Narrative Competence in
Monolingual and Bilingual
School Children
Barbara Zurer Pearson
University of Massachusetts
University of Miami Bilingualism Study Group
1988-1998
With special thanks to RAs:
Patti Ortega, Ana Ferrer, Yael
Wiesner, Esperanza Rodriguez,
and a host of UM students
(all of whom spoke much better Spanish than I did--and
without whom I could not have studied narrative).
Questions:
ML > BL
Two-way > EI
? Linguality:
? IMS:
?for Hi-SES
?for Lo-SES
? Interdependence:
L1 predict L2
The Core Design
Replicated at Kindergarten, 2nd and 5th Grades
Monolinguals
Bilinguals
Two-Way
English Immersion
ESH
Hi
Lo
Hi
Lo
OSH
Hi Lo
SES
ESH
Hi
Lo
OSH
Hi Lo
Contribution of Narrative Analyses:
1. Can HEAR the children. (They’re not just
“scores.”)
2. A SINGLE task combining both oral
language and the demands of literate
language.
3. Encourage longer responses: more
revealing than single word or short phrases
of the Woodcock-Johnson.
Findings from Narrative Analyses:
ML > BL
(mostly)
Two-way = EI (partly)
in English Two-way = EI
in Spanish Two-way > EI
L1 predict L2
oral language No
literacy
Yes
SAMPLE STORIES (4):
Handout from page 144-145 of LLBC
(and on CHILDES archive and in SALT, U WI)
400+ stories
10 each from all cells of 2nd and 5th graders
80 ML
160 BL in English
160 BL in Spanish
+ 24 “2nd stories” from MLs (to test the effect of
telling the story twice).
Frog, Where are You?
By M. Mayer (Dial 1969)
Story 1
1.
2.
3.
4.
The dog – looked in – the bottle and looked at
the frog.
And the boy was sitting on a chair, and his –
sock and his shirt was laying on the floor.
And the light was on, and the window was
opened….
When – {the} the boy and the dog were
sleeping – the frog – stuck his head out with his
head and his arm – out of the bottle.
Story 2
1.
2.
3.
4.
5.
One day a boy and his dog had found a
frog.
They kept him in the big jar.
While the boy was asleep, the frog
climbed out of the jar and ran away.
When the boy woke up the next morning,
he was very upset to see his frog
missing.
He searched everywhere. In boots … and
he turned over tables.
Story 3
1.
2.
3.
4.
5.
6.
Once there was a little boy with his little dog.
It was already night time. They were looking at
the little frog.
The little boy – and his dog went to sleep.
The frog – wanted to go out to see {the w} the
world.
So he came out of the little – can.
It was morning already. The puppy and the boy
looked to the – can and saw {that their} that
the frog was not there.
Story 4
1.
2.
3.
4.
5.
One day there was a boy ‘n a frog and a
dog.
And then while the boy sleeps, {he}… the
frog came out.
And then when he wake up, the dog and
the boy, the frog was gone.
He found everywhere and the dog –
found – almost found in the bucket.
And the boy was calling to the frog.
So…..?



Which one is a good story?
Which one is not good?
Which are from bilinguals? monolinguals?
Motivation to split the task back up into
ORAL LANGUAGE (“Language” Score)
&
LITERATE LANGUAGE
(“Story” Score)
Oral Language Element Examples
1
2
3
fluency
-- … {} Pause for For
vocab
Bottle
example
Morpho- [they]
syntactic was
accuracy
laying
effect
wordfinding
vs Jar
vs Can
4
-- {}
vs
Bucket
Found xx
looked
Boy
TO the
sleeps,
can
he wake
(boy with up
his dog)
Narrative Element Examples
1
orientation none
2
3
4
One
day..had
found
Once
One day
there was
Affective/
cognitive
info
none
upset;
saw frog
missing;
(stuck
head out) searched
wanted to none
see the
world; saw
the frog
not there
Temporal
links
“when2”
clause
no clauses
“already”
“nightmorning”
“while”
“when1”
“while” +
wrong
tense;
“when1”
What about “Complex Syntax”?
Grammatical devices (language) to introduce
complexity and point of view (story)

In English
Complement clauses:
“he saw that the frog was not there.”
Non-finite verbs:
“was upset to see his frog missing.”
Relative Clauses:
“the boy who had the frog woke up”

In Spanish
Perfect tenses: “vio que habi’an salido otras ranitas”
[he saw that HAD COME OUT other frogs]
Subjunctive: “dijo que se callara”
[he said that he was-to-be-quiet]
“Complex Syntax” (cont’d)

“Between Clauses”
Causal conjunctions:
“in order to”; “so that he could...”
Adverbs of simultaneity:
“while”; “when2”
Retrospective reference:
“still”; “already”; “todavia”
“Language” or “Story” or both?
Findings from Narrative Analyses:
(review)
ML > BL
(mostly)
Two-way = EI (partly)
L1 predict L2
oral language No
literacy
Yes
Question 1: ML outperform BLs (mostly)
-- but not equally in all aspects of the
task.
Remake fig 7.1 (p. 154) to emphasize story score equivalence
and language score discrepancies.
First for 2nd grade.
Next page for 5th grade . (heading: language gap narrows by
5th grade.)
Next page (heading: Language gap closes at HiSES, ESH:
table 7.10 p. 158.)
Next page: most persistent ML BL difference is in MS accuracy
fig 7.6 (p. 160) (“kids that don’t talk so good can’t think
so good” (but we saw in the story exerpts that that is not
true, at least not in a circumstance of on-going “2nd
language learning”
Question 1a: MLs outperform BLs (mostly).
(ML red/ BL blue
2nd gr solid/ 5th grade bars
80
English Total (of 96)
70
60
Grade 2
Grade 5
50
40
30
MLHiSES
BLHiSES
MLLoSES
BLLoSES
Question 1b: MLs outperform BLs (mostly)–
Not equally in all aspects of the task
Story scores more equivalent;
Most discrepancy in the Language Score .
2nd grade
Story
vs
Language
Language Scores (English)--Grade 2
BL (line) vs ML (bars: Hi & Lo SES)
40
35
30
Points (of 48)
Points (of 48)
Story Scores (English)--Grade 2
BL (line) vs ML (bars: Hi & Lo SES)
25
20
15
HI-OSH HI-ESH LOEng.
OSH
Imm.
BL
LO- HI-OSH HI-ESH LOESH
TwoOSH
Way
BL
LOESH
40
35
30
25
20
15
HI-OSH HI-ESH LOEng.
OSH
Imm.
BL
LO- HI-OSH HI-ESH LOESH TwoOSH
Way
BL
LOESH
Question 1c: MLs outperform BLs (mostly)–
Less so in the long term
.
5th grade:
ML/BL gap narrows
Story
Language Score
Language Scores (English)--Grade 5
BL (line) vs ML (bars: Hi & Lo SES)
40
35
30
25
20
15
Points (of 48)
Points (of 48)
Story Scores (English)--Grade 5
BL (line) vs ML (bars: Hi & Lo SES)
HI-OSH HI-ESH LOEng.
OSH
Imm.
BL
LO- HI-OSH HI-ESH LOESH TwoOSH
Way
BL
LOESH
40
35
30
25
20
15
HI-OSH HI-ESH
Eng.
Imm.
BL
LOOSH
LO- HI-OSH HI-ESH
ESH TwoWay
BL
LOOSH
LOESH
Q1d
Most persistent ML/ BL gap in MS Accuracy
Com ple x S ynta x (Englis h)
Frog S tory Le x ic on
M S -Ac c ura c y
24
12
12
20
10
10
18
8
8
6
6
22
16
14
4
12
Gr ad e 2
Gr ad e 5
4
Gr ad e 2
Gr ad e 5
Gr ad e 2
Gr ad e 5
(Story Lexicon and
Complex Syntax
“catch-up”)
Question 1e: ML outperform BLs – but
Gap in Total Score closes at HiSES, ESH
Simple Effect -- Linguality
Frog Total
ML
n M
BL
n
M
F
p
All 2nd grade
40
63.0
79
51.3
35.6
.01*
All 5th grade
40
67.3
81
64.4
3.29
.07
5th gr; HiSES 20
70.6
40
67.0
2.59
.11
5:HiSes;ESH 20
70.6
20
68.0
0.98
.33
Hypothesis 2: 2-Way = EI (partly)
IN ENGLISH yes: Remake figure 7.4 p. 159 w/o MLs; make
companion graph for Story Score
Next page: look at MS Accuracy-Lex-and comp. syntax for 2way vs EI 2nd to 5th (from scratch??)
Next page: the interaction with SES and Language at home
(figure 7.8 p. 162)
Next page: IN SPANISH no: 2-Way > EI: figure 7.9 (p. 165)
bigger difference is in Home Language
Question 2a: EI = 2-way (partly)–
In ENGLISH: YES.
(EI green, 2-way blue)
Language
40
40
35
35
30
25
EI
20
2-way
15
10
Language Score (of 48)
Story Score (of 48)
Story
30
25
10
5
0
0
Grade 5
Two-way
15
5
Grade 2
EI
20
Grade 2
Grade 5
Question 2b: EI = 2-way (partly)–
In SPANISH: NO.
(EI green, 2-way blue)
IMS
(OSH orange, ESH brown)
LSH
56
52
English Imm.
48
Two-way
44
40
Narrative Total (of 96)
Narrative Total (of 96)
56
52
Sp Only
48
Sp & Eng
44
40
Grade 2
Grade 5
Grade 2
Grade 5
Q2 b (con’t)
That is,
 For English IMS has little effect. SES
the biggest factor. LLBC p. 156
 In Spanish, IMS has largest effect.
SES has little effect; LSH less potent
than IMS. LLBC p. 164
Question 3a:
L1 predicts L2
In Literacy Measures--YES
Cross Language Correlations
Pearson r
0.8
0.6
0.4
0.2
0
St
or
y
La
Sc
or
e
ng
u
Co
m
ag
e
pl
e
Sc
or
e
W
-J
x
Sy
#
Pa
s
nt
ax
sa
of
ge
Cl
a
M
LU
us
e
Co
m
p
s
Question 3b
L1 predicts L2
In Oral Language-- NO
Cross Language Correlations
Pearson r
0.8
0.6
0.4
0.2
0
La
Fr
M
W
Co
S
-J # C
o
ng
m
-A
g
la
Pi
pl
ua
Le
cc
u
c
e
t V ses
ge
xi
ur
x
oc
:
Sc Syn con acy
ab N ar
or
ta
rT
x
e
ot
al
Q3c A note about story length
Despite .43 correlation to Narrative Score, length is
NOT a clear measure of story quality. ie. Story
quality is better at 5th grade for everyone, BLs’
stories are getting longer; MLs’ are getting shorter.
MLs tell BETTER stories with fewer words (and presumably,
eventually the bilinguals will too.)
# of words per story
350
300
ML
250
BL
200
150
Grade 2
Grade 5
Summary of Narrative Analysis
Results
Mirror the results from the Standardized
tests:
 EI = 2 way in English; 2-way > EI in
Spanish
• language of the home advantage for ESH
(in English) disappears by 5th grade;
language of the home advantage for OSH
(in Spanish) persists for oral language,
NOT narrative skills (where IMS and SES
are more potent influences).
Key Contribution of Narrative
Analysis Results
Despite language deficits in lexicon and MSaccuracy relative to ML peers, BLs demonstrated
age-appropriate skill in these DIFFICULT narrative
tasks:
 Creating a unified plot
 Motivating events through reference to internal
states
 Providing narrator’s comments on the unfolding
story
 Using compound time-referencing
 Using embedded structures which distinguished
their own thoughts from those of the characters
Limitations of Narrative Analysis
Results


Not standardized; hard to replicate our
scoring system, which would need to be
simplified to be practical.
Our subjects not the best bilinguals,
possibly not representative of most
bilinguals. (Spanish surprisingly weak.)
•
All born in US (avoided Age-of-arrival variable),
but children speak a “contact” variety: most BLs in
Miami have non-native English-language models AND
non-native Spanish-language models.
Bp: check how lo-ses osh did wrt IMS.
Final word (from p. 172)


“By using the factorial design of the
larger study, which balanced the effect of
each factor, we have enhanced our
ability to generalize findings from the
children’s stories.
By expanding the performance demand
on the children through the story task,
we have provided an ‘auditory snapshot’
of each individual, to add to the
perspective provided by the standardized
scores.”
Final word (con’t)
“This snapshot
• enriches our ability to
[appreciate] what the test scores
are saying, and
• to have greater confidence in the
messages they convey.”
--LLBC p. 172
Descargar

Language and Literacy in Bilingual Children: The Miami