Thursday Lecture Series “The Brown Bag”

This semester, we will feature several brownbag meetings that are intended to demonstrate tools, platforms, and resources for conducting research, led by students and/or faculty who have used these resources in their own work. In this week's brownbag, Yoonseo Kim and Rickey Larkin will introduce these resources and briefly illustrate how they have used them in their own work.

The Korean language exhibits regional variation and increasing diversity with the growing number of L2 learners, yet recognition of its varieties remains limited. This presentation will introduce AI Hub, a useful source for L1 and L2 speech samples. In addition, I will show how I made use Gorilla Experiment Builder to create a survey using data from AI Hub which examined L1 Korean speakers’ attitudes toward five L1 and four L2 Korean varieties across 11 dimensions.

MAXQDA is a qualitative data analysis software, similar to NVIVO, Transana, and Atlas.ti. In this talk, I will give an overview of the software, some use-cases, and how I utilize the software in my own research on language policy in the Philippines. Limitations and alternatives will also be discussed.

Teacher Development and Reflective Practice on an International Teaching Practicum

Betsy Gilliland (Professor, UHM, SLS); Ayano Kawasaki, Alex Tang, Ruge Zhao (PhD Students, UHM, SLS)

Both novice and experienced teacher-learners develop new pedagogical skills through immersion in the unfamiliar teaching context provided by international TESOL practicums. We present findings from our research as supervisors of a study abroad practicum and offer recommendations for teacher educators interested in coordinating similar programs.

Educators’ Perspectives on Automated Writing Scoring in High-Stakes K-12 Assessments

Alicia Kim, Jieun Kim (UHM, SLS PhD Candidate), Mark Chapman, Jason A. Kemp, & Lynn Shafer Willner

Automated writing scoring (AWS) is widely used in standardized language tests for adults but remains less common in K-12. This presentation reports on a two-phase study examining K-12 educators’ perceptions of AWS. Focus group interviews with 14 educators identified 20 themes aligned with Kunnan’s (2018) fairness and justice principles, which were further explored through a survey of 739 educators across 32 U.S. states. Findings highlight key considerations for implementing AWS with young multilingual learners.

This presentation showcases research Jieun Kim contributed to during a summer internship at WIDA. Attendees will also have the opportunity to learn about WIDA assessments and the internship from researchers.

About the WIDA Presenters

Alicia Kim 

Ahyoung Alicia Kim is a researcher at WIDA where she conducts validation research of ACCESS for ELLs and other WIDA language assessments. Her research interests include language assessment, child bilingualism, second language literacy development, and computer-assisted language learning. Prior to joining WIDA, Alicia was a post-doctoral fellow with a focus on child bilingualism at Cornell University. She has also led research on large-scale English proficiency exams at Seoul National University in Seoul, Korea. Over the years, she has taught TESOL and ESL courses in the U.S. and abroad. Alicia holds an Ed.D. in Applied Linguistics and an Ed.M. in TESOL from Teachers College, Columbia University.

Mark Chapman 

Mark Chapman is the Senior Innovation Researcher at WIDA. His work focuses on researching innovative assessment practices to enhance language tests for young multilingual learners. Mark also conducts research into the validation of speaking and writing assessments for young multilingual learners. He publishes his work in journals and edited volumes and regularly presents at domestic and international conferences. Mark previously served as the Associate Director of Assessment with the University of Michigan’s English as a Second Language testing program. He also spent more than a decade teaching English in Japan. He has a master’s degree in Teaching English as a Foreign Language from the University of Birmingham (U.K.) and a Ph.D. in assessment from the Center for Research in English Language Learning and Assessment (CRELLA) at the University of Bedfordshire (U.K.).

Jason A. Kemp 

Jason Kemp is an assessment researcher at WIDA. He participates in research projects that support K-12 multilingual learners and their educators. His duties also include contributing to the development of English language proficiency assessments. For example, Jason evaluates test items and graphics during annual bias, sensitivity, and content reviews. He has taught Spanish in K-20 contexts and trained future world language educators. Jason serves on the WIDA Social Justice Change Team and leads the Assessment Task Force. Jason completed his PhD in Curriculum & Instruction at the University of Wisconsin-Madison. He also holds a MA in Spanish from the University of Central Florida and BAs in Spanish and sociology from the University of North Florida.

Lynn Shafer Willner 

Dr. Lynn Shafer Willner designs language standards for multilingual learners, digital tools, and accessibility/accommodations research and guidelines. She is a member of the WIDA Assessment Team. Her current work focuses on the integration of the WIDA English Language Development (ELD) Standards Framework into the WIDA suite of assessments. Most recently, Lynn developed the alignment architecture for the WIDA English Language Development Standards Framework (2020 Edition). She has also authored a variety of articles on the WIDA Standards Framework, standards alignment, digitalization, and accessibility for multilingual learners. Previous projects completed while working at WestEd included lead authorship of the 2014 ELPA21 English Language Proficiency Standards and Proficiency Level Descriptors and management of the project to revise the PARCC K–2 Model Content Framework for English language arts and mathematics. A former Fairfax County (VA) Public Schools ESOL teacher, Lynn has a Ph.D. in Education from George Mason University with specializations in multilingual/multicultural education and instructional technology.

Multimodal Action Formation:  An L1 English Teacher’s Use of the Gassho Gesture in ESL Classroom Interaction

Teaching is an embodied accomplishment, jointly and contingently achieved with students through the complex management of various action sequences and larger instructional goals (Hall & Looney, 2019). A growing body of literature on the embodied work of teaching has shown how teachers use gestures to explain subject-specific content, such as the progressive aspect (Kimura & Kazik, 2017) and temporal concepts (Matsumoto & Dobs, 2017) of L2 English grammar, vocabulary (Lazaraton, 2014), and mathematical (Alibali et al., 2014) and physics concepts (Kääntä et al., 2018). As practices of instructional action, teachers’ gestures are recipient designed (Sacks et al., 1974) to address knowledge asymmetries between teachers and students (Lilja & Piirainen-Marsh, 2022), thereby enhancing students’ understanding.

In contrast to students’ content knowledge, a much less studied dimension of recipient design is students’ cultural membership and identity. Therefore, this study examines how an L1 English teacher with advanced Japanese linguistic and cultural knowledge employs the gassho gesture—a culturally specific Japanese gesture of placing the palms of both hands together (Greer, 2022)—in ways that are recipient-designed for the cultural membership of his students. The data come from 28 hours of video-recorded interaction collected over three semesters in lower-intermediate level ESL classes, consisting almost exclusively of students from Japan. The findings reveal a variety of instructional (e.g., facilitating students’ understanding) and social (e.g., apologizing, showing appreciation) actions achieved by the gestures. This study contributes to the under-investigated research on cultural specific gestures by expanding its context to an institutional setting and illuminating multiple dimensions of the gestures, including recipient designing and gestural transfer (Greer, 2022).

Instructional Recipient Design: Corrections of Embodied Procedures in Chado

Lin Chen, PhD student, Second Language Studies, UH Mānoa

Given the centrality of error correction in learning contexts, there is a large and growing body of applied linguistic research investigating corrections in various settings, including language, art, and science classrooms. Corrections in situated learning contexts are consistently, albeit to varying degrees, recipient-designed. However, recipient design is often seen but unnoticed, as analysts (and participants) might take the intersubjectivity of interaction for granted. Additionally, practices of recipient design generally do not surface until intersubjectivity becomes problematic. Adopting the framework of multimodal conversation analysis (Mondada, 2014), this study examines instructional corrections based on a corpus of video-recorded chado (Japanese Tea Ceremony) lessons. It aims to reveal the kinds of work instructors routinely and competently accomplish in their day-to-day activities by examining practices of recipient design (Sacks et al., 1974) in correction sequences. Participants include the chado instructor and her students studying chado in Hawaii.

The findings reveal a recurrent order where two sets of corrective practices (see also De Stefani & Gazin, 2014) are sequentially deployed in chado corrections. The first set of corrections is regularly achieved with minimal resources, such as verbalization of an object name or a single pointing gesture. Through minimal formulation, the instructor other-positions the students as competent members (Deppermann, 2015) who require no more than minimal reminders (Kornfeld & Rossi, 2023). However, if the reminders are not immediately and successfully complied with, another set of corrections follows. In the second attempt, the instructor recipient designs the correction based on their ongoing assessment of the student’s competence by calibrating its specificity and indexicality. This study contributes to praxeological studies of instruction and correction in performative art activities (Keevallik, 2010). It also highlights the continuous work of assessment and recipient design that instructors unceasingly accomplish in instructional activities.

Joonhee Kim, PhD Student in Second Language Studies, UH Mānoa; Carolyn Siegman, AGC student in SLS and PhD student in Linguistics, UH Mānoa; Dr. Tomomi Nishikawa, Professor in Languages and Culture, Ochanomizu University; Ces Jocson, MA student in SLS, UH Mānoa; Dr. Theres Grüter, Professor in SLS, UH Mānoa

We present HALA Online, a freely available lexical retrieval task designed to assess bilingual language strength through picture naming. Participants name body parts shown as black-and-white line drawings, and responses are scored for accuracy and response time (speech onset latency). Based on O’Grady et al. (2009) and updated by Kang (2011) and Kim & Kim (2022), this new version runs on the PCIbex online platform, allowing for flexible in-person or remote administration. HALA Online is designed for use in diverse linguistic and social contexts, including indigenous, minority, heritage, and endangered language communities. It provides insight into language dominance and potential early signs of language shift. A pilot study with 22 Korean-English bilinguals in Hawai‘i showed that HALA accuracy ratios strongly correlated with self-assessed proficiency ratios and C-test scores, while RT ratios did not. This suggests that HALA accuracy may serve as a coarse proxy for proficiency, especially where other assessments are unavailable (see also Roos, 2023).

Rickey Larkin, Jr. and Michol Malia Miller, PhD Students, Second Language Studies, UH Mānoa

In this brown bag talk, we will present two of the most widely-used tools for corpus-based approaches to language research, teaching and learning: AntConc and the vast collection of corpora hosted on English-Corpora.org. In this talk, we will give an overview of the two tools and provide examples of how we have utilized them in our research.

AntConc is a freeware corpora data analysis software for carrying out corpus linguistics research and implementing data-driven language learning. Similar tools exist in Natural Language Processing, but these require technical background knowledge. We will demonstrate best practices for compiling your own corpus data to use in AntConc and the primary features of the software for corpus analysis, along with examples from our own work.

English-Corpora.org “is the most widely used collection of corpora anywhere in the world” (Davies, 2020, p. 1); the collection is used by over 130,000 people each month. Some of the major features of the Corpus for Contemporary American English (COCA) will be outlined, along with an overview of some possible applications for language research, teaching, and data-driven language learning. 

Michelle Kunkel, PhD Student, Second Language Studies, UH Mānoa

In this presentation, I will discuss teacher research (TR), an approach to practitioner inquiry in which teachers study themselves and/or their own classrooms. This talk is geared towards prospective teachers who might wish to undertake TR for their own professional purposes or for completing an academic requirement in the SLS program (e.g., an SP or QP).

After presenting a brief overview of TR, I discuss the most important research instrument in a TR project: the teacher’s research journal. Although this instrument is essential, my previous reviews of the literature have revealed very few practical guidelines for keeping a journal. That is, very few resources explain how/what to write or what "good quality" journal entries look like. Therefore, the majority of this talk will focus on specific strategies for journal writing, from building a journaling habit and overcoming anxiety, to example reflection and meta-reflection questions, and the utility of multimodal and multilingual resources for journaling. I also provide several illustrations of how journal entries can become data in written reports of TR.

The data supporting this talk come primarily from my dissertation research (i.e. my own journal entries and those of my participants), from my previous research projects in SLS, and from previous SLS students who have successfully published their TR projects.