Comparing the Performance and Preference of Students Experiencing a Reading Aloud Accommodation to those who do not on a Virtual Science Assessment

Abstract

Many United States secondary students perform poorly on standardized summative science assessments. Situated Assessments using Virtual Environments (SAVE) Science is an innovative assessment project that seeks to capture students' science knowledge and understanding by contextualizing problems in a game-based virtual environment called Scientopolis. Within Scientopolis, students use an "avatar" to interact with non-player characters (NPCs), artifacts, embedded clues and "sci-tools" in order to help solve the problems of the townspeople. In an attempt to increase students' success on assessments, SAVE science places students in an environment where they can use their inquiry skills to solve problems instead of reading long passages which attempt to contextualize questions but ultimately cause construct-irrelevant variance. However, within these assessments reading is still required to access the test questions and character interactions. This dissertation explores how students' in-world performances differ when exposed to a Reading Aloud Accommodation (RAA) treatment in comparison to a control group. Student perceptions of the treatment are also evaluated. While a RAA is typically available for students with learning disabilities or English language learners, within this study, all students were randomly assigned to either the treatment or control, regardless of any demographic factors or learning barriers. The theories of Universal design for learning and brain-based learning advocate for multiple ways for students to engage, comprehend, and illustrate their content knowledge. Further, through providing more ways for students to interact with content, all students should benefit, not just those with learning disabilities. Students in the experimental group listened to the NPCs speak the dialogue that provides them with the problem, clues, and assessment questions, instead of relying on reading skills to gather the information. Overall, students in the treatment group statistically outperformed those in the control. Student perceptions of using the reading aloud accommodation were generally positive. Ideas for future research are presented to investigate the accommodation further.