A web site for the developing language teacher

A Profile of Dumindi - a Sri Lankan Learner
by Sharon Buddemeier
- 3

The Samples

The speaking samples have been transcribed and comprise Appendices 1 (individual speaking tasks) and 2 (interactive speaking task) but the line numbering is continuous to avoid confusion when referring to examples in the analysis.

The writing samples have been re-typed and comprise Appendices 3 (a letter giving directions) and 4 (description of a typical dish). Again, the line numbering is continuous to ease the burden on the reader.

The entire original test can be found in Appendix 5. These appendices will be referred to as necessary in the analysis.

Analysis of Data

General findings

According to the FCE marking criteria on the Cambridge ESOL website, a pass grade (C) “corresponds to about 60% of the total marks.” Dumindi scored 66% in Reading, 52% in Use of English, and 77% in Listening. One of her writing papers fulfilled the requirements of the task to a satisfactory level, but one did not. Her speaking paper was quite reasonably effective overall; as an FCE oral examiner and I would give her a ‘global achievement’ score of ‘3.5’ which is a pass. All of these areas are analyzed in much greater detail below.

Overall strengths include comprehensible pronunciation including tone softening, a wide variety of lexical chunks including multiword verbs, good interactive communication skills such as responding and turn-taking, and the quality and extent of her ideas when speaking.

Prominent weaknesses are problems with modality that obscure meaning, the lack of grammatical cohesion in both spoken and written discourse, grammatical accuracy for her needs and level, and frequent repetition even though her range is clearly evident at times.

A different type of interference

English is one of the official languages in Sri Lanka and so ‘mother tongue’ interference is two-fold when analyzing errors. In addition to ‘normal’ L1 interference, there is also Sri Lankan English. While the errors originally came from the mother tongue, they have become accepted in the Sri Lankan English. ‘Errors’ can be found in newspapers, official documents, and in English textbooks like Dumindi’s.

Differences include certain uses of present continuous, use of present and past perfect, use of some modal verbs, and the appearance of antiquated and excessively formal language. Hence, some errors were systematically fossilized at a young age as she was taught them at home and in school. Specific examples are noted in the detailed analysis of these and other errors while strengths are primarily just referred to by line number in order to respect word limit.


The following criteria were used to assess Dumindi’s speaking:
• Range, accuracy, and appropriacy of grammar and lexis
• Discourse management
• Pronunciation
• Interactive communication

* Reminder: the following line numbers are from Appendices 1 and 2

Range and accuracy of grammar


• Some range of verb tenses:
o present continuous including negative form (1, 5),
o present passive (10, 11, 23)
o zero conditional (16-17)
o first conditional (44, 48)

• Range and use of multiword verbs such as advised to, keep her eye on (6, 23, 41, 45, 48)

• Use of adj + prep: interested in (4), good for (25)

• Use of direct speech to talk about her parents scolding her (21-22).

• Use of ‘like’ to mean ‘for example’ (23).


• Modality:
o she don’t have enough time (42)- fails to use ‘will’ for prediction
o misuse of ‘must’ instead of ‘will/might’: so the interviewers must misjudge (37)- this is a problem with meaning and use- she uses ‘will’ correctly in the same situation (47)

Systematic errors. Her father told me that Dumindi really wants to attend Colombo University but they “must not let her in” [probably will not].

• Errors with present continuous are confusing and strain the listener:
o with time words: after I am doing A levels (2)
o with state verbs- they are telling (21), she won’t be having (45).

(L1/Sri Lankan English interference; also grammar of say/tell)

• over-generalizes ‘made by’ + pronoun to ‘made by’ + noun (10, 12)

Range, accuracy, and appropriacy of lexis


• Wide range and variety of lexis:
o lexis related to houses and materials
o lexis related to giving advice about appearance
o good use of: prefer (16), deceive (28), waste of money (29), misjudge (38)


• inability to substitute pronouns or allow the listener to infer through use of ellipses leads to frequent repetition (dressing, 23-24) and elders (35, 36, 37).

• repetition of ‘good’ (27), perfectly (26, 29) and ‘suitable’ (18, 20, 21, 22, 30, 32) are evidence of a lack of range of adjectives.

• Misuse of vocabulary:
o ‘dress’ instead of ‘wear’ (19, 22, 23)
o ‘elders’ for ‘the elderly’
o ‘post’ for ‘position/job’ (correct in Sri Lankan English)
o ‘perfectly’ to describe a way of dressing (26, 29).
o ‘endless’ money (31)
o ‘patterns’ to mean appearance/style (39)

Discourse management (coherence, extent, relevance)


• Extends responses appropriately:
o when asked about interest in other languages (4-5).
o which house she would prefer to live in (16-17)
o importance of dressing fashionably (29-32)

• Relevant and mature ideas:
o the importance and implications of dress and appearance in the workplace
o how money should be wisely spent
o sensitivity to ‘elders’ feelings about clothing

• able to sustain a long turn without major pauses

• able to self correct- no one can say . . . no one can recognize (19)


• problems with grammatical cohesion:
o excessive repetition of words and phrases- suitable (18, 20-22), we are dressing (23-24) and good (27) due to avoidance or inability to use substitution or ellipsis as well as lack of vocabulary.
o misuse of pronouns (all), problems with plurality, and referencing difficulties (a flat?are apartment, a town?a big building) when describing houses makes utterances ambiguous.

• relevance:
o fails to use language for giving advice during interactive speaking task (Eg, should/n’t) and instead focuses on the reaction of the interviewer. This again probably relates to her lack of understanding about the meaning and use of modal verbs [see objectives].
o her utterance about not having time to put on make-up is a bit far-fetched (44).

Pronunciation (stress and rhythm, intonation, individual sounds)


• Utterances are almost always understandable so communication is not impeded.

• Problems with staccato rhythm combined with and ‘sing song’ intonation rarely interfere with meaning or listener comprehension because she speaks and she puts the primary stress in the correct place on words and puts stronger emphasis (although not with the appropriate tone) on the important parts of the sentences.


• Speech is often ‘sing song’ (4, 13, 42, 44) due to mother tongue being syllable-timed as opposed to stress-timed. Also lack of weak forms and contractions.

• some words which caused difficulty were ‘personality’ (25), deceives (28), and misjudge (38) due to uncertainty of stress placement. Also, ‘interviewers’ (34) because of the /v/ and /w/ sounds being so close together, and ‘through’ (45) because of the consonant cluster /?r/.

Interactive communication (initiating and responding, hesitation, turn-taking)


• Appropriate responses and ability to politely soften her tone when expressing disagreement.

• Appropriate turn-taking and interaction particularly evident when she helps Asha to complete her thought when she couldn’t find the right words (43).


• Many brief but noticeable hesitations in the form of ‘uh’ even in short utterances sometimes put a strain on the listener. This may related to stress-timing as many of the ‘hesitations’ appear to be put in to create a rhythm that feels right to her.

To page 4 of 7

To the print friendly version

Back to the articles index

Back to the top

Tips & Newsletter Sign up —  Current Tip —  Past Tips 
Train with us Online Development Courses    Lesson Plan Index
 Phonology — Articles Books  LinksContact
Advertising — Web Hosting — Front page

Copyright 2000-2016© Developing