Ordinal Factor Research Articles

Factor analysis of the feeding and eating in AutiSm Together Assessment

BackgroundAutistic children experience high rates of feeding challenges, negatively affecting their.health, eating patterns, and social eating experiences. The Feeding and Eating in AutiSm Together; FEAST, a caregiver-report questionnaire, was developed to characterize feeding challenges in autistic children ages 2 to 12 years. The purpose of this study is to determine the construct validity of the FEAST assessment using a large national sample of autistic children. MethodA large national sample (N = 396) of caregivers completed the online survey. An ordinal confirmatory factor model using a robust weighted least squares approach validated the four predicted groupings of feeding challenges (sensory processing; behavior; oral.motor skills; and gastrointestinal) with generally strong loadings onto the four distinct.factors. ResultsModel fit statistics indicated good model fit (Robust Root Mean Square Error of.Approximation=.080, Robust Comparative Fit Index =.909, Robust Tucker-Lewis.Index=.901) Findings provide preliminary confirmation of the construct validity of the.FEAST assessment to characterize feeding difficulties for autistic children ages 2–12. Small to medium correlations between all factors indicate items successfully targeted.four unique areas of feeding challenge. For all factors, the largest correlation was with.behavior. ConclusionAssessments developed to address the complex feeding and mealtime challenges for.autistic children are essential to identify and integrate the most appropriate intervention.approaches. FEAST offers researchers and clinicians with a comprehensive and psychometrically sound assessment tool to address autistic children’s unique feeding.challenges.

Automated recommendation model using ordinal probit regression factorization machines

Automated recommendation model using ordinal probit regression factorization machines

Factor analyzing ordinal items requires substantive knowledge of response marginals.

In the social sciences, measurement scales often consist of ordinal items and are commonly analyzed using factor analysis. Either data are treated as continuous, or a discretization framework is imposed in order to take the ordinal scale properly into account. Correlational analysis is central in both approaches, and we review recent theory on correlations obtained from ordinal data. To ensure appropriate estimation, the item distributions prior to discretization should be (approximately) known, or the thresholds should be known to be equally spaced. We refer to such knowledge as substantive because it may not be extracted from the data, but must be rooted in expert knowledge about the data-generating process. An illustrative case is presented where absence of substantive knowledge of the item distributions inevitably leads the analyst to conclude that a truly two-dimensional case is perfectly one-dimensional. Additional studies probe the extent to which violation of the standard assumption of underlying normality leads to bias in correlations and factor models. As a remedy, we propose an adjusted polychoric estimator for ordinal factor analysis that takes substantive knowledge into account. Also, we demonstrate how to use the adjusted estimator in sensitivity analysis when the continuous item distributions are known only approximately. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

The Instrumental Motivation of Teachers: Implications of High-Stakes Accountability for Professional Learning

ABSTRACT This article considers the motivations of teachers to pursue ongoing professional learning. During recent decades, the international policy context has been characterised by high-stakes accountability, but the implications of this agenda for teachers’ motivations toward professional learning remains under-explored. In this mixed methods study, combining a large teacher survey and in-depth teacher interviews, a new and significant concept of ‘instrumental motivation’ is generated to capture how high-stakes performance management policies damage the motivation of teachers to learn professionally. This innovative approach, employing ordinal factor analysis and inductive/deductive hybrid thematic analysis inclusive of self-determination theory, reveals how the ‘instrumental motivation’ of teachers should be constrained and argues for the adoption of alternative motivational strategies to support effective professional learning in schools.

Cross-sectional evaluation of an asynchronous multiple mini-interview (MMI) in selection to health professions training programmes with 10 principles for fairness built-in

ObjectivesWe aimed to explore the psychometric properties of the first known online asynchronous multiple mini-interview (MMI) designed for fairness with subgroup analyses by key characteristics, usability and acceptability.DesignCross-discipline multimethod evaluation.SettingOne...

An analysis of urban residents’ efficiency in paying for water services delivered by municipalities: A case of Chinhoyi Town, Zimbabwe

Chinhoyi urban local authority is failing to deliver obligated services such as water provision to residents, citing unavailability of funds due to non-payment by the residents as the main cause. This research aims at establishing whether or not this deterioration in water provision is indeed due to non-payment by citizens. Ten households were randomly selected from each of the sixteen residential areas in Chinhoyi. Each of the households was given a questionnaire to complete and the responses were analysed using Limdep and SPSS. The efficiency of residents in paying for water services was estimated using the Stochastic Frontier method. The factors that were used in the model were either quantitative or ordinal-qualitative. A likert-scale with ratings 1 to 5 was used for the ordinal factors. Results show that residents are paying for the water services provided by the municipality, suggesting that the poor supply of water by the Chinhoyi municipality is not due to non-payment by citizens. These results imply that revenue being collected from residents is either not sufficient for the provision of water or has a part that is being directed elsewhere. It was also established that water quantity, amount charged, household income, opportunism and inability significantly affect the amount of revenue paid to the municipality by the residents. The effect of opportunism and inability was, however, negative. Implying that an increase in each of the two factors reduces both the amount that a household will pay and the efficiency of the individual household.

A direct comparison of two measures of ordinal knowledge among 8-year-olds

Children’s knowledge of the ordinal relations among number symbols is related to their mathematical learning. Ordinal knowledge has been measured using judgment (i.e., decide whether a sequence of three digits is in order) and ordering tasks (i.e., order three digits from smallest to largest). However, the question remains whether performance on these two ordinal tasks tap into similar cognitive processes. Canadian children (N = 87; Age M = 8.7 years, Grade 3) completed symbolic number tasks (i.e., number comparison, ordering, and order judgment) and measures of arithmetic fluency (i.e., addition and subtraction) and working memory (i.e., digit span backward). For both ordinal tasks, there was a reverse distance effect for ordered sequences such that children responded faster to adjacent than to non-adjacent sequences (e.g., 2 3 4 vs. 4 7 9) and a canonical distance effect for unordered sequences such that children responded faster to non-adjacent than to adjacent sequences (e.g., 4 2 3 vs. 4 9 7). Working memory and number comparison each predicted unique variance in the ordinal measures (ordering, order judgment, and a latent ordinal factor based on the two measures). Furthermore, ordinal skills superseded the role of number comparison as the key predictor of arithmetic, controlling for children’s gender and working memory skills. In summary, although both ordering and order judgment tasks index ordinal knowledge, a latent factor that excludes task-specific error may be a better index than either task separately.

Feasibility, validity and reliability of the ASCOT-Proxy and ASCOT-Carer among unpaid carers of people living with dementia in England

BackgroundPeople with dementia living at home represent a growing group of social care services users in England. Many are unable to complete questionnaires due to cognitive impairment. The ASCOT-Proxy is an adapted version of an established measure, ASCOT, which was developed as a way of collecting social care-related quality of life (SCRQoL) data from this group of service users, either alone or alongside the ASCOT-Carer, a measure of SCRQoL for unpaid carers. The ASCOT-Proxy includes two perspectives, the proxy-proxy perspective (‘My opinion: What I think’) and proxy-person perspective (‘What I think the person I represent thinks’). We aimed to establish the feasibility, construct validity and reliability of the ASCOT-Proxy and ASCOT-Carer, with unpaid carers of people with dementia living at home unable to self-report. We also aimed to establish structural characteristics of the ASCOT-Proxy.MethodsCross-sectional data were collected using self-administered questionnaire (paper or online) among unpaid carers living in England between January 2020 and April 2021. Unpaid carers could take part if they supported someone living with dementia who was unable to self-complete a structured questionnaire. The person living with dementia or their unpaid carer had to use at least one social care service. We used the proportion of missing data to establish feasibility, ordinal exploratory factor analysis to establish structural characteristics, Zumbo’s ordinal alpha for internal reliability, and hypothesis testing for construct validity. We also conducted Rasch analysis.ResultsWe analysed data for 313 carers (62.4(± 12.0) years, 75.7% (N=237) females). We were able to calculate the ASCOT-Proxy-proxy overall score for 90.7% of our sample, the ASCOT-Proxy-person overall score for 88.8% of our sample and in case of the ASCOT-Carer for 99.7% of our sample. As there was an issue with structural characteristics of the ASCOT-Proxy-proxy we conducted Rasch, reliability and construct validity analysis for the ASCOT-Proxy-person and ASCOT-Carer only.ConclusionsThis was a first study to explore psychometric characteristics of the ASCOT-Proxy and ASCOT-Carer with unpaid carers of people with dementia living at home unable to self-report. There are some aspects of the psychometric characteristics of the ASCOT-Proxy and ASCOT-Carer that warrant further investigation in future.Trial registrationNA

Ordinal consistency based matrix factorization model for exploiting side information in collaborative filtering

In designing modern recommender systems, item feature information (or side information) is often ignored as most models focus on exploiting rating information. However, the side information is equally essential for capturing users' interests in items. Also, the recommender systems that use side information partially process the feature information by ignoring the locality-preserving property of item features. This study proposes an approach for collaborative filtering by applying an ordinal consistency-based matrix factorization (MF) model to maintain the locality-preserving property of item features to counter this problem. The ordinal consistency condition is implied using a loss function to the item features. Using MF removes the redundancy and inconsistency in item features, producing good results in calculating similarity information for recommendations. We have used five benchmark datasets to evaluate and compare the proposed model. Results obtained using the experiments suggest significant improvement in performance compared to related baselines.

Evaluating Close Fit in Ordinal Factor Analysis Models With Multiply Imputed Data.

Multiple imputation (MI) is one of the recommended techniques for handling missing data in ordinal factor analysis models. However, methods for computing MI-based fit indices under ordinal factor analysis models have yet to be developed. In this short note, we introduced the methods of using the standardized root mean squared residual (SRMR) and the root mean square error of approximation (RMSEA) to assess the fit of ordinal factor analysis models with multiply imputed data. Specifically, we described the procedure for computing the MI-based sample estimates and constructing the confidence intervals. Simulation results showed that the proposed methods could yield sufficiently accurate point and interval estimates for both SRMR and RMSEA, especially in conditions with larger sample sizes, less missing data, more response categories, and higher degrees of misfit. Based on the findings, implications and recommendations were discussed.

Management of material resources in the context of ensuring the sustainability of the agroindustrial complex

Substantiated are approaches to improving the processes of managing material resources, focused on the formation of an integrated management system in the context of increasing the efficiency of agricultural production. The provisions of the scientific and theoretical (the target model of material resources management in agriculture, taking into account the interrelation of the subjects of the national economy, the evolution of the national economy and the conditions for making managerial decisions in terms of material resources; the general scheme of ordinal factors of state regulation of direct and indirect influence on the effectiveness of the use of material resources in agriculture) and methodological (improvement blocks the current practice of managing material resources in the context of ensuring the financial and investment stability of the agroindustrial complex) character.

Heywood Cases in Unidimensional Factor Models and Item Response Models for Binary Data.

Heywood cases are known from linear factor analysis literature as variables with communalities larger than 1.00, and in present day factor models, the problem also shows in negative residual variances. For binary data, factor models for ordinal data can be applied with either delta parameterization or theta parametrization. The former is more common than the latter and can yield Heywood cases when limited information estimation is used. The same problem shows up as non convergence cases in theta parameterized factor models and as extremely large discriminations in item response theory (IRT) models. In this study, we explain why the same problem appears in different forms depending on the method of analysis. We first discuss this issue using equations and then illustrate our conclusions using a small simulation study, where all three methods, delta and theta parameterized ordinal factor models (with estimation based on polychoric correlations and thresholds) and an IRT model (with full information estimation), are used to analyze the same datasets. The results generalize across WLS, WLSMV, and ULS estimators for the factor models for ordinal data. Finally, we analyze real data with the same three approaches. The results of the simulation study and the analysis of real data confirm the theoretical conclusions.

Reproducibility of the brief religious coping inventory with African athletes' sample using ordinal factor analytical approach.

Previous studies have revealed that religious coping strategy is common among athletes due to the stressful experiences before and during competitions as part of the mental preparations they go through, the uncertainty of sporting outcomes, and other organizational issues they encounter. This research assessed the reproducibility of the Brief Religious Coping (RCOPE) instrument in an African setting using athletes' samples from different countries. Particularly, the research sought to assess the (1) factor structure of Brief RCOPE with an African sample, (2) construct validity of the RCOPE measure, and (3) measurement invariance of the RCOPE instrument based on gender and nationality. The study surveyed a convenient sample of 300 athletes, including 164 male and 136 female athletes, from 3 African countries (Benin, Ghana, and Nigeria) who participated in the 2018 West African University Games. The Brief RCOPE instrument was administered to the athletes for validation purposes before the competition. Exploratory and confirmatory factor analyses were conducted using the ordinal factor analytic approach. This validation study confirmed the two-factor dimension (positive and negative religious coping) of the Brief RCOPE measure. Further, all items for each of the dimensions of the inventory contributed significantly to the measure of the Brief RCOPE domains. The positive and negative religious coping dimensions contributed more than half of the variance of their respective indicators. Measurement invariance across gender and nationality was confirmed. Sufficient evidence was gathered to support the interpretation and use of the Brief RCOPE measure. Coaches and sports psychologists could adopt the Brief RCOPE measure to understand the mental or thought patterns of religious athletes based on existential concerns or stress accrued from impending competitions to inform appropriate religious coping interventions. This notwithstanding, the Minimum Clinical Important Difference (MCID) of the Brief RCOPE should be further investigated to enhance the utility of the instrument for use in intervention-based studies.

The reward and punishment responsivity and motivation questionnaire (RPRM-Q): A stimulus-independent self-report measure of reward and punishment sensitivity that differentiates between responsivity and motivation.

Reward and punishment sensitivity seem important traits in understanding behavior in general and psychopathology in particular. Though the definitions used for reward and punishment sensitivity differentiate between responsivity and motivation, the measures thus far used to assess these constructs do not. Further, specificity of the type of reward (e.g., drugs) and punishment (e.g., spiders) in questionnaires might result in measurement bias especially when examining the relationship with psychopathology. Therefore, we developed a stimulus-independent multidimensional questionnaire of reward and punishment sensitivity that differentiates between responsivity and motivation. This study addresses the psychometric qualities of this newly developed reward and punishment responsivity and motivation questionnaire (RPRM-Q). On the basis of exploratory ordinal factor analysis (N = 273) that was used to examine the quality of the initial pool of 39 items, the number of items was reduced to 18. Confirmatory ordinal factor analysis on the remaining items in an independent sample (N = 328) supported a 18-item four-factor model, and showed acceptable to good internal reliability. The relationship between the subscales of the RPRM-Q and often used questionnaires was examined in the combined sample (N = 601), which showed some first support for the ability of the new questionnaire to differentiate between responsivity and motivation to approach/avoid. The findings indicate that the RPRM-Q might be a helpful instrument to further test the relevance of punishment and reward sensitivity in psychopathology.

Statistical Modelling of Road Traffic KSI Car Accidents in England (STATS19)

Several generalised linear models for counts (i.e., Poisson Model) as well as for binary response (i.e., Binary Logistic Model) and ordinal response (i.e., Ordinal Logistic Model) depending on selected multiple explanatory factors (discrete/ categorical) were developed for the road KSI car accidents in England based on STATS19 data (that were manipulated and several new factors were created), after exploratory exploration of discrete/ dichotomous/ nominal/ ordinal factors applied graphical EDA techniques followed by univariate ANOVA/ ANCOVA as well as MANOVA/ MANCOVA based on same selected multiple explanatory factors. Only the main effects as well as two-way interactions were investigated. Majority of main effects and several interaction effects in GLM models were found statistically significant with greater or lesser likelihood of having consequences. The statistically significant KSI car accident factors were identified and quantified for leading to aims to reduce as well as to prevent the car accident, particularly the killed or seriously injured car accidents. It also leads to inform the policymakers on how best to reduce the number and severity of car crashes.

On the Bias in Confirmatory Factor Analysis When Treating Discrete Variables as Ordinal Instead of Continuous

Confirmatory factor analysis is some of the most widely used statistical techniques in the social sciences. Frequently, variables (i.e., items) stemming from questionnaires are analyzed. Two competing approaches for estimating confirmatory factor analysis can be distinguished. First, ordinal variables could be treated as in the case of continuous variables using Pearson correlations, and maximum likelihood estimation method would be applied. Second, an ordinal factor analysis based on polychoric correlations can be fitted. In the majority of the psychometric literature, there is a preference for the ordinal factor analysis based on polychoric correlations because the continuous treatment of variables results in biased factor loadings and biased factor correlations. This article argues that it is not legitimate to speak about bias when comparing the two competing factor analytic approaches because it depends on how true model parameters are defined. This decision can be made individually by a researcher. It is shown in simulation studies and analytical derivations that treating variables ordinally using polychoric correlations instead of continuous using Pearson correlations can also lead to biased estimates of factor loadings and factor correlations. Consequently, it should only be stated that different model parameters are defined in a continuous and an ordinal treatment, and one approach should not generally be preferred over the other.

Missed Opportunities? An Observational Analysis of Lung Cancer Screening Utilization Amongst Patients With Lung Cancer.

Lung cancer (LC) is the leading cause of cancer-related deaths worldwide. The U.S. Preventive Services Task Force (USPSTF) and National Comprehensive Cancer Network (NCCN) recommend annual low-dose CT chest (LDCT) for LC screening in high-risk adults who meet appropriate criteria, which primarily focus on age and smoking history. Despite this, screening rates remain low and patients with LC are typically diagnosed at a later stage.We conducted a single-center retrospective analysis of patients with an established diagnosis of lung cancer to evaluate if screening guidelines were appropriately followed before the cancer diagnosis.Patients diagnosed with LC between 2016 and 2019 were included in the analysis. Charts were reviewed for demographics, detailed smoking history, as well as histology and stage of LC. Associations between categorical factors and screening were examined using the chi-square test. Associations between continuous and ordinal factors and screening were examined using the Mann–Whitney test.A total of 530 charts were reviewed, of which 52% met NCCN criteria and 35% met USPSTF criteria. Only 4.0% and 4.8% of patients who met NCCN and USPSTF criteria, respectively, underwent screening. There was a significant association between staging at diagnosis and screening with LDCT. All the patients who had screening CT scans were diagnosed at localized stages of lung cancer in both NCCN and USPSTF groups compared to 49.1% and 48% in eligible subjects that did not undergo screening, respectively.Our study showed that despite established guidelines for LC screening and insurance coverage, a vast majority of screening-eligible LC patients have never had LDCT. We found that patients who underwent screening as per guidelines were diagnosed at earlier stages of the disease. Ongoing efforts to increase awareness and adherence to LC screening guidelines are needed to improve early detection and reduce LC mortality.

The Measurement Properties and Acceptability of a New Parent-Infant Bonding Tool ('Me and My Baby') for Use in United Kingdom Universal Healthcare Settings: A Psychometric, Cross-Sectional Study.

IntroductionThe National Institute for Health and Care Excellence (NICE) guidelines acknowledge the importance of the parent–infant relationship for child development but highlight the need for further research to establish reliable tools for assessment, particularly for parents of children under 1 year. This study explores the acceptability and psychometric properties of a co-developed tool, ‘Me and My Baby’ (MaMB).Study designA cross-sectional design was applied. The MaMB was administered universally (in two sites) with mothers during routine 6–8-week Health Visitor contacts. The sample comprised 467 mothers (434 MaMB completers and 33 ‘non-completers’). Dimensionality of instrument responses were evaluated via exploratory and confirmatory ordinal factor analyses. Item response modeling was conducted via a Rasch calibration to evaluate how the tool conformed to principles of ‘fundamental measurement’. Tool acceptability was evaluated via completion rates and comparing ‘completers’ and ‘non-completers’ demographic differences on age, parity, ethnicity, and English as an additional language. Free-text comments were summarized. Data sharing agreements and data management were compliant with the General Data Protection Regulation, and University of York data management policies.ResultsHigh completion rates suggested the MaMB was acceptable. Psychometric analyses showed the response data to be an excellent fit to a unidimensional confirmatory factor analytic model. All items loaded statistically significantly and substantially (>0.4) on a single underlying factor (latent variable). The item response modeling showed that most MaMB items fitted the Rasch model. (Rasch) item reliability was high (0.94) yet the test yielded little information on each respondent, as highlighted by the relatively low ‘person separation index’ of 0.1.Conclusion and next stepsMaMB reliably measures a single construct, likely to be infant bonding. However, further validation work is needed, preferably with ‘enriched population samples’ to include higher-need/risk families. The MaMB tool may benefit from reduced response categories (from four to three) and some modest item wording amendments. Following further validation and reliability appraisal the MaMB may ultimately be used with fathers/other primary caregivers and be potentially useful in research, universal health settings as part of a referral pathway, and clinical practice, to identify dyads in need of additional support/interventions.

Deciphering Feature Effects on Decision-Making in Ordinal Regression Problems: An Explainable Ordinal Factorization Model

Ordinal regression predicts the objects’ labels that exhibit a natural ordering, which is vital to decision-making problems such as credit scoring and clinical diagnosis. In these problems, the ability to explain how the individual features and their interactions affect the decisions is as critical as model performance. Unfortunately, the existing ordinal regression models in the machine learning community aim at improving prediction accuracy rather than explore explainability. To achieve high accuracy while explaining the relationships between the features and the predictions, we propose a new method for ordinal regression problems, namely the Explainable Ordinal Factorization Model (XOFM). XOFM uses piecewise linear functions to approximate the shape functions of individual features, and renders the pairwise features interaction effects as heat-maps. The proposed XOFM captures the nonlinearity in the main effects and ensures the interaction effects’ same flexibility. Therefore, the underlying model yields comparable performance while remaining explainable by explicitly describing the main and interaction effects. To address the potential sparsity problem caused by discretizing the whole feature scale into several sub-intervals, XOFM integrates the Factorization Machines (FMs) to factorize the model parameters. Comprehensive experiments with benchmark real-world and synthetic datasets demonstrate that the proposed XOFM leads to state-of-the-art prediction performance while preserving an easy-to-understand explainability.

SNPxE: SNP-environment interaction pattern identifier

BackgroundInteractions of single nucleotide polymorphisms (SNPs) and environmental factors play an important role in understanding complex diseases' pathogenesis. A growing number of SNP-environment studies have been conducted in the past decade; however, the statistical methods for evaluating SNP-environment interactions are still underdeveloped. The conventional statistical approach with a full interaction model with an additive SNP mode tests one specific interaction type, so the full interaction model approach tends to lead to false-negative findings. To increase detection accuracy, developing a statistical tool to effectively detect various SNP-environment interaction patterns is necessary.ResultsSNPxE, a SNP-environment interaction pattern identifier, tests multiple interaction patterns associated with a phenotype for each SNP-environment pair. SNPxE evaluates 27 interaction patterns for an ordinal environment factor and 18 patterns for a categorical environment factor. For detecting SNP-environment interactions, SNPxE considers three major components: (1) model structure, (2) SNP’s inheritance mode, and (3) risk direction. Among the multiple testing patterns, the best interaction pattern will be identified based on the Bayesian information criterion or the smallest p-value of the interaction. Furthermore, the risk sub-groups based on the SNPs and environmental factors can be identified. SNPxE can be applied to both numeric and binary phenotypes. For better results interpretation, a heat-table of the outcome proportions can be generated for the sub-groups of a SNP-environment pair.ConclusionsSNPxE is a valuable tool for intensively evaluate SNP-environment interactions, and the SNPxE findings can provide insights for solving the missing heritability issue. The R function of SNPxE is freely available for download at GitHub (https://github.com/LinHuiyi/SIPI).