Engineering | Slate Lab

Automated Writing Evaluation

ongoing research

Can a computer “read” and “grade” your writing?

We are conducting research to develop methods and algorithms that allow computer-based tools to automatically assess writing and provide useful feedback. For example, we have conducted to studies to test new ways of detecting strengths and weaknesses in written essays, detect revisions, and using natural language processing (NLP) tools to understand diverse types of successful writing.

Roscoe, R. D., Snow, E., Allen, L., & McNamara, D. S. (2015). Automated detection of essay revising patterns: applications for intelligent feedback in a writing tutor. Technology, Instruction, Cognition, and Learning, 10, 59-79.

Crossley, S. A., Roscoe, R. D., & McNamara, D. S. (2014). What is successful writing? An investigation into the multiple ways writers can write successful essays. Written Communication, 31, 184-215.

View a poster on automated detection of revisions (presented at EDM 2016).

Writing Instruction and Assessment

ongoing research

How can we teach and support the development of writing proficiency?

We conduct research on how students write and revise effectively. In addition, we are developing and testing computer-based tools for teaching writing strategies and scaffolding successful writing.

Roscoe, R. D., Jacovina, M. E., Harry, D., Russell, D., & McNamara, D. S. (2015). Partial verbal redundancy in multimedia presentations for writing strategy instruction. Applied Cognitive Psychology, 29, 669-679.

Roscoe, R. D., Varner, L., Weston, J., Crossley, S., & McNamara, D.S. (2014). The Writing Pal Intelligent Tutoring System: usability testing and development.Computers and Composition, 34, 39-59.

View a poster on evaluative misalignment (presented at AERA 2016).

Online Learning and Decision Making

ongoing research

How do we search for information online and use that information to make decisions?

Learning online is often self-initiated and self-directed (i.e., answering your own questions), but is also a cognitively complex task. We are studying how search strategies and behaviors affect the quality of information obtained. We are also investigating how naturalistic searches affect decision making about purchases and behaviors.

Roscoe, R. D., Grebitus, C., O’Brian, J., Johnson, A. C., & Kula, I. (2016). Online information search and consumer choice: effects of web search stance.Computers in Human Behavior, 56, 103-118.

View a poster on online search and decision making (presented at the ISTL Learning Innovation Showcase 2016).

Human Systems Engineering Education

emerging research

Can we improve student engagement and performance via “the human side” of engineering?

Human Systems Engineering at Arizona State University uniquely bridges applied psychology and cognitive science, human-computer interaction and human systems integration, human factors, and user-centered design. By teaching engineering students about the people they are engineering for (e.g., clients and customers) and with (e.g., coworkers and interdisciplinary teams), and perhaps themselves as engineers, we believe that students will be more engaged, successful, and effective.

Dr. Roscoe was recently selected as a Tooker Professor for 2017-2019 to pursue this research.

Rethinking Our Relationships to Educational Technology

emerging research

How can we extend, repurpose, or reimagine how we learn with educational technologies?

The traditional “role” of many educational technology users is as a “learner.” Students are expected to use educational technologies to gain knowledge and skills in the target subject. One challenge is that many students enact this “learner” role in a rather passive way. However, different roles–such as “designer,” “teacher,” or even “marketer”–may activate more constructive ways of thinking with greater agency. In this research, we investigate how inviting students to take on different roles may change or improve how they learn with educational technology.

Dr. Roscoe received a KEEN Professorship award in 2016-2017 to conduct research on this topic, and to design a new course (“Designing for Learning”) that teaches about learning and educational technology.

Would you like to collaborate with SLATE Lab? Are you a student seeking research experience? Contact us!

10 + 11 =