Scoring behavior of language instructors

Abstract EN

This experimental study made use of different qualitative and quantitative analyses to uncover instructors' scoring behaviors of Reading-to-Write tasks in Egypt.

Egyptian instructors working in the field of higher education in Egypt were requested to score Reading-to-Write essays following the Think Aloud Protocol, assign scores to the different writing features on the analytic rubric, and be interviewed.

Based on the quantitative and qualitative data, it was clear that instructors paid most of their attention to judgment strategies, and most specifically to rhetorical/ ideational aspects with little attention to language focus.

The central tendency in grading was mostly observed in the scoring process.

All raters followed a pattern of having a mental image representation of scores once they started scoring and either confirmed this or changed it once reading was completed.

Instructors did not follow the order of the analytic rubric used and instead assigned scores to the features that stood out the most because they were either too good or too bad.

No clear pattern of severity or leniency was observed due to the limited number of participants although it was noticed that the instructor who was mostly severe almost always made negative comments when scoring, whereas the rater who was mostly lenient was the one who was generally sympathetic with the students when scoring.