I just failed your online test and would love to know what I got wrong as I’ve been writing and editing professionally for more than 25 years.

13 August 2016 | | Anders Schepelern

We don’t have the means to comment individually on each test, but below is a brief summary of the marking process. We are marking on two criteria: accuracy (the main one) and intervention. By intervention, we mean whether sensible queries are raised, in what quantity and whether any inaccuracies are introduced. Candidates must perform acceptably on both scores.

  • Candidates must find and correct at least 36 errors per test (i.e. 90 per cent).
  • There are a few ‘must haves’ in each test, e.g. principle/principal, and missing one of these counts as two errors. If a candidate gets 35–36, there is room for discretion depending on how they do in the next category.
  • Anything introduced that is plain wrong is an instant fail.
  • Sensible amendments are acceptable and have neither a positive nor a negative effect on the final outcome unless the candidate shows signs of tinkering too much (say more than three interventions/queries per test).
  • If we have reason to think someone is a non-native speaker, we will note this.