Classical Item Research Articles

Improved Diversity-Promoting Collaborative Metric Learning for Recommendation.

Collaborative Metric Learning (CML) has recently emerged as a popular method in recommendation systems (RS), closing the gap between metric learning and collaborative filtering. Following the convention of RS, existing practices exploit unique user representation in their model design. This paper focuses on a challenging scenario where a user has multiple categories of interests. Under this setting, the unique user representation might induce preference bias, especially when the item category distribution is imbalanced. To address this issue, we propose a novel method called Diversity-Promoting Collaborative Metric Learning (DPCML), with the hope of considering the commonly ignored minority interest of the user. The key idea behind DPCML is to introduce a set of multiple representations for each user in the system where users' preference toward an item is aggregated by taking the minimum item-user distance among their embedding set. Specifically, we instantiate two effective assignment strategies to explore a proper quantity of vectors for each user. Meanwhile, a Diversity Control Regularization Scheme (DCRS) is developed to accommodate the multi-vector representation strategy better. Theoretically, we show that DPCML could induce a smaller generalization error than traditional CML. Furthermore, we notice that CML-based approaches usually require negative sampling to reduce the heavy computational burden caused by the pairwise objective therein. In this paper, we reveal the fundamental limitation of the widely adopted hard-aware sampling from the One-Way Partial AUC (OPAUC) perspective and then develop an effective sampling alternative for the CML-based paradigm. Finally, comprehensive experiments over a range of benchmark datasets speak to the efficacy of DPCML.

Adaptation and Validation of a Scale of Self-Regulation of Learning in Mathematical Problem Solving

[Objective] This work aims to adapt and validate a scale that measures the level of self-regulation in mathematical problem-solving contexts of teachers in initial training in primary education. [Methodology] For sample selection, non-probabilistic convenience or incidental sampling was used. A total of 269 first-year primary education undergraduate degree students at the University of the Basque Country (UPV/EHU) in the 2020–2021 academic year participated in the validation process of the adapted scale. The type of research conducted is descriptive. Data were analyzed using the SPSS statistical software. [Results] The Cronbach's alpha coefficient value of 0.884 revealed a consistent internal reliability. Likewise, to test the factor structure of the scale, an exploratory factor analysis was conducted, obtaining a Kaiser-Meyer-Olkin (KMO) index of 0.836 > 0.7 and a p= 0.000<0.05 in the Bartlett test of sphericity. It also indicated a seven-factor structure: students' perceptions of their ability and how these perceptions influence self-regulation of the resolution process; ethics, problem-solving and personal growth; attitude towards the statement; negative self-efficacy beliefs; external causal attribution; problem-solving methods; and the social environment. [Conclusions] The latent structure provided by the exploratory factor analysis agrees with the classification of items established a priori and allows for the orientation of the confirmatory factor analysis that provides continuity to the research.

Adaptation of the Diet Quality Questionnaire as a global public good for use in 140 countries

BackgroundFood group consumption data is useful for measuring and monitoring diet quality. To collect valid data across contexts, consistent and rigorous adaptation of survey questions is needed. ObjectivesThe objective of this research is to adapt food group consumption survey questions for 140 countries, by identifying the most common (sentinel) food items in each food group using a structured, participatory process and global standards for classification. MethodsSurvey questions were adapted for 29 food groups of the Diet Quality Questionnaire (DQQ) and for additional questions for infant and young child feeding (IYCF) indicators. For each country, adaptation comprised: (1) review of existing questionnaires, dietary intake data, and other information to draft food lists; (2) key informant (KI) interviews with 5-12 experts to identify and prioritize sentinel items including terminology; (3) comparison of items across countries within the same region to identify discrepancies, and follow-up with KIs to resolve them. Results1,016 KIs contributed to the adapted DQQs for 140 countries and IYCF DQQs for 96 countries, amounting to approximately 9,550 hours of collective effort (68 person-hours/country on average) from 2020-2024. The process revealed numerous challenges and decisions to ensure consistent classification of items and valid question formulation. ConclusionCountry-specific questions adhering to global standards, and adapted through cumulative and iterative input of local experts, enable the collection of food group consumption data that is valid and comparable across time and geographies. The adapted survey questions have been implemented in the Demographic and Health Surveys (DHS) and Gallup World Poll in 94 countries, generating the first cross-country data on minimum dietary diversity (MDD) and other diet quality indicators. The finalized country-adapted DQQs and IYCF DQQs were translated to 143 national languages and are published online as a global public good for population-level diet quality measurement.

Riverbank plastic distributions and how to sample them

As plastic pollution exists in aquatic ecosystems globally, monitoring its abundance and distribution has become crucial for understanding transport pathways, sources, sinks, and impacts. Riverbanks are accumulation zones for plastic, but the selection of monitoring methods is constrained by research goals, available resources, and site-specific conditions. This diversity in approaches has led to disparate datasets, highlighting the need for standardized monitoring protocols. Here, we study the spatial distribution of plastic at the riverbank scale, quantify the uncertainty of existing riverbank methods, and provide recommendations for improved monitoring based on the balance between uncertainty loss and increase in effort. We measured riverbank plastic abundance at eight Dutch riverbanks, categorizing the items using 108 item categories (River-OSPAR). For every riverbank, an area of 100 by 25 meters was subdivided into five-by-five-meter squares, resulting in 100 individual monitored sub-areas. We found riverbank plastic exhibited high spatial variability, with deposition patterns ranging from parallel to the waterline to clustered, random, or uniform (Moran’s I between -0.050 and 0.301). Individual measurements from diverse sampling protocols are 5-49 times less accurate than estimates derived from extensive sampling, highlighting the diminishing impact of specific methods with increased data collection. Lastly, our findings suggest that increasing the sampling area quickly reaches diminishing returns in terms of accuracy. Reducing the sampled area by 80% only increases the uncertainty in estimating the true plastic density by 20%. While standardized protocols are essential for data comparability, a rigid, uniform sampling approach may be less efficient and resource-intensive than a flexible (step-wise) strategy that adapts to local conditions. By demonstrating that extensive sampling can mitigate the differences between unique sampling protocols, this study promotes a shift towards flexible and efficient riverbank plastic monitoring, ultimately accelerating global efforts to combat plastic pollution.

A Multicriteria Customer Classification Method in Supply Chain Management

Since Kraljic’s strategic matrix was applied to supply chain management, classification of items, suppliers, and customers has become of increasing interest to research and companies. The aim of this research is to develop an easily interpretable multicriteria classification matrix method and validate it in real-world scenarios with a robustness analysis. This method assigns alternatives to one of four classes defined by critical dimensions that integrate several evaluation criteria. Initially, a global search pre-classifies the alternatives using the PROMETHEE net flows. Then, two local searches are carried out that make use of the discriminant properties of the net flow signs to improve the quality of the assignments. This approach is specifically applied to pre-classified alternatives near the boundary between two or more categories. The method has been validated by segmenting thousands of customers. Four customer segments were identified: strategic, collaborative, transactional, and non-preferred. A comparison was made between the results and those derived from an alternative method. Through an extensive sensitivity analysis, the proposed method was shown to be robust to parameter variation, highlighting its reliability in real dynamic contexts. The method provides valuable, easily interpretable information, which constitutes the basis for developing personalised strategies to enhance customer relationship management.

Category-guided multi-interest collaborative metric learning with representation uniformity constraints

Multi-interest collaborative metric learning has recently emerged as an effective approach to modeling the multifaceted interests of a user in recommender systems. However, two issues remain unexplored. (1) There is no explicit guidance for the matching of an item against multiple interest vectors of a user. (2) The desired property of item representations with respect to their categories is overlooked, resulting in that different categories of items are mixed up in the latent space. To overcome these issues, we devise a Category-guided Multi-interest Collaborative Metric Learning model (CMCML) with representation uniformity constraints. CMCML is designed as a novel category-guided Mixture-of-Experts (MoE) architecture, where the gating network leverages the item category to guide the matching of an item against multiple interest vectors of a user, encouraging items with the same category to approach the same interest vector. In addition, we design a user multi-interest uniformity loss and a category-aware item uniformity loss: The former aims to avoid representation degeneration by enlarging the difference among multiple interest vectors of the same user; the latter is tailored to push different categories of items apart in the latent space. Quantitative experiments on Ciao, Epinions and TaFeng demonstrate that our CMCML improves the value of NDCG@20 by 12.41%, 10.89% and 10.39% respectively, compared to other state-of-the-art collaborative metric learning methods. Further qualitative analyses reveal that our CMCML yields a better representation space where items from distinct categories are arranged in different regions with high density.

Impact of discounts on sales in Delhi NCR: a survey report

Background: Delhi NCR, which includes Delhi and its neighbouring areas, has experienced considerable growth in the shopping sector over the past decade. A major factor that draws customers to these shopping destinations is the opportunity for bargaining and discounts. This paper examines the impact of discounts on the sales of various item categories. Methods: A survey has been conducted to collect data. Along with focus on responses to a structured questionnaire, the study also analyses sales across various groups of items, with each group divided into two categories: one offering discounts and the other not. Results: findings of this study have the potential to shed light on the fact that offering discounts is beneficial and finally leads to increase in sales as it attracts more customers. For most the groups an increase in sales has been witnessed. Conclusions: From the findings of the study, it can be concluded that if discounts are offered, more customers are attracted which eventually leads to increase in sales day by day.

Reclassification of items in the Leicester Cough Questionnaire: Correlation analysis

BackgroundThe Leicester Cough Questionnaire (LCQ) is a validated tool for assessing cough that has three domains (physical, psychological, and social), with eight, seven, and four items, respectively. However, the assigned domain may not accurately reflect the characteristics of an item. This study aimed to reclassify the items in the Korean version of the LCQ (K-LCQ) to improve the coherence in each domain. MethodsData of patients with chronic cough from 16 centers who completed the K-LCQ were retrospectively analyzed. Spearman’s rank correlation analysis was used to assess the correlations between items and their domains. Principal component analysis was performed to recategorize the K-LCQ items. ResultsThe correlation analysis of the data from 255 patients demonstrated that certain items such as tiredness, embarrassment, and interference with daily work or overall life enjoyment showed strong or very strong correlations with all three domains. Cough bout frequency showed the weakest correlation with the physical domain, despite being included in that domain, and had stronger correlations with the psychological and social domain. The principal component analysis led to the reclassification of six items: one from the physical to psychological, two from the social to psychological, and three from the psychological to social domain. The within-domain correlation coherence was higher in the new classification than in the original. Validation using an independent cohort of 203 patients yielded similar results. ConclusionsThe new classification of the K-LCQ items showed improved within-domain correlation coherence.

Slow and steady—small, but insufficient, changes in food and drink availability after four years of implementing a healthy food policy in New Zealand hospitals

BackgroundA voluntary National Healthy Food and Drink Policy (the Policy) was introduced in public hospitals in New Zealand in 2016. This study assessed the changes in implementation of the Policy and its impact on providing healthier food and drinks for staff and visitors in four district health boards between 1 and 5 years after the initial Policy introduction.MethodsRepeat, cross-sectional audits were undertaken at the same eight sites in four district health boards between April and August 2017 and again between January and September 2021. In 2017, there were 74 retail settings audited (and 99 in 2021), comprising 27 (34 in 2021) serviced food outlets and 47 (65 in 2021) vending machines. The Policy’s traffic light criteria were used to classify 2652 items in 2017 and 3928 items in 2021. The primary outcome was alignment with the Policy guidance on the proportions of red, amber and green foods and drinks (≥ 55% green ‘healthy’ items and 0% red ‘unhealthy’ items).ResultsThe distribution of the classification of items as red, amber and green changed from 2017 to 2021 (p < 0.001) overall and in serviced food outlets (p < 0.001) and vending machines (p < 0.001). In 2021, green items were a higher proportion of available items (20.7%, n = 815) compared to 2017 (14.0%, n = 371), as were amber items (49.8%, n = 1957) compared to 2017 (29.2%, n = 775). Fewer items were classified as red in 2021 (29.4%, n = 1156) than in 2017 (56.8%, n = 1506). Mixed dishes were the most prevalent green items in both years, representing 11.4% (n = 446) of all items in 2021 and 5.5% (n = 145) in 2017. Fewer red packaged snacks (11.6%, n = 457 vs 22.5%, n = 598) and red cold drinks (5.2%, n = 205 vs 12.5%, n = 331) were available in 2021 compared to 2017. However, at either time, no organisation or setting met the criteria for alignment with the Policy (≥ 55% green items, 0% red items).ConclusionsIntroduction of the Policy improved the relative healthiness of food and drinks available, but the proportion of red items remained high. More dedicated support is required to fully implement the Policy.

Delphi-driven consensus definition for MSCs and clinical reporting guidelines for MSC-based therapeutics

BackgroundDespite promising results in preclinical studies, mesenchymal stromal cells (MSCs) face significant challenges in clinical translation. A scoping review by our group highlighted two key issues contributing to this gap: i) the lack of clear and consensus definition for MSCs and ii) under reporting of critical parameters in MSC clinical studies. To address these issues, we conducted a modified Delphi study to establish and implement a consensus definition for MSCs and develop reporting guidelines for MSC clinical studies. MethodsA steering committee of 22 international experts including stakeholders from different MSC research fields participated in the 3 Delphi rounds. For the first round, to obtain a broad perspective, additional investigators recommended by the steering committee were invited to participate. The first 2 rounds consisted of online surveys, while the 3rd round took the form of a virtual meeting. Participants were asked to rate a series of potential defining characteristics of MSCs and items for reporting guidelines. Consensus was defined as at least 80% of the participants rating the item in the same category of importance. ResultsEighty-seven international participants participated in the first-round survey (Spring 2023), 17 participants in the second online survey (Fall 2023) and 15 participants in the final virtual consensus meeting (January 2024). For the MSC definition, 20 items were considered and 9 reached consensus. Items included terminology (1 item), cell marker expression (5 items), tissue origin (1 item), stemness (1 item) and description of critical quality attributes (1 item). For the reporting guidelines, on the 28 initial items and with additional items suggested during round 1, a total of 33 items to report were included. This included items on MSC intervention group and control (i.e., MSC product, dose, and administration, etc.), MSC characteristics (i.e., MSCs provenance, “fitness and viability, immune compatibility, etc.), and MSC culture conditions (i.e., oxygen environment, culture medium, use of serum, etc.). ConclusionBy applying a Delphi method to establish a consensus definition for MSCs and reporting guidelines for MSC-based clinical, this work represents a significant advance in improving transparency and reproducibility in the conduct and reporting of MSC research.

Breast-cancer specific comprehensive archive of Patient-Reported Outcome Measures (PROMs) for clinical research and clinical practice in oncology: Results from the PRO4All project

BackgroundInclusion of patient-reported outcomes (PROs) in oncology clinical trials is strongly recommended. However, selecting the most appropriate patient-reported outcome measures (PROMs) is not easy. This study aimed to develop a breast cancer (BC) specific comprehensive archive of PROMs. MethodsAs part of the PRO4All project, we identified available PROMs in oncology by searching facit.org, eortc.org, eprovide.mapi-trust.org, PubMed, ema.europa.eu (European Public Assessment Reports) and published reviews. For this analysis, only BC tools were extracted. We described information about PROM name, type of questionnaire, questionnaire variant(s), recall period, number of items, and presence of minimum clinically important difference (MCID) reference in literature. Then, we assigned each item to a specific domain according to a predefined taxonomy of 38 items for outcome classification. ResultsWe identified and analyzed 383 PROMs. Of these, 29 were BC specific, but 2 were excluded because the questionnaires description was not available. 6 (22.2 %) were variants of another questionnaire. All questionnaires were self-reported. In 6 cases (22.2 %) the recall period to consider was the “last week”. The mean number of items per questionnaire was 25.81 (range 6–71). 602 items were assigned to an outcome domain: emotional functioning/wellbeing in 26.6 % of cases, physical functioning in 14.1 %, delivery of care in 10.8 %, and general outcomes in 10.5 %. MCID reference was found only in 4 (14.8 %) cases. ConclusionsThe newly developed archive represents a useful tool to optimize the use of PROMs in the evaluation of treatments in BC patients, promoting a patient-centered approach both in clinical research and practice.

Multi-Granularity and Multi-Interest Contrast-Enhanced Hypergraph Convolutional Networks for Session Recommendation

Aiming at the problems of the incomplete recommendation and sparsity of session data in session recommendation, a new multi-granularity and multi-interest contrast-enhanced hypergraph convolutional network (MGMI-CEHCN) model for session recommendation is proposed. The session data are modeled as a heterogeneous hypergraph, the information is embedded by the hypergraph using two granularities of item price and category, and then the information between different granularities is fused, while the final item embedding is obtained through multi-layer convolution. Finally, an interest perceptron is used to detect multiple potential interests for each item, and a decentralized interest extraction network based on a gated recurrent unit (GRU) is used to integrate the user’s final interests and obtain a global session representation through a soft attention mechanism; a local session representation is generated with the help of a weighted line graph convolutional network. A further joint contrast enhancement strategy is used to maximize the mutual information between the global session representation and local session representation, to improve the recommendation performance. Experiments on several real datasets showed that the recommendation performance of the MGMI-CEHCN model outperformed the current mainstream models. On Cosmetics, the P@20 reached up to 55.25% and M@20 reached up to 38.26%, improvements of 3.06% and 3.09%, respectively; on Diginetica-buy, the P@20 reached up to 65.60% and M@20 reached up to 27.47%, improvements of 2.47% and 6.64%, respectively, which proved the validity of the model.

Development of an evidence-based nursing practice course framework for undergraduate nursing students from a perspective of academic-practice partnerships: A Delphi study

AimTo develop an evidence-based nursing practice course framework for undergraduate nursing students from a perspective of academic-practice partnerships. BackgroundAcademic-practice partnerships could play a crucial role in bridging theory-practice gap in evidence-based nursing education. However, there is a lack of evidence-based nursing practice course framework for undergraduate nursing students from a perspective of academic-practice partnerships. DesignA Delphi study. MethodsThe Delphi study was conducted by literature analysis, qualitative interview and Delphi expert consultation to develop this course framework. 15 and 12 experts were involved in round one and round two of Delphi expert consultation, respectively. The consensus threshold of two-round Delphi expert consultation was described by expert's positive coefficient (≥ 50 %), positive coefficient (≥ 70 %), expert authority coefficient (≥ 0.7), coefficient of variation (<0.25) and full score frequency (>20 %). ResultsThe final course framework includes learning objectives (17 items); teaching chapters of theoretical classes (five items); teaching contents (30 items), methods (12 items) and class hours (9 items) of theoretical classes; and the responsibilities of academic teachers (11 items) and clinical teachers (11 items) in practical classes. ConclusionThe course framework developed by this study could provide guidance for evidence-based nursing education of undergraduate nursing students. The effectiveness of this course framework should be verified through further experimental studies in the future.

Cross-cultural adaptation of the awareness and beliefs about cancer measure for Hispanics/Latinos living in the United States.

The purpose of this study is to culturally adapt the Awareness and Beliefs about Cancer (ABC) measure for use in the Hispanic/Latino population living in the United States (US). In accordance with Patient Reported Outcomes (PRO) Consortium guidelines for cross-cultural adaptation of measures for content and linguistic validity, we conducted: two forward-translations, reconciliation, two back-translations, revision and harmonization, six cognitive interviews, revision, external expert review, and finalization of the version. We used a mixed methods approach, conducting cognitive interviews with Hispanic/Latino community members while also convening an expert panel of six clinicians, health professionals, and community representatives and including the in the entire process. After cross-culturally adapting the ABC measure, we assessed the psychometric properties of the instrument using item response theory analysis. Item parameters, discrimination and category thresholds, and standard errors were calculated. For each of the adapted subdomains, we used item information curves to report the graphical profile of item effectiveness. Twenty-two Hispanic/Latino community members were enrolled in cognitive interviews, and Hispanics/Latinos fluent in Spanish completed the measure to assess its psychometric properties. Cognitive interviews revealed opportunities to improve items. Key changes from the original measure include the inclusion of gender inclusive language and an inquiry into e-cigarette use on items related to smoking habits. Psychometric property analyses revealed that the anticipated delay in seeking medical help, general cancer beliefs, and cancer screening beliefs and behaviors subdomains had some slope parameters that were < 1; this implies that those items were not able to adequately discriminate the latent trait and had poor performance. The adapted ABC measure for US Hispanics/Latinos meets content and linguistic validity standards, with construct validity confirmed for cancer symptom recognition and barriers to symptomatic presentation subdomains, but revisions are necessary for others, highlighting the need for ongoing refinement to ensure the cultural appropriateness of instruments.

Evidence-based guidelines for the interpretation of the 9-item Concise Health Risk Tracking - Self-Report (CHRT-SR9) measure of suicidal risk.

The 9-item Concise Health Risk Tracking - Self-Report (CHRT-SR9) is a widely used patient-reported outcome measure of suicidal risk. The goal of this article is to provide an evidence-based interpretation of the CHRT-SR9 total score in terms of four clinically actionable categories of suicidal risk (none, mild, moderate, and severe). Data from two large programs involving adolescents and adults were combined in this paper. In these studies, the CHRT-SR9 was anchored against an independent measure of suicidal risk, the suicide item (Item #9) in the Patient Health Questionnaire (PHQ-9), with categories 0 (none), 1 (mild), 2 (moderate), and 3 (severe). In the combined data (n = 1945), we calculated the cumulative percentage of data across these four categories and the percentile score of the CHRT-SR9 total score that corresponded to these percentages; from this, we developed ranges of the CHRT-SR9 total score that corresponded to the four categories of Item #9 of PHQ-9. We also calculated similar ranges for two broad subscales of the CHRT-SR9 total score; Propensity and Suicidal Thoughts. To assess the robustness of our findings, we repeated the analysis at another timepoint across studies. Findings indicated that the CHRT-SR9 total score (range: 0-36) can be categorized as none (0-14), mild (15-21), moderate (22-26), and severe (27-36). Similar categories were calculated for the Propensity and Suicidal Thoughts subscales. The findings were the same when repeated at another timepoint. This categorization of the CHRT-SR9 total score can place patients into clinically meaningful and actionable categories of suicidal risk.

교착어적 특성을 활용한 한국어문법교육 연구 - 서울대 한국어 초·중급 교재를 중심으로

This study is developed by proceeding with the contents of “A Study on the Composition and Content Analysis of Korean Textbooks” (2021). Previous studies have shown that the characteristics of the structure of textbooks centered on oral language have led to a lack of linkage with the system of grammar education, and this study is linked to the fact that the problem of grammar education is found in the characteristics of Korean language called agglutinative language. The reason we paid attention to the agglutinative language characteristics was that it was sometimes pointed out as a major type of error by foreign learners, and it was considered suitable for understanding the basic principles. Therefore, by analyzing the target grammar for each level 1 to 4 in units of grammatical morpheme, a grammar category that revealed agglutinative language characteristics was set and classified. The classification items are grammar in which postpositions are combined, grammar in which endings are combined, and other vocabulary, which is based on the grammatical role of the ending or postposition, which is the minimum unit that constitutes grammar. Additionally, the grammar content was investigated by dividing the postposition combination grammar and the ending combination grammar into elementary and intermediate levels. The grammar related to ending combination occupied the highest proportion in the entire levels, confirming that Korean is predicate-oriented and the importance of the ending. The grammar content required for basic sentence composition at the beginner level was a final ending, a connection ending, prefinal ending, and some of transformative ending and auxiliary connective ending, In intermediate level, transformative ending that organizes sentences, auxillary connective ending that adds meaning, and various subordinate connective ending were required as grammar content. For grammar related to postposition combination, a case-marking postpositional particle that constitutes a basic sentence at the beginner level and a auxiliary postpositional particle that adds special meaning at the intermediate level were required as grammar content. It is hoped that the agglutinative language, typological characteristics of the Korean language, will help to select and organize the contents of Korean grammar education.

MACHINE LEARNING ALGORITHM FOR RAPID OBJECT DETECTION BASED ON COLOR FEATURES

The identification of objects is of utmost importance in a wide range of computer vision applications, such as surveillance systems, autonomous cars, and environmental monitoring. Accurate and efficient object recognition methods are crucial in pastoral environments, characterized by the prominent presence of cattle and other objects, to provide effective analysis and decision-making processes. The present study paper introduces an innovative methodology for efficient identification of objects in pastoral landscapes through the utilization of a Colour Feature Extreme Learning Machine (CF-ELM). The CF-ELM method integrates color characteristics with the ELM algorithm to attain enhanced object detection accuracy while preserving computational economy. The experimental findings provide empirical evidence supporting the efficacy and efficiency of the suggested approach in the detection of items within pastoral landscapes. In addition to the CF-ELM, an algorithm for desktop-based categorization of items within a pastoral environment is provided, with individual speeds ranging from 0.05 s to 0.17 s for a single image, evaluated in each color space. The algorithm is intended for usage in scenarios with challenging and variable terrain, making it appropriate for application in agricultural or pastoral settings.

Career Adapt-Abilities Scale: Reliability Generalization Meta-Analysis.

The concept of career adaptability emerged from the broader field of career development theory and has received increasing attention in the past two decades. To measure career adaptability, there are different scale development studies in the literature, but the most widely used and preferred one is the Career Adapt-abilities Scale developed by Savickas and Porfeli. Therefore, in the present study, the general reliability of the Career Adapt-abilities Scale was measured through meta-analysis. One hundred forty nine study (N = 82519) were included in the analyses. For the CAAS overall score, the average reliability coefficient among the 171 reliability estimations was high. However, reliability estimates in the studies included in the research show high heterogeneity. As a result of the moderator analysis, it was concluded that reliability estimates of these scores differ by item level, culture, language, category of items, different forms of CAAS, age, and SD. Regarding the application of the CAAS in research, the consequences of these findings are discussed in light of the relevant literature.

Classification of Known and Unknown Study Items in a Memory Task Using Single-Trial Event-Related Potentials and Convolutional Neural Networks.

This study examines the feasibility of using event-related potentials (ERPs) obtained from electroencephalographic (EEG) recordings as biomarkers for long-term memory item classification. Previous studies have identified old/new effects in memory paradigms associated with explicit long-term memory and familiarity. Recent advancements in convolutional neural networks (CNNs) have enabled the classification of ERP trials under different conditions and the identification of features related to neural processes at the single-trial level. We employed this approach to compare three CNN models with distinct architectures using experimental data. Participants (N = 25) performed an association memory task while recording ERPs that were used for training and validation of the CNN models. The EEGNET-based model achieved the most reliable performance in terms of precision, recall, and specificity compared with the shallow and deep convolutional approaches. The classification accuracy of this model reached 62% for known items and 66% for unknown items. Good overall accuracy requires a trade-off between recall and specificity and depends on the architecture of the model and the dataset size. These results suggest the possibility of integrating ERP and CNN into online learning tools and identifying the underlying processes related to long-term memorization.

Evidence for separate backward recall and n-back working memory factors: a large-scale latent variable analysis

ABSTRACT Multiple studies have explored the factor structure of working memory (WM) tasks, yet few have done so controlling for both the domain and category of the memory items in a single study. In the current pre-registered study, we conducted a large-scale latent variable analysis using variant forms of n-back and backward recall tasks to test whether they measured a single underlying construct, or were distinguished by stimuli-, domain-, or paradigm-specific factors. Exploratory analyses investigated how the resulting WM factor(s) were linked to fluid intelligence. Participants (N = 703) completed a fluid reasoning test and multiple n-back and backward recall tasks containing memoranda that varied across (spatial or verbal material) and within (verbal digits or letters) domain, allowing the variance specific to task content and paradigm to be assessed. Two distinct but related backward recall and n-back constructs best captured the data, in comparison to other plausible model constructions (single WM factor, two-factor domain, and three-factor materials models). Common variance associated with WM was a stronger predictor of fluid reasoning than a residual n-back factor, but the backward recall factor predicted fluid reasoning as strongly as the common WM factor. These data emphasise the distinctiveness between backward recall and n-back tasks.