LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 48
La enseñanza de lenguas basada en tareas y el desarrollo de la
expresión oral de nivel A2 en estudiantes de octavo grado
Task-Based Language Teaching and A2 Speaking Development in
Eighth-Grade Learners
FRANKLIN DAVID MOYA DAVILA*
https://orcid.org/0009-0003-7023-8188
franklin.moya@casagrande.edu.ec
Universidad Casagrande
Ecuador –Guayaquil
RAFAEL MANUEL FUNES GARCIA
https://orcid.org/0009-0002-2095-7208
rfunesg@est.ups.edu.ec
Universidad Politecnica Salesiana
Ecuador –Guayaquil
Article received: (22/06/2026) - Accepted for publication: 28/07/2026
Conflicts of interest: None to declare.
*Corresponding author
Suggested citation (APA, seventh edition)
Moya Davila, F., Funes Garcia, R. (2026). Task-based language teaching and A2
speaking development in eighth-grade learners. Linguatech, 2(2), pages 48 - 63.
RESUMEN
La enseñanza de lenguas basada en tareas (TBLT) organiza la instrucción en torno a resultados
con un propósito que requieren que los estudiantes intercambien significado. Este estudio
examinó los cambios en el desempeño oral de nivel A2 después de un programa basado en tareas
de ocho semanas. Se utilizó un diseño cuantitativo de un solo grupo con pretest-postest con 28
estudiantes de octavo grado de 12 a 13 años en un colegio privado urbano de Ecuador. Los
participantes completaron ocho sesiones de dos horas sobre presentaciones personales, rutinas,
descripciones personales, hechos pasados, predicciones, pedidos de comida, direcciones e
información de viaje. La expresión oral se evaluó antes y después de la instrucción con tareas de
formato A2 Key y tres escalas analíticas de 0 a 5: Gramática y Vocabulario, Pronunciación e
Interacción Comunicativa. Gramática y Vocabulario aumentó de M = 3.32 (DE = 0.94) a M =
4.25 (DE = 0.70); Pronunciación aumentó de M = 3.29 (DE = 0.98) a M = 4.29 (DE = 0.66); e
Interacción Comunicativa aumentó de M = 3.25 (DE = 0.80) a M = 4.25 (DE = 0.70). Las pruebas
de rangos con signo de Wilcoxon fueron significativas en las tres dimensiones tras la corrección
de Holm (todas con p ajustada < .001). Todos los participantes mejoraron en el puntaje compuesto
de 15 puntos, que pasó de M = 9.86 a M = 12.79. Los resultados indican una mejora intragrupo
considerable asociada con la secuencia instructiva. Sin embargo, la ausencia de un grupo de
comparación, la falta de reporte de la confiabilidad entre evaluadores y la ausencia de una
evaluación diferida impiden establecer conclusiones causales y de retención definitivas.
Palabras clave: enseñanza de lenguas basada en tareas, desempeño oral, estudiantes de nivel A2,
interacción oral, educación secundaria
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 49
ABSTRACT
Task-Based Language Teaching (TBLT) organizes instruction around purposeful
outcomes that require learners to exchange meaning. This study examined changes in A2
speaking performance after an eight-week task-based program. A quantitative one-group
pretest-posttest design included 28 eighth-grade learners aged 12-13 at an urban private
school in Ecuador. Participants completed eight two-hour lessons on introductions,
routines, personal descriptions, past events, predictions, food ordering, directions, and
travel information. Speaking was assessed before and after instruction with A2 Key-
format tasks and three analytic scales ranging from 0 to 5: Grammar and Vocabulary,
Pronunciation, and Interactive Communication. Grammar and Vocabulary increased from
M = 3.32 (SD = 0.94) to M = 4.25 (SD = 0.70); Pronunciation increased from M = 3.29
(SD = 0.98) to M = 4.29 (SD = 0.66); and Interactive Communication increased from M
= 3.25 (SD = 0.80) to M = 4.25 (SD = 0.70). Wilcoxon signed-rank tests were significant
for all dimensions after Holm correction (all adjusted p < .001). Every participant
improved on the 15-point composite score, which rose from M = 9.86 to M = 12.79. The
results indicate strong within-group improvement associated with the instructional
sequence. Nevertheless, the absence of a comparison group, unreported interrater
reliability, and lack of delayed assessment prevent definitive causal and retention claims.
Keywords: task-based language teaching, speaking performance, A2 learners, oral
interaction, secondary education
INTRODUCTION
Speaking is a central objective of foreign-language education because it requires
learners to coordinate linguistic knowledge with real-time processing. A speaker must
retrieve vocabulary, assemble grammatical forms, produce intelligible speech, interpret
an interlocutor, and respond under temporal pressure. These processes are difficult for
elementary learners, whose available language is limited and not yet automatic.
Classroom instruction can therefore produce an apparent contradiction: students may
recognize forms on a worksheet while remaining hesitant when they need to use the same
resources in an exchange. Addressing this gap requires instruction that preserves explicit
support while creating repeated reasons to communicate.
The Common European Framework of Reference for Languages describes A2
learners as basic users who can manage simple, routine exchanges and communicate
about familiar matters when support is available (Council of Europe, 2020). At this level,
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 50
successful performance does not require sophisticated discourse. It requires sufficient
control of simple language, intelligible pronunciation, and an ability to initiate or maintain
brief interaction. These components are closely related but not interchangeable. A learner
may have useful vocabulary yet be difficult to understand, or may pronounce familiar
words clearly but depend heavily on prompts. A balanced speaking program should
therefore integrate form, intelligibility, and interaction rather than equate oral ability with
grammatical accuracy alone.
Task-Based Language Teaching (TBLT) offers a framework for creating this
integration. In TBLT, tasks function as the primary unit of planning and require learners
to use language to achieve an outcome (Ellis, 2003; Nunan, 2004). A task may involve
finding differences, planning an itinerary, solving a practical problem, reaching
agreement, obtaining information, or performing a service encounter. The defining
feature is not the topic or resource but the relationship between language and purpose.
Learners must understand and convey meaning in order to complete the activity, and
success can be judged through the outcome as well as through the language produced.
This orientation distinguishes a task from an exercise. Exercises normally direct
attention toward reproducing a target feature accurately, whereas tasks direct attention
toward accomplishing something through language. The distinction does not mean that
form-focused work is excluded. Ellis et al. (2020) describe TBLT as capable of
incorporating preparation, feedback, and attention to form within a meaning-centered
sequence. A teacher can preteach essential vocabulary, provide planning time, observe
recurring difficulties, and conduct a post-task language focus. What changes is the
organizing logic: linguistic support serves a communicative goal rather than replacing it.
The task cycle is particularly useful for A2 learners. A pre-task phase can activate
topic knowledge, clarify the outcome, model useful expressions, and reduce uncertainty.
During task performance, learners use available resources to exchange information or
create a product. Planning and reporting then encourage them to organize what they want
to say, while post-task feedback draws attention to forms that became relevant during
communication (Willis & Willis, 2007). This sequence protects meaning while providing
structure. For adolescents who are reluctant to speak publicly, rehearsal in pairs or small
groups can also reduce the social risk of immediate whole-class performance.
Task difficulty must be carefully calibrated. Skehan (1998) argues that limited
attentional capacity creates trade-offs among fluency, accuracy, and complexity. If a task
contains unfamiliar content, many steps, and severe time pressure, elementary learners
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 51
may devote most of their attention to finding words and lose control of grammar or
pronunciation. Conversely, a task that is fully scripted may produce accurate output
without meaningful choice. Teachers can manage demand through familiar topics, visual
support, role cards, planning time, repetition, and progressive increases in complexity.
These adjustments do not remove the task outcome; they make the outcome attainable
with the learners' developing resources.
Interaction is one mechanism through which tasks can support development.
Collaborative work obliges learners to listen to one another, signal misunderstanding,
request clarification, reformulate messages, and confirm decisions. Foster and Ohta
(2005) showed that peer interaction involves more than formal negotiation routines; it
also includes assistance, co-construction, and mutual support. Such exchanges can
expand participation and make gaps in knowledge noticeable. For A2 learners, even short
sequences such as asking for directions or ordering food can require multiple turns,
responsive listening, and adjustment to a partner's contribution.
Repetition and planning can further stabilize oral performance. Repeating a
communicative task reduces the conceptual burden because learners already understand
the procedure and content, allowing more attention to formulation and delivery. Research
with child foreign-language learners has shown that task repetition can influence
negotiation strategies and pair dynamics (García Mayo & Imaz Agirre, 2016). Effective
repetition should preserve a genuine purpose while changing a condition, partner,
information set, or audience. The learner is then able to reuse language without simply
reciting an identical script.
TBLT also aligns with an action-oriented view of language proficiency. The CEFR
Companion Volume treats learners as social agents who use language to act with others,
not merely as recipients of linguistic content (Council of Europe, 2020). This view is
relevant to adolescent classrooms because familiar life domains can be transformed into
manageable tasks. Introductions, daily routines, descriptions, past experiences, future
plans, food, directions, and travel information provide accessible content while
supporting different interactional purposes. These topics can be sequenced from
predictable exchanges toward decisions and short narratives.
Assessment must reflect the intended construct. The A2 Key Speaking assessment
uses analytic criteria for Grammar and Vocabulary, Pronunciation, and Interactive
Communication on a 0-5 scale (Cambridge Assessment English, 2020). The criteria
provide a useful classroom framework because they separate linguistic range and control
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 52
from intelligibility and responsiveness. At the same time, scores depend on rating
judgments, task sampling, and administration conditions. Bachman and Palmer (2010)
emphasize that interpretations are strongest when assessment use is aligned with purpose
and supported by evidence about consistency. When pretest and posttest ratings involve
different evaluators, documentation of training and interrater agreement becomes
especially important.
Classroom evaluations of TBLT often operate under practical constraints. Intact
groups cannot always be randomly assigned, sample sizes may be small, and the same
teacher may design and implement the intervention. East (2017) notes that the translation
of TBLT from research into practice is shaped by teachers' beliefs, curriculum demands,
and contextual realities. These conditions do not make classroom evidence uninformative,
but they require claims to match the design. A one-group pretest-posttest study can
identify systematic within-group change and generate a basis for further testing; it cannot,
by itself, rule out maturation, test familiarity, concurrent instruction, or rater effects.
Research with school-age learners remains valuable because much task-based
literature has focused on adults, university students, or tightly controlled performance
tasks. Children and adolescents differ in cognitive maturity, classroom dependence,
motivation, and experience with collaborative learning. The scholarship synthesized by
Ahmadian and Long (2021) indicates that task-supported interaction can benefit school-
age learners when tasks, support, and expectations are developmentally appropriate.
Evidence from ordinary secondary classrooms can therefore clarify whether a structured
task cycle is feasible and which speaking dimensions appear most responsive over a short
instructional period.
The present study examined an eight-week speaking program for A2 eighth-grade
learners. Eight lessons placed communicative outcomes at the center of work on familiar
topics and were aligned with analytic speaking criteria. The objective was to determine
whether learners demonstrated improvement in Grammar and Vocabulary, Pronunciation,
and Interactive Communication from pretest to posttest. The null hypothesis stated that
the median paired difference would be zero for each dimension; the alternative predicted
positive change. A supplementary composite score was calculated to summarize
performance across all three dimensions while retaining dimension-specific analyses.
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 53
METHODS
Design
A quantitative one-group pretest-posttest design was used. All participants
completed an A2 speaking assessment before the intervention, received the same eight-
week task-based program, and completed a posttest using the same analytic criteria. The
design is commonly described as preexperimental or single-group quasi-experimental
because exposure is manipulated but no randomized or non-equivalent comparison group
is present. Accordingly, the analysis focuses on within-group change and avoids
attributing every observed difference exclusively to TBLT.
The independent condition was participation in eight task-based speaking lessons.
The primary dependent outcomes were scores for Grammar and Vocabulary,
Pronunciation, and Interactive Communication. Each dimension was scored from 0 to 5.
A composite speaking score, ranging from 0 to 15, was derived by summing the three
dimension scores for each learner at each measurement point. This composite was
supplementary because the analytic dimensions remain pedagogically distinct.
Context and participants
The study was conducted in an urban private school in Ecuador. The accessible
population comprised one intact eighth-grade cohort, and all 28 enrolled learners
participated. Students were 12-13 years old and had been placed at approximately A2
proficiency through the school's English program. Because the full cohort was included,
no random sampling or allocation was performed. Sex-disaggregated participant
information was not available in the study records used for this analysis.
Table 1
Study design and participant profile
Characteristic
Description
Design
One-group pretest-posttest
Participants
28 eighth-grade learners
Age range
12-13 years
Initial level
Approximately CEFR A2
Intervention
Eight task-based lessons across eight weeks
Instructional time
Two hours per lesson; 16 hours total
Outcome scale
Three analytic dimensions, each scored 0-5
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 54
Note. The intact cohort constituted the full accessible population.
Instructional intervention
The program consisted of eight two-hour lessons delivered over approximately eight
weeks. Each lesson was organized around a familiar communicative domain and a
practical outcome. Activities used flashcards, images, maps, short videos, worksheets,
story prompts, menus, role-play materials, and travel information. The sequence
progressed from predictable interpersonal exchanges toward narration, decision making,
and information exchange.
Across lessons, the instructional cycle included orientation to the outcome, activation
or modeling of useful language, pair or group task performance, and feedback. Grammar,
vocabulary, and pronunciation support were embedded where needed, but learners were
expected to use language to complete an exchange rather than only reproduce isolated
forms. Interactive communication was present throughout the program and was the
principal target in lessons on introductions, directions, and travel information.
Table 2
Eight-week task-based speaking sequence
Wee
k
Communicative
focus
Principal outcome
Targeted dimensions
1
Greetings and
introductions
Exchange personal details and
introduce another person
Interactive
communication
2
Daily routines
Compare routines and describe
current activities
Grammar/vocabulary;
interaction
3
Personal
appearance
Identify and describe people from
visual information
Grammar/vocabulary;
interaction
4
Past events
Construct and share a short past-
time narrative
Grammar/vocabulary;
interaction
5
Predictions and
plans
Discuss future intentions and
agree on likely outcomes
Grammar/vocabulary;
interaction
6
Food and meals
Complete a restaurant role-play
and express preferences
Grammar/vocabulary;
interaction
7
Directions
Guide a partner to a destination
using a map
Interactive
communication
8
Travel information
Request and provide information
about places and travel
Interactive
communication
Note. Each lesson lasted two hours. Pronunciation was addressed through modeling, rehearsal, performance,
and feedback across the sequence.
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 55
Speaking assessment
Speaking performance was assessed with tasks modeled on the A2 Key for Schools
Speaking test and evaluated with the published A2 analytic scales. The scale includes six
score points from 0 to 5, with descriptors anchored at selected bands. Grammar and
Vocabulary addresses control of simple grammatical forms and the appropriateness and
range of language used. Pronunciation addresses control of phonological features and
overall intelligibility. Interactive Communication addresses the ability to maintain
exchanges, respond, and require an age- and level-appropriate amount of prompting or
support (Cambridge Assessment English, 2020).
Baseline scores originated from a mock-examination process conducted under
standardized conditions and involving certified examination personnel. Posttest
performances were recorded and rated by experienced teachers using the same published
rubric. The available records did not report whether the same raters scored both occasions,
whether ratings were blind to measurement time, or whether a second independent rating
was obtained. These omissions are treated as limitations because changes in rater severity
could influence observed score differences.
Procedure and ethics
The baseline assessment was completed before the instructional sequence. Learners
then participated in the eight lessons during regular English instruction. The posttest
followed the intervention and used the same three analytic criteria. Scores were recorded
without names in the analytic dataset. The instructional program was reviewed before
implementation for alignment with language-learning theory, CEFR A2 expectations, and
the intended assessment criteria.
Institutional authorization and informed permission procedures were completed
before participation. Learners received information about the instructional activities,
confidentiality, and voluntary participation. They could decline specific activities without
academic penalty. Results were handled in aggregate, and no learner, teacher, or
institution is identified in this article. Classroom activities were designed to provide equal
access, culturally respectful content, and additional scaffolding when required.
Data analysis
Participant-level scores were reconstructed from the pretest-posttest dataset. Means,
standard deviations, medians, interquartile ranges, minimums, maximums, and
percentages were calculated for each dimension. The percentage of learners scoring in
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 56
Bands 4-5 was used as a descriptive indicator of movement toward the upper part of the
classroom rating scale. Change direction was summarized as improved, unchanged, or
decreased.
Because the outcomes were bounded ordinal ratings and the paired difference
distributions departed from normality, Wilcoxon signed-rank tests were used. Tests were
two-sided and excluded zero differences from the signed-rank calculation. The family of
three dimension-specific tests was controlled with the Holm procedure; adjusted
probabilities remained below .001. Standardized paired change was reported
descriptively as dz, calculated by dividing the mean difference by the standard deviation
of individual difference scores. A supplementary Wilcoxon test was also applied to the
15-point composite. Statistical significance was evaluated at alpha = .05, but
interpretation emphasized score patterns, improvement counts, and design limitations
rather than p values alone.
RESULTS AND DISCUSSION
Descriptive change
Performance increased in all three analytic dimensions. Grammar and Vocabulary
rose from M = 3.32 (SD = 0.94) to M = 4.25 (SD = 0.70), a mean gain of 0.93 points.
Pronunciation rose from M = 3.29 (SD = 0.98) to M = 4.29 (SD = 0.66), a one-point gain.
Interactive Communication also gained one point, increasing from M = 3.25 (SD = 0.80)
to M = 4.25 (SD = 0.70). The median moved from 3 to 4 for every dimension.
Dispersion decreased after instruction. Standard deviations fell by 0.24 for Grammar
and Vocabulary, 0.32 for Pronunciation, and 0.10 for Interactive Communication. This
pattern indicates that higher posttest means were accompanied by somewhat greater
concentration in the upper score bands. No dimension had a posttest score below 3,
whereas baseline scores extended to 2. The data therefore show both upward movement
and reduced representation at the lower end of the scale.
Table 3
Pretest and posttest descriptive statistics by speaking dimension
Dimension
M
SD
Media
n
IQR
Mean
gain
Grammar and
Vocabulary
3.32
0.94
3
3-4
Grammar and
Vocabulary
4.25
0.70
4
4-5
0.93
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 57
Dimension
M
SD
Media
n
IQR
Mean
gain
Pronunciation
3.29
0.98
3
2.75-4
Pronunciation
4.29
0.66
4
4-5
1.00
Interactive
Communication
3.25
0.80
3
3-4
Interactive
Communication
4.25
0.70
4
4-5
1.00
Composite (0-15)
9.86
1.65
9
9-11
Composite (0-15)
12.7
9
1.26
13
12-14
2.93
Note. n = 28. IQR = interquartile range. The composite is the sum of the three 0-5 analytic ratings.
Individual improvement patterns
The direction of change was uniformly favorable. In Grammar and Vocabulary, 20
learners (71.4%) improved and 8 (28.6%) remained unchanged. In both Pronunciation
and Interactive Communication, 21 learners (75.0%) improved and 7 (25.0%) remained
unchanged. No participant decreased in any analytic dimension. Because different
learners could improve in different dimensions, all 28 participants increased their
composite score by at least one point.
The proportion of learners scoring 4 or 5 also increased substantially. Grammar and
Vocabulary moved from 42.9% at baseline to 85.7% after instruction. Pronunciation
increased from 42.9% to 89.3%, and Interactive Communication increased from 46.4%
to 85.7%. These shifts are pedagogically meaningful because they show that the higher
means did not result from a few exceptional gains. Most learners moved into or remained
within the upper portion of the classroom scale.
Table 4
Direction of change and upper-band performance
Dimension
Improved
n (%)
Unchanged
n (%)
Decreased
n (%)
Bands 4-5
pretest
Bands 4-5
posttest
Grammar and
Vocabulary
20 (71.4%)
8 (28.6%)
0 (0.0%)
12 (42.9%)
24 (85.7%)
Pronunciation
21 (75.0%)
7 (25.0%)
0 (0.0%)
12 (42.9%)
25 (89.3%)
Interactive
Communication
21 (75.0%)
7 (25.0%)
0 (0.0%)
13 (46.4%)
24 (85.7%)
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 58
Dimension
Improved
n (%)
Unchanged
n (%)
Decreased
n (%)
Bands 4-5
pretest
Bands 4-5
posttest
Composite score
28
(100.0%)
0 (0.0%)
0 (0.0%)
N/A
N/A
Note. Bands 4-5 apply only to the analytic scales. Composite improvement indicates a higher posttest sum.
Inferential results
Wilcoxon signed-rank tests rejected the null hypothesis of zero median change for
every dimension. For Grammar and Vocabulary, W = 0 and the unadjusted two-sided p
value was .000042. Pronunciation and Interactive Communication each produced W = 0
and p = .000026. Holm adjustment for the three primary tests did not alter the conclusion;
every adjusted p value remained below .001. The zero W values reflect the absence of
negative differences among learners whose scores changed.
Standardized paired changes were large and nearly identical across the dimensions:
dz = 1.30 for Grammar and Vocabulary, 1.30 for Pronunciation, and 1.30 for Interactive
Communication after rounding. The composite increased by M = 2.93 points (SD of gain
= 1.51), with W = 0, p = .000003, and dz = 1.93. These statistics describe strong within-
group change. They should not be interpreted as controlled treatment effects because the
design cannot estimate how much change would have occurred without the program.
Table 5
Wilcoxon signed-rank tests and standardized paired change
Outcome
Mean
gain
Positi
ve
Zero
Neg.
W
Two-
sided p
dz
Grammar and
Vocabulary
0.93
20
8
0
0
< .001
1.30
Pronunciation
1.00
21
7
0
0
< .001
1.30
Interactive
Communication
1.00
21
7
0
0
< .001
1.30
Composite (0-15)
2.93
28
0
0
0
< .001
1.93
Note. Primary dimension tests remained p < .001 after Holm correction. Positive, zero, and negative refer to
paired posttest-pretest differences. dz = mean paired difference divided by its standard deviation.
Interpretation of the findings
The results show a consistent upward pattern across linguistic control, intelligibility,
and interaction. This breadth matters because a narrow intervention might improve
rehearsed vocabulary without changing pronunciation or responsiveness. Here, the mean
gains were similar across the three scales, and no dimension showed a decline for any
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 59
learner. The outcome is compatible with the program's integrated design: familiar
communicative tasks required learners to prepare language, listen to partners, formulate
messages, and respond within the same lesson cycle.
The sequence from introductions to travel information may have supported
development through graduated demand. Early lessons relied on predictable personal
content and short exchanges. Later lessons required narration, prediction, role
performance, route explanation, and information exchange. This progression is consistent
with Skehan's (1998) argument that task conditions shape learners' allocation of attention.
Familiarity and support can preserve participation at the beginning, while additional
reasoning or interactional requirements can be introduced once learners understand the
task cycle.
Pronunciation produced the highest posttest mean, although the difference from the
other dimensions was small. Repeated oral performance may have increased
opportunities to monitor intelligibility and receive feedback. Modeling, rehearsal, and
partner response can draw attention to whether a message is understandable without
turning pronunciation into isolated mimicry. Nevertheless, the analytic rating does not
reveal which phonological features changed. Claims about segmental accuracy, stress,
rhythm, or intonation would require recorded-speech analysis or criterion-level rater
comments that were not available.
The gain in Interactive Communication is also coherent with task-based principles.
Directions, restaurant exchanges, introductions, plans, and travel questions require
reciprocal turns rather than individual display. Learners must respond to content, not only
produce a memorized sentence. Foster and Ohta's (2005) account of peer assistance helps
explain how pair work can make elementary interaction possible: learners co-construct
utterances, repeat, prompt, and provide time for one another. Such support should be
interpreted as part of developing interactional competence rather than as evidence that a
learner failed to perform independently.
Grammar and Vocabulary improved by 0.93 points, slightly less than the one-point
gains in the other dimensions. This result challenges the assumption that meaning-
centered instruction necessarily neglects form. The lesson sequence repeatedly recycled
simple structures and high-frequency vocabulary within different outcomes. After
learners encountered a need during task performance, feedback could connect form to a
communicative consequence. Ellis et al. (2020) emphasize that TBLT can include
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 60
planned and incidental focus on form, particularly when attention is brief and tied to
meaning.
The movement into Bands 4-5 is noteworthy but should not be equated directly with
certification outcomes. The analytic rubric supports classroom judgment about
performance qualities, while an official qualification result depends on standardized
administration, trained examiners, test specifications, and score conversion procedures.
The present study used A2 Key-format tasks and published descriptors as an
instructionally aligned measure. It therefore provides evidence of classroom speaking
development on those criteria, not a substitute for an official examination result.
The composite score adds a useful whole-performance summary. Every learner
improved, and the mean rose by nearly three points on a 15-point range. Yet summing
the dimensions assumes that each contributes equally and that distances between rating
bands can be treated as approximately comparable. For this reason, the composite was
supplementary and did not replace the analytic results. Reporting both views is more
informative: the composite shows broad movement, while the dimensions identify the
areas in which that movement occurred.
The very small p values require contextual interpretation. Statistical significance
reflects the consistency of paired improvement, not proof that the intervention alone
caused it. Without a comparison group, alternative explanations include maturation,
increased familiarity with the speaking format, concurrent English instruction, seasonal
changes in motivation, and differences between pretest and posttest raters. The fact that
no learner declined also raises a measurement question: true broad improvement is
plausible, but changes in scoring severity cannot be excluded when interrater reliability
and rater blinding are undocumented.
A stronger future evaluation would retain the same task-based sequence while adding
a comparison condition. Intact classes could be matched on baseline proficiency,
instructional time, and teacher experience. If random assignment were not feasible, an
analysis of covariance could compare posttest scores while controlling for baseline
differences. Audio recordings should be independently scored by at least two trained
raters who are blind to testing occasion, and interrater agreement should be reported for
each analytic dimension. Parallel task forms would reduce simple test familiarity.
Retention also remains unknown. The posttest indicates performance after the eight-
week sequence, but no delayed measure determined whether gains persisted. A follow-
up after four to eight weeks could examine whether learners maintained the language
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 61
without continued rehearsal. Transfer tasks would be equally valuable. Students could
complete an unfamiliar service encounter or collaborative decision task that was not
practiced during the program. Successful performance under new conditions would
provide stronger evidence that learning extended beyond the taught scenarios.
The intervention nevertheless has practical value. Eight lessons and 16 hours
constitute a feasible unit that can be embedded within a secondary English course.
Materials were accessible and did not depend on specialized technology. Teachers can
adapt the content to local routines and student interests while preserving the core
architecture: a clear outcome, preparation, meaningful exchange, reporting, and feedback.
This adaptability is important because TBLT works as a design framework rather than a
fixed collection of activities (Bygate, 2020).
For implementation, participation structures deserve deliberate attention. Pair
rotation, assigned roles, individual planning notes, time limits, and visible success criteria
can prevent confident learners from controlling the task. Brief self-assessment can ask
whether the learner conveyed the message, responded to a partner, used target
expressions, and repaired misunderstanding. Feedback should identify one or two high-
value language points instead of interrupting every error. At A2, communicative
confidence and intelligibility can grow when learners experience successful exchanges
while still receiving precise guidance.
Overall, the findings support TBLT as a promising approach for elementary
adolescent speaking development. The strongest evidence is not any single mean but the
convergence of three patterns: most learners improved in every analytic dimension,
upper-band participation nearly doubled, and every composite score increased. The
design limits causal inference, but the consistency of change justifies a controlled
replication and supports continued use of carefully scaffolded speaking tasks in
comparable classrooms.
CONCLUSIONS
An eight-week task-based speaking program was associated with substantial
improvement in the A2 classroom speaking performance of 28 eighth-grade learners.
Mean scores increased from 3.32 to 4.25 in Grammar and Vocabulary, from 3.29 to 4.29
in Pronunciation, and from 3.25 to 4.25 in Interactive Communication. The median rose
from Band 3 to Band 4 in every dimension, and the proportion scoring in Bands 4-5
increased to at least 85.7% across all three scales.
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 62
Paired analyses confirmed that the pattern was highly consistent. Between 71.4% and
75.0% of learners improved in each analytic dimension, no learner declined, and all 28
improved on the 15-point composite. Wilcoxon tests remained significant after correction
for multiple comparisons, and standardized paired changes were large. The results
therefore reject the null hypotheses of zero median change and identify the program as a
credible instructional approach for further evaluation.
The pedagogical contribution lies in the integration of meaningful outcomes with
structured support. Familiar topics enabled learners to practice introductions, routines,
descriptions, narratives, plans, service encounters, directions, and travel exchanges.
Preparation, modeling, rehearsal, interaction, and feedback linked linguistic form with
communicative purpose. The similar gains across grammar/vocabulary, pronunciation,
and interaction suggest that task-based work can support a balanced conception of
speaking rather than prioritize fluency at the expense of accuracy or intelligibility.
The findings must remain bounded by the design. A single intact cohort, absence of
a control group, short duration, unreported rater agreement, and lack of a delayed posttest
prevent definitive causal or retention claims. Future studies should use matched or
randomized conditions, parallel speaking tasks, blinded dual ratings, interrater reliability
statistics, delayed measurement, and transfer tasks. Within these limits, the program
provides a feasible model for expanding purposeful oral language use among A2
secondary learners.
REFERENCES
Ahmadian, M. J., & Long, M. H. (Eds.). (2021). The Cambridge handbook of task-
based language teaching. Cambridge University Press.
https://doi.org/10.1017/9781108868327
Bachman, L. F., & Palmer, A. S. (2010). Language assessment in practice. Oxford
University Press.
Bygate, M. (2016). Sources, developments and directions of task-based language
teaching. The Language Learning Journal, 44(4), 381-400.
https://doi.org/10.1080/09571736.2015.1039566
Bygate, M. (2020). Some directions for the possible survival of TBLT as a real-world
project. Language Teaching, 53(3), 275-288.
https://doi.org/10.1017/S0261444820000014
LINGUATECH –REVISTA INTERNACIONAL, VOL. II, NUM. II pág. 63
Cambridge Assessment English. (2020). A2 Key for Schools handbook for teachers.
https://www.cambridgeenglish.org/images/504506-a2-key-for-schools-handbook-
2020.pdf
Council of Europe. (2020). Common European Framework of Reference for Languages:
Learning, teaching, assessment - Companion volume. Council of Europe
Publishing. https://rm.coe.int/common-european-framework-of-reference-for-
languages-learning-teaching/16809ea0d4
East, M. (2017). Research into practice: The task-based approach to instructed second
language acquisition. Language Teaching, 50(3), 412-424.
https://doi.org/10.1017/S0261444816000409
Ellis, R. (2003). Task-based language learning and teaching. Oxford University Press.
Ellis, R., Skehan, P., Shintani, N., Lambert, C., & Révész, A. (2020). Task-based
language teaching: Theory and practice. Cambridge University Press.
https://doi.org/10.1017/9781108643689
Foster, P., & Ohta, A. S. (2005). Negotiation for meaning and peer assistance in second
language classrooms. Applied Linguistics, 26(3), 402-430.
https://doi.org/10.1093/applin/ami014
García Mayo, M. P., & Imaz Agirre, A. (2016). Task repetition and its impact on EFL
children's negotiation of meaning strategies and pair dynamics: An exploratory
study. The Language Learning Journal, 44(4), 451-466.
https://doi.org/10.1080/09571736.2016.1185799
Long, M. H. (2015). Second language acquisition and task-based language teaching.
Wiley Blackwell.
Nunan, D. (2004). Task-based language teaching. Cambridge University Press.
Shintani, N. (2015). The incidental grammar acquisition in focus on form and focus on
forms instruction for young beginner learners. TESOL Quarterly, 49(1), 115-140.
https://doi.org/10.1002/tesq.166
Skehan, P. (1996). A framework for the implementation of task-based instruction.
Applied Linguistics, 17(1), 38-62. https://doi.org/10.1093/applin/17.1.38
Skehan, P. (1998). A cognitive approach to language learning. Oxford University Press.
Van den Branden, K. (2016). The role of teachers in task-based language education.
Annual Review of Applied Linguistics, 36, 164-181.
https://doi.org/10.1017/S0267190515000070
Willis, J., & Willis, D. (2007). Doing task-based teaching. Oxford University Press.