The aim of this investigation was to measure the intelligibility of educated Nigerian speakers of English to British listeners and to analyse the main causes of intelligibility failure. The test material consisted of the following: I- Connected Speech, II - Reading Passage, III - Phonemes, IVA - Stress, IVB - Intonation. The speech of 24 Nigerian first-year university students - 12 Yoruba and 12 Hausa speakers - was recorded. An RP speaker was also recorded. The-recordings were played to 240 British listeners, each Nigerian speaker being assessed by 10 British listeners. A scoring system was devised for the tests of Connected Speech. The intelligibility scores ranged from 92.7% to 29.9%, with a mean score of 64.4%. The RP speaker's score (based on all 240 listeners) was 99.4%. Listeners' impressionistic judgements of the speakers' intelligibility correlated closely with the scores obtained on Test I. The most intelligible Nigerian speaker was 93% as efficient as the RP speaker, the mean Nigerian speaker was 65% as efficient, and the least intelligible Nigerian speaker 30% as efficient as the RP speaker. Test I- Connected Speech - was taken as the criterion of fundamental importance in assessing intelligibility. The other tests were regarded as subsidiary. - It was found that Connected Speech was significantly correlated with Reading and Stress, but not with the tests of Phonemes and Intonation. Partial correlation analyses showed that stress is the major component of all aspects of intelligibility. The errors leading to intelligibility failure were categorised into four groupings: - rhythmic/stress, segmental, phonotactic, lexical/syntactic. Rhythmic/stress errors (38.2% for all speakers) were the major cause of intelligibility failure. This was closely followed by segmental errors (33.0%). Phonotactic errors (20.00) were of lesser importance, while lexical/syntactic errors (8.8%) were of minor importance. Details of the actual phonetic errors are summarised in Chapters 11-13. The study concludes with some observations on the testing and teaching of oral English in the light of these findings.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:475219 |
| Date | January 1974 |
| Creators | Tiffen, B. W. |
| Publisher | University College London (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://discovery.ucl.ac.uk/1349554/ |
Page generated in 0.0117 seconds