2.

Task-Based Language Teaching for Beginner-

Page 2
Task-Based Language Teaching for Beginner-Level Learners of L2
French: An Exploratory Study
This study was motivated by a teacher development program for school teachers of
foreign languages in New Zealand that aims to equip them with the skills they need
to implement a task-based approach. It also drew on theoretical considerations.
Page 3
Pedagogical background
The study we report was designed to address (a) and (b) by investigating how
performing tasks can lead to learning in a real classroom with the students' usual
teacher.
Input-based tasks are viable in the case of beginners, who lack knowledge of the
language and therefore cannot be expected to engage in free production. They are
premised on two assumptions: learners can comprehend the input and they can pick
up specific linguistic features embedded in the input.
Page 4
The tasks were focused on facilitating the acquisition of specific, predetermined
language. In the case of a production-based task, a focused task is designed to elicit
production of the target language, whereas in the case of an input-based task, it is
designed to attract the learners' attention to the target features.
Focused input-based tasks must still meet the criteria for tasks in general, and must
encourage learners to treat language as a tool for communicating rather than as an
object of study.

Theoretical background
We drew on several key theoretical constructs in second language acquisition to
frame the study. These constructs include receptive knowledge, form - function
mapping, and default processing strategies.
Page 5
Receptive knowledge precedes production
Lightbown (2016, p.193) summarized her lifelong work in classroom language
learning by commenting that comprehension is the first requirement for second
language learning. She found that a comprehension-based program for young
learners in a New Brunswick school was effective at improving their understanding
and producing spoken language.
Vocabulary studies have shown that receptive knowledge outstrips productive
knowledge and that receptive knowledge of new words is established before
productive knowledge. An input-based approach to teaching can potentially facilitate
productive as well as receptive knowledge of new vocabulary.
In comparison with vocabulary, research on grammar has paid little attention to the
acquisition of receptive knowledge. However, it is generally accepted that receptive
knowledge must precede productive knowledge, and that receptive knowledge of
morphological features such as markers of plurality must be established before
production.
Page 6
Necessity of attention
Schmidt (2001) claimed that attention is constrained by working memory, and that
learners need to focus their attention on specific linguistic forms for learning to occur.
Tomlin and Villa (1994) suggested that attention need not involve consciousness.
There are two views of the role of attention in language learning: implicit learning and
incidental acquisition. Implicit learning occurs without intention and without
consciousness, whereas incidental acquisition occurs with intention and
consciousness.
Page 7
Focused input tasks expose learners to multiple exemplars of the target language
and cater to detection and frequency-driven implicit learning. However, such tasks
potentially do more than this, because they create a functional need for learners to
process the target features.
Overcoming default processing strategies
Learners are naturally predisposed to attend to those aspects of language that can
be processed most easily and that pay the greatest dividends in enabling them to
comprehend. Therefore, learners are unlikely to attend to the plural markers in a
sentence such as Les deux chats sont sous la table.
Processing instruction involves structured input activities that force attention to key
grammatical features and the meanings they convey. Task-based language teaching
(TBLT) similarly aims to attract attention to linguistic form by means of focus on form,
but there are some fundamental differences.
Page 8
Previous research
Previous research has focused on production-based tasks, but relatively few studies
have investigated input-based tasks, which make it easier to investigate the
acquisition of specific linguistic features.
Several studies have investigated the effects of performing input-based tasks on
learners' comprehension and the acquisition of new words. These studies involved
post-beginner learners, but Shintani's research involves complete beginners and
shows that input-based tasks may work for grammar acquisition only if they succeed
in helping students create a form meaning mapping.
Page 9
Listen-and-do tasks, such as selecting the correct picture or performing an action,
can be effective in helping learners incidentally acquire new words and at least some
grammatical structures.

Research questions
The acquisition of receptive knowledge and productive knowledge were measured
by tests that required the learners to produce target words orally.

Participants
The study took place in a girls' school in one of New Zealand's major cities with 34
students. One class was chosen as the experimental group and the other class took
part in testing only.
Page 10
Most students spoke English as their first language, and had studied French in the
previous two years in three one-hour lessons per week in "taster" courses. However,
they had only 20 weeks of exposure to French in these "taster" courses.

Research methods
The experimental group completed a pre-test, 1.5 hours of teaching spread out over
two lessons on two separate days, and a post-test and delayed post-test were
administered. The control group completed all tests but otherwise received their
normal class instruction.
Page 11
Target structure
The target structure was markers of plurality in L2 French. The instruction was
entirely oral, so the study focused on plural marking on determiners and the common
irregular verb sont.
Page 12
Instructional treatments
The researchers designed a variety of input-based tasks and met with the
experimental group teacher to go over how each task was to be taught. The teachers
agreed to avoid explicit focus on the target structures during the study.
Pre-task
Students listened to 22 statements about clothes belonging to three adolescents and
chose the picture that best matched the statement. They were given no grammatical
explanation.
Task 1a: bingo game
Each student had a board with nine squares that contained pictures of clothes. The
teacher read out a clothing item and students crossed out the corresponding item.
Page 13
Task 1b: bingo moche
Students worked with the same boards used in Task 1a, but this time they heard
phrases such as "les casquettes sont moches" (the caps are ugly) and crossed off
the pictures that depicted these phrases.
Page 14
Task 2a: shopping task
Students were shown a table of clothing items and listened to sentences describing
what each person bought. They entered their purchases in a table according to who
made the purchase.
Task 3: cool or not?
Students referred to the same table of clothing items as in Task 2a and decided
whether each item was cool or moche. There was no correct answer to this task.
Satisfying the criteria for a task
The shopping task had a primary focus on meaning, and students had to rely on their
own knowledge of clothes vocabulary and plurality markers to make correct choices
for each stimulus.
Page 15
Feedback in relation to task completion
Students were given immediate feedback by the teacher for each picture choice in
the pre-task. They also received feedback about their learning when they achieved
the task outcome.
This test aimed to assess receptive learning of vocabulary. Students selected the
picture that best matched the word they had heard.
Grammar Listening Test
This test was designed to assess students' receptive knowledge of markers of
plurality. Students had to circle the picture that best matched the statement "Je
voudrais des tomates".
The reliability of this test was lower than desirable, possibly because students had a
50% chance of getting each item correct.
Page 16
Elicited Imitation Test
This test was designed to assess learners' productive knowledge of markers of
plurality. It consisted of 24 statements, and students had to select the picture that
best corresponded to the statement they had heard.
Students had to make the correct picture choice for an item to be scored. This was to
increase the possibility that the Elicited Imitation Test was reconstructive.
Students heard and repeated the statement "Il mange des chocolats" (he eats some
chocolates). Those who chose the correct picture received one mark for their
repetition of des.
Page 17
Analysis
Descriptive statistics were calculated for both groups on all tests. Parametric t tests
were used to compare the experimental and control groups' scores, and effect sizes
were calculated to establish between-groups differences or within-group differences
over time.
Vocabulary Listening Test
The high scores on this test were due to several items of vocabulary being cognates,
so non-parametric tests were performed to test for statistical significance. The
experimental group made greater gains than the control group.
Page 18
Grammar Listening Test
Independent-samples t tests showed that the experimental group made statistically
significant gains over the control group from pre- to post-test.
Teacher interview
The teacher thought that using input-based tasks was a good idea, but
acknowledged that using only production-based tasks would not have been
successful because the students did not have the knowledge to produce the target
structures.
Page 19
Although she said that the production tasks were teacher based and repetitive, she
did not provide grammatical explanations about new grammatical structures, and
said that she started to do so with Year 9 students.

Discussion
The study was designed to investigate whether input-based tasks were effective in
enabling learners to acquire new vocabulary and grammatical features.
Page 20
The results indicate that the students were able to acquire receptive knowledge of
the target vocabulary and that the gains were maintained over time. The mean gains
from pre- to post-tests for the experimental group were small.
The results show that the learners acquired receptive knowledge of the target
structure, and that the gain from pre- to immediate posttest was large. However, the
gain was not maintained from pre- to delayed posttest.
Input-based instruction contributed to the development of productive knowledge of
the plural markers in 11 of the students, but the difference between the experimental
and control groups did not reach statistical significance.
Page 21
The study shows that input-based tasks for beginner-level learners can promote
incidental acquisition of both vocabulary and grammar, by creating a functional need
for learners to attend to target features, consciously or subconsciously.
The second purpose of the study was to examine the teacher's experience of using
input-based tasks in her own classroom. The teacher reported that she had not used
input-based tasks of this kind previously, but could see their value with beginner-
level learners.

Conclusion
This was a small-scale study conducted in a real classroom. It may have some
ecological validity, but the researcher was present in the experimental group
classroom.
Page 22
The results demonstrate that incidental acquisition takes place when input-based
tasks are designed to create a functional need for learners to attend to target items.
Some limitations have already been acknowledged, such as the small number of
students taking part in the Elicited Imitation Test and the short period of instruction.
We also failed to point out the need to provide instant feedback on students'
responses to each stimulus.
Future improvements to the study include a longer period of instruction, more input-
based tasks, incorporating suggestions from the teacher, and using a research
assistant to assist with one-on-one testing.