Journal of Educational Measurement最新文献

Scoring high-dimensional assessments (e.g., > 15 traits) can be a challenging task. This paper introduces the multilabel neural network (MNN) as a scoring method for high-dimensional assessments. Additionally, it demonstrates how MNN can score the same test responses to maximize different performance metrics, such as accuracy, recall, or precision, to suit users' varying needs. These two objectives are illustrated with an example of scoring the short version of the College Majors Preference assessment (Short CMPA) to match the results of whether the 50 college majors would be in one's top three, as determined by the Long CMPA. The results reveal that MNN significantly outperforms the simple-sum ranking method (i.e., ranking the 50 majors' subscale scores) in targeting recall (.95 vs. .68) and precision (.53 vs. .38), while gaining an additional 3% in accuracy (.94 vs. .91). These findings suggest that, when executed properly, MNN can be a flexible and practical tool for scoring numerous traits and addressing various use foci.

While significant attention has been given to test equating to ensure score comparability, limited research has explored equating methods for rater-mediated assessments, where human raters inherently introduce error. If not properly addressed, these errors can undermine score interchangeability and test validity. This study proposes an equating method that accounts for rater errors by utilizing item response theory (IRT) observed-score equating with a hierarchical rater model (HRM). Its effectiveness is compared to an IRT observed-score equating method using the generalized partial credit model across 16 rater combinations with varying levels of rater bias and variability. The results indicate that equating performance depends on the interaction between rater bias and variability across forms. Both the proposed and traditional methods demonstrated robustness in terms of bias and RMSE when rater bias and variability were similar between forms, with a few exceptions. However, when rater errors varied significantly across forms, the proposed method consistently produced more stable equating results. Differences in standard error between the methods were minimal under most conditions.

Although there exists an extensive amount of research on subscores and their properties, limited research has been conducted on categorical subscores and their interpretations. In this paper, we focus on the claim of Feinberg and von Davier that categorical subscores are useful for remediation and instructional purposes. We investigate this claim by examining (a) the agreement between true and observed subscore classifications and (b) the agreement between subscore classifications across parallel forms of a test. Results show that the categorical subscores of Feinberg and von Davier are often inaccurate and/or inconsistent, pointing to a lack of justification for using them for remediation or instructional purposes.

Eye-tracking procedures generate copious process data that could be valuable in establishing the response processes component of modern validity theory. However, there is a lack of tools for assessing and visualizing response processes using process data such as eye-tracking fixation sequences, especially those suitable for young children. This study, which explored student responses to a spatial reasoning task, employed eye tracking and social network analysis to model, examine, and visualize students' visual transition patterns while solving spatial problems to begin to elucidate these processes. Fifty students in Grades 2–8 completed a spatial reasoning task as eye movements were recorded. Areas of interest (AoIs) were defined within the task for each spatial reasoning question. Transition networks between AoIs were constructed and analyzed using selected network measures. Results revealed shared transition sequences across students as well as strategic differences between high and low performers. High performers demonstrated more integrated transitions between AoIs, while low performers considered information more in isolation. Additionally, age and the interaction of age and performance did not significantly impact these measures. The study demonstrates a novel modeling approach for investigating visual processing and provides initial evidence that high-performing students more deeply engage with visual information in solving these types of questions.

Educational tests often have a cluster of items linked by a common stimulus (testlet). In such a design, the dependencies caused between items are called testlet effects. In particular, the directional testlet effect (DTE) refers to a recursive influence whereby responses to earlier items can positively or negatively affect the scores on later items. This study aims to introduce an innovative measurement model to describe DTEs among multiple polytomouslyscored open-ended items. Through simulations, we found that (1) item and DTE parameters can be accurately recovered in Latent GOLD^®, (2) ignoring positive (or negative) DTEs by fitting a standard item response theory model can result in the overestimation (or underestimation) of test reliability, (3) collapsing multiple items of a testlet into a super item is still effective in eliminating DTEs, (4) the popular multidimensional strategy of adding nuisance factors to describe item dependencies fails to account for DTE adequately, and (5) fitting the proposed model for DTE to testlet data involving nuisance factors will observe positive DTEs but will not have a better fit. Moreover, using the proposed model, we demonstrated the coexistence of positive and negative DTEs in a real history exam.

Group differences in test scores are a key metric in education policy. Response time offers novel opportunities for understanding these differences, especially in low-stakes settings. Here, we describe how observed group differences in test accuracy can be attributed to group differences in latent response speed or group differences in latent capacity, where capacity is defined as expected accuracy for a given response speed. This article introduces a method for decomposing observed group differences in accuracy into these differences in speed versus differences in capacity. We first illustrate in simulation studies that this approach can reliably distinguish between group speed and capacity differences. We then use this approach to probe gender differences in science and reading fluency in PISA 2018 for 71 countries. In science, score differentials largely increase when males, who respond more rapidly, are the higher performing group and decrease when females, who respond more slowly, are the higher performing group. In reading fluency, score differentials decrease where females, who respond more rapidly, are the higher performing group. This method can be used to analyze group differences especially in low-stakes assessments where there are potential group differences in speed.

Pretrained large language models (LLMs) have gained popularity in recent years due to their high performance in various educational tasks such as learner modeling, automated scoring, automatic item generation, and prediction. Nevertheless, LLMs are black box approaches where models are less interpretable, and they may carry human biases and prejudices because historical human data have been used for pretraining these large-scale models. For these reasons, the prediction tasks based on LLMs require scrutiny to ensure that the prediction models are fair and unbiased. In this study, we used BERT—a pretrained encoder-only LLM for predicting response accuracy using action sequences extracted from the 2012 PIAAC assessment. We selected three countries (i.e., Finland, Slovakia, and the United States) representing different performance levels in the overall PIAAC assessment. We found promising results for predicting response accuracy using the fine-tuned BERT model. Additionally, we examined algorithmic bias in the prediction models trained with different countries. We found differences in model performance, suggesting that some trained models are not free from bias, and thus the models are less generalizable across countries. Our results highlighted the importance of investigating algorithmic fairness in prediction models utilizing algorithmic systems to ensure models are bias-free.

Hurtz, G.M., & Mucino, R. (2024). Expanding the lognormal response time model using profile similarity metrics to improve the detection of anomalous testing behavior. Journal of Educational Measurement, 61, 458–485. https://doi.org/10.1111/jedm.12395

We apologize for this error.

The emergence of sophisticated AI tools such as ChatGPT, coupled with the transition to remote delivery of educational assessments in the COVID-19 era, has led to increasing concerns about academic integrity and test security. Using AI tools, test takers can produce high-quality texts effortlessly and use them to game assessments. It is thus critical to detect these nonauthentic texts to ensure test integrity. In this study, we leveraged keystroke logs—recordings of every keypress—to build machine learning (ML) detectors of nonauthentic texts in a large-scale writing assessment. We focused on investigating the fairness of the detectors across demographic subgroups to ensure that nongenuine writing can be predicted equally well across subgroups. Results indicated that keystroke dynamics were effective in identifying nonauthentic texts. While the ML models were slightly more likely to misclassify the original responses submitted by male test takers as consisting of nonauthentic texts than those submitted by females, the effect sizes were negligible. Furthermore, balancing demographic distributions and class labels did not consistently mitigate detector bias across predictive models. Findings of this study not only provide implications for using behavioral data to address test security issues, but also highlight the importance of evaluating the fairness of predictive models in educational contexts.