Back

School-leavers' knowledge about grammar

• Background
• The test instrument
• History of the project

Background

This webpage is a temporary collecting place for material related to the question: How much do school-leavers know about grammar? It uses a simple test instrument which was first used in 1986 and then more recently, as explained in the history. The primary concern till now has been the state of affairs in the UK, but it would also be good to know more about the situation in other countries. This international dimension has already started, with a study in Spain which produced extremely interesting results.

The test instrument

The test instrument for use in 2010 is a cut-down version of the instrument used in 2009, which in turn was reduced from the 1986 instrument. The new instrument consists of:

• a one-page questionnaire in .doc format, which you can expand if you want to find further information about the student concerned - e.g. whether or not they studied a language in their last years at school.
• correct answers and report form for this questionnaire.

The method is also simple: distribute the questionnaire to any group of incoming, first-year undergraduates (i.e. school leavers with no experience of grammar teaching at university) and ask them to fill it in on the spot. It should only take about ten minutes. If possible make sure they don't collaborate - this is a test of individual knowledge. If you feel you should collect more information such as whether their schooling was in your country or elsewhere, do; but don't worry if this would only exclude a handful of students. Remember: this is a very rough assessment of what a typical student learns at school in your country. If you can distinguish 'linguists' (those who have chosen a university course related to language) and 'non-linguists', please do.

The text and questions should be suitable for any English-speaking country, but you may want to translate both if your students have a different language. On the other hand, the trial in Spain showed that Spanish-speaking students who know English can out-perform English-speaking students even when the text and questions are in English (and include grammatical terms that have no corresponding terms in Spanish!), so you may prefer to leave it in English. If you report your results to me, please tell me whether or not you translated the questionnaire.

The results are reported as the mean (average) number of errors or gaps per student. Clearly this is an extremely crude measure, but it is at least indicative. I would be pleased if you reported the results to me so that I can add them to an international database.

History of the project

• In 1986, Tom Bloor (Aston University) constructed a questionnaire about grammatical knowledge and reported the results of applying it to students at Aston and at UCL.
  • A short published report.
  • A longer report, which includes details of the test of grammatical knowledge.
  • His questionnaire about grammatical knowledge.
  • His questionnaire about general KAL.
• In 1997, Charles Alderson and colleagues (Lancaster University) applied a similar questionnaire, including part of Bloor's, to a larger number of modern language undergraduates at Lancaster.
  • Their published report: Alderson, J. C., Clapham, Caroline, and Steel, David 1997. 'Metalinguistic knowledge, language aptitiude and language proficiency.', Language Teaching Research 1: 93-121.
  • One of their findings was: "Any instruction which assumes that students know more than 'noun' or 'verb' will cause problems for many students."
• In 2009, Dick Hudson and Charles Alderson encouraged colleagues in other departments to conduct a 'grammar audit' using the main questions from Bloor's 1986 questionnaire about grammatical knowledge:
  • The audit questionnaire (very short!)
  • The mark scheme and report form
  • The results from 13 departments.
  • A report on the project by Charles Alderson.
• In 2010, Isabel Corona and Pilar Mur-Dueñas used the same test (in English!) with 148 Spanish-speaking students at Zaragoza University:
  • 73 linguists (studying English): mean 2 errors on parts of speech (cf the best UK linguists: 3.02, in 1986)
  • and 75 non-linguists (studying engineering and nursing): mean 3.1 errors (cf the best UK non-linguists: 7.45 in 1986)
• Their report is here.
• These figures are striking because:
  • Zaragoza university is a typical Spanish university (officially ranked tenth out of 50).
  • the Spanish students were tested on their knowledge of English grammatical terminology, which in some cases has no equivalent in Spanish (e.g. 'finite verb').
  • in spite of this handicap, the Spanish results are consistently better than any comparable UK figures.
  • they show relatively little difference between linguists and non-linguists (just 1.1, compared with a difference of 5 in the UK 1986 sample), showing that all school children master grammatical terminology in Spain, regardless of their later specialisation. (The UK figures for 2009 show a similarly small difference between linguists and non-linguists, showing that not even linguistic specialists learn much grammar.)
• In 2012 a paper was accepted for publication.

Visits since 20 Sept 2010: View My Stats