Search

In relational event networks, the tendency for actors to interact with each other depends greatly on the past interactions between the actors in a social network. Both the volume of past interactions and the time that has elapsed since the past interactions affect the actors' decision-making to interact with other actors in the network. Recently occurred events may have a stronger influence on current interaction behavior than past events that occurred a long time ago--a phenomenon known as "memory decay". Previous studies either predefined a short-run and long-run memory or fixed a parametric exponential memory decay using a predefined half-life period. In real-life relational event networks, however, it is generally unknown how the influence of past events fades as time goes by. For this reason, it is not recommendable to fix memory decay in an ad-hoc manner, but instead we should learn the shape of memory decay from the observed data. In this paper, a novel semi-parametric approach based on Bayesian Model Averaging is proposed for learning the shape of the memory decay without requiring any parametric assumptions. The method is applied to relational event history data among socio-political actors in India and a comparison with other relational event models based on predefined memory decays is provided.

In recent years, some models for binary and graded format responses have been proposed to assess unipolar variables or "quasi-traits." These studies have mainly focused on clinical variables that have traditionally been treated as bipolar traits. In the present study, we have made a proposal for unipolar traits measured with continuous response items. The proposed log-logistic continuous unipolar model (LL-C) is remarkably simple and is more similar to the original binary formulation than the graded extensions, which is an advantage. Furthermore, considering that irrational, extreme, or polarizing beliefs could be another domain of unipolar variables, we have applied this proposal to an empirical example of superstitious beliefs. The results suggest that, in certain cases, the standard linear model can be a good approximation to the LL-C model in terms of parameter estimation and goodness of fit, but not trait estimates and their accuracy. The results also show the importance of considering the unipolar nature of this kind of trait when predicting criterion variables, since the validity results were clearly different.

The Expectancy-Value model has been extensively used to understand students' achievement motivation. However, recent studies propose the inclusion of cost as a separate construct from values, leading to the development of the Expectancy-Value-Cost model. This study aimed to adapt Kosovich et al.'s ("The Journal of Early Adolescence", 35, 5-6, 2015) instrument for assessing achievement motivation in science and math among Spanish-speaking students aged 9-12. The psychometric evaluation encompassed examining validity, reliability, and measurement invariance for gender. The findings support the applicability of the expectancy-value-cost model in elementary school students, providing a valid and reliable tool for assessing achievement motivation in the science and math domain. However, the findings for the measurement invariance for gender are less conclusive and warrant future psychometric evaluations. This study opens avenues for future research on achievement motivation in Spanish-speaking contexts and populations.

Unidimensionality is an important assumption of measurement but it is violated very often. Most of the time, tests are deliberately constructed to be multidimensional to cover all aspects of the intended construct. In such situations, the application of unidimensional item response theory (IRT) models is not justifieddue to poor model fit and misleading results. Multidimensional IRT (MIRT) models can handle several dimensions simultaneously and yield person ability parameters on several dimensions which is helpful for diagnostic purposes too. Furthermore, MIRT models use the correlation between the dimensions to enhance the precision of the measurement. In this study, a reading comprehension test is modeled with the multidimensional Rasch model. The findings showed that a correlated 2-dimensional model has the best fit to the data. The bifactor model revealed some interesting information about the structure of reading comprehension and the reading curriculum. Implications of the study for the testing and teaching of reading comprehension are discussed.

Despite the numerous endeavours made to develop questionnaires to assess learners' strategic behaviour in general and learning/use strategies across different language areas and skills, one can surprisingly find no inventories to address learners' grammar learning /use strategies. This study aims to validate a measure of additional language learners' grammar strategies. The validation of the instrument, grammar learning/use strategy questionnaire, involved three phases. To generate an item pool as the first phase, the researcher drew on Oxford's instructional modes of L2 learning, Oxford's strategic self-regulated model of learning, Larsen-Freeman's three-dimension model of grammar, grammar strategies in the inventories of language skills and Brigg's classroom task-based grammar learning strategies. The second phase of the study involved submitting the instrument to extensive piloting in an adequate sample, followed by the third phase, which comprised evaluating the psychometric properties of the revised instrument through exploratory and confirmatory factor analysis. The results indicate that the developed instrument has satisfactory psychometric characteristics and that the hypothesized theoretical model has a great fit with the data. The article closes with some tentative pedagogical recommendations and implications as well as guidelines on how grammar learning strategies could be utilized and investigated.

Today, school bullying is considered an important social problem that causes developmental injuries and health issues in bullies and victims. The present study investigated the mediating role of brain-behavioural systems in relationships between school climate and emotional self-regulation with school bullying in high school students. This study had a cross-sectional correlational design. The sample size was 257, and the participants were selected using cluster sampling. They completed the Olweus Bully/Victim Questionnaire, Organizational Climate Index, Emotion Regulation Questionnaire, and Behavioural/Brain Systems Questionnaire. The data were analysed using the Pearson correlation coefficient and structural equation modelling (SEM). The findings revealed that the school bullying model based on the school climate and emotional self-regulation with the mediating role of brain-behavioural systems had a good fit. They also demonstrated that school climate had a direct negative effect on school bullying, whereas emotional self-regulation had a direct positive effect on it (p < 0.01). The school climate and emotional self-regulation were found to have a positive indirect effect on school bullying through the mediating role of brain-behavioural systems (p < 0.01). Therefore, it is necessary to develop plans to improve school climate, emotional self-regulation, and brain-behavioural systems to reduce school bullying.

Purpose: This study investigated the reading profiles of middle school Spanish-speaking emergent bilinguals (EBs) with significantly below grade level reading comprehension and whether these profiles varied in their reading comprehension performance over time. Method: Latent profile analyses were used to classify Grade 6 and 7 Hispanic EBs (n = 340; 39% female) into subgroups based on their word reading and vocabulary knowledge. Growth models were then fit within each profile to evaluate reading comprehension performance over time. Results: Analyses revealed four latent profiles emerged: (a) very low word reading and low vocabulary (10%), (b) low word reading and low vocabulary (71%), (c) average word reading and low vocabulary (16%), and (d) high word reading and low vocabulary (3%). Subgroups varied in their reading comprehension initially and over one year. Students in the subgroup marked by very low word reading and low vocabulary showed the lowest reading comprehension performance initially; however, they also showed the greatest growth over one year. Conclusion: These findings suggest that there is heterogeneity in the reading skill profiles of Spanish-speaking EBs with reading comprehension difficulties. They also underscore the prevalence of word reading difficulties among these students. These may be important factors to consider when developing interventions to prevent and remediate these difficulties.

Confirmatory factor analyses (CFA) are often used in psychological research when developing measurement models for psychological constructs. Evaluating CFA model fit can be quite challenging, as tests for exact model fit may focus on negligible deviances, while fit indices cannot be interpreted absolutely without specifying thresholds or cutoffs. In this study, we review how model fit in CFA is evaluated in psychological research using fit indices and compare the reported values with established cutoff rules. For this, we collected data on all CFA models in "Psychological Assessment" from the years 2015 to 2020 (N[subscript Studies]=221). In addition, we reevaluate model fit with newly developed methods that derive fit index cutoffs that are tailored to the respective measurement model and the data characteristics at hand. The results of our review indicate that the model fit in many studies has to be seen critically, especially with regard to the usually imposed independent clusters constraints. In addition, many studies do not fully report all results that are necessary to re-evaluate model fit. We discuss these findings against new developments in model fit evaluation and methods for specification search.

Coefficient omega indices are model-based composite reliability estimates that have become increasingly popular. A coefficient omega index estimates how reliably an observed composite score measures a target construct as represented by a factor in a factor-analysis model; as such, the accuracy of omega estimates is likely to depend on correct model specification. The current paper presents a simulation study to investigate the performance of omega-unidimensional (based on the parameters of a one-factor model) and omega-hierarchical (based on a bifactor model) under correct and incorrect model misspecification for high and low reliability composites and different scale lengths. Our results show that coefficient omega estimates are unbiased when calculated from the parameter estimates of a properly specified model. However, omega-unidimensional produced positively biased estimates when the population model was characterized by unmodeled error correlations or multidimensionality, whereas omega-hierarchical was only slightly biased when the population model was either a one-factor model with correlated errors or a higher-order model. These biases were higher when population reliability was lower and increased with scale length. Researchers should carefully evaluate the feasibility of a one-factor model before estimating and reporting omega-unidimensional.

Applications of polytomous IRT models in applied fields (e.g., health, education, psychology) are abound. However, little is known about the impact of the number of categories and sample size requirements for precise parameter recovery. In a simulation study, we investigated the impact of the number of response categories and required sample size for successful parameter recovery and also examined model fit. Our results suggested that sample size requirements of approximately 250 or larger are typically sufficient for reasonable recovery of either person or item parameters (in particular discrimination). Further, we found that in many cases, the conditions of 10 items yielded more favorable results than those conditions when only 5 items were considered. With respect to the number of response categories, results were inconsistent, suggesting more complex recommendations in selecting the number of categories when developing scales. We discuss limitations of the current study, identify future directions, and provide recommendations for applied researchers who engage in scale development.

The commonly used Vineland Adaptive Behavior Scales (VABS-3) divides adaptive behavior into three domains comprising three subdomains. The validity of this three-factor structure has not been explored in autistic samples, which are often heterogeneous with respect to language and IQ. Furthermore, although there are two comparable forms, Comprehensive Interview Form and Comprehensive Parent/Caregiver Form (a questionnaire), the original validation was based on interview data. Considering the widespread use of the VABS-3 in autism research, and the increased feasibility of online-administered questionnaires, it is necessary to establish the validity of the Parent/Caregiver form in autistic individuals across a range of abilities. This study aimed to investigate the measurement invariance of the VABS-3 Parent/Caregiver form between a minimally verbal group and verbal group of autistic people; however, poor overall fit of the three-factor structure precluded invariance analyses. Subsequent analyses suggested poor fit in both language and age groups, as well as a unidimensional model. The results of the current study suggest that neither the three-factor or unidimensional model fit the VABS-3 Parent/Caregiver Form, thereby cautioning against interpretation of domain or overall adaptive behavior composite scores in autistic individuals and further encouraging careful consideration of administration format.

A frequent criticism of exploratory factor analysis (EFA) is that it does not allow correlated residuals to be modelled, while they can be routinely specified in the confirmatory (CFA) model. In this article, we propose an EFA approach in which both the common factor solution and the residual matrix are unrestricted (i.e., the correlated residuals need not be specified a priori). The estimation procedures are two-stage and based on the unweighted least squares principle. Procedures for judging the solution appropriateness (including goodness of fit) are also proposed. The simulation studies and illustrative example suggest that the approach works quite well Although the proposal is based on existing results, most of the developments can be considered to be new contributions, and are expected to be particularly useful in the earlier stages of item calibration. The whole procedure has been implemented in both R language and a well-known non-commercial EFA program.

The Bayesian piecewise growth model (PGM) is a useful class of models for analyzing nonlinear change processes that consist of distinct growth phases. In applications of Bayesian PGMs, it is important to accurately capture growth trajectories and carefully consider knot placements. The presence of missing data is another challenge researchers commonly encounter. To address these issues, one could use model fit and selection indices to detect misspecified Bayesian PGMs, and should give care to the potential impact of missing data on model evaluation. Here we conducted a simulation study to examine the impact of model misspecification and missing data on the performance of Bayesian model fit and selection indices (PPP-value, BCFI, BTLI, BRMSEA, BIC, and DIC), with an additional focus on prior sensitivity. Results indicated that (a) increasing the degree of model misspecification and amount of missing data aggravated the performance of indices in detecting misfit, and (b) different prior specifications had negligible impact on model assessment. We provide practical guidelines for researchers to facilitate effective implementation of Bayesian PGMs.

A standard assumption of latent class (LC) analysis is conditional independence, that is the items of the LC are independent of the covariates given the LCs. Several approaches have been proposed for identifying violations of this assumption. The recently proposed likelihood ratio approach is compared to residual statistics (bivariate residuals [BVR] and expected parameter change [EPC] statistics) for identifying nonuniform direct effect of covariates on the items of the LC model. The simulation study results show that the likelihood ratio (LR) test correctly identifies direct effects more often than the BVR statistics, showing comparable results to the EPC statistic in many situations- this at the price of having also a higher false positive rate than BVR. A real data example illustrates the use of the three procedures. Overall the combined use of residual statistics and LR testing is recommended for applied research.

Aim/Purpose: This mixed-methods study aims to examine factors influencing academicians' intentions to continue using AI-based chatbots by integrating the Task-Technology Fit (TTF) model and social network characteristics. Background: AI-powered chatbots are gaining popularity across industries, including academia. However, empirical research on academicians' adoption behavior is limited. This study proposes an integrated model incorporating TTF factors and social network characteristics like density, homophily, and connectedness to understand academics' continuance intentions. Methodology: A qualitative study involving 31 interviews of academics from India examined attitudes and the potential role of social network characteristics like density, homophily, and connectedness in adoption. Results showed positive sentiment towards chatbots and themes on how peer groups accelerate diffusion. In the second phase, a survey of 448 faculty members from prominent Indian universities was conducted to test the proposed research model. Contribution: The study proposes and validates an integrated model of TTF and social network factors that influence academics' continued usage intentions toward AI chatbots. It highlights the nuanced role of peer networks in shaping adoption. Findings: Task and technology characteristics positively affected academics' intentions to continue AI chatbot usage. Among network factors, density showed the strongest effect on TTF and perceived usefulness, while homophily and connectedness had partial effects. The study provides insights into designing appropriate AI tools for the academic context. Recommendations for Practitioners: AI chatbot designers should focus on aligning features to academics' task needs and preferences. Compatibility with academic work culture is critical. Given peer network influences, training and demonstrations to user groups can enhance adoption. Platforms should have capabilities for collaborative use. Targeted messaging customized to disciplines can resonate better with academic subgroups. Multidisciplinary influencers should be engaged. Concerns like plagiarism risks, privacy, and job impacts should be transparently addressed. Recommendation for Researchers: More studies are needed across academic subfields to understand nuanced requirements and barriers. Further studies are recommended to investigate differences across disciplines and demographics, relative effects of specific network factors like size, proximity, and frequency of interaction, the role of academic leadership and institutional policies in enabling chatbot adoption, and how AI training biases impact usefulness perceptions and ethical issues. Impact on Society: Increased productivity in academia through the appropriate and ethical use of AI can enhance quality, access, and equity in education. AI can assist in mundane tasks, freeing academics' time for higher-order objectives like critical thinking development. Responsible AI design and policies considering socio-cultural aspects will benefit sustainable growth. With careful implementation, it can make positive impacts on student engagement, learning support, and research efficiency. Future Research: Conduct longitudinal studies to examine the long-term impacts of AI chatbot usage in academia. Track usage behaviors over time as familiarity develops. Investigate differences across academic disciplines and roles. Requirements may vary for humanities versus STEM faculty or undergraduate versus graduate students. Assess user trust in AI and how it evolves with repeated usage, and examine trust-building strategies. Develop frameworks to assess pedagogical effectiveness and ethical risks of conversational agents in academic contexts.

Testing programs have explored the use of technology-enhanced items alongside traditional item types (e.g., multiple-choice and constructed-response items) as measurement evidence of latent constructs modeled with item response theory (IRT). In this report, we discuss considerations in applying IRT models to a particular type of adaptive testlet referred to as a branching item. Under the branching format, all test takers are assigned to a common question, and the assignment of the next question relies on the response to the first question through deterministic rules. In addition, the items at both stages are scored together as one polytomous item. Real and simulated examples are provided to discuss challenges in applying IRT models to branching items. We find that model-data misfit is likely to occur when branching items are scored as polytomous items and modeled with the generalized partial credit model and that the relationship between the discrimination of the routing component and the discriminations of the subsequent components seemed to drive the misfit. We conclude with lessons learned and provide suggested guidelines and considerations for operationalizing the use of branching items in future assessments.

A C-Test is a gap-filling test for measuring language competence in the first and second language. C-Tests are usually analyzed with polytomous Rasch models by considering each passage as a super-item or testlet. This strategy helps overcome the local dependence inherent in C-Test gaps. However, there is little research on the best polytomous Rasch model for C-Tests. In this study, the Rating Scale Model (RSM) and the Partial Credit Model (PCM) for analyzing C-Tests were compared. To this end, a C-Test composed of six passages with both RSM and PCM was analyzed. The models were compared in terms of overall fit, individual item fit, dimensionality, test targeting, and reliability. Findings showed that, although the PCM has a better overall fit compared to the RSM, both models produce similar test statistics. In light of the findings of the study, the choice of the best Rasch model for C-Tests will be discussed.

The acquisition of skills amongst learners is integral for their survival. This is the reason why most of the pedagogies integrate them as part of the core learning objectives. However, the advent of technology has revolutionized teaching approaches as well as skills training. Many of the skills in 21st century education highlight ingenuity, critical analysis and IT literacy. Though it is understood that these prepare learners to be efficient in the future, there are significant underlying issues and challenges in how schools partake in the training process or how skills acquisition amongst learners are monitored and promoted in the schools' advocacies. This research is aimed at evaluating primary school administration how 21st century skills should be reflected in the school system. Identified factors that help promote 21st century skills were analyzed using content analysis and focus group discussion of ten (10) experts and later on evaluated and confirmed statistically through Confirmatory Factor Analysis (CFA). Four hundred respondents: two hundred (200) school directors and two hundred (200) academic head teachers from 322 schools in southern Thailand were considered as samples. A 5-rating scale questionnaire was used for data collection. Through second-order confirmatory factor analysis, the results confirmed eight (8) factors that promote 21st century skills in primary school administration. Arranged in descending order by goodness of fit, these factors were: (1) teacher quality development (0.97), (2) educational evaluation (0.92), (3) teaching and learning management (0.90), (4) learning environment (0.88), (5) curriculum, learning media, and technology (0.87), (6) budget for education (0.84), (7) community cooperation (0.66), and (8) volunteer supervision (0.64). The measurement of all factors (R[superscript 2]) also indicated positive reliability ranging between 0.41-0.94.

The purpose of this study is to develop Education Value Perception Scale (EVPS) based on Bronfenbrenner's Ecological Theory and to investigate its psychometric properties according to Classical Test Theory (CTT) and Item Response Theory (IRT). The data were collected from 2872 secondary school students by stratified purposeful sampling method. Measurement invariance of EVPS was tested by multigroup confirmatory factor analysis based on gender, and scalar invariance was observed to have been provided. The estimations based on IRT were conducted based on Graded Response Model. While high positive correlations were found between the item discriminations estimated according to different test theories, high negative correlations were identified between item means. McDonald's Omega was calculated to be 0.79 according to CTT from reliability estimation methods, marginal reliability coefficient was determined to be 0.77 according to IRT. In the test-retest applications performed at 20-day intervals, the stability coefficient was found to be 0.81.

This research aimed to investigate the innovative leadership factors and leader characteristics of school administrators in affecting teachers' involvement in the professional learning community of primary education schools in Bangkok and its vicinity of Thailand. Hence, the researcher would shed light on a linear structural relationship model to examine the impacts of innovative leadership factors and leader characteristics of primary school administrators on teachers' involvement in the professional learning community. A quantitative approach survey design was employed in this research. A total of 840 respondents responded to questionnaires in a proportional of two teachers to one school administrator from 280 primary schools. The respondents participated in a survey utilizing a multi-stage sampling technique. The researcher planned to test whether the identified innovative leadership factors and leader characteristics are fitting with empirical data as the key research output. The findings indicated that there was a total of five innovative leadership factors and three leader characteristics in a professional learning community model. The linear structural relationship model was supported to the empirical data, with X[superscript 2] = 42.321, df = 31, X[superscript 2]/df = 1.3652, CFI = 0.998, TLI = 0.997, RMSEA = 0.021, and SRMR = 0.01, p = 0.0845. In conclusion, the linear structural relationship model for primary school administrators has a goodness of fit with the attained data. Finally, the findings of this research have successfully proposed a linear structural relationship model that would be guidelines for a primary school administrator to develop his capabilities to promote a professional learning community.

High stakes testing, such as IELTS, is designed to select individuals for decision-making purposes (Fulcher, 2013b). Hence, there is a slow-growing stream of research investigating the subskills of IELTS listening and, in feedback terms, its effects on individuals and educational programs. Here, cognitive diagnostic assessment (CDA) performs it through breaking down the abstract listening concepts into fine-grained subcomponents, and applying it to in-class assessment and teaching. Therefore, the strengths and weaknesses of language learners concerning various subsections of the IELTS listening test were explored with use of three CDMs, i.e., DINA, DINO and mixed DINA-DINO. As a result, the analysis of the participants' (N=463) performances revealed that the mixed-DINA-DINO model using the original Q-matrix was the most effective model. On the other hand, all three models indicated that gap-filling was cognitively less demanding than the other subsections. However, they did not show any agreement on multiple-choice sub-competency. Accordingly, it can be argued that the CDM-driven diagnostic information related to the sub-sections of the IELTS listening test can be used in educational systems to explore the underlying structure of a test and detect the learners' potential strengths and weaknesses with mastery and non-mastery of the intended items.

The aim of this study was to investigate the comparability of feedback across culturally diverse countries by assessing the measurement invariance in PISA 2015 data. A multi-group confirmatory factor analysis showed that the feedback scale implemented in PISA 2015 was not invariant across countries. The intercepts and residuals of the factor model were clearly not the same, and the factor loadings also differed. Model fit slightly improved when the more individualism-oriented countries were separated from the more collectivism-oriented countries, but not to an acceptable level. This implies that the feedback results from PISA 2015 have a different meaning across countries, and it is necessary to be careful when making cross-cultural comparisons. However, the absence of measurement invariance did not affect the relationship between feedback and science achievement scores. This means that feedback, as measured by PISA 2015, can be compared across culturally different countries, although the current form of this scale lacks important, culturally specific elements.

As beliefs are well-known antecedents of teachers' practices, including assessment practices, sound measurement of teacher beliefs is critical for scholarly research as well as practical purposes. The present study examined the validity of inferences derived from the Conceptions of Assessment III--Abridged (COA-IIIA) instrument with US PK-12 pre-service (N = 554) and in-service (N = 341) teachers. Prior research on the COA-IIIA in diverse contexts and with different teacher populations has yielded highly divergent findings concerning this instrument's internal score structure. The study relies on confirmatory factor analyses to investigate the COA-IIIA's score structure among US pre-service and in-service teachers and to test the measurement invariance of select COA-IIIA measurement models across these groups. Findings indicate poor fit of most prior COA-IIIA measurement models with US pre-service and in-service teachers, though we were able to specify several simplified models that were at least metrically invariant across these populations.

The present paper introduces a general multidimensional model to measure individual differences in learning within a single administration of a test. Learning is assumed to result from practicing the operations involved in solving the items. The model accounts for the possibility that the ability to learn may manifest differently for correct and incorrect responses, which allows for distinguishing different types of learning effects in the data. Model estimation and evaluation is based on a Bayesian framework. A simulation study is presented that examines the performance of the estimation and evaluation methods. The results show accuracy in parameter recovery as well as good performance in model evaluation and selection. An empirical study illustrates the applicability of the model to data from a logical ability test.

In recent years, harmonic regression models have been applied to implement quality control for educational assessment data consisting of multiple administrations and displaying seasonality. As with other types of regression models, it is imperative that model adequacy checking and model fit be appropriately conducted. However, there has been no literature on how to perform a comprehensive model adequacy evaluation when applying harmonic regression models to sequential data with seasonality in the educational assessment field. This paper is intended to fill this gap with an illustration of real data from an English language assessment. Two types of cross-validation, leave-one-out and out-of-sample, were designed to measure prediction errors and check model validation. Three types of R-squared (R[superscript 2], R[superscript 2][subscript adj], and R[superscript 2][subscript pred]) and various residual diagnostics were applied to check model adequacy and model fitting.

The study revisited the community of inquiry (CoI) instrument for construct revalidation. To that end, the study used confirmatory factor analysis (CFA) to examine four competing models (unidimensional, correlated-factor, second-order factor, and bifactor models) on model fit statistics computed using parameter estimates from a statistical estimator for ordinal categorical data. The CFA identified as the optimal structure the bifactor model where all items loaded on their intended domains and the existence of the general factor was supported, essentially evidence of construct validity for the instrument. The study further examined the bifactor model using mostly model-based reliability measures. The findings confirmed the contributions of the general factor to the reliability of instrument scores. The study concluded with validity and reliability evidence for the bifactor model, supported the model as a valid and reliable representation of the CoI instrument and a fuller representation of the CoI theoretical framework, and recommended its use in CoI-related research and practice in online education.

The University Student Depression Inventory (USDI; Khawaja and Bryden 2006) is a 30- item scale that is used to measure depressive symptoms among university students. Its psychometric properties have been widely investigated under the classical test theory (CTT). This study explored the application of the polytomous Rasch partial credit model (PCM) in evaluating the USDI using response data from a sample of Filipino university students (n = 441). Using sequential tests under the Rasch measurement framework, model fitting was performed through item- and person-fit analyses to detect and address possible sources of measurement noise, followed by tests of local independence and differential item functioning (DIF). Results revealed that the original scale contained five misfitting items (6, 7, 10, 12, 20); hence, the deletion of such items was proposed to provide a new but psychometrically sound measure of student depression. Further analysis of the data detected person misfits whose responses were removed in subsequent analyses of local independence and DIF. One pair of locally dependent items (25, 26) and three gender-biased items (1, 3, 8) were detected, which necessitates further item review for possible idiosyncratic meanings. This study showed that Rasch analysis of self-reported questionnaires like the USDI can complement factor analytic approaches, especially in the detection of multiple sources of measurement errors that may undermine the quality of survey data.

School victimization issues remain largely unresolved due to over-reliance on unidimensional conceptions of victimization and data from a few developed OECD countries. Thus, support for cross-national generalizability over multiple victimization components (relational, verbal, and physical) is weak. Our substantive-methodological synergy tests the cross-national generalizability of a three-component model (594,196 fifteen-year-olds; nationally-representative samples from 77 countries) compared to competing (unidimensional and two-component) victimization models. We demonstrate the superior explanatory power of the three-component model--goodness-of-fit, component differentiation, and discriminant validity of the three components concerning gender differences, paradoxical anti-bullying attitudes (the Pro-Bully Paradox) whereby victims are more supportive of bullies than of other victims, and multiple indicators of well-being. For example, gender differences varied significantly across the three components, and all 13 well-being indicators were more strongly related to verbal and particularly relational victimization than physical victimization. Collapsing the three components into one or two components undermined discriminant validity. Cross-nationally, systematic differences emerged across the three victimization components regarding country-level means, gender differences, national development, and cultural values. These findings across countries support a tripartite model in which the three components of victimization--relational, verbal, and physical--relate differently to key outcomes. Thus, these findings advance victimization theory and have implications for policy, practice, and intervention. We also discuss directions for further research: the need for simultaneous evaluation of multiple, parallel components of victimization and bullying, theoretical definitions of bullying and victimization and their implications for measurement, conceptual bases of global victimization indices, cyberbullying, anti-bullying policies, and capitalizing on anti-bullying attitudes.

The purpose of this study was to evaluate the degree of classification quality in the basic latent class model when covariates are either included or are not included in the model. To accomplish this task, Monte Carlo simulations were conducted in which the results of models with and without a covariate were compared. Based on these simulations, it was determined that models without a covariate better predicted the number of classes. These findings in general supported the use of the popular three-step approach; with its quality of classification determined to be more than 70% under various conditions of covariate effect, sample size, and quality of indicators. In light of these findings, the practical utility of evaluating classification quality is discussed relative to issues that applied researchers need to carefully consider when applying latent class models.

Students who are self-regulated leaners have been reported to perform more successfully in higher education institutions (HEIs). Therefore, instruments that can monitor students' motivation and learning have been devised and implemented. The aims of this work were to investigate the dimensionality of the motivated strategies for learning questionnaire (MSLQ) and compare the validation of the three competing models. Three competing measurement models (1-factor, 2-factor, and second-order factor) were selected as candidates. To reveal which model explained the original MSLQ most effectively and meaningfully, the original 15 indicators and 81 items were used, for which data was gathered from 945 participating engineering students in Thailand. The results of confirmatory factor analysis revealed that all three of the competing models fitted the data quite well, as all standardized factor loadings of these models were statistically significant. It appeared that two-factor and second-order factor models yielded a better overall fit to the data in comparison to one-factor model. These results confirmed that the original MSLQ is a reliable and valid measurement instrument, particularly the second-order factor model, which was the best model.

Although Diagnostic Classification Models (DCMs) were introduced to education system decades ago, it seems that these models were not employed for the original aims upon which they had been designed. Using DCMs has been mostly common in analyzing large-scale non-diagnostic tests and these models have been rarely used in developing Cognitive Diagnostic Assessment (CDA) from scratch. Despite the prevalence of "retrofitting" CDA studies, "true" applications of CDA are believed to be rare since, firstly, a coherent framework to conduct such studies had not been available and, secondly, researchers were not able to analyze various DCMs according to the same model fit indices and criteria. This paper presents a summary of different types of DCMs and reviews "true" and "retrofitting" CDA studies. Having examined the limitations of previous CDA studies, the present study argues for the implication and application of Ravand and Baghaei's (2019) framework to conduct "true" CDA studies. This framework is of importance since not only does it fit into prominent frameworks in education assessment such as Cognitive Design System and Assessment Triangle, but also it can provide test-developers with practical steps in conducting valid cognitive diagnostic tests.

This research aimed at creating a structural model of the indicators of Educational Leadership for Primary School Principals in Thailand, which is considered to be a theoretical model that has been used to test for coherence with the empirical data collected from a sample group of 580 participants, who were selected from 30,719 Primary School Principals from across the country. To create this theoretical structural model, a study of the suitability of the indicators was carried out so that it could be further used in the selection within the model, as well as in the model's coherence test with the empirical data and in the investigation of the factor loading. The results of the research were as follows: Firstly, all indicators, which had been applied in the research were selected and were then placed into the theoretical structural model because the average and distribution coefficient values were as set in the criteria. Secondly, the theoretical model is coherent with the empirical data as the values of relative Chi-square, Root Mean Square Error of Approximation, Goodness-of-Fit Index, Adjusted Goodness-of-Fit Index, Comparative Fit Index, and Normed Fit Index were as set in the criteria. Finally, the factor loadings of the key elements, sub-elements, and the indicators were as set in the criteria. This showed that the theoretical model from this research can be beneficial for the research population with construct validity.

In this study, the correct classification level of whether forgiving oneself, others and the situation is abusing their partners was determined by logistic regression analysis. There are 221 young adults ranging from 19-30 in this study, which was designed in the scanning model. Heartland Forgiveness Scale and Information Form were used in the study. In the initial model of the analysis, all participants were classified in the group that exploited their partner, with a classification percentage of 62.9%. The biggest contribution to the initial model comes from the variable of forgiving others, respectively, the variables of self-forgiveness and forgiveness. Cox & Snell RSquare value for the final model was calculated as 0.10. This finding shows that when the predictor variables are included in the model, 10% of the predicted variable is explained. Accordingly, the model has a good fit. Of the 82 individuals who did not abuse the result model, 34 were classified correctly and 48 were incorrect, with a percentage of correct classification of 41.5%. Of 139 individuals abused, 120 were correct and 19 were incorrectly classified, with an accurate classification percentage of 86.3%. One-unit increase in the self-forgiveness variable is 7.90% in the odds of abuse [(1-0.921)0.100]; One-unit increase in forgiveness to others causes an 8.10% [(1-0.919)0.100] increase in exploit odds. Findings show that the variables of self-forgiveness and forgiveness for others make significant contributions to classifying individuals who abuse and do not. The variables of forgiving yourself and others increase the predictive power of the model created.

The current study, guided by the cross-sectional survey method, assessed interface design quality, and its predictive ability on E-learners' post-adoption behavior in E-learning course environments. DeLone and McLean's Information Systems Success Model, Khan's E-learning Framework, and Bhattacherjee's Information System Continuance Model formed the current study's theoretical underpinning. Data for the study were collected from 232 E-learners in selected Ugandan higher learning institutions, using a 38-item self-administered questionnaire. Principal Components Analysis produced a four-factor structure of interface design quality that comprised of content interactivity, accessibility design, system navigation, and visual-aesthetics design, which were found to be valid and reliable using Confirmatory Factor Analysis. The Structural Equation Model fit indices revealed that the hypothesised model achieved adequate goodness-of-fit to the data. Regarding the structural relationships, the four factors were found to be statistically significant predictors of E-learners' satisfaction; and in-turn, satisfaction impacted learning agility. The results have clearly aligned with the study's theoretical framework, buttressing existing empirical data on interface designs and end-user post adoption with E-learning interventions. The current study is crucial for making evidence-based pedagogical and design decisions by key E-learning stakeholders for the successful implementation and continued use of digital learning solutions in higher education contexts.

In recent years, proponents of configurational comparative methods (CCMs) have advanced various dimensions of robustness as instrumental to model selection. But these robustness considerations have not led to computable robustness measures, and they have typically been applied to the analysis of real-life data with unknown underlying causal structures, rendering it impossible to determine exactly how they influence the correctness of selected models. This article develops a computable criterion of "fit-robustness," which quantifies the degree to which a CCM model agrees with other models inferred from the same data under systematically varied threshold settings of fit parameters. Based on two extended series of inverse search trials on data simulated from known causal structures, the article moreover provides a precise assessment of the degree to which fit-robustness scoring is conducive to finding a correct causal model and how it compares to other approaches of model selection.

The recognition memory receiver operating characteristic (ROC) is typically asymmetric with a characteristic elevation of the left-hand portion. Whereas the unequal variance signal detection model (uvsd) assumes the asymmetry results because old item evidence is noisier than new item evidence, the dual process signal detection model (dpsd) assumes it results because old items convey more useful information than new items. To test these assumptions, the models were fit to old/new recognition data and their evidence parameters were used to predict performance on a novelty, three-alternative forced-choice (N3AFC) recognition task. Critically, under the uvsd model, increased old item variance (sigma) predicts poorer N3AFC performance, whereas under the dpsd model, increased recollection rates (Ro) predict better N3AFC performance. Hence, the asymmetry parameters of the two models make divergent predictions. In two experiments the dpsd model's predictions were supported, whereas the uvsd model yielded unpredicted (from that model's perspective) patterns. Through simulation it was also shown that the dpsd model predicted the uvsd model's mispredictions, which resulted because increases in old item noise markedly depress the upper portion of the ROC. Overall, the data demonstrate that increasing ROC asymmetry is not a function of increasingly noisier target evidence, but instead increasingly informative target evidence. These findings invalidate the uvsd model, which heretofore has been primarily supported by its post-hoc fitting ability, not its construct validity.

Multiple group modeling is one of the methods to address the measurement noninvariance issue. Traditional studies on multiple group modeling have mainly focused on item responses. In computer-based assessments, joint modeling of response times and action counts with item responses helps estimate the latent speed and action levels in addition to latent ability. These two new data sources can also be used to further address the measurement noninvariance issue. One challenge, however, is to correctly model action counts which can be underdispersed, overdispersed, or equidispersed in real data sets. To address this, we adopted the Conway-Maxwell-Poisson distribution that accounts for different types of dispersion in action counts and incorporated it in the multiple group joint modeling of item responses, response times, and action counts. Bayesian Markov Chain Monte Carlo method was used for model parameter estimation. To illustrate an application of the proposed model, an empirical data analysis was conducted using the Programme for International Student Assessment (PISA) 2015 collaborative problem-solving items where potential measurement noninvariance issue existed between gender groups. Results indicated that Conway-Maxwell-Poisson model yielded better model fit than alternative count data models such as negative binomial and Poisson models. In addition, response times and action counts provided further information on performance differences between groups.

In selected-response assessments such as attitude surveys with Likert-type rating scales, examinees often select from rating scale categories to reflect their locations on a construct. Researchers have observed that some examinees exhibit "response styles," which are systematic patterns of responses in which examinees are more likely to select certain response categories, regardless of their locations on the construct (Baumgartner & Steenkamp, 2001; Paulhus, 1991; Roberts, 2016; Van Vaerenbergh & Thomas, 2013). To identify and minimize construct-irrelevant impacts of response styles, researchers have proposed tools such as the Partial Credit Model -- Response Style (PCMRS; Tutz et al., 2018) as an extension of the Partial credit model (PCM; Masters, 1982) to model the tendency for examinees to exhibit response styles. The PCMRS directly models response styles as a person-specific gamma parameter and corrects estimates of item difficulty for the presence of response styles. In this study, the authors describe details about the PCMRS model parameters and explore the correspondence between indicators of response styles from the PCMRS and person fit statistics that reflect the Rasch measurement framework to identify examinees who exhibit midpoint and extreme response styles. Findings suggest that researchers and practitioners who aim to identify response styles as a type of construct-irrelevant variance can do so using measurement models such as the PCM.

Purpose: This article provides a tutorial introduction to ordinal pattern analysis, a statistical analysis method designed to quantify the extent to which hypotheses of relative change across experimental conditions match observed data at the level of individuals. This method may be a useful addition to familiar parametric statistical methods including repeated measures analysis of variance and generalized linear mixed-effects models, particularly when analyzing inherently individual characteristics, such as perceptual processes, and where experimental effects are usefully modeled in relative rather than absolute terms. Method: Three analyses of increasing complexity are demonstrated using ordinal pattern analysis. An initial analysis of a very small data set is designed to explicate the simple mathematical calculations that make up ordinal pattern analysis, which can be performed without the aid of a computer. Analyses of slightly larger data sets are used to demonstrate familiar concepts, including comparison of competing hypotheses, handling missing data, group comparisons, and pairwise tests. All analyses can be reproduced using provided code and data. Results: Ordinal pattern analysis results are presented, along with an analogous linear mixed-effects analysis, to illustrate the similarities and differences in information provided by ordinal pattern analysis in comparison to familiar parametric methods. Conclusion: Although ordinal pattern analysis does not produce familiar numerical effect sizes, it does provide highly interpretable results in terms of the proportion of individuals whose results are consistent with a hypothesis, along with individual and group-level statistics, which quantify hypothesis performance.

Survey score interpretations are often plagued by sources of construct-irrelevant variation, such as response styles. In this study, we propose the use of an IRTree Model to account for response styles by making use of self-report items and anchoring vignettes. Specifically, we investigate how the IRTree approach with anchoring vignettes compares to traditional approaches that either do not include anchoring vignettes or do not account for response styles. We analyze secondary data using four different models: 1) total score; 2) graded response model; 3) IRTree without the consideration of anchoring vignettes, and 4) IRTree considering anchoring vignettes. We found significant differences in trait estimates from models that account for response styles compared to those that do not. Additionally, we found differences in trait estimates between the IRTree Models when considering anchoring vignettes and when not. Model comparisons suggest that trait differences are due to adjusting for acquiescence response style.