National Institute for Testing and Evaluation

The relentless drive toward smaller semiconductor nodes demands advanced lithographic materials with exceptional resolution, sensitivity, line-edge roughness (LER), and defect control. Negative Tone Development (NTD) processes offer unique advantages for specific patterning applications. To circumvent the RLS trade-off, co-optimization of both the exposure and the development spaces is essential. We systematically explore the NTD developer formulation dimension besides the photoresist-formulation-related exposure space, leveraging advanced computational and machine learning tools. In this presentation, we will present our recent progress and results on the development of high-performance NTD developers via high-throughput experimentation (HTE) utilizing Hansen-solubility parameters (HSP) as a pivotal gauge.

Background/Objectives: Scoliosis is a prevalent comorbidity in Rett syndrome (RTT), often necessitating surgical intervention. This study investigated the impact of a 10-month individualized home exercise program (HEP) on scoliosis progression and gross motor function in girls aged six to 16 years with RTT. Methods: A multiple-baseline single-case design (AABA) was employed with 20 participants. A remotely supervised HEP, based on established principles focused on posture and physical activity, was implemented daily for at least one hour. The primary outcome was the rate of scoliosis progression assessed through the Cobb angle change measured via spinal radiographs at baseline, pre-intervention, and post-intervention. The secondary outcome was the gross motor function. Results: The HEP did not significantly reduce the rate of scoliosis progression. However, individual responses varied, with three participants showing scoliosis reduction. Significant improvements were observed in gross motor function, particularly in standing, walking, and stair-climbing abilities. Conclusions: The HEP did not significantly impact overall scoliosis progression, but a significant improvement was found in gross motor function. Further research into larger sample sizes is needed to confirm the effectiveness of exercise interventions in people with RTT.

This study investigates empirically whether and how the use of gender-neutral language affects the performance of women and men in real high-stakes exams.We make use of a natural experiment in which the institute administering Israel's standardized college admission tests amended the language used in its exams, making test language more gender neutral.We find that the change to a more gender-neutral language was associated with a significant improvement in the performance of women on quantitative questions, which meaningfully reduced the gender gap between male and female performance on these questions.However, the change did not affect female performance on verbal questions nor male performance on either quantitative or verbal questions.Our findings are consistent with the hypothesis that gendered language may introduce a "stereotype threat" that adversely affects women's performance in tasks in which they are stereotypically perceived to underperform.Our findings have significant implications for the ongoing academic and policy discussions regarding the use and effects of gender-neutral language.

In most academic institutions, students with learning impairments (LIs) are entitled to various accommodations as a means of compensating for their impairment. Ensuring that the appropriate accommodations were selected requires an intelligent support tool to track their effectiveness. In this article, we regard the effectiveness of such accommodations in terms of their quality and reliability. High-quality accommodations allow students with LIs equal access and equal opportunity to demonstrate their knowledge compared to their peers who do not have LIs. Highly reliable accommodations mean a significant performance difference between students with LIs who actually use the accommodations compared with students with LIs who do not use the accommodations, given similar student characteristics. Previous literature is inconclusive regarding the evidence of the effectiveness of such accommodations since different accommodations may have a different effect on different subgroups of exams and students. This article proposes a methodology, based on ordinal interpretable models, that produce practical insights to professionals who are responsible for students with LIs, to address the problem of exploring the effectiveness of learning accommodations. The suggested models use the ordinal information of the target variables for evaluating student performance and yield practical insights for designing the most suitable accommodation for a student with LIs based on their characteristics. The ordinal interpretable models are evaluated using a database of tens of thousands of engineering students. The results demonstrate that the suggested interpretable models perform significantly better than the compared algorithms on all measures.

ABSTRACT The purpose of this study is to evaluate how learning disabilities (LDs), in combination with accommodations, affect the performance of a decision-tree to predict the stability of academic behaviour of undergraduate engineering students. Additionally, this study presents several examples to illustrate how a college could use the resultant model to choose the appropriate accommodation to give a student with a learning disability, from among a set of possible accommodations. The findings show that: (1) The models yield superior performance in predicting the stability category for a given student when the LD and accommodation factors are included; (2) Different types of accommodation action have different effects on the stability of academic behaviour, depending on the student pattern. Such a model could be useful for engineering faculties, as it would allow them to predict the stability of academic behaviour and to provide early intervention for students who are likely to need additional support.

Abstract Much research shows that judgmental estimation could be improved by combining estimates from independent judges as well as within judges. These results have been obtained mostly with judgments about matters of fact, that is, for which there are objective truth criteria. In the present research, we extend these findings to performance evaluations. In a controlled field study, expert judges provided evaluations of a large number of essays written by college applicants taking college entrance tests. The judges were asked to evaluate each essay twice—on two occasions, a week apart. This design allowed us to assess the benefits of two methods of combining evaluations: within rater and across raters. Accuracy gains were obtained with both methods. Although the within‐rater combinations yielded fewer gains than the across‐rater ones, they were still appreciable in comparison with the across rater ones. Our findings extend the class of judgments to which the “wisdom of many” could be applied. These findings are potentially applicable to performance evaluations in social, educational, and employment settings.

ABSTRACTIn the current study, two pools of 250 essays, all written as a response to the same prompt, were rated by two groups of raters (14 or 15 raters per group), thereby providing an approximation to the essay’s true score. An automated essay scoring (AES) system was trained on the datasets and then scored the essays using a cross-validation scheme. By eliminating one, two, or three raters at a time, and by calculating an estimate of the true scores using the remaining raters, an independent criterion against which to judge the validity of the human raters and that of the AES system, as well as the interrater reliability was produced. The results of the study indicated that the automated scores correlate with human scores to the same degree as human raters correlate with each other. However, the findings regarding the validity of the ratings support a claim that the reliability and validity of AES diverge: although the AES scoring is, naturally, more consistent than the human ratings, it is less valid.

A frequent topic of psychological research is the estimation of the correlation between two variables from a sample that underwent a selection process based on a third variable. Due to indirect range restriction, the sample correlation is a biased estimator of the population correlation, and a correction formula is used. In the past, bootstrap standard error and confidence intervals for the corrected correlations were examined with normal data. The present study proposes a large-sample estimate (an analytic method) for the standard error, and a corresponding confidence interval for the corrected correlation. Monte Carlo simulation studies involving both normal and non-normal data were conducted to examine the empirical performance of the bootstrap and analytic methods. Results indicated that with both normal and non-normal data, the bootstrap standard error and confidence interval were generally accurate across simulation conditions (restricted sample size, selection ratio, and population correlations) and outperformed estimates of the analytic method. However, with certain combinations of distribution type and model conditions, the analytic method has an advantage, offering reasonable estimates of the standard error and confidence interval without resorting to the bootstrap procedure's computer-intensive approach. We provide SAS code for the simulation studies.