Important keystroke features for classifying at-risk writers differ by stage: early models emphasize focus shifts and initial pause, later models emphasize inter-keystroke-interval variance¶

Claim · All claims
Evidence · 1 study · associational r2 · q2
Currency · newest study 2022 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? For the first five minutes, number of focus shifts to task, variance in long pauses, and initial pause time were most important; SD of IKI between words became important only late in the process. → Rianne Conijn 2022

Evidence¶

Rianne Conijn 2022¶

Rianne Conijn, Christine Cook, Menno van Zaanen, Luuk Van Waes. (2022). Early prediction of writing quality using keystroke logging. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-021-00268-w

q2 · i? · associational · r2

Feature-importance analysis (RQ3) across classification models at each timepoint (Figs. 3–4). Importance varied by model type and timeframe; e.g., "the SD IKI between words was not important" early but was among the most important late. No effect size printed.

"For the first five minutes of keystroke data, the number of focus shifts to task, the distance between 30 s windows with more than one keystroke (variance in long pauses), and the initial pause time were the most important features. When more data were considered (from 10 up to 30 min), these features were less important."

Discussion¶

Related Claims¶

• Which keystroke features correlate with writing grade changes across stages of the writing process: pause-variance and focus-shift features matter early, sentence-transition and event features later — related
• Over the full writing process, only one of 54 keystroke features (percentage of long pauses between words) correlates significantly with final writing grade in ESL academic summaries — related