Fluency Versus Accuracy in the Role-play
In a speaking assessment, a student performs a role-play: explaining to a host family why she will arrive late for dinner. The prompt gives three required elements: the reason, an apology, and a proposed alternative time. The student speaks for 90 seconds. She produces the reason clearly, apologises twice, and proposes a time, but in doing so makes several case-ending errors and one verb-aspect error that briefly confuses the host's imagined response. She self-corrects one error, hesitates twice for about three seconds, and fills the gaps with 'па' and 'значи'. She never asks the host a question. The criteria weight communicative effectiveness and interaction alongside accuracy.
1. How should the student's performance be judged against the three required elements?
Read the worked answer
All three required elements are present: reason, apology, and alternative time. That satisfies the task's communicative core, which is the first thing to establish before weighing accuracy. The double apology is not a problem in itself; it may even read as natural politeness. The proposed time is delivered, so the role-play's purpose is achieved. The judgement should therefore start from 'task complete' and then deduct or comment on the linguistic and interactive weaknesses, rather than treating the errors as evidence that the task failed. Separating task completion from language quality is the key judgement here.
2. The verb-aspect error briefly confuses meaning. How much should this weigh, and how does self-correction change the picture?
Read the worked answer
An error that momentarily misleads the listener weighs more than an error that is merely non-standard, because communicative effectiveness is affected. However, the student self-corrected, which shows monitoring and repairs the confusion, so the penalty should be moderated rather than maximal. The case-ending errors, by contrast, rarely block understanding and should be noted as accuracy issues without dominating the judgement. The principle: rank errors by their effect on the listener, credit repair, and avoid averaging all errors into a single impression. A student who makes several harmless errors and repairs one harmful one is in a different position from a student whose errors go unnoticed and uncorrected.
3. Hesitations and the missing question: what do they reveal, and what should the feedback prioritise?
Read the worked answer
The three-second hesitations with fillers suggest the student is composing rather than interacting; the fillers keep the turn alive but do not move it forward. The missing question is the more significant gap, because a role-play with a host family is a two-way exchange, and interaction criteria reward inviting a response. Feedback should prioritise two things: first, a small repertoire of turn-ending questions the student can deploy under pressure, and second, strategies for buying time without long pauses, such as a short reformulating phrase. Accuracy work on case endings should continue, but it is not the top priority given that the task was completed and meaning was largely preserved.