Transcript Learningfromreadingsummary file

Incidental / intentional teaching
of vocabulary from reading
Schmitt & Sonbul
Direct vocabulary teaching after
reading: Is it worth the effort?
Previous research
• Various studies; both incidental and
intentional needed
• Studies showing that explicit focus on
vocabulary after reading resulted in better
learning
Looked at:
• Incidental learning from reading only (readonly)
• Combination of incidental learning from
reading plus explicit instruction (read plus)
Kinds of knowledge
Given
Meaning
Form
Tested
Recall
Form recall (supply
the L2 item)
Meaning recall
(supply definition,
L1 meaning, picture
etc.)
Recognition
Form recognition
(select the L2 item)
Meaning
recognition (select
definition, L1
meaning, picture
etc.)
Questions
1. Is read-plus more effective than read-only?
2. To what degree do the read-only and readplus input conditions facilitate learning at the
three levels of mastery of the form-meaning
link (form-recall, meaning recall and formmeaning recognition)?
Population: 40 female medical students
Process
Input conditions
• Read-only condition: read the text,
comprehension questions
• Read-plus: words explained,
Tests (immediately and after one week)
• Form recall: definition and first three letters.
• Meaning recall: full word, students translated
• Meaning recognition: m/c definitions
Results (post-test)
• Read-only condition: very little recall, 38%
recognition
• Read-plus: higher scores for all 3, 9% for form
recall, 19 for meaning recall, 50% for
recognition
• Very little attrition, probably because the ss
reread the article.
• In most cases meaning recognition > meaning
recall > form recall
Conclusions
• Intentional vocab teaching is worth the effort.
• Rereading the text within a week reinforces
learning.
Batia Laufer: Vocab acquisition in a
foreign language: do learners really
acquire most vocab by reading: some
empirical evidence.
Acquisition through reading?
• Native speakers acquire most of their
vocabulary through reading because couldn’t
possibly learn all through direct teaching.
• High school graduates – about 20,000 families
maybe more.
• True of L2? Krashen says yes.
Assumptions
1. Noticing: reader notices unfamilar words.
Noticing necessary for learning (Schmidt)2. Guessing ability assumption: can and will guess
right meaning
3. Guessing-retention link: having guessed –
learner will retain it.
4. Cumulative gain: even if not remembered the
first time, additional encounters will make more
likely – eight or twelve. If learn 1/20 and a
million words exposure, will learn 1000 words a
year.
BUT
1. Learners don’t necessarily identify words they don’t know
– overestimate what they know
2. Can’t usually guess, or guess wrongly
3. If new words unimportant, will be ignored. If important,
may be guessed, but may be guessed wrongly. Words
guessed with difficulty remembered better.
(But in extensive reading, SS don’t want to hang about
working on difficult guessing.)
4. Enormous amount of reading needed: To acquire108 new
words, need to read 9 books (assuming at this level about
22.600 words per book). And in the conditions described
earlier; and before they’ve forgotten them.
Experiment process
First experiment
• First group reading plus glosses plus m/c comprehension questions;
• Second group list of word with explanations and translations, and
had to write sentences.
• Sentence-writing group had higher scores.
Second experiment:
• same idea, but composition rather than sentences. Same result.
Third experiment
• Reading group plus looking up unknown words in a dictionary;
• Second group words without text, told to write sentences;
• Third group: sentence completion: given sentences, asked to look
up meanings and complete sentences. Reading group worse than
the other two.
Conclusion
• If a word is practised in a productive task, it
has a better chance of being remembered
than if it is just encountered in a text.
• Generally, research shows very small gains
from reading.
• Word-focus activities make the reader ‘notice’
Waring and Takaki, 2003, At what rate
do learners learn and retain new vocab
from reading a graded reader?
Study process
•
•
•
•
•
25 words in The little princess .
‘Substitute’ words used, e.g. . house = windle
15 young adult learners.
High percentage of known vocabulary (96%)..
Followed by tests of various kinds: word-form
recognition, multiple choice, meaning by
translation
Result
• Overall: longterm unprompted learning – one
word was learnt per one hour’s reading.
• Learn better if meet several times
• About half the words were forgotten after 3
months (but weren’t encountered again)
• Only 42% of the words occurring more than
15 times were translated correctly, back to
less than one word after 3 months.
Comment
• Multiple choice is not a good test of if the
learner knows the item or not. Translation
better.
• In order to have 50% chance of recognizing a
form 3 months later, need at to encounter at
least 8 times.
• But for unprompted meaning knowledge,
even 18 times was not enough.
Why?
• The learners were focusing on understanding,
may not have noticed words that didn’t need for
general gist..
• Coverage rate may have been too high.
• Too few chances to learn the words. May need 20
or 30 encounters: may need to read several
thousand words. (the rarer the word, the more
text needs to be read in order to encounter it.
• Substitute words may have been confusing,
because the learners already knew the real
English word.
Further comments
• Speed of reading didn’t make a difference.
• Some items were easier to learn than others.
• Some items were confused because they
sounded or looked like words they knew.
Conclusions
• Few new words learnt from reading.
• Test type makes a big difference.
• Multiple choice-based research may have overestimated
gains.
• But other types of lexical learning may have taken place:
deepening of lexical knoweldge of items already known,
familiarization, collocation and chunks, increased ability to
guess from context, etc. ER mainly consolidates alreadyknown language• Various implications for graded readers design, e.g. making
sure that basic words are recycled within a level, not
worrying too much about introducing new words.
Questions
• If there had been more encounters, would
learning have improved?
• If learners were forced to focus on new words,
would they have learnt them better?
Pigada, M., & Schmitt, N. (2006).
Vocabulary acquisition from extensive
reading: a case study
Introductory comments
• Process of incidental vocab acquisition is not really
understood.
• This study looks particularly at the number of times a word
appears in a text, and relates it to the kinds of knowledge of
a word that are acquired, other than meaning.
• Advantages of extensive reading: pedagogically efficient
(Huckin and Coady), as enhances both reading fluency and
vocabulary acquisition at the same time; facilitates learner
autonomy; motivating; increases sight vocab. May be the
only out-of-class source of vocabulary acquisition.
• But there is no direct evidence as to its efficiency.
• Difference between being able to inference and retaining as
a result of that inference.
• Richness of context can sometimes lead to the learner
skipping the word, as they do not need it.
Studies on vocabulary growth through
reading
• Incidental: mostly reputed to be small, roughly
1/12.
• But: measuring instruments didn’t test partial
knowledge.
• Number of words small.
• Didn’t control text difficulty.
• No long-term test.
The present study
•
•
•
•
•
•
•
•
A single learner
One-month period of extensive reading
133 target words
Measurement procedures that are sensitive to partial
knowledge, and different types of knowledge other than
reading. (orthographic, phonological, morphological,
syntactic, collocational and pragmatic).
Present study looked also at form (spelling), meaning (formmeaning relationship) and use (grammatical functions).
So even if can’t guess meaning, other aspects of the word
learnt e.g. spelling.
Importance of consolidation of previously learnt words –
other research has focused on new words.
Importance of frequency (e.g. Waring and Takaki study).
Questions
• Does extensive reading lead to increased word
knowledge in terms of spelling, meaning and
grammatical behaviour of words?
• Are all these three equally affected by
extensive reading?
• Are words that occur more frequently in the
texts more affected in these three aspects of
word knowledge?
Method
• Simplified easy readers (at least 95% coverage)
• Learner read one reader per week, for a month. 228 pages
in total.
• Pre-test and post test. Based on interview and very detailed
and exhaustive.
• Spelling: dictation, had to write down.
• Meaning and grammar: the learner was given a list of the
words, asked to say anything he knew about the meaning.
Grammar: asked to say le/la for nouns, and which
preposition followed the verb.
• 4 days after pre-test, started reading.
• Guesses not accepted.
Results
• 66/133 enhanced in one way, 13 in two ways, 8 in all
three
• 87 /133 (about 2/3) were enhanced in at least one
way• Spelling most, then articles (grammar).
• Frequency: overall, only when 20+ was there a definite
increase in uptake rates.
• But also depended on the word.
• Many misinterpretations because of L1 interference or
similarity to other words,and confused with other
French words with similar meanings.
• The learner commented that sometimes guessed,but
forgot later.
Overall conclusions
• Learning new vocabulary incidentally through
extensive reading is very slow.
• Extensive reading has value for reading
fluency, motivation, re-encounter with
previously-learnt words.
• We need a lot of focused, intentional work on
introducing and reviewing new vocabulary.
• BUT…
BUT…
• No way can we possibly teach in lessons the
5,000 – 8,000 word families that we would like
our students to learn by 12th grade.
• So…?
My conclusions
• Teach and review as much as we can
• Make students aware of the importance of
learning new vocabulary.
• Teach them strategies for learning on their
own.
Autonomous strategies
• Vocabulary notebooks
• Look out for, and note down, useful words or
chunks you encounter
• Review regularly (5 minutes a day?)