ArticlePDF Available

Abstract and Figures

Evaluated the effectiveness of 3 types of French immersion programs: (a) early total immersion, (b) 1-yr late immersion, and (c) 2-yr late immersion on students from Grades 7, 8, and 9. Proficiency in French was assessed using reading comprehension, writing, language arts, dictation, listening comprehension, and oral production tests. Statistical analyses at each grade level revealed that the early immersion and 2-yr late immersion Ss had achieved generally higher levels of proficiency in French with few differences in performance between the 2 groups. There were indications that the differential between the 1-yr late group and the other 2 groups was narrowing in the higher grades. Findings are discussed in terms of their practical educational significance and theoretical implications for the critical period hypothesis for 2nd language learning. (French abstract) (30 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Content may be subject to copyright.
A comparison of early and late second language learning
FRED GENESEE
McGill University
The comparative effectiveness of three types of French immersion programs was evaluated: (a) early
total immersion, (b) one-year late immersion, and (c) two-year late immersion. Samples of students
from grades 7, 8, and 9 of each program were included in the evaluations. Their proficiency in French
was assessed using the following kinds of language tests: reading comprehension, writing, language
arts,
dictation, listening comprehension, and oral production. Statistical analyses of the results at each
grade level revealed that the early immersion and two-year late immersion students had achieved
generally higher levels of proficiency in French than the one-year late immersion students, whereas
there were few differences in performance between the former two
groups.
There were also indications
from the results that the differential between the one-year late group and the other two immersion
groups was narrowing in the higher grades. The findings are discussed in terms of their practical
educational significance, particularly with respect to the role of follow-up programs, and in terms of
their theoretical implications with respect to the critical period hypothesis for second language
learning.
A number of Canadian studies have demonstrated the general pedagogical
effectiveness of French immersion school programs in which a second language is
used as the sole or major medium of curriculum instruction for English-speaking
children (Genesee, 1978a; Genesee, Polich, & Stanley, 1977; Lambert & Tucker,
1972;
Swain, 1978). These studies indicate that students who participate in French
immersion programs acquire competence in the second language that is far
superior to that of comparable students in programs where French is taught as a
separate subject for a limited part of the school day. At the same time, immersion
students attain the same level of English language development and academic
achievement as students educated solely in English. Findings indicate further that
immersion programs are effective with students from lower as well as middle and
upper socioeconomic backgrounds (Bruck, Tucker, & Jakimik, 1975), with
students with low levels of intellectual ability as well as those of average or above
average ability (Genesee, 1976), and even with students who have disabilities in
English (Bruck, 1978).
A number of alternative forms of immersion have been developed, varying with
respect
to
the proportion of curriculum instruction provided in the second language
and the grade level(s) during which the second language is used extensively. Thus,
This research was made possible by research grants to the Instructional Services Department of the
Protestant School Board of Greater Montreal from le Ministers de l'£ducation du Qu6bec.
I am grateful to La Commission des Ecoles Catholiques de Montreal for permission to use the Tests
de rendement. I would also like
to
thank Drs M. Bruck, McGill-Montreal Children's Hospital Learning
Centre, G.R. Tucker, Center for Applied Linguistics, and Jyotsna Vaid, McGill University, for helpful
comments on a draft of this report.
CANAD. J. BEHAV. SCI./REV. CANAD. Set. COMP.
13(2),
1981
116 GENESEE
in early total French immersion (Genesec, 1978a, for a full description), all
curriculum instruction is presented via French during the primary grades
(kindergarten, 1, 2, and sometimes even 3 or 4). Subsequently, a follow-up
program is provided in which varying proportions of select course material
continue to be taught via French at the same time that other subjects are taught via
English. In contrast, so called late immersion programs delay use of the second
language as the main medium of instruction until the end of elementary or the
beginning of secondary school, i e., grades 7 or
8.
At this time usually all courses,
except English, are taught via the second language. A follow-up program, varying
in intensity and consisting of instruction in a number of select courses via French,
is provided in the higher grades (Genesec et al., 1977, for
an
evaluation of this type
of program). The results of a number of evaluations indicate that early total
immersion students usually acquire significantly higher levels of second language
competence than do one-year late immersion students, and at the same time
demonstrate equivalent English language development and academic achievement
(Bruck, Lambert, & Tucker, 1976; Cziko, Holobow, & Lambert, Note 1;
Genesee, Note 2; Swain, 1978; Troue, Note 3).
The purpose of the present research was to evaluate the second language
achievement of students in a two-year (grades 7 and 8) late immersion program in
comparison with that of students in a one-year (grade 7) late immersion program
and an early total immersion program. The two-year program consists of two years
of immersion in grades 7 and 8 during which all curriculum instruction, except
English language arts, is in French. Students who take this option have already had
7 years (K to 6) of regular French-as-a-second language. The two years of
immersion are followed in grades 9 to 11 by a follow-up program consisting of
one intensive course in French language arts and one other subject taught in French
each year. This option is a recent innovation irrthe use of immersion and,
in
fact, it
had not been evaluated at the time that this research was begun. Thus, this study is
of some pedagogical significance in contributing to our knowledge of
the
relative
effectiveness of alternative second language programs. In view of previous
comparisons of early and late alternatives, one would expect early immersion
students to outperform two year late immersion students.
In addition to their educational significance, the results of this study also lend
themselves to an evaluation of the so-called critical period or optimal age
hypothesis for second language learning. A number of theoretical perspectives,
including the notion of language learning as an innate ability that dissipates with
age (Chomsky, 1972; McNeill, 1970), and that of neural plasticity which also
decreases with age and concomitantly reduces one's language learning ability
(Lennebcrg,
1967;
Penfield & Roberts, 1959), have advocated early instruction in
a second language. While both arguments are ostensibly presented in terms of first
language acquisition, they apply equally to second language learning. Thus,
according to
a
simple version of the hypothesis, one would expect second language
EARLY AND LATE SECOND LANGUAGE LEARNING 117
teaming to be more rapid and complete if learning occurs prior to a certain critical
age,
usually argued to occur around puberty (Lenneberg, 1967), than if learning
occurs after the critical age. In accord with the hypothesis, a small number of
empirical studies which have examined second language achievement over the
long term in natural contexts suggest that individuals who begin learning a second
language during childhood ultimately attain higher levels of proficiency than those
who begin learning in adolescence or adulthood (Asher & Garcia, 1969; Oyama,
1976,
1978; Seliger, Krashen, & Ladefoged, 1975). However, contrary to the
hypothesis, a number of other studies report that older learners attain higher levels
of second language achievement than younger learners given equivalent short term
exposure. These latter studies have examined language learning both in natural
contexts (Ervin-Tripp, 1974; Fathman, 1975; Snow & Hoefnagel-Hohle, 1978)
and in formal or school settings which involved relatively traditional instructional
techniques, such as drills or audiolingual methods that emphasized acquisition of
language structure (Asher & Price, 1967; Burstall, 1974; Grinder, Otomo, &
Toyota, 1962; Olson & Samuels, 1973).
In contrast with other school settings in which this issue has been examined, the
early immersion program provides extended, naturalistic second language
exposure. Consequently, one might expect greater second language achievement
among students
who
begin immersion in kindergarten than among those who begin
in secondary school. Comparison of early immersion students with two-year late
immersion students is particularly pertinent here since the latter alternative
provides an approximately equivalent exposure to total immersion, namely two
years,
as does the former alternative. Thus, discussion of the critical period
hypothesis will focus on the results of this specific comparison.
METHOD
The
Samples
The following groups of students participated in the evaluations (see Table 1 for sample sizes):
1.
Early
immersion.
These students had been in an early immersion program since kindergarten and
were following a
40%
follow-up program in French in grades 7, 8, and 9.
2.
One-year late immersion. These students had followed a regular English curriculum with
French-as-a-second language
(FSL)
instruction for 20 to 30 minutes/day from kindergarten to grade 6,
and had been in
a
one-year late immersion program in grade 7. In grades 8 and 9, they followed
a
40%
follow-up program in French; the follow-up options available to these students were not the same as
those for the early immersion students. The one-year late immersion students were divided into two
subgroups, depending upon whether the immersion program they attended was in
a
centre that offered
only grade 7 immersion (the Centre group) or
in
a wing of
a
regular English school
that also
offered
FSL
instruction (the Wing group). This contrast was included initially to evaluate the effectiveness of these
two
models,
the
expectation being
that
the Centre students might achieve higher levels of proficiency in
French than the Wing students owing to the greater "Frenchncss" of the centres where all
communication within the school
and
between
the
school and home was
in
French.
This distinction was
dropped in the grade 9 evaluation in order to streamline testing somewhat. The grade 9 one-year late
immersion sample included an equal number of Wing and Centre students.
118 GENESEE
TABLE 1
Summary of sample sizes
Early
Grade immersion
Grade 7
immersion
Centre WingGrade 7/8
immersionFrench
controlEnglish
control
(a) Excluding oral production evaluation
Grade 7 31 38 37 a 15 37
Grade 8 33 40 40 36 14 35
Grade 9 38 54 62 24 47
(b) For oral production evaluations
Grade 7 31 38 37 a 15 37
GradeS 33 40 40 36 14 35
Grade 9 22 31 29 16 30
a Same as Grade 7 immersion.
3.
Two-year late immersion. These students had been in a two-year grade 7-8 late immersion
program preceded by
FSI.
instruction from kindergarten to grade 6. The grade 9 students were taking
one intensive French language arts course and one subject in French, i.e., 40% of their course load.
Grade 7 students from this program were not evaluated since this year of the program is identical to
grade 7 of the one-year late immersion program.
4.
English
control.
These students had been following the regular English curriculum with FSL.
None of the students in the immersion or English Control groups spoke French as
a
native language,
nor did they have close relatives who spoke French. Furthermore, the results of surveys carried out
with similar early immersion, one-year late immersion and FSL students suggest that none of these
groups has appreciably greater exposure to French outside school than the others (Genesee, 1978b),
indicating
that
extra-curricular exposure
to
French is not likely
to
play a
major role in
test performance.
5. French
control.
These students all spoke French as a native language and were attending regular
all-French schools.
All samples were drawn from schools in predominantly middle class neighborhoods by selecting
intact class groups. Selection of
the
English samples was based on the students'
IQ
as measured by the
Canadian Lorge Thomdikc Test of Intelligence. Although 10 information was not available for all of
the grade 7 and 8 students, the information that was available indicated
that
they were comparable and
included average to above average students. Information on ig was available for the entire grade 9
anglophone group. Analyses of variance of their io scores, using language program as the main
variable, failed to reveal statistically significant differences between the groups, TO, 189) = .81,
p > .05.
Test
Battery
With the exception of the Tests de rendement, identical tests were administered to the grade 7 and S
samples. Alternative forms of the same types of tests were administered to the grade 9 samples.
1.
Test de rendement en francais. This test assesses a variety of French language arts skills,
including spelling,
parts
of speech, word knowledge, logical sequencing,
and
comprehension of prose.
Appropriate levels of the test were administered to each grade level. Statistical analyses were
performed on stanine scores which are based on the performance of native French-speaking students
attending French schools in Montreal.
2.
Reading: Test de lecture. The grade 7 and 8 reading test consisted of two cloze passages from
which
every seventh word
had been
deleted
and
replaced by
a
dash.
The
students had to
fill
in the
spaces
by supplying the missing words. The total acceptable response method was used to score the tests.
EARLY AND LATE SECOND LANGUAGE LEARNING 119
The grade 9 students were administered the Test de comprehension de
l'ccrit
(Ontario Institute for
Studies in Education, 1977a) which consists of 10 separate items (e.g., newspaper articles, a brief
story, a
recipe)
followed by a
number
of multiple-choice questions. Total correct scores were analyzed
3.
Dictation: La dictie. The students were asked to transcribe a text presented orally by the exa-
miner. The text was
read
three times - first without interruption, while the students listened; a second
time, in short segments, as the students transcribed it; and a final time, again without interruption, to
allow the students to make corrections. Each student's score comprised the total number of words that
were misspelled, added, or omitted.
4.
Writing:
Composition.
The grade 8
and
9 students were asked to write
a
composition of 75 to 100
words based on one of several topics provided by the examiner. The compositions were written in one
45-minute class period. Each composition was subsequently assigned a global rating based on the
independent judgments of two raters, both native French-speakers. The ratings ranged from 5
(superior) to
1
(very poor) and took into account the ideas expressed in the composition, the form and
quality of presentation, richness of vocabulary, and complexity of sentence structure. Detailed error
analyses of sub-samples of the compositions were also carried out but are not reported here (see
Genesee & Morcos, Note 4).
5.
Listening
comprehension:
Test
de
comprehension
auditive.
The grade 7 and 8 students were
read
a
story extracted from a secondary school reader
and
then answered multiple-choice questions
related
to
it.
The story was
read
twice with the relevant questions following the second reading.
The grade 9 listening comprehension test (Ontario Institute for Studies in Education, 1977b)
consisted of a series of items excerpted from French radio broadcasts, e.g., news items, a weather
forecast, and an interview, followed by a number of multiple-choice questions. Both tests were
administered by means of recl-to-reel tape recorders. Total correct scores were analyzed statistically.
6.
Speaking:
Production orale. A sub-sample of
each
student group (see Table 1) was interviewed
individually for 5-10 minutes by one of two native French-speaking interviewers. The interviews,
which were initiated by having each student describe
a cartoon picture and
were continued
by
engaging
the student in a short conversation on matters of interest to him/her, were tape recorded for later
analyses. The interviewers were "blind" as to the school group of each student.
Each student's interview was later rated blind by two native French-speaking evaluating, working
independently, on five different dimensions: comprehension, pronunciation, grammar, vocabulary,
and communicativeness. The communicativeness dimension was intended to assess the student's
willingness
to
speak French
and
his/her
fluency
while doing so.
The ratings ranged
from
a
maximum of
5, indicating excellent or nativelike skill, to a minimum of /, indicating very poor skill. The ratings of
both evaluators were combined to form an average score for each dimension. The French Control
students' interviews were included during the evaluation of the speech samples to provide a basis of
comparison for the other groups, but they were not included in the statistical analyses because of their
small numbers.
RESULTS
Test results have been analyzed by means of analysis of variance using language
program (early immersion, one-year late immersion, divided into centre and wing
subgroups
in
the grade
7
and
8
analyses, two-year
late
immersion, English Control,
and French Control) as the basis of classification. Newman-Keuls Tests of
Multiple Comparisons were carried out whenever the analyses of variance
revealed significant main effects. The test results and statistical analyses are
summarized in Tables 2 to 4.
120GENESEE
TABLE 2
Summary of Grade 7 test results
Tests
1.
Test de rendement
en francais (9)
2.
Test de lecture (29)
3.
Ladictec
4.
Test de comprehension
auditive (20)
5. Production oraleb
(a) Comprehension (5)
(b) Pronunciation (5)
(c) Grammar (5)
(d) Vocabulary (5)
(e) Communicativeness (S)
French
control
N.A.*
25.95
5.24
15.02
5.00
4.86
4.05
4.95
4.81
Early
immersion
4.80
23.88
4.75
13.50
4.48
3.78
3.75
3.76
3.63
Centre
3.70
19.92
10.40
11.08
4.02
3.67
3.56
3.45
3.30
Wing
3.55
18.54
11.02
10.02
3.88
3.52
3.44
3.34
3.26
English
control
1.90
9.88
27.72
N.A."
2.90
2.72
2.72
2.54
2.46
F
ratio
39.74*
57.37*
70.58*
31.51*
34.41*
22.45*
26.04*
22.70*
22.28*
Note. Maximum test scores are presented in parentheses.
Means
joined by a solid line do not differ
significantly from one another according to the Newman-Keuls Multiple Comparison procedure.
*N.A. Not administered.
'Trench control results were not analyzed statistically.
*p< .001.
Grade7
The Early Immersion students scored significantly higher than the one-year late
immersion students, both Centre and Wing, on all of the tests except the
pronunciation and grammar scales of the oral production evaluation, in which
cases there were no significant differences. There were no significant differences
between the Centre and Wing students on any of the tests.
Comparisons between the immersion and English Control groups revealed that
the latter scored significantly lower than the former on all of the
tests.
Comparisons
with the French Control group revealed that they scored significantly higher than
all three immersion groups, except in the case of reading comprehension and
dictation, where the Early Immersion students scored at the same level as the
French Controls. The English Control students scored significantly lower than the
French Control students on all of the tests.
The Early Immersion students scored in the average stanine range (4-5-6) for
francophone students on the Test de rendement en francais; the Centre and Wing
students scored somewhat below average; and the English Control group scored
well below average.
TABLE 3
Summary of Grade 8 test results
Tests
1.
Test de rendement
en
franc,
ais (9)
2.
Test de lecture (29)
3.
Ladictee
4.
Test de comprehension
auditive (20)
S. Composition (5)
6. Production oraleb
(a) Comprehension (5)
(b) Pronunciation (5)
(c) Grammar (5)
<d) Vocabulary (5)
(e) Communicativeness (5)
French
control
N.A.a
26.69
2.51
15.40
3.30
5.00
4.79
4.71
4.79
4.82
Early
immersion
5.27
24.50
7.05
12.63
2.49
4.69
3.57
3.34
3.30
3.16
Grade 7/8
immersion
5.57
24.60
4.00
13.15
2.56
4.55
3,60
3.35
3.36
3.34
Centre
4.29
20.63
11.07
10.86
2.00
4.49
3.15
3.02
2.88
2.89
Wing
3.61
18.77
11.18
9.59
1.81
4.04
2.79
2,72
2.56
2.58
English
control
2.85
17.78
18.54
10.13
1.66
3.35
2.18
2.02
1.97
2.33
F
ratio
24.67*
34.01*
41.44*
27,99*
40.04*
13.07*
17.89*
15.86*
17.19*
11.57*
Note. Maximum test scores are presented in parentheses. Means joined by a solid line do not differ significantly from one another
according to the Newman-Keuls Multiple Comparison procedure.
*N.A. Not administered.
'"French control results were not analyzed statistically.
*p< .001,
TABLE 4
Summary of Grade 9 test results
Tests
1,
Test de rendement
en franc,ais (9)
2.
Test de comprehension
de
1'ecrit
(20)
3.
Ladictee
4.
Test de comprehension
auditive (17)
5.
Composition (5)
6. Production oraleb
(a) Comprehension (5)
(b) Pronunciation (5)
(c) Grammar (5)
(d) Vocabulary (5)
(e) Communicativeness (5)
French
control
N.A."
16.05
7.69
14.12
i 4.40
5.00
5.00
4.94
5.00
5.00
Early
immersion
5.47
16.97
10.58
13.20
2.59
4,96
3.98
3,70
3.44
3.69
Grade 7/8
immersion
5,31
17,71
11,63
11.16
2,13
4.98
4,12
4.01
3,78
3.81
Grade 7
immersion
4.73
15.61
20.54
9.75
2.53
4.92
3.84
3.39
3.16
3.18
English
control
2.02
8.98
38.72
5.85
1.23
4.32
2.85
2,00
1.98
2.13
F
ratio
69.52*
56.99*
79.74*
56.42*
86.45*
7.70*
62.92*
75.19*
67.96*
58.97*
Note, Maximum test scores are presented in parentheses. Means joined by a solid line do not differ significantly from one
another according to the Newman-Keuls Multiple Comparison procedure.
"N.A. Not administered.
'"French control results were not analyzed statistically.
*p< .001.
EARLY AND LATE SECOND LANGUAGE LEARNING 123
Grade 8
There were no significant differences between the Early Immersion and two-year
late immersion students on any of the tests except the dictation in which case the
Early Immersion group scored higher. The Early Immersion and two-year late
immersion groups scored significantly higher than both one-year late immersion
groups on all tests except the comprehension and grammar scales of the oral
production evaluation in which cases there were no significant differences between
the two-year immersion group and the Centre group. The one-year Centre
immersion students scored at the same level as the one-year Wing immersion
students on
a
number of the tests-dictation, listening comprehension, composition
and the pronunciation, grammar and communication scales of
the
oral production
evaluation. The Centre students scored significantly higher than
the
Wing students
on the other tests.
Comparisons involving the English Control students revealed that they scored
significantly lower than the Early Immersion and two-year late immersion students
on all of the tests, and lower than the Centre students on all tests except listening
comprehension and the composition where there were no significant differences.
The English Control students also scored significantly lower than the Wing stu-
dents on most of the tests, the exceptions being reading, listening comprehension,
writing, and the communicativeness scale of the oral production evaluation.
Comparisons involving the French Control group revealed that they scored
significantly higher than all of the other groups on all of the tests except dictation in
which case there was no difference between the French Control group and the
two-year late immersion group.
The Early Immersion, two-year late immersion, and Centre students scored in
the average stanine range on the Test de rendement en franc.ais, while the Wing
and English Control students scored in the below average range.
Grade 9
The Early Immersion students scored at the same level as the two-year late
immersion students on all tests except listening comprehension in which case the
former scored significantly higher than the latter. The Early Immersion group
scored significantly higher than the one-year late immersion group on dictation,
listening comprehension and the communicativeness scale of the oral production
evaluation; otherwise, there were no statistically significant differences between
these two groups. In contrast, the two-year late immersion students scored
significantly higher than the one-year late immersion students on all of the tests
except the Test de rendement en franc.ais and the comprehension and pronuncia-
tion scales of the oral production evaluation, in which cases there were no
significant differences.
The
English Control group scored significantly lower than all other groups on all
of the tests. The French Control group scored significantly higher than the
124
GENESEE
one-year late immersion group on all of the tests, higher than the two-year late
immersion group on writing and listening comprehension, and higher than the
Early Immersion group only on writing. Other comparisons with the French
Control group were not significantly different.
All three immersion groups scored in the average stanine range on the Test de
rendement en francais, while the English Control students scored in the below
average range.
GENERAL SUMMARY AND DISCUSSION
The results can be summarized in terms of the following statements:
1.
The Early Immersion students scored significantly higher than the one-year
late immersion students on every test administered at the grade 7 level, on most of
the tests administered at the grade
8
level, but on only two of the tests administered
at the grade 9 level (La dictee and Le test de comprehension auditive). A similar
pattern of results characterized comparisons between the two-year and one-year
late immersion groups.
2.
The English Control groups scored generally lower than all other groups at all
three grade levels.
3.
The Early Immersion students scored at the same level as the two-year late
immersion students on most tests. These results have been replicated by Genesee
and Morcos (Note 4) and by Adiv and Morcos (Note 5).
4.
The Early Immersion groups and the two-year late immersion groups scored
at the same level
as
the French Control students on
the
dictations and on the reading
tests at all three grade levels and they scored in the average stanine range on the
Test de rendement at all three grade levels- Their performance on the listening
comprehension test varied, sometimes being equivalent to that of the French
Controls and sometimes being poorer. They were rated consistently lower than the
French Control students on the oral production evaluations. The one-year late
immersion groups scored generally lower than the French Control groups at all
three grade levels.
Findings 1 and 2 corroborate previous studies in demonstrating the superior
second language achievement of students in immersion programs in comparison to
that of students in traditional second language programs, and the superior second
language achievement of early immersion students in comparison to one-year late
immersion students (Genesee, 1978a; Genesee et al., 1977, Swain, 1978; for
example). Of particular interest is the additional finding that the two-year late
immersion and early immersion student groups performed quite comparably. That
the early immersion groups scored as well as the two-year late groups despite the
fact that the former had been in total immersion during kindergarten, grades
1
and
2,
six to seven years prior to the time of testing, attests to the robustness of early
acquired second language competence. The findings of a comparative evaluation
EARLY AND LATE SECOND LANGUAGE LEARNING 125
of similar early and late immersion programs in the Ottawa area suggest that if the
early immersion follow-up program is more extensive than that provided in the
programs in the present evaluation, then early immersion students may actual-
ly show consistent and marked superiority in French when compared to students
who have been in a two-year late immersion program (Morrison, Note
6).
It would
appear from a consideration of the Montreal and Ottawa programs that
considerable follow-up time is necessary if this is to happen. The Ottawa early
immersion program maintained a French-to-English ratio of 80:20 in grades 3,4,
and
5
and 50:50 in grades 6 to 8, while the Montreal program maintained a ratio of
60:40 in grade 3 and 40:60 in grades 4 to 8; both programs provided all instruction
in kindergarten, grades 1 and 2 in French.
That the two-year late immersion students achieved parity with the early
immersion students despite the fact that the former had had considerably less
cumulative exposure to French than the latter at the time of evaluation implies
relatively faster learning on the part of the older students and suggests, therefore,
that beginning intensive second language instruction early in school is not
necessarily advantageous, other things being equal. Because the early immersion
students had received considerably more exposure to French in school than the late
immersion students by the time of testing it is not possible from these data to isolate
the effect of age per
se.
However, the Test de rendement provides a suitable basis
for comparing groups of students at different age levels but with the same length of
second language exposure since it has been standardized at all grade levels, and it
has,
in fact, been administered to immersion students in Montreal at all grade
levels.
If one
takes
into account FSL instruction during kindergarten to grade 6, the
two-year late immersion students had had almost the same amount of exposure to
French at the end of grade 8 as the early immersion students had had at the end of
grade 2, namely, 2'/2 years. When a post hoc comparison is made between grade 2
early immersion students (Polich, Note 7) and grade 8 two-year late immersion
students, it is found that whereas the latter scored in the average stanine range,
(i.e.,
5) on the grade 8 test, the former scored in the below average range on the
grade 2 test. Thus, these results along with the present findings corroborate those
from studies of short term second language learning in conventional instructional
programs in demonstrating that young children are not necessarily more effective
learners than older children (Asher& Price, 1967;Burstall,
1974;
Fathman, 1975;
Price, 1978; Stern, 1963) and, in fact, may make slower progress.
Krashen, Long, and Scarcella (1979) claim in a review of the literature that
older learners make rapid progress particularly in learning the syntactic and
morphological aspects of a second language. However, given the preponderance
of syntactic and morphological tests used in the research to date, this generaliza-
tion may be somewhat artifactual. In fact, the results from the few communica-
tively-oriented tests, such as listening comprehension, that have been used,
evince superiority on the part of older learners in learning these types of skills as
126 GENESEE
well (see Krashen et al., 1979, for a review of relevant studies). The oral produc-
tion and listening comprehension results from the present evaluation also indicate
that interpersonal communication skills may be acquired relatively quickly by
older students.
It is important to point out, before
concluding,
that not all the variables that might
influence the relative effectiveness and/or desirability of early versus late intensive
second language instruction have been considered (see
Genesee,
1978c,
for a more
thorough discussion of this issue). For example, while the factor of exposure has
been discussed, it has been considered only within the context of the school
curriculum. A corollary aspect of this factor is the possibility of extracurricular
second language exposure, which, to the extent that it actually occurs, is
tantamount to extending the learning
experience.
While this possibility is available
to students in both early and late immersion programs, it is greater in the case of
early second language programs by virtue of their longer time frame. This type of
exposure is likely to be particularly important since it has been found that
individuals who begin learning a second language in childhood generally achieve
higher levels of proficiency in the long term than those who begin in adolescence or
adulthood if exposure occurs in natural settings (Asher & Garcia,
1969;
Oyama,
1976,
1978;
Seliger et al., 1975). Therefore, the possibility of extracurricular use
of the second language and its associated benefits in terms of eventual language
proficiency should be a major consideration in bilingual communities where real
opportunities to use the language exist.
Evaluation comparee de trois types de programmes d'immersion en langue franchise: (a) immersion
totale precocc, (b) immersion retardee d'un an, el (c) immersion rctardee de deux ans. Lcs sujets
participant a chacun des programmes sont dex £]£ves de 7e, 8c et 9e anncc scolaire. Les epreuves
suivantes de langue servenl a evaluer le progres realise' en francais: lecture, rendement en francais,
composition, dictee, comprehension auditive, et production orale. L'analyse statistique des r£sultats,
pour chacunc des armies scolaire;,, revile que I'immersion precocc ct
1'immersion
retardee
dc
deux ans
donne des meilleurs resultats que
1'immersion
rctardee d'un an, sans difference importante entre ces
deux premiers types de
programme.
Les resultats indiquent cgalement que la difference observte entre
1'immersion
retardee d'un an ct les deux autrcs types d'immersion s'amenuise
dans
les annees scolaires
les plus avancces. La discussion releve la signification pratique de ccs resullats, notamment en ce qui
concerne les programmes de relance, ct en releve £galemenl les implications thcoriques relatives k
l'hypothese
d'une
pfiriode critique dans l'apprentissage
d'une
langue secondc.
REFERENCE NOTES
1.
Cziko, G., Holobow, N., & Lambert, WE. Early and late French
immersion:
A compari-
son of children at grade seven. Mimeo, Department of Psychology, McGill University, April
1977.
2.
Genesee, F. Comparative evaluation of the early French immersion, grade 7 French immer-
sion and FSL programs: A follow-up study. Instructional Services Department. Protestant
School Board of Greater Montreal, November 1976.
... Students enrolled in early immersion programmes have generally been found to achieve greater target language proficiency than their counterparts who start later, especially in oracy (Genesee, 1987;Turnbull, Lapkin, Hart, & Swain, 1998). However, it is possible for students enrolled in late immersion programmes to perform just as well in their reading and writing as their early immersion counterparts, even though they have less exposure to the language (Genesee, 1981). Some reasons given for this finding are that the older students can transfer the knowledge and skills acquired in their dominant language to the target language, tend to be self-selecting at this age and are more committed and motivated to learn the language (Lindholm-Leary & Genesee, 2014). ...
... ucation in the 21st century: A global perspective. Malden, MA: Wiley-Blackwell. García, O., Ibarra Johnson, S., & Seltzer, K. (2017). The translanguaging classroom: Leveraging student bilingualism for learning. Philadelphia, PA: Caslon. García, O., & Li, W. (2014). Translanguaging: Language, bilingualism and education. New York: Palgrave Macmillan.Genesee, F. (1981). A comparison of early and late second language learning.Canadian Journal of Behavioral Science, 13, 115-127. https://doi. org/10.1037/h0081168Genesee, F. (1983). Bilingual education of majority-language children: The immersion experiments in review. Applied Psycholinguistics, 4(1), 1-46.Genesee, F. (1987). Learning through two language ...
Book
"This timely volume is an important contribution to the fast expanding literature on Translanguaging. The coverage of mainstream school education, TESOL, CLIL and EMI is most impressive. The rich data from the Australian context have significant implications for school management, pedagogy and the learning outcomes of bi/multilinguals worldwide." --Li Wei, Chair of Applied Linguistics, University College London, UK This book explores multilingualism as a resource and goal at school in contexts of student diversity and institutional monolingualism. Combining translanguaging theory and sociocultural theory, the author proposes a framework for the learning and use of both foreign and heritage languages across the curriculum in mainstream schools. By clearly linking language practices to teaching and learning objectives, the book aims to support school leaders and practitioners make informed decisions about how best to promote multilingualism in their school, as well as to enhance the learning outcomes of bi/multilinguals. In addition to school leaders and practitioners, it will be of interest to students and academics in the fields of bilingual education and TESOL, as well as applied linguistics and language teaching more broadly. Marianne Turner is Senior Lecturer in Bilingual Education and TESOL at Monash University, Australia. She researches context-sensitive approaches to the integration of language and content in English as an Additional Language (EAL), foreign and heritage language settings.
... An opposite prediction could be that older learners may benefit from increased cognitive abilities and more advanced L1 academic proficiency (Jaekel et al., 2017;Lorenzo et al., 2010;Muñoz, 2015). North American research found an 'older-is-better' trend in full immersion programs (Genesee, 1981), with students in later immersion programs performing as well as students in early immersion classes (cf. Wesche, Toews-Janzen, & MacFarlane, 1996). ...
Article
Full-text available
This study reveals hitherto overlooked effects of age of onset (AO) in immersive school contexts, using multiple measures over time so as to focus on fluctuations and trends in individual data. The second language (L2) English development was studied in 91 children who received 50–50 content and language integrated learning (CLIL) instruction in German and English and varied in their AO (5, 7, or 9). Data collection occurred 4 times annually for up to 8 school years (ages 5–12), via oral and written production tasks, motivation questionnaires, and interviews. Meaningfully integrating quantitative analyses (GAMM) and qualitative analyses, the study focuses not only on the process itself and on quantification of change but also on the underlying environmental and psychological reasons for change. Results reveal that (a) slightly later CLIL beginners (AO 7) turn out to show similar L2 development to that of the earlier beginners (AO 5), (b) besides external states and events, many internal states at any given moment contribute to significant L2 growth, (c) learners show significant improvement in the last 2.5 years of primary school, starting from age 10, and (d) there are significant differences between L2 oral and written performance in terms of height and shape of learner trajectories across (pre)primary school.
... Studies have compared the English-language development of students enrolled in oneway and two-way programs that devote more instructional time to the minority language and less to English (at least initially) with those that devote approximately 50% to each language throughout the duration of the program. They have found no difference in English-language achievement despite less exposure to English (e.g., Christian, Genesee, Lindholm-Leary, & Howard, 2004;Genesee, 1981;Lindholm-Leary, 2001). ...
... Kimilerine göre ise dil öğrenme merkezinin gelişiminin, yani yanallaşmanın, daha erken yaşta gerçekleştiği, bu gerçekleşmenin 5 yaş civarında tamamlandığı varsayılmaktadır. Bu ekolde olanlara göre, çocuklar için çevresinde konuşulan dili edinmenin en uygun evrenin 0-5 yaş arasındaki zaman dilimi; ebeveynlerin evde veya çocukların yakın çevresinde konuşulmayan dili edinmenin en uygun zaman dilimimin 2-7 yaş aralığı; ikinci bir dili öğrenerek edinmenin en uygun zaman diliminin ise 10-13 yaş aralığı olduğu iddia edilmektedir ( Penfield & Robert, 1959;Krashen,1979;Long & Scarsella, 1979;Genesee, 1981;Harley, 1989;Newport, 1990). ...
... In keeping with previous research by Canale and Swain (1980), we are of the view that successful mastery of an L2 involves not only knowledge of grammatical and phonological forms, but also knowledge of formal variants, informal variants and local norms. Mastery of such L1 speaker norms is important for students who lament not being able to interact with L1 speakers in a natural way (Auger, 2002;Genesee, 1978Genesee, , 1981MacFarlane, 2001;Tarone & Swain, 1995;Thibault & Sankoff, 1993). This goal further reflects the fact that the Ontario Ministry of Education (2000) clearly stated that one of the aims of French immersion programmes is to produce students who are able to incorporate colloquialisms and idiomatic expressions into their speech and debate issues both formally and informally. ...
Article
Our study examines the extent to which French immersion students use lax /ɪ/ in the same linguistic context as native speakers of Canadian French. Our results show that the lax variant is vanishingly rare in the speech of immersion students and is used by only a small minority of individuals. This is interpreted as a limitation of French immersion students’ sociolinguistic competence. Within the group of students who do use both variants, we document a positive correlation between female and middle-class students and use of the lax variant and suggest these speakers are generally more sensitive to sociolinguistic variation. A reverse correlation between English cognates and laxing was found. This is taken as evidence that the learning of laxing is lexically mediated.
... The rationale behind the rubric requirements for specific amounts of time in the target language comes from research demonstrating that the more time spent in TL, the higher TL proficiency attained (Genesee, 1987;Lindholm-Leary, 2001;Lindholm-Leary & Howard, 2008;Turnbull, Lapkin, & Hart, 2001). It is important to note that Genesee (1981) and Lindholm-Leary (2014) found no differences are attained in first language (L1) development in dual language programs regardless of the amount of time spent in L1; however, the length of time spent in the TL had a profound effect on TL proficiency. There is even research that shows that despite reduced exposure to L1, immersion students L1 skills exceed those of non-immersion monolingual students (Björklund & Mård-Miettinen, 2011;Lindholm-Leary & Howard, 2008;Lambert, Genesee, Holobow, & Chartrand, 1993). ...
Article
Full-text available
At the time of the demise of bilingual education in California via Proposition 227, Chueng and Drabkin (1999) discussed the unchecked prevalence of poorly run bilingual programs and the apathetic administrators who provided inadequate oversight in terms of ensuring program quality. Proponents of dual language immersion programs hope to avoid this apathy and poor implementation. Program effectiveness depends on the quality of its implementation (Li, Steele, Slater, Bacon, & Miller, 2016). While immersion practitioners and researchers are acutely aware of the list of non-negotiables that are the cornerstone of a successful immersion pathway (Fortune, 2009, de Jong, 2016), they often encounter difficulties in engendering respect of these principles among site administrators. These administrators are confused by some of the counterintuitive aspects of the immersion model. In Louisiana, Louisiana Act 196 (Session 2014) was the solution found for this issue. Louisiana Act 196 requires that the state design a template reflecting these non-negotiables, create a review rubric, attach a state certification to it, and make it mandatory for schools to obtain their certification by 2016- 17. This literature review examines the rationale for each element of the immersion state site certification rubric.
Article
Two-way immersion (TWI) is a variant of the increasingly popular bilingual instruction. Most TWI research lacks longitudinal data or the consideration of background variables to control for possible selection effects. This article examines the development of German reading comprehension of TWI students (N = 984) from fourth to sixth grade compared to conventionally taught students (N = 992). The latent growth curve models showed that immersion students reached the same level of German reading comprehension over the three measurement points, even if background variables like first language, socioeconomic background, and cognitive ability were included. Despite reduced instruction in German, TWI students showed the same reading comprehension level as students in regular instruction while having the advantage of learning an additional language. Although the level of reading comprehension differed between language groups (L1 German speakers, L1 partner language speakers, simultaneous bilinguals), the learning trajectories of reading comprehension were similar.
Article
Previous research has established that early second language (L2) learners in classroom immersion may not ultimately produce all L2 morphosyntactic features as first language (L1) speakers of the language do, whereas L2 comprehension outcomes are reported to be less divergent from those of L1 speakers. However, immersion learners’ L2 comprehension is typically assessed using tasks of holistic understanding, and therefore, little is known about fine-grained comprehension of specific morphosyntactic constructions. To address this, the present study examined online comprehension of English plural–singular marking by Mandarin-speaking, English-immersion learners in Taiwan. This semantically transparent feature differs from the L1 grammar and is a notable area of difficulty for Mandarin-speaking L2-English learners. The present study assesses middle school-aged immersion learners’ comprehension using a visual-world eye-tracking task combined with a picture decision task, comparing results to age-matched English-monolingual controls. After more than 8 years of L2 exposure, the immersion participants showed similarities and differences to monolinguals in plural–singular marking comprehension as measured by eye-tracking, and were less accurate in their interpretations on the picture decision task. This study shows that comprehension differences for a semantically transparent morphosyntactic construction can be apparent even after many years for learners who started immersion at an early age.
Article
Full-text available
In the last two decades, the implementation of bilingual education programs has become commonplace in Spain. This article reviews three of the myths about bilingual education that most concern families and that have led to debate within the Spanish context. The literature shows that the benefits of learning two languages are much greater than the disadvantages. Keywords: bilingual education, bilingualism, myths, realities
Chapter
The chapter considers ways in which languages have been structured into everyday teaching and learning across the curriculum. Use of languages is considered in different contexts and includes programmes that target the learning of a language at school and English-medium settings where students are able to leverage a language they speak at home to assist in their learning. My aim in the chapter is to provide some background on how bilingual education and the education of (emergent) bi/multilinguals has been conceptualised in relation to language use, and also to discuss language-related pedagogies. Cross-curricular distribution of languages is discussed first in relation to instructional language separation, language flexibility and incorporation of students’ languages into English-medium classes. Everyday language-related pedagogies then focus on both learning through a target language and cross linguistic learning.
Book
The coming of language occurs at about the same age in every healthy child throughout the world, strongly supporting the concept that genetically determined processes of maturation, rather than environmental influences, underlie capacity for speech and verbal understanding. Dr. Lenneberg points out the implications of this concept for the therapeutic and educational approach to children with hearing or speech deficits.
Article
This report presents the second evaluation of the progress of two groups of children from working class, English-speaking Canadian homes who have participated in a French immersion program for the first two and three years of schooling. The children in the experimental program are compared with control groups conventionally educated in French or in English at the appropriate grade level. The findings suggest that the children in the Experimental group have begun to master basic French and English language skills and are able to acquire content material taught via French without detrimental side effects. The progress of these children is comparable to that of their middle-class peers who are also in immersion programs.
Article
The assumption that younger children can master the phonological system of a second language more easily than those who are older was tested under laboratory conditions. This assumption is based on observations of immigrant children in natural settings and findings related to the ability of various age groups to recover full use of speech function following trauma to the dominant cerebral hemisphere. In the study, each of three groups of twenty elementary, twenty junior high, and twenty college students received ten sessions, each 15-25 minutes in length, of pre-taped German phoneme pronunciation instruction. A total of thirty-three phonemes were taught in 2 weeks using various mimicry drills. The students were pretested and post tested and given the Raven Progressive Matrices Test. Analysis of variance and covariance on the pretest indicated no difference in pronunciation. Contrary to common belief, on the posttest the junior high and college groups were significantly (p .01) better at pronunciation than the elementary group.
Article
On a évalué tous les ans, jusqu'à la fin de leur 3e année, des enfants qui éprouvent des difficultés de langage inscrits dans des classes maternelles d'immersion en français, afin de déterminer si ces programmes leur conviennent. Leurs résultats à plusieurs épreuves académiques, cognitives et linguistiques sont comparés à ceux d'un groupe d'enfants inscrits dans des programmes d'anglais ayant aussi des difficultés de langage, ainsi qu'à deux groupes-contrôles d'enfants normaux. Les comparaisons ont indiqué que ces enfants acquièrent les bases de leur langue maternelle et apprennent les aspects fondamentaux de la lecture, de l'orthographe et des mathématiques; ils réussissent également dans l'étude de la langue seconde. On interprète les données dans le sens d'une continuation du programme d'immersion avec une aide appropriée en ce qui concerne le rattrapage et non d'un retrait des enfants de ces programmes d'immersion.