1. Foreign language learning and assessment

A recent three-year study examined the conventional wisdom that homestay placements for American students studying abroad promote oral language proficiency gains due to the opportunities for target language input that living with a native host family affords. Following up on studies questioning the richness of student-host family interactions and findings that homestay students do not make greater language gains than learners in other living arrangements, this study explored a way to optimize the homestay experience for oral proficiency gain by providing training to study abroad host families. The 152 student participants, who were enrolled in semester study abroad programs and living in homestays in Chile, China, Russia, and Peru, took a Simulated Oral Proficiency Interview (SOPI) and a survey at the beginning and end of the semester abroad. Homestay families participated in a training session with a program staff member who provided strategies and advice on drawing their student into more meaningful conversation in the target language. Families who participated in the training also took surveys at the beginning and end of the semester. Analysis showed that the entire group of students made significant gains in oral proficiency as measured by the SOPI and rated using the American Council on the Teaching of Foreign Languages (ACTFL) Proficiency Guidelines – Speaking. There was no significant difference, however, between gains made by students whose families participated in training and students whose families did not receive training. Analysis of holistic proficiency gains with regard to other factors of the study abroad experience suggested that students who engaged in language partner exchanges with native speakers were more likely to make language gains as measured by the SOPI. Participant surveys revealed that students and host families generally held positive views of the homestay, with significant differences by target language. Spanish-speaking families and students were most enthusiastic about all aspects of the homestay and their relationships, and students of Russian tended to be least positive (Di Silvio, Donovan & Malone Reference Di Silvio, Donovan and Malone2014). Future analysis will analyze dinner-table conversations recorded by participating students and their host families.

In a previous Research in Progress article (Moore & Sugarman Reference Moore and Sugarman2010), CAL reported on a study of the perceptions, beliefs, and attitudes of users of the TOEFL® Internet-based Test (iBT), which is used to assess the academic English proficiency of students seeking to study in English-speaking universities. The goal of this study was to gather information from university administrators, instructors, and students about the TOEFL® iBT as a measure of academic language ability. Findings indicated that instructors generally agree that the TOEFL® iBT can show how well students can read, write, listen, and speak in English, while most students do not agree that all sections of the assessment measure their ability to speak English. CAL is undertaking a second research study on the TOEFL® iBT, this time investigating whether non-native English-speaking students’ performance on TOEFL® iBT writing tasks is comparable to their performance on actual academic writing tasks assigned in required university writing courses in the United States. Student participants will complete two TOEFL® iBT writing tasks for this study and will submit two graded academic writing assignments, and their performance on these tasks will be compared by using a common analytic rubric to rate various dimensions of writing quality. In addition, test takers and their instructors will share their perspective on the comparability of TOEFL® iBT and classroom writing tasks through surveys and interviews.

Since 2007, CAL has worked with the National Foreign Language Center (NFLC) and the Center for Advanced Study of Language (CASL) to provide technical assistance, evaluation, and research support to the STARTALK program, part of the US government's National Security Language Initiative. This initiative aims to increase the number of Americans learning, speaking, and teaching critical need foreign languages by offering summer language classes to students, and professional development to teachers, at sites throughout the United States (www.startalk.umd.edu). Between 2009 and 2012, CAL and CASL engaged in a research project to examine the reliability and validity in the STARTALK context of LinguaFolio Online, a self-assessment tool that was originally developed by the National Council of State Supervisors for Languages and put online by the Center for Applied Second Language Studies (CASLS) at the University of Oregon. Two other assessments were used for comparison purposes: the Computerized Assessment of Proficiency (CAP), an objective test of reading, writing, listening, and speaking developed by CASLS; and the Teacher Retrospective Observation of Student Progress (TROSP), a paper-based checklist that was developed by CAL. Data were collected over three years from over 1,100 students enrolled in 45 different Arabic, Hindi, and Mandarin STARTALK programs. Findings indicated that LinguaFolio showed evidence of reliability and validity as it was used in the STARTALK summer programs, but that there was significant variation across programs in terms of the strength of the relationship between LinguaFolio results and CAP and TROSP scores. Qualitative data suggested that the implementation of, and student training for, LinguaFolio may have contributed to this variability; therefore, the second and third summers of data collection also included the development and dissemination of two training videos. However, the data did not suggest that these training videos had an impact on the accuracy or validity of the self-assessment instrument, although that conclusion is attenuated by limitations in tracking which study participants had viewed the videos.

2. English language learner (ELL) education

The study ‘Implementing technology for the instruction of English language learner students’ was prompted by the growing number of districts with emerging ELL student populations and by the increased national focus on transforming instruction through the use of technology. This study, funded by the Regional Educational Laboratory (REL) Appalachia, sought to provide information to administrators and educators in that region of the US on the use of technology-based resources for instruction of ELL students, including academic content resources (such as websites on specific academic topics), digital productivity tools (such as slide presentation software), digital communication/networking tools (such as email), and digital classroom devices (such as interactive whiteboards). Site visits were conducted in eight districts selected from the four REL Appalachia states (Kentucky, Tennessee, Virginia, and West Virginia) for the purpose of interviewing administrators, coaches, and instructors using technology with ELLs in English as a second language and general education settings. Findings indicated that most academic content resources used across settings were designed for mainstream students, not for ELLs; that the use of technology to support content learning was more consistent among mainstream teachers than teachers of English as a second language; and that successful implementation of new technological resources is supported through professional coaching that is tailored to each instructional context.

The ‘Survey of Dual Enrollment Programs for English Learner Students’ was the first study to investigate the participation of ELLs in dual enrollment programs (these programs allow students to earn university-level course credits while still enrolled in high school). The ten participating US localities included several cities with among the highest numbers of ELLs in the country as well as some with rapidly increasing numbers of ELLs. Surveys were distributed to representatives at the high school and community college level in each locality. Findings demonstrated that dual enrollment programs do not typically target ELL students as participants nor include institutionalized support needed by ELLs. Among those respondents who provided enrollment information, only two indicated that their program enrolled more than 100 ELLs. Further, only one program indicated that current and former ELL students were a target population, and two reported that professional development on teaching ELLs was provided to instructors. Knowing that few programs target ELLs but that dual enrollment programs have been shown to improve college completion and success rates for the broader population of low-achieving, ‘at-risk’ youth, the survey's findings point to a need for further investigation of what policies and practices might promote the involvement of ELL and immigrant youth in dual enrollment programs.

In 2014 CAL began a four-year research project funded by the US Department of Education's Investing in Innovation (i3) initiative entitled ‘Promoting Educator Networks: Standards-Based Instruction for English Learners’. The project is being conducted in partnership with the TESOL International Association, the University of Oregon's Global and Online Education Department, SRI International, and two partner school districts. The goal of the project is to improve educational outcomes for ELLs by facilitating teacher networking and professional development through an online platform. Professional learning will focus on implementation of the Common Core State Standards, the Next Generation Science Standards and other state-based standards in order to provide instruction for ELLs that prepares these students to be college- and career-ready. The project will employ an experimental design that compares the outcomes of students whose teachers participate in the online network with those who do not. Additional data will be collected through online program analytics and on-site interviews and observations that will provide insight into teacher participation in network activities and strategies that teachers in treatment and control conditions employ to deliver standards-based instruction to ELLs. Results will indicate whether participation in the network improves student outcomes and if higher levels or different types of participation in the online network affects instruction and student achievement.

3. Language policy

A recent study undertaken in collaboration with the UCLA Civil Rights Project/Proyecto Derechos Civiles and the Educational Testing Service (ETS) investigated the relationship between Spanish-English bilingualism, biliteracy, and employability among Latino immigrants in the United States based on analysis of data available from the 2007–2011 American Community Survey (ACS) published by the US Census Bureau. The study's goals involved developing a more informed understanding of the extent and impact of bilingualism and biliteracy among the US workforce and discussing implications of policies regarding immigrant language and literacy maintenance and development. ACS data were analyzed to focus specifically on Spanish-English bilingualism, level of English fluency, and Spanish literacy in relation to employment levels, occupational patterns, and education among immigrants of Mexican, Cuban, Puerto Rican, Central, and South American origin. Though the ACS does not directly capture native language and literacy skills, researchers constructed composite variables from the ACS to measure respondents’ bilingualism and native language literacy using variables such as home country years of schooling, highest degree achieved, and English proficiency. After researchers identified surrogate variables likely reflecting levels of bilingualism and biliteracy, they were cross-tabulated with employment and occupational variables to investigate the relationships between bilingualism, biliteracy, employment, and occupation. Findings indicate that literacy in any language translates to economic advantages and that Spanish-English bilingual ability does in some instances correspond to higher levels of employability and income (Moore et al. in press).

A research project being conducted on behalf of the International Baccalaureate Organization (IBO) will investigate the development and implementation of language policies in IB World Schools. Through case studies of eight IB World Schools in eight countries, researchers will seek to understand how diverse schools have established school-wide language policies that outline their approach to the use and learning of multiple languages. This study will employ in-depth multiple-site case studies utilizing document analysis and interviews at four sites, and document analysis, interviews, and observations in four other schools. The research will investigate how IBO policies and philosophies are implemented at the local level, how school language policies are developed and implemented, and how language policies vary across settings – in different world regions, with different funding structures, and at different grade levels.