There are several existing automatic writing evaluation (AWE) software tools for supporting free-text responses in online courses. However, the existing AWE tools are typically used for summative purposes like generating scores (e.g., EASE from edX), based on linguistic assessment (e.g., LSA from Educational Texting Service [ETS]), and/or implemented mostly in English language. My research is to develop a formative structural AWE tool applicable to any language (English, Korean, Chinese, Dutch, and Arabic so far) by integrating two standalone offline software tools from different research traditions, ALA-Reader (Clariana & Wallace, 2007) from education and Pathfinder KNOT (Tossell, Schvaneveldt, & Branaghan, 2010) from a graph-theoretic cognitive science approach – innovation through integration. The fully developed software will be able to immediately convert students’ writings into network graphs, Pathfinder Networks, which are hypothesized to represent the “knowledge structure (KS)” related to the text content. Based on our empirical evidence carried out in various kinds of learning environments across several languages, this “integration” approach is a useful and usable way to capture and visually represent the most salient content KS inherent in writings, regardless of which language is used; for example, we revealed the implicit KS in several narrative and expository text passages across five languages and found that all the expository texts in five languages have fairly similar hierarchical KS structure, visually confirming previous linguistic studies.
With support from the Center for Online Innovation in Learning (COIL), my research team will (1) develop an integrated web-based version of the offline, called GISK (Graphical Interface of Structural Knowledge), that automatically provide individualized KS feedback of online learners’ writing assignments to indicate specific areas of their current knowledge strengths and weaknesses, (2) implement the online as a formative structural assessment tool to empower online learners – namely, through reflection on their KS, and (3) evaluate the effectiveness and appropriateness of its automated evaluation and KS feedback for online learners’ academic success. I am now working as a Principal Investigator (PI) on this COIL project, Exploratory Development of a Tool to Measure Learners’ Knowledge Structure in Online Learning Courses: Formative Structural Assessment.
Further, to extend the accessibility of the visually represented KS graphs to blind learners, my team is now developing a mobile device to be able to convert automatically viewable KS graphs to touchable KS graphs so that the blind can haptically (vibratory) feel their current KS derived from their writings on the touchscreen or tactually on swell touch paper. This project is also supported from the COIL, and I am now working as a PI on this second COIL project, Providing Touchable Knowledge Structure Graphic Feedback for Blind Online Learners: Tablet-Based Haptic Feedback and Paper-Based Tactile Feedback.
Thus, the fully implemented system can promote both sight and blind online learners’ active engagement in the development of their KS during learning online by providing individualized KS feedback, and also benefit instructor’s understanding of students’ KS and thinking, which may lead to using improved pedagogy and individualized instructional strategies.
|Attribution-NonCommercial-NoDerivs 3.0 United States
|February 26, 2016
This resource is currently not in any collection.