The fluent essay that answers a different question
At the June session, a Language 1 Swedish candidate chooses the essay prompt: 'Evaluate the claim that social media has damaged democratic debate.' The paper states a 600-word limit and a 20-mark essay. The candidate writes 690 words. The script opens with an anecdote about a family WhatsApp group, argues that social media is 'mostly good because I keep in touch with cousins', includes one paragraph on misinformation, offers no counterargument, and ends with a rhetorical question. The Swedish is fluent: varied sentence structure, almost no grammar errors, but the register slips into slang such as 'super annoying' and 'likes'. The marking grid separates task response and content (8 marks), organisation (4), language range and accuracy (6), and register and audience (2). The candidate's teacher predicts a high language score. The examiner must decide the overall band and write feedback that is fair but not misleading.
1. Which marking criteria are most damaged by the script's choices, and which are relatively strong? Justify with evidence from the setup.
Read the worked answer
Task response and content are the weakest. The prompt asks for evaluation of a claim about democratic debate; the script offers a personal anecdote and a one-sided assertion. 'Mostly good because I keep in touch with cousins' is not an evaluation of democratic debate, and the single misinformation paragraph is not weighed against a counterargument. Organisation is also weak: the script follows narrative chronology rather than an argument structure, and the rhetorical question does not conclude. Language range and accuracy are strong: varied sentences and near-error-free Swedish fit a high language band. Register and audience are weak because slang and 'likes' clash with a formal essay. The key judgement is not to let fluent Swedish lift the content band. A high language score and a low task-response score can coexist because the overall mark is a weighted sum, not a global impression.
2. How should the 90-word overrun be handled under a 600-word limit? Consider fairness, double-penalisation, and whether extra text can earn marks.
Read the worked answer
First check the rubric. If it specifies a penalty for overrun, apply it once. A 90-word overrun is 15% over the limit, which is substantial. If the rubric says material beyond the limit is not read, mark only the first 600 words; if it says a deduction applies, deduct once from organisation or task response, not both. Do not invent a penalty if none is stated. Avoid double-penalising: if the missing counterargument falls after the cut-off, do not also mark it as a content failure. The fairest approach is to read the whole script for coherence, base the mark on the first 600 words, and apply a single prescribed overrun deduction if one exists. Record the decision so that moderation is consistent.
3. Write the core of the feedback: two strengths and two priorities, tied to the criteria.
Read the worked answer
Strengths: first, lexical range and grammatical accuracy are strong enough for a high language band; second, the family anecdote is vivid and could work as a short hook if framed as an example rather than the whole argument. Priorities: first, state a thesis in the opening paragraph and evaluate it with at least one counterargument and a response; second, rewrite two sentences in formal register and cut the script to 600 words by deleting repetition, not evidence. Feedback should quote the rubric: evaluation requires weighing, not merely asserting. The candidate should practise planning a 600-word argument in five minutes, with one paragraph for the opposing view and one for rebuttal, before drafting.