Exploring Writing Analytics and Postsecondary Success Indicators
Saved in:
| Title: | Exploring Writing Analytics and Postsecondary Success Indicators |
|---|---|
| Language: | English |
| Authors: | Burstein, Jill, McCaffrey, Daniel, Beigman Klebanov, Beata, Ling, Guangming, Holtzman, Steven |
| Source: | Grantee Submission. 2019Paper presented at the International Conference on Learning Analytics & Knowledge (9th, Tempe, AZ, Mar 2019). |
| Peer Reviewed: | Y |
| Page Count: | 4 |
| Publication Date: | 2019 |
| Sponsoring Agency: | Institute of Education Sciences (ED) |
| Contract Number: | R305A160115 |
| Document Type: | Reports - Research Speeches/Meeting Papers |
| Education Level: | Higher Education Postsecondary Education |
| Descriptors: | Undergraduate Students, Writing (Composition), Writing Evaluation, Learning Analytics, Writing Research, Grade Point Average, Writing Achievement, Natural Language Processing, Predictor Variables |
| Abstract: | Writing is a challenge and a potential obstacle for students in U.S. 4-year postsecondary institutions lacking prerequisite writing skills. This study aims to address the research question: Is there a relationship between specific features (analytics) in coursework writing and broader success predictors? Knowledge about this relationship could contribute to more immediate personalized learning support for students. To investigate, we collected authentic coursework writing from students enrolled at one of six 4-year colleges. We then extracted natural language processing (NLP) writing features (analytics) from the writing samples and examined relationships between the analytics and college grade point average (GPA). Consistent with Burstein et al. (2017), findings suggest that NLP writing analytics may contribute to college GPA prediction. Our findings imply that real-time NLP writing analytics from authentic coursework writing could be used to efficiently track success and flag potential obstacles during students' college careers. [This paper was published in: "Companion Proceedings 9th International Conference on Learning Analytics & Knowledge" (LAK19), p.213-214, 2019.] |
| Abstractor: | As Provided |
| IES Funded: | Yes |
| Entry Date: | 2019 |
| Accession Number: | ED598690 |
| Database: | ERIC |
| Abstract: | Writing is a challenge and a potential obstacle for students in U.S. 4-year postsecondary institutions lacking prerequisite writing skills. This study aims to address the research question: Is there a relationship between specific features (analytics) in coursework writing and broader success predictors? Knowledge about this relationship could contribute to more immediate personalized learning support for students. To investigate, we collected authentic coursework writing from students enrolled at one of six 4-year colleges. We then extracted natural language processing (NLP) writing features (analytics) from the writing samples and examined relationships between the analytics and college grade point average (GPA). Consistent with Burstein et al. (2017), findings suggest that NLP writing analytics may contribute to college GPA prediction. Our findings imply that real-time NLP writing analytics from authentic coursework writing could be used to efficiently track success and flag potential obstacles during students' college careers. [This paper was published in: "Companion Proceedings 9th International Conference on Learning Analytics & Knowledge" (LAK19), p.213-214, 2019.] |
|---|
