Abstract

The United States is a culturally and linguistically diverse country. Few assessments used in educational settings have been developed that can be used appropriately with culturally and linguistically diverse students, and research has shown that static, norm-referenced assessments are often biased against these students. One alternative method that shows great promise for overcoming the limitations of static, norm-referenced tests is dynamic assessment. Despite the research and recommendations that support the use of dynamic assessments, relatively few dynamic assessments have been formally developed for the purpose of identifying language disorder. Therefore, the current study sought to determine the validity of a narrative-based, dynamic assessment for diagnostic purposes. A total of 230 students, including those from culturally and linguistically diverse backgrounds, were administered the Dynamic Measure of Oral Narrative Discourse (DYMOND) by trained examiners. Several hypothesis-driven and exploratory models based on different dynamic assessment variables were analyzed using logistic regression analysis and receiver operating curve analysis to determine the diagnostic accuracy of the DYMOND. The presence or absence of a formal individualized education program for language was used as the reference standard to which DYMOND scores were compared. Descriptive statistics on the models and individual variables were also analyzed to provide preliminary normative data for the entire sample as well as for several demographic subgroups. Additionally, interrater reliability and interrater agreement estimates were obtained and reported using a one-way, random intraclass correlation analysis and Many Facet Rasch Model analysis. Administration fidelity was calculated by determining the proportion of steps an examiner correctly followed in the administration procedure. Results indicate that the DYMOND is capable of accurately differentiating between those with and without language disorder. However, different models have various degrees of effectiveness in making this distinction. Additionally, preliminary normative data using descriptive statistics for both the sample and for different demographic subgroups were obtained to which students' individual performances can be compared. Results also suggest that the DYMOND can be scored reliably and administered faithfully by a large number of examiners. Ultimately, these results demonstrate that the DYMOND can be a useful tool for making diagnostic decisions when used as prescribed.

Degree

PhD

College and Department

David O. McKay School of Education; Educational Inquiry, Measurement, and Evaluation

Rights

https://lib.byu.edu/about/copyright/

Date Submitted

2023-08-14

Document Type

Dissertation

Handle

http://hdl.lib.byu.edu/1877/etd13362

Keywords

language impairments, evaluation, culturally relevant education

Language

english

Download

Included in

Education Commons

Share

COinS