Developing critical thinking skills is an important goal of higher education, as it can lead to better academic outcomes, improved problem-solving abilities, and greater success in the workforce. In this research project, we designed multiple critical thinking training modules to assess and support the development of critical thinking skills and collected empirical evidence from 84 college students with various backgrounds. Our pilot results showed that the tasks demonstrated acceptable reliability for assessing students’ critical thinking skills, students’ GPAs were significantly correlated with their critical thinking task scores, and students performed better on the second set of Research module tasks than on the first set. Students’ survey responses highlighted overall positive perceptions on the critical thinking training modules. Students viewed the modules as a useful tool for developing critical thinking skills, recognizing their relevance to academic work and appreciating their engaging content and appropriate difficulty level. Students also valued the feedback provided within the modules, which they perceived as constructive and helpful for supporting their understanding and performance. This research has implications on designing tools that support the development and assessment of critical thinking skills in higher education. Suggested citation: Song, Y., Bridgeman, B., Yang, Z., & Li, C. (in press). Cultivating critical thinking: Design and evaluation of training modules for college students. ETS Research Report Series. https://doi.org/10.64634/emhbze14.
This longitudinal study examines patterns of linguistic changes in young English-as-an-additional-language (EAL) learners’ written responses across tasks, score levels, and test administrations over a 1-year period. Writing samples were collected from 114 learners ages 8 through 14 using a descriptive task and an email task. The written responses were scored holistically by trained raters, and five linguistic features were extracted using automated text analysis tools. Results showed that higher scoring responses were generally longer, exhibited greater lexical diversity, included words acquired later in life, contained less imageable vocabulary, and demonstrated greater syntactic variety. Task type also influenced the response characteristics. The email responses tended to show greater linguistic complexity than the descriptive responses, particularly at the lower score levels, although the differences between tasks diminished at higher score levels. Analyses across administrations revealed systematic differences in response characteristics over time, with larger changes observed between the first and second administrations than between the second and third administrations. Among the linguistic features examined, syntactic variety showed the strongest association with both score differences and temporal variation, whereas lexical sophistication measures displayed more nuanced and nonlinear patterns. These findings highlight score-, task-, and time-sensitive variations in young EAL learners’ written responses and the importance of examining different linguistic dimensions when characterizing young learners’ writing quality in classroom-based longitudinal data. Hsieh, C.-N., & Cai, H. (2026). Linguistic features in young English-as-an-additional-language learners’ writing: A classroom-based longitudinal study (TOEFL Research Report No. TOEFL-RR-110). ETS. https://doi.org/10.64634/0t8dx119
This study evaluates the score reliability, classification decision accuracy, and interrater reliability of TOEFL Primary® test scores. Using large operational datasets from Reading, Listening (Step 1 and Step 2), and Speaking tests, reliability was estimated via Cronbach’s alpha and standard error of measurement (SEM). Classification decision accuracy for CEFR levels was estimated using the Livingston and Lewis (1995) method. Interrater reliability for Speaking was evaluated through percentage agreement, correlation, and quadratic weighted kappa based on a double-scoring design. Results indicate strong internal consistency across all test sections (α = .83–.90) and relatively small SEMs (1.19–1.61), supporting score precision. Classification decision accuracy was high for Reading and Listening (approximately .78–.82) and somewhat lower for Speaking (.76), reflecting greater complexity in performance-based assessment. Interrater reliability for Speaking demonstrated moderate to strong agreement, with correlations and weighted kappa values ranging from .68 to .88 and adjacent agreement exceeding 96%. Overall, findings provide strong evidence supporting the reliability and interpretability of TOEFL Primary scores for reporting CEFR-aligned proficiency levels and making educational decisions about young English learners. Suggested citation: Li, F. (2026). Score reliability, classification decision accuracy, and interrater reliability of TOEFL Primary® tests (Research Memorandum No. RM–26-09). https://doi.org/10.64634/ys398g44
Higher education admissions have been undergoing significant disruption, with a variety of social forces impacting it including the rise of generative AI and the decline of race-based affirmative action. In this report, we reflect upon the current state of change in graduate admissions, drawing upon insights from ETS’s February 2025 pulse survey of members of NAGAP, The Association for Graduate Enrollment Management (GEM). We also present survey data on factors influencing graduate application requirements across institutions, including the role of the U.S. Supreme Court’s Students for Fair Admissions affirmative action decisions. We conclude with a discussion on the implications of these findings for GEM policies and practice. Suggested citation: Gooch, R. M., & Haviland, S. B. (in press). Graduate admissions testing perceptions: Results from the ETS/NAGAP February 2025 survey (Research Notes). ETS. https://doi.org/10.64634/37f75641
The TOEFL® Teaching Framework: Guidelines for Teaching Lessons and Courses (Norris & Sasayama, 2022) is designed, “to provide general recommendations regarding the design of curriculum, materials, and pedagogic activities that are intended to help learners develop their academic English language proficiency” (p. 2). The Framework’s primary focus is on recommended pedagogical practices drawn from both empirical research and expert guidance. It is distinct from other language teaching frameworks in that it draws from principles of task-based language teaching (TBLT; Ellis, 2025; Long, 2015) in conceptualizing best practices for instructed second language acquisition in EAP contexts. There is a clear need for EAP professional development (Ding & Campion, 2016). One application of the Framework is that it can guide professional learning for EAP instructors. In this report, we present partial results from a study about the use of the Framework in a continuing professional development program with 25 EAP instructors working in Chinese higher education. Through the analysis of qualitative data collected during the course and in post-program interviews and questionnaires, we seek to understand how the Framework’s principles are understood and locally adapted by participants. This study contributes to a better understanding of continuing professional development for EAP instructors in China as well as the usefulness of language teaching frameworks as resources for instruction and professional learning. Suggested citation: Montee, M., Sun, E., & Bushton, J. (in press). The TOEFL® teaching framework in China: Effective application in EAP courses (Research Report). ETS.
