Poster #106 - MECO-LAB: A Mandarin-English Child Online Language Assessment Tool for Researchers, Clinicians, and Parents

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Language, Communication

Integrative Statement

Early language is the foundation for future academic and vocational success. Developmental language disorder is a pervasive but hidden disability that is difficult to diagnose in culturally and linguistically diverse groups due to the lack of large-scale normative data and valid assessment tools. Children of Chinese descent is one of the most rapidly growing segments in the US but access to the population poses great challenges for the development of language assessment tools. Advances in web-based experimentation suggest a novel solution that could overcome this participation bottleneck and make the problem tractable. Scott, Chu and Schulz (2017) adapted 3 well-known developmental studies to the online environment and recruited parents of 11-60-month-old children to self-administer the protocols in their homes. The web studies obtained similar results as laboratory studies but had notable advantages in population diversity and lowered recruitment and testing cost. Inspired by these findings, the current study aims to 1) evaluate the usability of a web-based receptive language testing platform that targets 4-7-year-old Mandarin-English Dual Language Learners, and 2) examine the comparability between data collected from parent-supervised testing and experimenter-supervised testing.
The current testing platform (https://udllblab.wixsite.com/meco-lab) consists of a Mandarin module and an English module, each of which contains a 4-item practice, and 64 test items divided into 2 tasks of 32 items each. The items utilize a forced-choice format with 2 to 4 alternatives. Children have to select one picture from the array that best matches the sentence played in the audio. The tasks can be completed using a computer or a tablet. To minimize parental interference, the platform: 1) includes an introduction page to explain the importance of not interfering and provide instructions on browser requirements and test environment; 2) incorporates practice trials for parent-child dyads to learn the task; 3) has an audio button so parents can pause the task until the child is attentive; 4) integrates a completion progress bar so parents can track progress and avoid unnecessary interruptions, and 5) has a report function that displays the child’s accuracy level upon completion of all tasks. In phase 1 of the project, we will conduct small iterative usability tests (n=5 per test) with Chinese families with 4-5-year-old children. We will videotape parent-child dyads as they use the platform at home, ask parents to fill out a usability survey and answer interview questions. New features and functions will be implemented based on parent feedback and coding of video-recorded interactions before each new round of usability test, until the usability ratings reach excellence. In phase 2 of the project, we will test 45 4-7-year-old children twice, once through parent supervision and once through experimenter supervision using the revised platform. Phase 2 tests the hypothesis that parent- and experimenter-supervised testing will yield comparable data.
This poster presentation will report the preliminary results from the study. This project will provide proof-of-concept information on a novel assessment model scalable to other minority groups. The improved platform yielded from phase 1 will be open for use by researchers, clinicians, and parents.

Authors

• Li Sheng, University of Delaware
  Presenting Author

• Danyang Wang, University of Delaware
  Non-Presenting Author

• Audrey Yang, Archmere Academy
  Non-Presenting Author

• Ning Hsu, University of Delaware
  Non-Presenting Author