Different Sets Of Items Research Articles

A narrative review of adaptive testing and its application to medical education.

Adaptive testing has a long but largely unrecognized history. The advent of computer-based testing has created new opportunities to incorporate adaptive testing into conventional programmes of study. Relatively recently software has been developed that can automate the delivery of summative assessments that adapt by difficulty or content. Both types of adaptive testing require a large item bank that has been suitably quality assured. Adaptive testing by difficulty enables more reliable evaluation of individual candidate performance, although at the expense of transparency in decision making, and requiring unidirectional navigation. Adaptive testing by content enables reduction in compensation and targeted individual support to enable assurance of performance in all the required outcomes, although at the expense of discovery learning. With both types of adaptive testing, candidates are presented a different set of items to each other, and there is the potential for that to be perceived as unfair. However, when candidates of different abilities receive the same items, they may receive too many they can answer with ease, or too many that are too difficult to answer. Both situations may be considered unfair as neither provides the opportunity to demonstrate what they know. Adapting by difficulty addresses this. Similarly, when everyone is presented with the same items, but answer different items incorrectly, not providing individualized support and opportunity to demonstrate performance in all the required outcomes by revisiting content previously answered incorrectly could also be considered unfair; a point addressed when adapting by content. We review the educational rationale behind the evolution of adaptive testing and consider its inherent strengths and limitations. We explore the continuous pursuit of improvement of examination methodology and how software can facilitate personalized assessment. We highlight how this can serve as a catalyst for learning and refinement of curricula; fostering engagement of learner and educator alike.

EVALUATING THE PERCEPTION OF MARKETING COMMUNICATION IN POLITICAL MARKETING FROM A GENERATIONAL PERSPECTIVE

The objective of this study is to examine whether there are disparities in the perception of marketing communication in political marketing among female voters based on different generations. This research distinguishes itself from previous studies in the field by focusing on assessing the perceptions of female voters according to their generational cohorts. To achieve this aim, a research model was proposed using a scale derived from existing literature, aligning it with the study's objective. The collected data was analyzed using the SPSS 22 statistical package, employing descriptive statistics to ascertain the socio-demographic characteristics of the participants. Various statistical analyses, including reliability testing, the KMO test, factor analysis, correlation analysis, ONEWAY ANOVA, and post hoc tests, were conducted on the dataset. The results of the factor analysis revealed the presence of three distinct factors, consisting of different sets of items, labeled as “Marketing Communication,” “Advertisement,” and “Political Product.” Based on the conducted analyses, it was concluded that the Baby Boomer generation may exhibit a higher sensitivity compared to other generations towards various advertising channels such as print and digital media, political leaders, and party rallies.

Optimizing inventory control through a data-driven and model-independent framework

Machine learning has shown great potential in various domains, but its appearance in inventory control optimization settings remains rather limited. We propose a novel inventory cost minimization framework that exploits advanced decision-tree based models to approximate inventory performance at an item level, considering demand patterns and key replenishment policy parameters as input. The suggested approach enables data-driven approximations that are faster to perform compared to standard inventory simulations, while being flexible in terms of the methods used for forecasting demand or estimating inventory level, lost sales, and number of orders, among others. Moreover, such approximations can be based on knowledge extracted from different sets of items than the ones being optimized, thus providing more accurate proposals in cases where historical data are scarce or highly affected by stock-outs. The framework was evaluated using part of the M5 competition’s data. Our results suggest that the proposed framework, and especially its transfer learning variant, can result in significant improvements, both in terms of total inventory cost and realized service level.

Graphemes are used when reading: Evidence from Monte Carlo simulation using word norms from mega-studies.

Recently, Chetail (Journal of Memory and Language, 2020) has claimed there is no strong evidence that multi-letter graphemes are used in reading tasks with proficient adult readers, with most studies being statistically weak or having confounds in the stimuli used. Here, I used Monte Carlo simulation with data from reading mega-studies to examine the extent to which the number of multi-letter graphemes matters in words when letter length is held constant. This was done by simulating thousands of experiments using different sets of items for each of a small number of comparisons (e.g., words with only single-letter graphemes versus words with one multi-letter grapheme). The results showed that words with two multi-letter graphemes tended to cause slower reaction times than words with one or no multi-letter graphemes, with effects found in both naming and lexical decision tasks. Interestingly, when words with no multi-letter graphemes were compared with words with one multi-letter grapheme, the differences were much weaker. Simulations of naming results using two computer models, the connectionist dual-process (CDP) model and the dual-route cascaded (DRC) model, showed only CDP predicted this pattern. Since CDP learns simple associations between graphemes and phonemes whereas DRC uses a set of grapheme-phoneme rules, this suggests that the results may have been caused by simple associations between spelling and sound being relatively easy to learn with words with one compared with two multi-letter graphemes. More generally, the results suggest that graphemes are used when reading, but they often produce relatively weak effects and thus differences in some studies may not have been found due to a lack of power.

Hybrid Maximum Clique Algorithm Using Parallel Integer Programming for Uniform Test Assembly

Educational assessments often require uniform test forms, for which each test form has equivalent measurement accuracy but with a different set of items. For uniform test assembly, an important issue is the increase of the number of assembled uniform tests. Although many automatic uniform test assembly methods exist, the maximum clique algorithm (MCA)-based method is known to assemble the greatest number of uniform tests with the highest measurement accuracy based on the item response theory. In that method, the graph is constructed by sequentially adding a randomly formed test as a vertex without considering the graph structure. However, an important difficulty is its high space complexity, which interrupts search cliques with more than a hundred thousand vertices. To overcome this difficulty, this article proposes a new uniform test assembly algorithm: hybrid maximum clique algorithm using parallel integer programming. The first step searches a maximum clique that is as large as possible up to computer memory limitations using a state-of-the-art MCA with low time complexity but with high space complexity. The second step repeatedly searches a vertex connected with all vertices of the current maximum clique from the remaining vertices using integer programming with low space complexity but with high time complexity. The proposed method constructs a larger number of tests than the traditional methods do. Finally, we use simulation and actual data experiments to demonstrate the effectiveness of the proposed method. Results show that our method assembles a 1.5–2.7 times greater number of uniform tests than traditional methods can.

An exact algorithm for two-dimensional vector packing problem with volumetric weight and general costs

The volumetric weight of a package has become an essential factor in calculating the delivery cost of shipments in the international logistics market. In this work, we extend the two-dimensional vector packing problem by considering a more realistic cost structure, which is a general function of volumetric weight. The problem is to pack a set of different items into some identical bins without violating weight limits and volume capacities so that the total delivery cost is minimized. We develop an exact approach based on a branch-and-price algorithm and subset-row inequalities for the problem. To efficiently solve the pricing problem in column generation, a label-setting algorithm with an effective label dominance rule and a bounding procedure is presented. A stronger label dominance rule is derived for the case where the cost function is convex. The computational results show that the exact method is effective in solving the various test instances of the problem. If the volumetric weight is not considered, the exact method can be adapted to solve the two-dimensional vector packing problem with piecewise linear cost function and outperformed the existing exact algorithm by computing 27 optimal solutions for previously open instances.

Partial Measurement Invariance: Extending and Evaluating the Cluster Approach for Identifying Anchor Items.

When measurement invariance does not hold, researchers aim for partial measurement invariance by identifying anchor items that are assumed to be measurement invariant. In this paper, we build on Bechger and Maris’s approach for identification of anchor items. Instead of identifying differential item functioning (DIF)-free items, they propose to identify different sets of items that are invariant in item parameters within the same item set. We extend their approach by an additional step in order to allow for identification of homogeneously functioning item sets. We evaluate the performance of the extended cluster approach under various conditions and compare its performance to that of previous approaches, that are the equal-mean difficulty (EMD) approach and the iterative forward approach. We show that the EMD and the iterative forward approaches perform well in conditions with balanced DIF or when DIF is small. In conditions with large and unbalanced DIF, they fail to recover the true group mean differences. With appropriate threshold settings, the cluster approach identified a cluster that resulted in unbiased mean difference estimates in all conditions. Compared to previous approaches, the cluster approach allows for a variety of different assumptions as well as for depicting the uncertainty in the results that stem from the choice of the assumption. Using a real data set, we illustrate how the assumptions of the previous approaches may be incorporated in the cluster approach and how the chosen assumption impacts the results.

Evaluating a Computerized Adaptive Testing Version of a Cognitive Ability Test Using a Simulation Study

This study evaluated the psychometric quality of a computerized adaptive testing (CAT) version of the general cognitive ability test (GCAT), using a simulation study protocol put forth by Han, K. T. (2018a). For the needs of the analysis, three different sets of items were generated, providing an item pool of 165 items. Before evaluating the efficiency of the GCAT, all items in the final item pool were linked (equated), following a sequential approach. Data were generated using a standard normal for 10,000 virtual individuals ( M = 0 and SD = 1). Using the measure’s 165-item bank, the ability value (θ) for each participant was estimated. maximum Fisher information (MFI) and maximum likelihood estimation with fences (MLEF) were used as item selection and score estimation methods, respectively. For item exposure control, the fade away method (FAM) was preferred. The termination criterion involved a minimum SE ≤ 0.33. The study revealed that the average number of items administered for 10,000 participants was 15. Moreover, the precision level in estimating the participant’s ability score was very high, as demonstrated by the CBIAS, CMAE, and CRMSE). It is concluded that the CAT version of the test is a promising alternative to administering the corresponding full-length measure since it reduces the number of administered items, prevents high rates of item exposure, and provides accurate scores with minimum measurement error.

RETRACTED: Beyond moral dilemmas: The role of reasoning in five categories of utilitarian judgment

This article has been retracted: please see Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy). The authors discovered from comments on their open data that Studies 1 and 2 of their manuscript suffered from a serious flaw. There was a mistake in the Python code they used, and the items (vignettes) were not properly randomized across conditions. In Study 1 (which investigated the effect of cognitive load on moral judgment), all participants in the Load condition received one set of items, whereas all participants in the No Load condition received a different set of items. This meant that they cannot determine whether the difference they observed between the Load and No Load conditions was due to the manipulation (Load vs. No Load) or to a simple difference between items. Given these problems, the authors cannot guarantee that the conclusions of their paper are valid. Because of this error, the authors have requested retraction of their manuscript.

Psycholinguistic properties of words on fluency testing relate to different aspects of subjective cognitive decline

AbstractBackgroundSubjective Cognitive Decline (SCD) is posited as a marker of preclinical Alzheimer’s disease. However, the nature of SCD and its relationship with objective cognition remain the subject of debate. More sensitive markers of objective cognition are likely needed to capture variability in SCD; one candidate is the psycholinguistic information embedded in language tasks, such as lexical frequency and concreteness, which is rich in cognitive information. In addition, attention to the nature of SCD is important for advancing our understanding of its relationship with objective cognition. Here we investigate how psycholinguistic information (lexical frequency and concreteness) relate to different sets of items on an SCD scale.MethodParticipants included 74 cognitively healthy older adults (mean age = 73.76 (7.20), education =16.91 (2.20), 66% female, 85% Caucasian, 10% African American, 4% Asian and 1% other). SCD was measured with a 20‐item questionnaire, using a 7‐point Likert scale to rate difficulty on each item compared with others their age. Principal Component Analysis (PCA) was conducted to derive underlying SCD factors. Participants also completed semantic fluency (animals) and phonemic fluency (CFL) tasks. Lexical frequency values were computed for each word generated on fluency tasks. Concreteness was computed for CFL only; this value would not vary within a semantic task. Nonparametric partial correlations between psycholinguistic measures and SCD factors were conducted, controlling for age, sex and education.ResultsPCA identified three SCD factors with Eigen values >1 that together explained 63% of the total variance. Factor 1 loaded primarily on memory, Factor 2 was a combination of memory and language, and Factor 3 related to attention/executive function. Lexical frequency (animals) was correlated with Factor 1 only; higher SCD related to higher frequency (more common) animals (r=.27, p=.025). Concreteness (CFL) was correlated with Factor 2 only; higher SCD was associated with more concrete, less abstract, words (r=.25, p=.035).ConclusionTwo measures of psycholinguistic information were uniquely related to different SCD items. This double dissociation reinforces the idea that mapping SCD to objective cognition requires a carefully refined measurement approach with regard to subjective experiences and the information collected from objective measures.

The same cloze for all occasions?

Abstract Target language proficiency assessment has become an integral part of Second Language Acquisition (SLA) research design, with cloze tests frequently serving this purpose for reasons of practicality. Assumptions underlying the interpretation of such cloze test scores, however, are often not examined. With the goal of providing researchers with better means for drawing inferences from cloze test scores, we present an analysis of a combined dataset comprised of scores from 1,724 test takers on a frequently used English cloze test (Brown 1980). We examine variation in score distributions and reliability estimates among L2 groups, between L2 and native-speaker (NS) examinees, and for different scoring methods, and investigate the degree to which different sets of items were effective for classifying low- vs high-proficiency L2 examinees and L2 vs NS test takers. Standardized scores are provided for each scoring method so that future researchers can reference their scores to this larger set.

EMERGENCE OF THE SKILLS THAT DEFINE NAMING IN CHILDREN WITH AUTISM

Individuals with autism lack flexibility to emit novel language. We explored the emergence of the skills that define the naming capacity in four young children with autism who received an early intensive behavioral intervention. They learned to tact objects or pictures (i.e., saying the name when seeing the object) then we probed, with no prompts or reinforcement, the emergence of selecting the same objects(pointing to the object or picture when hearing the name); they also learned to select a different set of items and we probed the emergence of tacts. Initially, two children did not demonstrate the emergence of any relation and the other two demonstrated the emergence of the selections. The procedure was repeated 15-21 months later. The first children demonstrated the emergence of the selections and the remaining children demonstrated the emergence of both selections and tacts. The results suggest that both selections and tacts can emerge in children with autism when they are taught with specific behavioral procedures.

Improving performances of Top-N recommendations with co-clustering method

Collaborative filtering has been widely used in many applications. The typical idea is to identify preferences of users by utilizing their interaction data over the whole items. However, this is sometimes inaccurate since users might share different preferences with different sets of items. In this paper, we put forward a new recommendation method based on collaborative filtering called User-Item Community Detection based Recommendation (UICDR) method. The new parameter-free and scalable community detection method is modified from our previous work. We derive a unipartite form of bipartite modularity and put forward a new network representation. By constructing a bipartite network with user-item interaction data, we first partitions users and items into several subgroups. After getting clusters with tightly linked users and items, traditional collaborative filtering models can be trained for each cluster. The results on four real-world data sets show that, the proposed UICDR significantly improves the performances of Top-N recommendations of several traditional collaborative filtering methods. In addition, UICDR is helpful to cold-start problem.

Is It the Same Country? Representations of Cambodia on Official Tourist Sites and in Tourists’ Comments

This paper explores how a certain country is described on official tourist sites whose main aim is to achieve more tourists (formal communication sources), and in informal tourists’ comments, who want to exchange their opinions. The paper suggests that the difference in aims of tourists impacts on the construction of country image, as a set of different items reflecting real aspects of a certain country. The semantic and contextual analyses of Russian and German formal and informal tourism related sources indicates a predominance of cognitive items (compared with affective ones) in both source types, but cultural items shown as positive prevail in formal sources (and thus an elaborated image is developed) whereas different cognitive and affective items are common for informal sources.

The Measurement of Green Workplace Behaviors: A Systematic Review

The literature on the green behaviors adopted by individuals in workplace settings has grown significantly over the past decade. Many studies have examined the factors associated with individual actions. However, the comparability of the studies conducted on the subject is a common concern, mainly because of the wide range of measurement tools based on different sets of items used in such research. Therefore, the aim of this study is to determine the degree of methodological maturity of green workplace behaviors based on a systematic review of research published on the subject between 1977 and 2016. Five major trends were identified from the 53 papers reviewed as part of this research: (a) the predominance of scales for measuring “green office” behaviors, (b) the redundancy of certain items, (c) the limited efforts devoted to measuring counterproductive green behaviors, (d) the emergence of new subcategories of proenvironmental behaviors, and (e) and the abundance of scales measuring voluntary green behaviors (extra-role). Through an analysis of existing measurement tools, this article proposes a decision tree designed to help scholars choose appropriate items for their studies. This may, in turn, contribute to the literature on green workplace behaviors by reducing bias and limiting the unnecessary creation of new measurement scales.

Hierarchical Item Response Models for Analyzing Public Opinion

Opinion surveys often employ multiple items to measure the respondent’s underlying value, belief, or attitude. To analyze such types of data, researchers have often followed a two-step approach by first constructing a composite measure and then using it in subsequent analysis. This paper presents a class of hierarchical item response models that help integrate measurement and analysis. In this approach, individual responses to multiple items stem from a latent preference, of which both the mean and variance may depend on observed covariates. Compared with the two-step approach, the hierarchical approach reduces bias, increases efficiency, and facilitates direct comparison across surveys covering different sets of items. Moreover, it enables us to investigate not only how preferences differ among groups, vary across regions, and evolve over time, but also levels, patterns, and trends of attitude polarization and ideological constraint. An open-source R package, hIRT, is available for fitting the proposed models.

Hybrid Method for Multicriteria Items Ordering

A hybrid method for ordering multiple criteria items according to a Decision Maker’s (DM’s) preferences is proposed for inverse multicriteria bin-packing problem solving. If such a problem arises repeatedly for different sets of items, estimated upon the same qualitative criteria with the same scales, the STEPCLASS method is applied at the first stage for preliminary multicriteria sorting on the Cartesian product of the criteria scales. Consequently, it allows defining a class for any item without DM input. In the case of the necessity to order the objects from a particular set within a single class more precisely, the UniComBOS method is applied at the second stage. It is shown that it is possible to use a DM’s preferences extracted at the stage of multicriteria sorting as source information for the UniComBOS method.

No evidence of binding items to spatial configuration representations in visual working memory

When detecting changes in visual features (e.g., colour or shape), object locations, represented as points within a configuration, might also be automatically represented in working memory. If the configuration of a scene is represented automatically, the locations of individual items might form part of this representation, irrespective of their relevance to the task. Participants took part in a change-detection task in which they studied displays containing different sets of items (shapes, letters, objects), which varied in their task relevance. Specifically, they were asked to remember the features of two sets, and ignore the third set. During the retention interval, an audio cue indicated which of the to-be-remembered sets would become the target set (having a 50% probability of containing a new feature). At test, they were asked to indicate whether a new feature was present amongst the target set. We measured binding of individual items to the configuration by manipulating the locations of the different sets so that their position in the test display either matched or mismatched their original location in the study display. If items are automatically bound to the configuration, location changes should disrupt performance, even if they were explicitly instructed not to remember the features of that particular set of items. There was no effect on performance of changing the locations of any of the sets between study and test displays, indicating that the configural representation did not enter their decision stage, and therefore that individual item representations are not necessarily bound to the configuration.

Business Analytics for Streamlined Assort Packing and Distribution of Fashion Goods at Kolon Sport

Kolon Sport (K/S), a leading outdoor fashion brand in South Korea, must deal with a large variety of items during each selling season. In doing so, the company has encountered a challenging operational problem—the assort-packing and distribution problem. The problem involves making decisions on the optimal method to use in packing a set of different items in a box and allocating the boxes to stores to meet the stores’ demands. In this paper, we introduce an analytics project initiated to improve the assort-packing and distribution process, and we describe the formulation and solution approach we developed to solve this industrial problem in a timely manner. We validated the proposed approach by computational and onsite pilot testing, which demonstrated that the decisions made using this approach are superior to those made with the manual method K/S used previously. Inventory is distributed to all stores in a more balanced way by considering the demands of the stores. K/S, which implemented the proposed approach into its internal system in July 2015, estimates that the new system improves sales by approximately eight percent.

Parameter Estimation in Rasch Models for Examinee‐Selected Items

AbstractThe examinee‐selected‐item (ESI) design, in which examinees are required to respond to a fixed number of items in a given set of items (e.g., choose one item to respond from a pair of items), always yields incomplete data (i.e., only the selected items are answered and the others have missing data) that are likely nonignorable. Therefore, using standard item response theory models, which assume ignorable missing data, can yield biased parameter estimates so that examinees taking different sets of items to answer cannot be compared. To solve this fundamental problem, in this study the researchers utilized the specific objectivity of Rasch models by adopting the conditional maximum likelihood estimation (CMLE) and pairwise estimation (PE) methods to analyze ESI data, and conducted a series of simulations to demonstrate the advantages of the CMLE and PE methods over traditional estimation methods in recovering item parameters in ESI data. An empirical data set obtained from an experiment on the ESI design was analyzed to illustrate the implications and applications of the proposed approach to ESI data.