies with this sample have shown that at 4 years of age, the hard-to-manage children displayed elevated frequencies of violent pretend play (J. Dunn & C. Hughes, 2001), coupled with poor performance on tests of executive function and theory of mind (C. Hughes, J. Dunn, & A. White, 1998). In this study, 4-year-olds whose pretend play indicated a preoccupation with violence were more likely to respond negatively to the threat of losing a competitive game at age 5 and at age 7. Four-year-olds who performed poorly on tests of theory of mind and executive function showed higher rates of negative behavior at age 5 but not age 7. These findings highlight just a few of the multiple paths leading to peer problems among children with disruptive behavior problems.     

Evidence-Based Assessment in Children and Adolescents with Obsessive–Compulsive Disorder

The assessment of varied psychiatric disorders, including obsessive–compulsive disorder (OCD), is shifting towards the use of evidence-based assessments (EBAs). This shift has fostered the development, validation and adaptation of several measures to rate obsessive–compulsive symptoms and other related problematic areas such as functional impairment or family attitudes among others. The aim of this paper is to present a systematic review of psychometric studies on pediatric OCD-specific measures to classify these according to assessment evidence-based criteria. Selection criteria that determined which studies were included in the review were: (1) analyzing an OCD measure and (2) including participants’ age being 18 years or younger. The literature search procedure was conducted in Medline, PsycINFO, PsycARTICLES, ERIC, Cochrane Library, and Scholar Google databases and enabled us to locate 42 studies which analyzed psychometric properties of 14 OCD measures studied in children and adolescents. Instruments were grouped into the following assessment areas: symptom presence and severity, functional impairment, family functioning and cognitive dimensions of OCD. Psychometric data regarding internal structure, internal consistency, reliability, validity and diagnostic precision were also reported. Further, measures were classified as well-established, approaching well-established and promising assessments in terms of reliability and validity. We concluded that the assessment of OCD in pediatric populations is a growing field that in a short-medium term could provide a wide variety of EBAs for the evaluation obsessive–compulsive symptoms and other OCD-related dimensions. The paper concludes by highlighting directions for future research.     

Maternal responsivity to infants in the “High Chair” assessment: Longitudinal relations with toddler outcomes in a diverse,low-income sample

Infant–parent interactions occur across many situations, yet most home-based assessments of parenting behaviors are conducted under conditions of low stress, such as free play. In this study, low-income mothers from the Early Head Start Research and Evaluation Project were observed at home interacting with their 14-month-olds in the mildly stressful “High Chair” assessment (n = 1718 dyads). This methodological study tested whether High Chair maternal responsiveness and detachment predicted later toddler cognitive and emotion outcomes, over and above equivalent maternal predictors during free play. High Chair responsiveness and detachment were significant, although modest, predictors of child cognitive and emotion outcomes, over and above maternal responsiveness and detachment during free play; except High Chair responsiveness did not predict the emotion outcome. There were no significant differences between ethnic groups in prediction of outcomes. Results are discussed in terms of the methodological value of assessing parenting behaviors across diverse situations and populations.     

Towards a “Sophisticated” Model of Belief Dynamics. Part I: The General Framework

It is well-known that classical models of belief are not realistic representations of human doxastic capacity; equally, models of actions involving beliefs, such as decisions based on beliefs, or changes of beliefs, suffer from a similar inaccuracies. In this paper, a general framework is presented which permits a more realistic modelling both of instantaneous states of belief, and of the operations involving them. This framework is motivated by some of the inadequacies of existing models, which it overcomes, whilst retaining technical rigour in so far as it relies on known, natural logical and mathematical notions. The companion paper (Towards a “sophisticated” model of belief dynamics. Part II) contains an application of this framework to the particular case of belief revision. Presented by Hannes Leitgeb