User Experience in Digital English Assessment: An Exploratory Comparison of Duolingo English Test and EnglishScore

Abstract

Digital language assessments expand access and speed up the reporting of results, but user experience can introduce variance that is unrelated to language constructs. This study compared user experiences of the Duolingo English Test (DET) and EnglishScore among Indonesian Professional Teacher Education (PPG) participants. A convergent mixed-methods design was used, combining a five-point Likert-scale survey and open-ended comments. The analysis included 264 completed responses, with 130 responses for the DET and 134 for EnglishScore. Twenty-five survey items were organized into five dimensions: access and ease of use, clarity of instructions, confidence in the score, usefulness of feedback, and technical readiness. Quantitative data were analyzed descriptively, and open-ended comments were examined thematically. The overall mean for EnglishScore (M = 4.34) was higher than for the DET (M = 3.99) across all dimensions. The largest gap was found in clarity of instructions (4.31 versus 3.89), especially in understanding the test sections. Participant comments highlighted four main sources of difficulty: instructions and navigation, time pressure, audio or recording issues, and connection or device glitches. Bringing together the findings suggests that technical accessibility alone is not enough; procedural clarity and the usability of score reports also shape the assessment experience. Since the two platforms were assessed by different groups and individual data are not available, these results are not meant to provide evidence of psychometric superiority for either test. The findings support the use of pre-test orientations, device checks, interactive sample tasks, and more diagnostic result reports when implementing digital assessments in EFL contexts.