Search for: All records

Writing quality is dependent upon the organization and sequencing of cognitive processes during writing. College students need writing-strategy advice that is tailored to their individual needs and is cognizant of their already-established writing processes. However, there is an obstacle to providing such advice: Both writing instructors and the writers lack awareness of the moment-by-moment actions by which text was produced. This is because switching between the processes of defining the task, coming up with ideas, outputting text, evaluating, and revising is largely regulated implicitly. To address this shortcoming, the present study uses a design-based research approach to develop and evaluate a minimally viable prototype of a system called “ProWrite” that uses novel biometric technology (concurrent keystroke logging and eye tracking) for providing real-time, individualized, automated, process-focused feedback to writers. This feedback is grounded in the analysis of each writer's individual needs and is presented in the context of a learning cycle consisting of an initial diagnostic, an intervention assignment, and a final follow-up. In two iterations, eight students used the system. Effects on student behavior were determined through direct analysis of biometric writing-process data before and after remediation and through changes in writing-process and written-product measures. Semi-structured interviews revealed that students generally considered the system useful, and they would try to use the newly learned strategies in their future writing experiences. The study demonstrated that individualized, real-time feedback informed by biometric technology can effectively modify writers' processes when writing takes place. 

Linguistic corpus analysis is often an overlooked research method in writing center studies. This methodology has the potential to reveal countless patterns in datasets, but frequently lacks important details. Pairing corpus analysis with inductive coding—a qualitative approach—provides a comprehensive view of both overarching themes and specific information. This paper utilized this mixed-methods approach to explore the types of feedback that writing consultants provide to students during sessions at Iowa State University’s writing center. Session notes, written by a consultant during a writing session, contain an abundance of information surrounding the inner workings of writing centers, but few studies have recognized them as viable data sources. For the quantitative analysis, this study utilized AntConc to derive frequencies of commonly occurring words and n-grams in session notes. The qualitative analysis consisted of a process of inductively coding the data to identify commonly occurring themes and define them based on their linguistic realizations. By creating an initial coding guide, completing several rounds of session note annotations, and adjusting the guide as needed, inductive coding provided a level of context and detail that was instrumental in understanding the characteristics of writing center session notes. 

Revision plays an important role in writing, and as revisions break down the linearity of the writing process, they are crucial in describing writing process dynamics. Keystroke logging and analysis have been used to identify revisions made during writing. Previous approaches include the manual annotation of revisions, building nonlinear S-notations, and the automated extraction of backspace keypresses. However, these approaches are time-intensive, vulnerable to construct, or restricted. Therefore, this article presents a computational approach to the automatic extraction of full revision events from keystroke logs, including both insertions and deletions, as well as the characters typed to replace the deleted text. Within this approach, revision candidates are first automatically extracted, which allows for a simplified manual annotation of revision events. Second, machine learning is used to automatically detect revision events. For this, 7120 revision events were manually annotated in a dataset of keystrokes obtained from 65 students conducting a writing task. The results showed that revision events could be automatically predicted with a relatively high accuracy. In addition, a case study proved that this approach could be easily applied to a new dataset. To conclude, computational approaches can be beneficial in providing automated insights into revisions in writing.