Vol. 13/ Núm. 2 2026 pág. 1210
https://doi.org/10.69639/arandu.v13i2.2255
Effects of AI-based educational applications on EFL students’
speaking and writing performance: A pre-experimental study
Efectos de las aplicaciones educativas basadas en IA en el desempeño oral y escrito de
estudiantes de inglés como lengua extranjera: Un estudio preexperimental
Cristian Santiago Mesias Masabanda
cristian.mesias@upec.edu.ec
https://orcid.org/0009-0008-3414-4412
Universidad Politécnica Estatal del Carchi
Ecuador – Tulcán
Elena Valeria Flores Borja
https://orcid.org/0009-0005-4800-8938
elena.flores@upec.edu.ec
Universidad Politécnica Estatal del Carchi
Ecuador – Tulcán
Artículo recibido: 10 abril 2026- Aceptado para publicación:16 mayo 2026
Conflictos de intereses: Ninguno que declarar.
ABSTRACT
In the Ecuadorian educational context, particularly at the secondary school level, traditional
methodologies centered on memorization and grammatical instruction of the English language
still predominate, which has constrained the development of communicative competencies,
especially oral and written expression. In response to this issue, the study was conducted in a
school setting where low levels of performance were observed in pronunciation, fluency, spelling,
and punctuation, as well as a limited use of artificial intelligence–based educational applications.
The aim of the study was to determine the impact of implementing practical exercises supported
by ELSA Speak and Writing & Improve on the enhancement of integrated oral and written
production skills in secondary school students. The research adopted a quantitative approach with
a pre-experimental pretest–posttest design. A validated structured questionnaire was administered
to a sample of 60 students. The data were recoded into comparative categories and converted into
proportions for inferential statistical analysis. Subsequently, a Shapiro–Wilk normality test and a
paired-samples t-test were conducted using SPSS software. The results demonstrated a significant
improvement across all assessed skills following the didactic intervention. Statistical analysis
confirmed significant differences between the pretest and posttest (p < 0.05). The findings
indicated that the pedagogically contextualized integration of tools such as ELSA Speak and
Writing & Improve promotes more active, autonomous, and meaningful learning. Nevertheless,
the study could be strengthened by including a control group and extending the intervention
period, thereby opening the possibility for future research.
Keywords: artificial intelligence, english language learning, speaking skills, technological
integration, writing skills
Vol. 13/ Núm. 2 2026 pág. 1211
RESUMEN
En el contexto educativo ecuatoriano, particularmente en el nivel de educación secundaria, aún
predominan metodologías tradicionales centradas en la memorización y en la instrucción
gramatical del idioma inglés, lo que ha limitado el desarrollo de las competencias comunicativas,
especialmente la expresión oral y escrita. En respuesta a esta problemática, la investigación se
llevó a cabo en un entorno escolar donde se evidenciaban bajos niveles de desempeño en
pronunciación, fluidez, ortografía y uso de signos de puntuación, así como un uso limitado de
aplicaciones educativas basadas en inteligencia artificial. El objetivo del estudio fue determinar
el impacto de la implementación de ejercicios prácticos apoyados en ELSA Speak y Writing &
Improve en el fortalecimiento de las habilidades integradas de producción oral y escrita en
estudiantes de secundaria. La investigación adoptó un enfoque cuantitativo con un diseño
preexperimental de pretest y postest. Se aplicó un cuestionario estructurado validado a una
muestra de 60 estudiantes. Los datos fueron recodificados en categorías comparativas y
transformados en proporciones para su análisis estadístico inferencial. Posteriormente, se
realizaron la prueba de normalidad de Shapiro-Wilk y la prueba t para muestras relacionadas
mediante el software SPSS. Los resultados evidenciaron una mejora significativa en todas las
habilidades evaluadas tras la intervención didáctica. El análisis estadístico confirmó diferencias
significativas entre el pretest y el postest (p < 0,05). Los resultados indicaron que la integración
pedagógicamente contextualizada de herramientas como ELSA Speak y Writing & Improve
promueve un aprendizaje más activo, autónomo y significativo. No obstante, el estudio podría
fortalecerse mediante la inclusión de un grupo de control y la ampliación del periodo de
intervención, abriendo la posibilidad a futuras investigaciones.
Palabras clave: inteligencia artificial, aprendizaje del idioma inglés, habilidades de
expresión oral, integración tecnológica, habilidades de escritura
Todo el contenido de la Revista Científica Internacional Arandu UTIC publicado en este sitio está disponible bajo
licencia Creative Commons Atribution 4.0 International.
RESUMEN
En el contexto educativo ecuatoriano, particularmente en el nivel de educación secundaria, aún
predominan metodologías tradicionales centradas en la memorización y en la instrucción
gramatical del idioma inglés, lo que ha limitado el desarrollo de las competencias comunicativas,
especialmente la expresión oral y escrita. En respuesta a esta problemática, la investigación se
llevó a cabo en un entorno escolar donde se evidenciaban bajos niveles de desempeño en
pronunciación, fluidez, ortografía y uso de signos de puntuación, así como un uso limitado de
aplicaciones educativas basadas en inteligencia artificial. El objetivo del estudio fue determinar
el impacto de la implementación de ejercicios prácticos apoyados en ELSA Speak y Writing &
Improve en el fortalecimiento de las habilidades integradas de producción oral y escrita en
estudiantes de secundaria. La investigación adoptó un enfoque cuantitativo con un diseño
preexperimental de pretest y postest. Se aplicó un cuestionario estructurado validado a una
muestra de 60 estudiantes. Los datos fueron recodificados en categorías comparativas y
transformados en proporciones para su análisis estadístico inferencial. Posteriormente, se
realizaron la prueba de normalidad de Shapiro-Wilk y la prueba t para muestras relacionadas
mediante el software SPSS. Los resultados evidenciaron una mejora significativa en todas las
habilidades evaluadas tras la intervención didáctica. El análisis estadístico confirmó diferencias
significativas entre el pretest y el postest (p < 0,05). Los resultados indicaron que la integración
pedagógicamente contextualizada de herramientas como ELSA Speak y Writing & Improve
promueve un aprendizaje más activo, autónomo y significativo. No obstante, el estudio podría
fortalecerse mediante la inclusión de un grupo de control y la ampliación del periodo de
intervención, abriendo la posibilidad a futuras investigaciones.
Palabras clave: inteligencia artificial, aprendizaje del idioma inglés, habilidades de
expresión oral, integración tecnológica, habilidades de escritura
Todo el contenido de la Revista Científica Internacional Arandu UTIC publicado en este sitio está disponible bajo
licencia Creative Commons Atribution 4.0 International.
Vol. 13/ Núm. 2 2026 pág. 1212
INTRODUCTION
In 21st-century education, the use of AI-based applications has become a key approach
for enhancing English language teaching. According to Li (2022), these tools provide immediate
feedback, personalized tasks and dynamic practice, enabling students to take an active role in the
learning process. According to Huang et al. (2022), AI-based applications are transforming
language teaching by facilitating adaptive learning processes in speaking and writing,
demonstrating that these tools can complement the work of teachers when integrated effectively
into pedagogical contexts. Schmidt and Strassner (2022), on the other hand, demonstrated that
AI-driven language learning tools promote learner autonomy and self-regulated learning.
Similarly, Guzmán and Esquivel (2025), in their study on AI in EFL teaching, found that
these tools positively impact on the integrated development of speaking and writing skills,
particularly in contexts that promote consistent and contextualized practice. According to Ayala
and Alvarado (2023), these tools offer a flexible approach that adapts to each student's pace and
learning style, enhancing communicative skills. Kolegova and Levina (2024), on the other hand,
demonstrated that AI-based teaching resources enable personalized learning, generate interactive
content and the promotion of communicative skills practice in simulated contexts. Fitria (2021),
for her part, analyzed the role of AI in language learning within the English Language Teaching
framework and concluded that artificial intelligence-based technologies generate more dynamic
learning environments, support the simultaneous development of speaking and writing skills and
improve student motivation.
Developing speaking and writing skills requires guidance, continuous feedback, and
interactive learning environments that lead to meaningful learning; that is, learning that is
applicable in everyday contexts (Mayorga and Tibán, 2024). However, as Syuhra et al. (2025)
note, many educational institutions still rely on grammar-based and memorization-focused
approaches, which hinder effective English language learning. Similarly, Armendáriz et al. (2024)
demonstrated that learning centred solely on grammatical structures limits fluency and
meaningful communication. Therefore, integrating Artificial Intelligence tools helps to overcome
these barriers by providing contextualized practice that strengthens communicative skills in a
more natural and functional manner.
The incorporation of Artificial Intelligence applications such as ELSA Speak, Duolingo
and ChatGPT has transformed English language teaching through immediate feedback, content
personalization and autonomous practice. The research conducted by Anggraini (2022) showed
that ELSA Speak notably improved students' pronunciation. Xiaofan and Annamalai (2025)
indicate that AI-based adaptive platforms promote motivation and performance in writing and
speaking. Furthermore, Al-khresheh (2024) demonstrated that ChatGPT contributes to increasing
learner autonomy and communicative fluency. Along these lines, Sari (2023) highlights that the
INTRODUCTION
In 21st-century education, the use of AI-based applications has become a key approach
for enhancing English language teaching. According to Li (2022), these tools provide immediate
feedback, personalized tasks and dynamic practice, enabling students to take an active role in the
learning process. According to Huang et al. (2022), AI-based applications are transforming
language teaching by facilitating adaptive learning processes in speaking and writing,
demonstrating that these tools can complement the work of teachers when integrated effectively
into pedagogical contexts. Schmidt and Strassner (2022), on the other hand, demonstrated that
AI-driven language learning tools promote learner autonomy and self-regulated learning.
Similarly, Guzmán and Esquivel (2025), in their study on AI in EFL teaching, found that
these tools positively impact on the integrated development of speaking and writing skills,
particularly in contexts that promote consistent and contextualized practice. According to Ayala
and Alvarado (2023), these tools offer a flexible approach that adapts to each student's pace and
learning style, enhancing communicative skills. Kolegova and Levina (2024), on the other hand,
demonstrated that AI-based teaching resources enable personalized learning, generate interactive
content and the promotion of communicative skills practice in simulated contexts. Fitria (2021),
for her part, analyzed the role of AI in language learning within the English Language Teaching
framework and concluded that artificial intelligence-based technologies generate more dynamic
learning environments, support the simultaneous development of speaking and writing skills and
improve student motivation.
Developing speaking and writing skills requires guidance, continuous feedback, and
interactive learning environments that lead to meaningful learning; that is, learning that is
applicable in everyday contexts (Mayorga and Tibán, 2024). However, as Syuhra et al. (2025)
note, many educational institutions still rely on grammar-based and memorization-focused
approaches, which hinder effective English language learning. Similarly, Armendáriz et al. (2024)
demonstrated that learning centred solely on grammatical structures limits fluency and
meaningful communication. Therefore, integrating Artificial Intelligence tools helps to overcome
these barriers by providing contextualized practice that strengthens communicative skills in a
more natural and functional manner.
The incorporation of Artificial Intelligence applications such as ELSA Speak, Duolingo
and ChatGPT has transformed English language teaching through immediate feedback, content
personalization and autonomous practice. The research conducted by Anggraini (2022) showed
that ELSA Speak notably improved students' pronunciation. Xiaofan and Annamalai (2025)
indicate that AI-based adaptive platforms promote motivation and performance in writing and
speaking. Furthermore, Al-khresheh (2024) demonstrated that ChatGPT contributes to increasing
learner autonomy and communicative fluency. Along these lines, Sari (2023) highlights that the
Vol. 13/ Núm. 2 2026 pág. 1213
inclusion of AI reduced dependence on rote learning and enhanced meaningful interaction in the
English language.
Alqaed (2024), on the other hand, demonstrated that the use of applications such as
ChatGPT in English as a Foreign Language teaching contexts has produced significant
improvements in error correction and learner autonomy. Marghany (2023) concluded that the
integration of AI applications for individualized feedback increased oral fluency and accuracy in
non-native speaking students. Sehlaoui (2024), in contrast, established that the use of adaptive AI
platforms in English language learning promoted greater motivation, engagement and progress in
writing. López et al. (2025) confirmed that AI-driven learning environments with voice assistants
enhanced communicative competence through simulated dialogue and immediate correction.
The simultaneous development of skills such as speaking, writing, reading and listening
facilitates more comprehensive English language learning, where each skill reinforces the others.
In this regard, Hipo et al. (2022) found that the Integrated Skills Approach increased students'
communicative competence. Mahapatra (2024) demonstrated that project-based activities
integrating listening and speaking increased student motivation and participation. Murillo et al.
(2021), in their analysis of technology-based learning environments, indicated that the integration
of different skills through technology reinforced authentic interaction in English. Similarly, Sapan
and Uzun (2024) demonstrated that the incorporation of ChatGPT in English language teaching
significantly improved the writing and vocabulary of English as a Foreign Language learners.
The inclusion of artificial intelligence-based educational applications facilitates a more
personalised and adaptive learning experience, which promotes the simultaneous development of
speaking and writing skills in English. According to Sanabria et al. (2023), Artificial Intelligence
tools allow for immediate feedback and autonomous practice, which notably improves
performance in productive skills. Chicaiza et al. (2025) have found that the consistent use of AI
platforms is associated with progress in speaking and writing skills among English language
students. This demonstrates that AI technologies increase student motivation and lead to better
outcomes in English language learning.
From this analyzed context, platforms such as ELSA Speak and Writing & Improve offer
learning opportunities to overcome the limitations presented by traditional English language
teaching methods that are still being implemented in classrooms within the Ecuadorian
educational system. Educational inclusion implies not only access, but also the implementation
of strategies that eliminate barriers and guarantee quality education, as supported by the
Constitution and the Ley Orgánica de Educación Intercultural (Román et al., 2025).
Background to the Problem
The problem addressed by this research relates to English language teaching at the Unidad
Educativa de las Fuerzas Armadas, Colegio Militar Nro. 3 Héroes del 41, located in the city of
Machala, in the province of El Oro. Learning takes place in an educational environment where
inclusion of AI reduced dependence on rote learning and enhanced meaningful interaction in the
English language.
Alqaed (2024), on the other hand, demonstrated that the use of applications such as
ChatGPT in English as a Foreign Language teaching contexts has produced significant
improvements in error correction and learner autonomy. Marghany (2023) concluded that the
integration of AI applications for individualized feedback increased oral fluency and accuracy in
non-native speaking students. Sehlaoui (2024), in contrast, established that the use of adaptive AI
platforms in English language learning promoted greater motivation, engagement and progress in
writing. López et al. (2025) confirmed that AI-driven learning environments with voice assistants
enhanced communicative competence through simulated dialogue and immediate correction.
The simultaneous development of skills such as speaking, writing, reading and listening
facilitates more comprehensive English language learning, where each skill reinforces the others.
In this regard, Hipo et al. (2022) found that the Integrated Skills Approach increased students'
communicative competence. Mahapatra (2024) demonstrated that project-based activities
integrating listening and speaking increased student motivation and participation. Murillo et al.
(2021), in their analysis of technology-based learning environments, indicated that the integration
of different skills through technology reinforced authentic interaction in English. Similarly, Sapan
and Uzun (2024) demonstrated that the incorporation of ChatGPT in English language teaching
significantly improved the writing and vocabulary of English as a Foreign Language learners.
The inclusion of artificial intelligence-based educational applications facilitates a more
personalised and adaptive learning experience, which promotes the simultaneous development of
speaking and writing skills in English. According to Sanabria et al. (2023), Artificial Intelligence
tools allow for immediate feedback and autonomous practice, which notably improves
performance in productive skills. Chicaiza et al. (2025) have found that the consistent use of AI
platforms is associated with progress in speaking and writing skills among English language
students. This demonstrates that AI technologies increase student motivation and lead to better
outcomes in English language learning.
From this analyzed context, platforms such as ELSA Speak and Writing & Improve offer
learning opportunities to overcome the limitations presented by traditional English language
teaching methods that are still being implemented in classrooms within the Ecuadorian
educational system. Educational inclusion implies not only access, but also the implementation
of strategies that eliminate barriers and guarantee quality education, as supported by the
Constitution and the Ley Orgánica de Educación Intercultural (Román et al., 2025).
Background to the Problem
The problem addressed by this research relates to English language teaching at the Unidad
Educativa de las Fuerzas Armadas, Colegio Militar Nro. 3 Héroes del 41, located in the city of
Machala, in the province of El Oro. Learning takes place in an educational environment where
Vol. 13/ Núm. 2 2026 pág. 1214
traditional methods based on grammar and the repetition of content through memorization
predominate. Although traditional methodology is useful for learning structure, it prevents
students from having opportunities to practice the language in real contexts, particularly affecting
their ability to speak and write with fluency. If this approach to teaching continues unchanged,
students will persist in facing communication difficulties, will display low confidence when
expressing themselves and will rely excessively on grammatical rules. This affects their academic
performance, reduces their motivation and limits their future opportunities in academic, social
and professional settings, where English is an essential tool for personal growth and career
development.
In light of this situation, the following research question arises: In what ways do artificial
intelligence educational applications impact the development of integrated English language
skills among second-year secondary school students at the Unidad Educativa de las Fuerzas
Armadas, Colegio Militar Nro. 3 Héroes del 41? The following guiding questions have also been
formulated for the study: a) What are students' perceptions regarding the usefulness of the ELSA
Speaking and Writing & Improve applications in strengthening oral and written expression in
English? b) What level of proficiency do students demonstrate in the basic skills of pronunciation,
fluency, spelling and punctuation? c) What is the design of a didactic process aimed at articulately
integrating oral and written production skills through the use of ELSA Speaking and Writing &
Improve in the English classroom? d) To what extent does the implementation of practical
exercises supported by ELSA Speaking and Writing & Improve contribute to the strengthening
of integrated speaking and writing skills among students?
Research Objectives
In this regard, the research sets out the following general objective: To analyze the impact
of artificial intelligence educational applications on the development of integrated English
language skills among second-year secondary school students at the Unidad Educativa de las
Fuerzas Armadas, Colegio Militar Nro. 3 Héroes del 41. In order to achieve this, the following
specific objectives were proposed: a) To describe students' perceptions of the usefulness of the
ELSA Speaking and Writing & Improve applications in strengthening oral and written expression
in English. b) To assess the level of proficiency demonstrated by students in the basic skills of
pronunciation, fluency, spelling and punctuation. c) To design a didactic process through which
oral and written production skills are articulately integrated via the use of ELSA Speaking and
Writing & Improve in the English classroom. d) To determine the contribution of the
implementation of practical exercises supported by ELSA Speaking and Writing & Improve to
the strengthening of integrated speaking and writing skills among students.
Research Hypothesis
In accordance with the research objectives, the following null hypothesis was formulated:
H0: Artificial intelligence educational applications do not have an impact on the development of
traditional methods based on grammar and the repetition of content through memorization
predominate. Although traditional methodology is useful for learning structure, it prevents
students from having opportunities to practice the language in real contexts, particularly affecting
their ability to speak and write with fluency. If this approach to teaching continues unchanged,
students will persist in facing communication difficulties, will display low confidence when
expressing themselves and will rely excessively on grammatical rules. This affects their academic
performance, reduces their motivation and limits their future opportunities in academic, social
and professional settings, where English is an essential tool for personal growth and career
development.
In light of this situation, the following research question arises: In what ways do artificial
intelligence educational applications impact the development of integrated English language
skills among second-year secondary school students at the Unidad Educativa de las Fuerzas
Armadas, Colegio Militar Nro. 3 Héroes del 41? The following guiding questions have also been
formulated for the study: a) What are students' perceptions regarding the usefulness of the ELSA
Speaking and Writing & Improve applications in strengthening oral and written expression in
English? b) What level of proficiency do students demonstrate in the basic skills of pronunciation,
fluency, spelling and punctuation? c) What is the design of a didactic process aimed at articulately
integrating oral and written production skills through the use of ELSA Speaking and Writing &
Improve in the English classroom? d) To what extent does the implementation of practical
exercises supported by ELSA Speaking and Writing & Improve contribute to the strengthening
of integrated speaking and writing skills among students?
Research Objectives
In this regard, the research sets out the following general objective: To analyze the impact
of artificial intelligence educational applications on the development of integrated English
language skills among second-year secondary school students at the Unidad Educativa de las
Fuerzas Armadas, Colegio Militar Nro. 3 Héroes del 41. In order to achieve this, the following
specific objectives were proposed: a) To describe students' perceptions of the usefulness of the
ELSA Speaking and Writing & Improve applications in strengthening oral and written expression
in English. b) To assess the level of proficiency demonstrated by students in the basic skills of
pronunciation, fluency, spelling and punctuation. c) To design a didactic process through which
oral and written production skills are articulately integrated via the use of ELSA Speaking and
Writing & Improve in the English classroom. d) To determine the contribution of the
implementation of practical exercises supported by ELSA Speaking and Writing & Improve to
the strengthening of integrated speaking and writing skills among students.
Research Hypothesis
In accordance with the research objectives, the following null hypothesis was formulated:
H0: Artificial intelligence educational applications do not have an impact on the development of
Vol. 13/ Núm. 2 2026 pág. 1215
integrated English language skills. Likewise, the following alternative hypothesis was
established: H1: Artificial intelligence educational applications do have an impact on the
development of integrated English language skills.
Justification of the Research
This research is justified in that it allows for an understanding, from a theoretical
standpoint, of how Artificial Intelligence applications can transform teaching by moving beyond
the traditional methods that have focused for years on grammar and memorization. On a practical
level, it offers a more dynamic teaching alternative that is better aligned with current
communicative needs. From a methodological perspective, it provides evidence on the use of a
pre-experimental design applied in real school contexts. Furthermore, it is socially relevant as it
supports students who need to strengthen their communicative skills in order to face increasingly
demanding academic and professional environments. It is also economically viable, given that the
majority of AI tools used are freely accessible and can be used on common devices.
Significance of the Research
The relevance of this study lies in the fact that it responds to the need to improve students'
ability to communicate effectively in English, overcoming the limitations generated by models
focused solely on rules and memorization. Its significance is also reflected in the methodological
innovation it introduces, by incorporating AI tools that offer immediate feedback, autonomous
practice and opportunities to develop speaking and writing skills in more realistic situations. The
direct beneficiaries are the students, who will be able to access more motivating and functional
learning experiences, and the teachers, who will have a complementary resource to enrich their
practice. The potential impact is considerable, as an improvement in communicative skills opens
doors to higher education and the professional world. Furthermore, this research is aligned with
UNESCO's Sustainable Development Goal 4, which seeks to guarantee inclusive and quality
education by promoting innovative methods that ensure relevant and equitable learning outcomes.
MATERIALS AND METHODS
Research Paradigm and Approach
In accordance with the research problem and the objectives set out, the study is framed
within the positivist paradigm and a quantitative approach. Castrillo (2024) establishes that this
paradigm considers reality to be observable, quantifiable and explainable through objective data,
allowing for the analysis of changes occurring within a given phenomenon. In turn, the
quantitative approach, according to Ramírez (2021), allows for the examination of relationships
between variables and the testing of hypotheses through systematic procedures, which is
consistent with the purpose of evaluating the effect of an educational strategy on a specific group
of students. In this way, the paradigm and approach adopted ensure the rigour and clarity
necessary to assess the progress achieved through an educational intervention.
integrated English language skills. Likewise, the following alternative hypothesis was
established: H1: Artificial intelligence educational applications do have an impact on the
development of integrated English language skills.
Justification of the Research
This research is justified in that it allows for an understanding, from a theoretical
standpoint, of how Artificial Intelligence applications can transform teaching by moving beyond
the traditional methods that have focused for years on grammar and memorization. On a practical
level, it offers a more dynamic teaching alternative that is better aligned with current
communicative needs. From a methodological perspective, it provides evidence on the use of a
pre-experimental design applied in real school contexts. Furthermore, it is socially relevant as it
supports students who need to strengthen their communicative skills in order to face increasingly
demanding academic and professional environments. It is also economically viable, given that the
majority of AI tools used are freely accessible and can be used on common devices.
Significance of the Research
The relevance of this study lies in the fact that it responds to the need to improve students'
ability to communicate effectively in English, overcoming the limitations generated by models
focused solely on rules and memorization. Its significance is also reflected in the methodological
innovation it introduces, by incorporating AI tools that offer immediate feedback, autonomous
practice and opportunities to develop speaking and writing skills in more realistic situations. The
direct beneficiaries are the students, who will be able to access more motivating and functional
learning experiences, and the teachers, who will have a complementary resource to enrich their
practice. The potential impact is considerable, as an improvement in communicative skills opens
doors to higher education and the professional world. Furthermore, this research is aligned with
UNESCO's Sustainable Development Goal 4, which seeks to guarantee inclusive and quality
education by promoting innovative methods that ensure relevant and equitable learning outcomes.
MATERIALS AND METHODS
Research Paradigm and Approach
In accordance with the research problem and the objectives set out, the study is framed
within the positivist paradigm and a quantitative approach. Castrillo (2024) establishes that this
paradigm considers reality to be observable, quantifiable and explainable through objective data,
allowing for the analysis of changes occurring within a given phenomenon. In turn, the
quantitative approach, according to Ramírez (2021), allows for the examination of relationships
between variables and the testing of hypotheses through systematic procedures, which is
consistent with the purpose of evaluating the effect of an educational strategy on a specific group
of students. In this way, the paradigm and approach adopted ensure the rigour and clarity
necessary to assess the progress achieved through an educational intervention.
Vol. 13/ Núm. 2 2026 pág. 1216
Research Design
This research adopts a design that allows for the observation of the effect of an
educational intervention based on artificial intelligence applications on integrated English
language skills. In this case, a pre-experimental design is used, understood as one that works with
a single study group, to which a pre-test is administered before the intervention and a post-test
after it, with the aim of identifying possible changes attributable to an educational treatment (Arias
and Covinos, 2021). This type of design is appropriate when the aim is to assess the influence of
a strategy without the availability of comparison groups, whilst still obtaining clear evidence of
the progress achieved. In this way, this design offers a relevant approach for analyzing advances
in speaking and writing following the implementation of artificial intelligence tools, as proposed
in the objectives of the study.
Scope of the Research
Furthermore, the descriptive-explanatory scope, described by Galarza (2020) as one that
first allows for the description of an initial situation and subsequently analyses the changes
generated following an intervention, guides this study. The descriptive level helps to characterize
the state of artificial intelligence application use and speaking and writing skills both before and
after the treatment. At the same time, the explanatory level allows for an assessment of whether
the educational strategy applied contributed to the improvement of these skills by comparing the
pre-test and post-test results. In this way, this scope combines a detailed view of reality with an
analysis that seeks to understand the concrete effects of the technological intervention.
Population and Sample
In order to adequately organize the research process, the population and sample of the
study were defined in terms of who forms part of the study and how the group was selected.
According to Hernández and Mendoza (2020), the population is the total set of people, objects or
elements that share characteristics relevant to the research, whilst the sample is a representative
part of that population chosen to obtain the data. In this case, the population corresponds to all
students at the Unidad Educativa de las Fuerzas Armadas, Colegio Militar N.º 3 "Héroes del 41",
whilst the sample comprises 60 second-year secondary school students.
In order to select the participants, it was necessary to opt for a sampling method that was
suited to the characteristics of the military educational context and to the actual availability of the
students and the provisions of the military educational authorities. In this case, non-probability
convenience sampling was used, understood as a method in which participants are chosen based
on their accessibility and presence in the environment where the research takes place, without
resorting to random procedures. According to González (2021), this type of sampling is
appropriate when the researcher needs to work with a group that is easily identifiable and available
for the study. Table 1 presents the population and sample of the research.
Research Design
This research adopts a design that allows for the observation of the effect of an
educational intervention based on artificial intelligence applications on integrated English
language skills. In this case, a pre-experimental design is used, understood as one that works with
a single study group, to which a pre-test is administered before the intervention and a post-test
after it, with the aim of identifying possible changes attributable to an educational treatment (Arias
and Covinos, 2021). This type of design is appropriate when the aim is to assess the influence of
a strategy without the availability of comparison groups, whilst still obtaining clear evidence of
the progress achieved. In this way, this design offers a relevant approach for analyzing advances
in speaking and writing following the implementation of artificial intelligence tools, as proposed
in the objectives of the study.
Scope of the Research
Furthermore, the descriptive-explanatory scope, described by Galarza (2020) as one that
first allows for the description of an initial situation and subsequently analyses the changes
generated following an intervention, guides this study. The descriptive level helps to characterize
the state of artificial intelligence application use and speaking and writing skills both before and
after the treatment. At the same time, the explanatory level allows for an assessment of whether
the educational strategy applied contributed to the improvement of these skills by comparing the
pre-test and post-test results. In this way, this scope combines a detailed view of reality with an
analysis that seeks to understand the concrete effects of the technological intervention.
Population and Sample
In order to adequately organize the research process, the population and sample of the
study were defined in terms of who forms part of the study and how the group was selected.
According to Hernández and Mendoza (2020), the population is the total set of people, objects or
elements that share characteristics relevant to the research, whilst the sample is a representative
part of that population chosen to obtain the data. In this case, the population corresponds to all
students at the Unidad Educativa de las Fuerzas Armadas, Colegio Militar N.º 3 "Héroes del 41",
whilst the sample comprises 60 second-year secondary school students.
In order to select the participants, it was necessary to opt for a sampling method that was
suited to the characteristics of the military educational context and to the actual availability of the
students and the provisions of the military educational authorities. In this case, non-probability
convenience sampling was used, understood as a method in which participants are chosen based
on their accessibility and presence in the environment where the research takes place, without
resorting to random procedures. According to González (2021), this type of sampling is
appropriate when the researcher needs to work with a group that is easily identifiable and available
for the study. Table 1 presents the population and sample of the research.
Vol. 13/ Núm. 2 2026 pág. 1217
Table 1
Research Population and Sample
Population: 1,732 students
Sample Frequency Percentage
Male 41 68 %
Female 19 32 %
Total 60 100%
Source: Data obtained from the educational institution
Data Collection
In order to obtain the information required for this research, the survey technique was
employed, as it allows data to be collected in a direct, organised and systematic manner from the
perspective of the participants themselves (Puente, 2020). This technique was implemented
through a closed structured questionnaire consisting of 18 questions with response options on a
four-level Likert scale (Always, Almost Always, Sometimes and Never), which facilitated the
quantification and comparison of results (Cisneros et al., 2022). The questions addressed, on the
one hand, the use and perception of artificial intelligence applications in the English classroom
and, on the other, the level of proficiency that students demonstrate in the skills of pronunciation,
fluency, vocabulary, grammatical structure, spelling and punctuation.
The instrument was designed on the basis of the variable operationalization process, with
dimensions and indicators defined beforehand in a manner consistent with the objectives of the
study. In this regard, the questionnaire allowed for the generation of sufficient information to
address both the general objective and the specific objectives set out. It should be noted that, in
order to assess the second specific objective, related to the level of proficiency in basic skills,
items 11 to 18 were used as pre-test and post-test, corresponding to the self-assessment of
performance in oral and written production. The remaining items were used to analyze
perceptions and the implementation of artificial intelligence use within the learning process.
Validity and Reliability of the Data Collection Instrument
The instrument was reviewed and validated by experts in research methodology, who
issued an approval document guaranteeing the relevance and coherence of the items. In addition,
the questionnaire was subjected to a reliability analysis using Cronbach's alpha coefficient, the
result of which, presented in Table 2, reached a value of 0.879, considered to represent a good
level of internal consistency.
Table 2
Reliability Statistics
Cronbach's Alpha Number of Items
0,878 17
Note: Value calculated using SPSS
Table 1
Research Population and Sample
Population: 1,732 students
Sample Frequency Percentage
Male 41 68 %
Female 19 32 %
Total 60 100%
Source: Data obtained from the educational institution
Data Collection
In order to obtain the information required for this research, the survey technique was
employed, as it allows data to be collected in a direct, organised and systematic manner from the
perspective of the participants themselves (Puente, 2020). This technique was implemented
through a closed structured questionnaire consisting of 18 questions with response options on a
four-level Likert scale (Always, Almost Always, Sometimes and Never), which facilitated the
quantification and comparison of results (Cisneros et al., 2022). The questions addressed, on the
one hand, the use and perception of artificial intelligence applications in the English classroom
and, on the other, the level of proficiency that students demonstrate in the skills of pronunciation,
fluency, vocabulary, grammatical structure, spelling and punctuation.
The instrument was designed on the basis of the variable operationalization process, with
dimensions and indicators defined beforehand in a manner consistent with the objectives of the
study. In this regard, the questionnaire allowed for the generation of sufficient information to
address both the general objective and the specific objectives set out. It should be noted that, in
order to assess the second specific objective, related to the level of proficiency in basic skills,
items 11 to 18 were used as pre-test and post-test, corresponding to the self-assessment of
performance in oral and written production. The remaining items were used to analyze
perceptions and the implementation of artificial intelligence use within the learning process.
Validity and Reliability of the Data Collection Instrument
The instrument was reviewed and validated by experts in research methodology, who
issued an approval document guaranteeing the relevance and coherence of the items. In addition,
the questionnaire was subjected to a reliability analysis using Cronbach's alpha coefficient, the
result of which, presented in Table 2, reached a value of 0.879, considered to represent a good
level of internal consistency.
Table 2
Reliability Statistics
Cronbach's Alpha Number of Items
0,878 17
Note: Value calculated using SPSS
Vol. 13/ Núm. 2 2026 pág. 1218
This calculated value ensures that the data obtained have the necessary rigour to support
subsequent analyses. Once the instrument was administered, the data obtained were organised
and processed using the Statistical Package for the Social Sciences (SPSS), which allowed for
reliable statistical analyses to be carried out.
RESULTS
The numerical results obtained are based on the administration of the questionnaire
consisting of 18 structured questions on a four-level Likert scale. The items addressed two main
dimensions: on the one hand, the use and perception of artificial intelligence-based applications
within the teaching-learning process of speaking and writing were addressed in questions 1 to 10;
and on the other hand, the level of proficiency that students demonstrate in basic oral and written
production skills, such as pronunciation, fluency, vocabulary, grammatical structure, spelling and
appropriate use of punctuation marks, were addressed in questions 11 to 18.
In order to interpret the results, the numerical values corresponding to the response
options were recoded into two groups to facilitate comparative analysis. The first group comprised
the options: Always (A=4) + Almost Always (AA=3), which grouped the favourable or higher-
frequency responses. The second group comprised the options: Sometimes (S=2) + Never (N=1),
which grouped the lower-frequency or unfavorable responses.
Of the 18 questions comprising the instrument, the first ten were considered in order to
address the first specific objective, related to perceptions of the use and usefulness of artificial
intelligence applications. Accordingly, Table 3 is presented below, the results of which allow for
the analysis corresponding to this objective to be developed.
Table 3
Consolidated Frequency Summary
No. Question S+CS AV+N
1 Does your teacher use artificial intelligence–based applications
as part of the instructional materials for English classes? 4 56
2 Does your teacher employ artificial intelligence as a tool to
correct grammatical writing errors in English? 3 57
3 Does the use of artificial intelligence applications motivate you
to learn the English language? 25 35
4 Does your teacher incorporate interactive English activities that
utilize artificial intelligence features to reinforce vocabulary
acquisition?
4 56
5 Do you receive immediate feedback on the quality of your
English writing through artificial intelligence applications? 12 48
6 Do you think that the use of artificial intelligence facilitates the
learning of writing skills in English? 43 17
7 Do you use artificial intelligence applications to perform
pronunciation exercises in English? 20 40
8 Do you use artificial intelligence applications as a tool to correct
pronunciation errors in English? 12 48
This calculated value ensures that the data obtained have the necessary rigour to support
subsequent analyses. Once the instrument was administered, the data obtained were organised
and processed using the Statistical Package for the Social Sciences (SPSS), which allowed for
reliable statistical analyses to be carried out.
RESULTS
The numerical results obtained are based on the administration of the questionnaire
consisting of 18 structured questions on a four-level Likert scale. The items addressed two main
dimensions: on the one hand, the use and perception of artificial intelligence-based applications
within the teaching-learning process of speaking and writing were addressed in questions 1 to 10;
and on the other hand, the level of proficiency that students demonstrate in basic oral and written
production skills, such as pronunciation, fluency, vocabulary, grammatical structure, spelling and
appropriate use of punctuation marks, were addressed in questions 11 to 18.
In order to interpret the results, the numerical values corresponding to the response
options were recoded into two groups to facilitate comparative analysis. The first group comprised
the options: Always (A=4) + Almost Always (AA=3), which grouped the favourable or higher-
frequency responses. The second group comprised the options: Sometimes (S=2) + Never (N=1),
which grouped the lower-frequency or unfavorable responses.
Of the 18 questions comprising the instrument, the first ten were considered in order to
address the first specific objective, related to perceptions of the use and usefulness of artificial
intelligence applications. Accordingly, Table 3 is presented below, the results of which allow for
the analysis corresponding to this objective to be developed.
Table 3
Consolidated Frequency Summary
No. Question S+CS AV+N
1 Does your teacher use artificial intelligence–based applications
as part of the instructional materials for English classes? 4 56
2 Does your teacher employ artificial intelligence as a tool to
correct grammatical writing errors in English? 3 57
3 Does the use of artificial intelligence applications motivate you
to learn the English language? 25 35
4 Does your teacher incorporate interactive English activities that
utilize artificial intelligence features to reinforce vocabulary
acquisition?
4 56
5 Do you receive immediate feedback on the quality of your
English writing through artificial intelligence applications? 12 48
6 Do you think that the use of artificial intelligence facilitates the
learning of writing skills in English? 43 17
7 Do you use artificial intelligence applications to perform
pronunciation exercises in English? 20 40
8 Do you use artificial intelligence applications as a tool to correct
pronunciation errors in English? 12 48
Vol. 13/ Núm. 2 2026 pág. 1219
9 In the English language teaching–learning process, does your
teacher facilitate knowledge acquisition through the use of
artificial intelligence to develop integrated oral production skills?
1 59
10 In the English language teaching–learning process, does your
teacher facilitate knowledge acquisition through the use of
artificial intelligence to develop integrated
1 59
Note: The options Always and Almost Always were grouped under the A+AA column, whilst the responses Sometimes
and Never were combined under the S+N column.
Likewise, Table 4 is presented below, which contains the results of the remaining eight
questions, which were used as a pre-test to assess the initial level of proficiency in the integrated
speaking and writing skills, in line with the second specific objective of the research.
Table 4
Pre-test Results for Integrated Skills
Nro. Question S+CS AV+N
11 Does the student correctly pronounce the English language? 2 58
12 Does the student communicate fluently in English? 10 50
13 Does the student use appropriate vocabulary when presenting a
topic in English? 4 56
14 Does the student correctly structure sentences when speaking in
English? 0 60
15 Does the student use appropriate vocabulary when writing texts
in English? 11 49
16 Does the student correctly structure grammatical sentences when
writing in English? 10 50
17 Does the student demonstrate correct spelling when writing in
English? 11 49
18 Does the student correctly use punctuation marks when writing
in English? 4 56
Note: The options Always and Almost Always were grouped under the A+AA column, whilst the responses Sometimes
and Never were combined under the S+N column.
Results Relating to Specific Objective 1: Describing students' perceptions of the usefulness
of the ELSA Speaking and Writing & Improve applications in strengthening oral and
written expression in English
In order to gain insight into students' perceptions of the usefulness of the ELSA Speaking
and Writing & Improve applications in strengthening oral and written expression in English, the
results corresponding to the first ten questions of the instrument were analyzed, recoded into two
comparative categories as presented in Table 3.
The data show that, in terms of actual implementation within the classroom, the majority
of students consider the use of Artificial Intelligence-based applications to be limited. In question
1, which addresses the incorporation of applications as part of teaching materials, in question 2,
concerning the use of artificial intelligence to correct grammatical errors, and in question 4,
regarding the inclusion of interactive activities to reinforce vocabulary, favourable responses were
9 In the English language teaching–learning process, does your
teacher facilitate knowledge acquisition through the use of
artificial intelligence to develop integrated oral production skills?
1 59
10 In the English language teaching–learning process, does your
teacher facilitate knowledge acquisition through the use of
artificial intelligence to develop integrated
1 59
Note: The options Always and Almost Always were grouped under the A+AA column, whilst the responses Sometimes
and Never were combined under the S+N column.
Likewise, Table 4 is presented below, which contains the results of the remaining eight
questions, which were used as a pre-test to assess the initial level of proficiency in the integrated
speaking and writing skills, in line with the second specific objective of the research.
Table 4
Pre-test Results for Integrated Skills
Nro. Question S+CS AV+N
11 Does the student correctly pronounce the English language? 2 58
12 Does the student communicate fluently in English? 10 50
13 Does the student use appropriate vocabulary when presenting a
topic in English? 4 56
14 Does the student correctly structure sentences when speaking in
English? 0 60
15 Does the student use appropriate vocabulary when writing texts
in English? 11 49
16 Does the student correctly structure grammatical sentences when
writing in English? 10 50
17 Does the student demonstrate correct spelling when writing in
English? 11 49
18 Does the student correctly use punctuation marks when writing
in English? 4 56
Note: The options Always and Almost Always were grouped under the A+AA column, whilst the responses Sometimes
and Never were combined under the S+N column.
Results Relating to Specific Objective 1: Describing students' perceptions of the usefulness
of the ELSA Speaking and Writing & Improve applications in strengthening oral and
written expression in English
In order to gain insight into students' perceptions of the usefulness of the ELSA Speaking
and Writing & Improve applications in strengthening oral and written expression in English, the
results corresponding to the first ten questions of the instrument were analyzed, recoded into two
comparative categories as presented in Table 3.
The data show that, in terms of actual implementation within the classroom, the majority
of students consider the use of Artificial Intelligence-based applications to be limited. In question
1, which addresses the incorporation of applications as part of teaching materials, in question 2,
concerning the use of artificial intelligence to correct grammatical errors, and in question 4,
regarding the inclusion of interactive activities to reinforce vocabulary, favourable responses were
Vol. 13/ Núm. 2 2026 pág. 1220
minimal compared to a marked majority who indicated Sometimes or Never. Similarly, when
asked whether the teacher facilitates the development of oral and written production through
artificial intelligence (questions 9 and 10), virtually all students indicated that this occurs with
low frequency.
When examining aspects related to feedback, a predominantly unfavorable perception
was also observed. In question 5, which referred to whether students received immediate feedback
on the quality of their writing through AI applications, the S+N category clearly predominated
with 48 responses, suggesting that such feedback was not a frequent experience for the majority.
Similarly, in question 8, focused on the use of applications as a support tool for correcting
pronunciation errors, the same pattern was repeated: few favourable responses and a majority of
48 students indicated that this type of correction occurred only sometimes or never.
However, the perception changes when the potential usefulness of these tools is analyzed.
In question 6, a considerable proportion of students indicated that artificial intelligence can indeed
facilitate the learning of written English. An intermediate tendency was also observed in aspects
such as motivation to learn (question 3) and the use of applications for pronunciation exercises
(question 7), where favourable responses, although not in the majority, demonstrate an openness
towards these technologies.
These results allow us to observe that, at an initial stage, the systematic use of applications
such as ELSA Speaking and Writing & Improve was not an established practice in the classroom;
however, students acknowledge their potential for strengthening both oral and written expression.
This gap between low implementation and a positive perception of usefulness constitutes a
relevant starting point for subsequently analyzing the impact of the application of a didactic
process and practical exercises supported by AI.
Results of Specific Objective 2. Assessing the level of proficiency demonstrated by students
in the basic skills of pronunciation, fluency, spelling and punctuation
In response to the second specific objective, aimed at assessing the level of proficiency
demonstrated by students in the basic skills of pronunciation, fluency, spelling and punctuation,
the results corresponding to items 11 to 18 of the questionnaire were analyzed. As in the previous
analysis, the responses were grouped into two categories: A+AA (Always and Almost Always)
as an indicator of a favourable perception of proficiency, and S+N (Sometimes and Never) as an
indication of limited or insufficient proficiency.
About pronunciation, the results show a low initial level. In question 11, only 2 students
considered that they pronounce English correctly, whilst 58 indicated that they do so only
sometimes or never. This tendency is maintained in item 12, referring to communicative fluency,
where 10 responses were favourable compared to 50 unfavorable. Likewise, question 13,
concerning the use of vocabulary when presenting a topic, recorded only 4 positive responses
compared to 56 negative ones. The most critical result was observed in item 14, related to the
minimal compared to a marked majority who indicated Sometimes or Never. Similarly, when
asked whether the teacher facilitates the development of oral and written production through
artificial intelligence (questions 9 and 10), virtually all students indicated that this occurs with
low frequency.
When examining aspects related to feedback, a predominantly unfavorable perception
was also observed. In question 5, which referred to whether students received immediate feedback
on the quality of their writing through AI applications, the S+N category clearly predominated
with 48 responses, suggesting that such feedback was not a frequent experience for the majority.
Similarly, in question 8, focused on the use of applications as a support tool for correcting
pronunciation errors, the same pattern was repeated: few favourable responses and a majority of
48 students indicated that this type of correction occurred only sometimes or never.
However, the perception changes when the potential usefulness of these tools is analyzed.
In question 6, a considerable proportion of students indicated that artificial intelligence can indeed
facilitate the learning of written English. An intermediate tendency was also observed in aspects
such as motivation to learn (question 3) and the use of applications for pronunciation exercises
(question 7), where favourable responses, although not in the majority, demonstrate an openness
towards these technologies.
These results allow us to observe that, at an initial stage, the systematic use of applications
such as ELSA Speaking and Writing & Improve was not an established practice in the classroom;
however, students acknowledge their potential for strengthening both oral and written expression.
This gap between low implementation and a positive perception of usefulness constitutes a
relevant starting point for subsequently analyzing the impact of the application of a didactic
process and practical exercises supported by AI.
Results of Specific Objective 2. Assessing the level of proficiency demonstrated by students
in the basic skills of pronunciation, fluency, spelling and punctuation
In response to the second specific objective, aimed at assessing the level of proficiency
demonstrated by students in the basic skills of pronunciation, fluency, spelling and punctuation,
the results corresponding to items 11 to 18 of the questionnaire were analyzed. As in the previous
analysis, the responses were grouped into two categories: A+AA (Always and Almost Always)
as an indicator of a favourable perception of proficiency, and S+N (Sometimes and Never) as an
indication of limited or insufficient proficiency.
About pronunciation, the results show a low initial level. In question 11, only 2 students
considered that they pronounce English correctly, whilst 58 indicated that they do so only
sometimes or never. This tendency is maintained in item 12, referring to communicative fluency,
where 10 responses were favourable compared to 50 unfavorable. Likewise, question 13,
concerning the use of vocabulary when presenting a topic, recorded only 4 positive responses
compared to 56 negative ones. The most critical result was observed in item 14, related to the
Vol. 13/ Núm. 2 2026 pág. 1221
correct structuring of sentences when speaking, where no student selected the higher-frequency
categories, with all responses concentrated in the S+N category. These data allow us to infer that
the self-perception of oral performance is considerably low, particularly with regard to syntactic
organization and structural accuracy now of expression.
Regarding written production, the results show slight variations, although the general
tendency remains predominantly unfavorable. In the appropriate use of vocabulary for writing
texts, addressed in question 15, 11 students indicated frequent levels of proficiency, whilst 49
reported difficulties. Similarly, in question 16, concerning the grammatical structuring of written
sentences, only 10 responses fell within the A+AA category compared to 50 in the S+N category.
Spelling, addressed in question 17, yielded 11 favourable responses but 49 unfavorable ones,
revealing insecurity in formal writing. Finally, regarding the correct use of punctuation marks,
addressed in question 18, only 4 students indicated that they apply them consistently, whilst 56
acknowledged doing so infrequently.
These results demonstrate that, at an initial stage of the research, students perceived a
limited level of proficiency in both oral and written skills. The greatest difficulties were
concentrated in the structuring of sentences when speaking and in the appropriate use of
punctuation in writing, followed by problems with pronunciation and grammatical accuracy.
These initial results reveal the need for pedagogical strategies aimed at the comprehensive
strengthening of communicative skills, which supports the relevance of designing a didactic
process and implementing practical exercises supported by ELSA Speaking and Writing &
Improve to strengthen students' integrated speaking and writing skills.
Results of Specific Objective 3. Designing a didactic process through which oral and written
production skills are articulately integrated via the use of ELSA Speaking and Writing &
Improve in the English classroom
Based on the results obtained in specific objective 1, where limited use of artificial
intelligence educational applications and difficulties in the integrated skills of speaking and
writing were evidenced, and in relation to specific objective 2, a didactic process was designed
and implemented in which practical exercises supported by ELSA Speaking and Writing &
Improve were applied.
The didactic process presented below, from Table 5 to Table 8, was planned to be
developed progressively over 3 sessions of 35 minutes each across 4 weeks, with activities per
session aimed at strengthening students' pronunciation, oral fluency and written production,
promoting more active, practical and meaningful learning.
correct structuring of sentences when speaking, where no student selected the higher-frequency
categories, with all responses concentrated in the S+N category. These data allow us to infer that
the self-perception of oral performance is considerably low, particularly with regard to syntactic
organization and structural accuracy now of expression.
Regarding written production, the results show slight variations, although the general
tendency remains predominantly unfavorable. In the appropriate use of vocabulary for writing
texts, addressed in question 15, 11 students indicated frequent levels of proficiency, whilst 49
reported difficulties. Similarly, in question 16, concerning the grammatical structuring of written
sentences, only 10 responses fell within the A+AA category compared to 50 in the S+N category.
Spelling, addressed in question 17, yielded 11 favourable responses but 49 unfavorable ones,
revealing insecurity in formal writing. Finally, regarding the correct use of punctuation marks,
addressed in question 18, only 4 students indicated that they apply them consistently, whilst 56
acknowledged doing so infrequently.
These results demonstrate that, at an initial stage of the research, students perceived a
limited level of proficiency in both oral and written skills. The greatest difficulties were
concentrated in the structuring of sentences when speaking and in the appropriate use of
punctuation in writing, followed by problems with pronunciation and grammatical accuracy.
These initial results reveal the need for pedagogical strategies aimed at the comprehensive
strengthening of communicative skills, which supports the relevance of designing a didactic
process and implementing practical exercises supported by ELSA Speaking and Writing &
Improve to strengthen students' integrated speaking and writing skills.
Results of Specific Objective 3. Designing a didactic process through which oral and written
production skills are articulately integrated via the use of ELSA Speaking and Writing &
Improve in the English classroom
Based on the results obtained in specific objective 1, where limited use of artificial
intelligence educational applications and difficulties in the integrated skills of speaking and
writing were evidenced, and in relation to specific objective 2, a didactic process was designed
and implemented in which practical exercises supported by ELSA Speaking and Writing &
Improve were applied.
The didactic process presented below, from Table 5 to Table 8, was planned to be
developed progressively over 3 sessions of 35 minutes each across 4 weeks, with activities per
session aimed at strengthening students' pronunciation, oral fluency and written production,
promoting more active, practical and meaningful learning.
Vol. 13/ Núm. 2 2026 pág. 1222
Table 5
Week 1: Familiarization and Practical Diagnosis
Element Description
Weekly
Objective
To introduce students to the guided use of the ELSA Speak and Writing &
Improve applications, identifying their initial level of proficiency in
pronunciation and writing.
Resources Mobile devices or computers, ELSA Speak, Writing & Improve, projector,
activity guide.
Methodology Guided learning with technological support and participatory reflection.
Activity 1
Initial pronunciation exercises with ELSA Speak
Time: 35 minutes
• The teacher guides students on how to access and use the ELSA Speak
application at a basic level.
• Students carry out initial pronunciation exercises using simple words and
phrases.
• Each student reviews their baseline score and identifies their main
phonetic difficulties.
Activity 2
Writing a short paragraph in Writing & Improve
Time: 35 minutes
• The teacher explains the basic structure of a simple paragraph in English.
• Students write a short text about their daily routine (approximately 80
words).
• The automatic feedback from the application is reviewed and the most
frequent errors are discussed.
Activity 3
Group reflection on the use of AI
Time: 35 minutes
• Students share the difficulties they encountered in pronunciation and
writing.
• The teacher guides a reflection on how AI can support learning.
• Initial perceptions of the use of these tools are recorded.
Note: Design and implementation carried out by the researcher.
Table 6
Week 2: Development of Specific Skills
Element Description
Weekly
Objective
To strengthen segmental pronunciation and basic grammatical structure
through the systematic use of AI applications.
Resources AI applications, grammar worksheets, student notebook.
Methodology Guided practice and immediate technology-assisted feedback.
Activity 1
Practice of problematic sounds in ELSA Speak
Time: 35 minutes
• The teacher identifies the English sounds that present the greatest
difficulty.
• Students practice these sounds repeatedly with automatic feedback.
• Students compare their current results with those obtained previously.
Activity 2 Writing of a descriptive paragraph in Writing & Improve
Time. 35 minutes
Table 5
Week 1: Familiarization and Practical Diagnosis
Element Description
Weekly
Objective
To introduce students to the guided use of the ELSA Speak and Writing &
Improve applications, identifying their initial level of proficiency in
pronunciation and writing.
Resources Mobile devices or computers, ELSA Speak, Writing & Improve, projector,
activity guide.
Methodology Guided learning with technological support and participatory reflection.
Activity 1
Initial pronunciation exercises with ELSA Speak
Time: 35 minutes
• The teacher guides students on how to access and use the ELSA Speak
application at a basic level.
• Students carry out initial pronunciation exercises using simple words and
phrases.
• Each student reviews their baseline score and identifies their main
phonetic difficulties.
Activity 2
Writing a short paragraph in Writing & Improve
Time: 35 minutes
• The teacher explains the basic structure of a simple paragraph in English.
• Students write a short text about their daily routine (approximately 80
words).
• The automatic feedback from the application is reviewed and the most
frequent errors are discussed.
Activity 3
Group reflection on the use of AI
Time: 35 minutes
• Students share the difficulties they encountered in pronunciation and
writing.
• The teacher guides a reflection on how AI can support learning.
• Initial perceptions of the use of these tools are recorded.
Note: Design and implementation carried out by the researcher.
Table 6
Week 2: Development of Specific Skills
Element Description
Weekly
Objective
To strengthen segmental pronunciation and basic grammatical structure
through the systematic use of AI applications.
Resources AI applications, grammar worksheets, student notebook.
Methodology Guided practice and immediate technology-assisted feedback.
Activity 1
Practice of problematic sounds in ELSA Speak
Time: 35 minutes
• The teacher identifies the English sounds that present the greatest
difficulty.
• Students practice these sounds repeatedly with automatic feedback.
• Students compare their current results with those obtained previously.
Activity 2 Writing of a descriptive paragraph in Writing & Improve
Time. 35 minutes
Vol. 13/ Núm. 2 2026 pág. 1223
• The teacher explains the use of descriptive vocabulary and text
organization.
• The student writes a descriptive paragraph applying what has been
explained.
• The student corrects their text using the suggestions provided by the
application.
Activity 3
Identification of correction patterns
Time: 35 minutes
• The student reviews the most frequent errors detected by the AI.
• The teacher clarifies common grammatical doubts.
• The student produces corrected examples based on the errors identified.
Note: Design and implementation carried out by the researcher.
Table 7
Week 3: Integration of Speaking and Writing
Element Description
Weekly
Objective
To integrate oral and written expression skills in contextualized
communicative activities.
Resources Mobile devices, AI applications.
Methodology Communicative language learning and task-based learning.
Activity 1
Reading aloud with ELSA Speak
Time: 35 minutes
• The teacher selects short texts that have been previously studied.
• The student reads the text aloud using the application.
• The student analyses the feedback related to fluency and pronunciation.
Activity 2
Role-play based on created texts
Time: 35 minutes
• The teacher organizes students into pairs or small groups.
• The student prepares dialogues based on their written texts.
• The student performs the dialogues whilst the teacher provides feedback.
Activity 3
Writing and oral explanation
Time: 35 minutes
• The teacher guides the writing of a short email in English.
• The student writes and corrects the text using Writing & Improve.
• The student records an audio explaining the content of the text using
ELSA Speak.
Note: Design and implementation carried out by the researcher.
• The teacher explains the use of descriptive vocabulary and text
organization.
• The student writes a descriptive paragraph applying what has been
explained.
• The student corrects their text using the suggestions provided by the
application.
Activity 3
Identification of correction patterns
Time: 35 minutes
• The student reviews the most frequent errors detected by the AI.
• The teacher clarifies common grammatical doubts.
• The student produces corrected examples based on the errors identified.
Note: Design and implementation carried out by the researcher.
Table 7
Week 3: Integration of Speaking and Writing
Element Description
Weekly
Objective
To integrate oral and written expression skills in contextualized
communicative activities.
Resources Mobile devices, AI applications.
Methodology Communicative language learning and task-based learning.
Activity 1
Reading aloud with ELSA Speak
Time: 35 minutes
• The teacher selects short texts that have been previously studied.
• The student reads the text aloud using the application.
• The student analyses the feedback related to fluency and pronunciation.
Activity 2
Role-play based on created texts
Time: 35 minutes
• The teacher organizes students into pairs or small groups.
• The student prepares dialogues based on their written texts.
• The student performs the dialogues whilst the teacher provides feedback.
Activity 3
Writing and oral explanation
Time: 35 minutes
• The teacher guides the writing of a short email in English.
• The student writes and corrects the text using Writing & Improve.
• The student records an audio explaining the content of the text using
ELSA Speak.
Note: Design and implementation carried out by the researcher.
Vol. 13/ Núm. 2 2026 pág. 1224
Table 8
Week 4: Consolidation and Conclusion of the Intervention
Element Description
Weekly
Objective
To consolidate the integrated speaking and writing skills developed
throughout the intervention and to conclude the instructional process by
demonstrating the progress achieved by the students.
Resources ELSA Speak, Writing & Improve, student notebook.
Methodology Guided reinforcement, supervised autonomous learning and reflective
conclusion.
Activity 1
Reinforcement of pronunciation and fluency with ELSA Speak
Time. 35 minutes
• The teacher revisits the main sounds and pronunciation patterns covered
in previous weeks.
• The student once again carries out pronunciation and guided reading
exercises using ELSA Speak.
• The student analyses the automatic feedback and recognizes the
improvements achieved in fluency and oral clarity.
Activity 2
Integrated oral expression practice (storytelling)
Time: 35 minutes
• The teacher guides an oral narration activity based on personal
experiences or everyday topics.
• The student prepares and delivers their story orally, applying the
recommendations previously received.
• The student adjusts their intonation, rhythm and pronunciation with the
support of the application's feedback.
Activity 3
Final reinforcement of written production with Writing & Improve
Time: 35 minutes
• The teacher recalls the basic criteria of coherence, vocabulary and
grammatical structure covered during the intervention.
• The student writes a final text of approximately 150 words, applying the
knowledge acquired throughout the intervention.
• The student reviews and improves their written work using Writing &
Improve, consolidating the corrections learnt throughout the process.
Note: Design and implementation carried out by the researcher.
Results of Specific Objective 4. Determining the contribution of the implementation of
practical exercises supported by ELSA Speaking and Writing & Improve to the
strengthening of integrated speaking and writing skills among students.
In order to determine the contribution of the implementation of practical exercises
supported by ELSA Speaking and Writing & Improve to the strengthening of integrated speaking
and writing skills among students, the questionnaire used in the initial phase of the research was
re-administered. On this occasion, only items 11 to 18 were considered, corresponding to the self-
assessment of performance in pronunciation, fluency, vocabulary, grammatical structure, spelling
and punctuation.
Table 8
Week 4: Consolidation and Conclusion of the Intervention
Element Description
Weekly
Objective
To consolidate the integrated speaking and writing skills developed
throughout the intervention and to conclude the instructional process by
demonstrating the progress achieved by the students.
Resources ELSA Speak, Writing & Improve, student notebook.
Methodology Guided reinforcement, supervised autonomous learning and reflective
conclusion.
Activity 1
Reinforcement of pronunciation and fluency with ELSA Speak
Time. 35 minutes
• The teacher revisits the main sounds and pronunciation patterns covered
in previous weeks.
• The student once again carries out pronunciation and guided reading
exercises using ELSA Speak.
• The student analyses the automatic feedback and recognizes the
improvements achieved in fluency and oral clarity.
Activity 2
Integrated oral expression practice (storytelling)
Time: 35 minutes
• The teacher guides an oral narration activity based on personal
experiences or everyday topics.
• The student prepares and delivers their story orally, applying the
recommendations previously received.
• The student adjusts their intonation, rhythm and pronunciation with the
support of the application's feedback.
Activity 3
Final reinforcement of written production with Writing & Improve
Time: 35 minutes
• The teacher recalls the basic criteria of coherence, vocabulary and
grammatical structure covered during the intervention.
• The student writes a final text of approximately 150 words, applying the
knowledge acquired throughout the intervention.
• The student reviews and improves their written work using Writing &
Improve, consolidating the corrections learnt throughout the process.
Note: Design and implementation carried out by the researcher.
Results of Specific Objective 4. Determining the contribution of the implementation of
practical exercises supported by ELSA Speaking and Writing & Improve to the
strengthening of integrated speaking and writing skills among students.
In order to determine the contribution of the implementation of practical exercises
supported by ELSA Speaking and Writing & Improve to the strengthening of integrated speaking
and writing skills among students, the questionnaire used in the initial phase of the research was
re-administered. On this occasion, only items 11 to 18 were considered, corresponding to the self-
assessment of performance in pronunciation, fluency, vocabulary, grammatical structure, spelling
and punctuation.
Vol. 13/ Núm. 2 2026 pág. 1225
The results obtained from the re-administration of the instrument, following the
consolidation of responses into two columns, A+AA (Always + Almost Always) and S+N
(Sometimes + Never), are presented below in Table 9.
Table 9
Post-test Results for Integrated Skills
Nro. Question S+CS AV+N
11 Does the student correctly pronounce the English language? 28 32
12 Does the student communicate fluently in English? 30 30
13 Does the student use appropriate vocabulary when presenting a
topic in English? 26 34
14 Does the student correctly structure sentences when speaking in
English? 24 36
15 Does the student use appropriate vocabulary when writing texts
in English? 32 28
16 Does the student correctly structure grammatical sentences when
writing in English? 29 31
17 Does the student demonstrate correct spelling when writing in
English? 34 26
18 Does the student correctly use punctuation marks when writing
in English? 27 33
Note: The options Always and Almost Always were grouped under the A+AA column, whilst the responses Sometimes
and Never were combined under the S+N column.
In order to compare the pre-test and post-test results, the frequencies of favourable
responses, always and almost always, were converted into a single numerical value for each
question. To this end, the number of favourable responses was divided by the total of 60 students,
thus obtaining a proportion ranging from 0 to 1. This procedure allowed for eight values to be
obtained in the pre-test and eight in the post-test, facilitating the statistical comparison between
both stages and enabling an objective identification of whether an improvement in the assessed
skills had occurred. These data are presented below in Table 10.
Table 10
Comparison of Favourable Pre-test and Post-test Proportions
Question 11 12 13 14 15 16 17 18 Mean
Pretest 0,03 0,17 0,07 0,00 0,18 0,17 0,18 0,07 0,11
Postest 0,47 0,50 0,43 0,40 0,53 0,48 0,57 0,45 0,48
Note: The values correspond to the proportion of favourable responses (A+AA) relative to the total number of students
(n = 60).
Prior to the comparative analysis between the pre-test and post-test results, it was
necessary to verify the statistical behaviour of the data obtained. To this end, a normality test was
applied in order to determine whether the distribution of values met the assumptions required for
the use of parametric tests (Sánchez et al., 2024). Given that 8 pairs of data were analyzed, the
The results obtained from the re-administration of the instrument, following the
consolidation of responses into two columns, A+AA (Always + Almost Always) and S+N
(Sometimes + Never), are presented below in Table 9.
Table 9
Post-test Results for Integrated Skills
Nro. Question S+CS AV+N
11 Does the student correctly pronounce the English language? 28 32
12 Does the student communicate fluently in English? 30 30
13 Does the student use appropriate vocabulary when presenting a
topic in English? 26 34
14 Does the student correctly structure sentences when speaking in
English? 24 36
15 Does the student use appropriate vocabulary when writing texts
in English? 32 28
16 Does the student correctly structure grammatical sentences when
writing in English? 29 31
17 Does the student demonstrate correct spelling when writing in
English? 34 26
18 Does the student correctly use punctuation marks when writing
in English? 27 33
Note: The options Always and Almost Always were grouped under the A+AA column, whilst the responses Sometimes
and Never were combined under the S+N column.
In order to compare the pre-test and post-test results, the frequencies of favourable
responses, always and almost always, were converted into a single numerical value for each
question. To this end, the number of favourable responses was divided by the total of 60 students,
thus obtaining a proportion ranging from 0 to 1. This procedure allowed for eight values to be
obtained in the pre-test and eight in the post-test, facilitating the statistical comparison between
both stages and enabling an objective identification of whether an improvement in the assessed
skills had occurred. These data are presented below in Table 10.
Table 10
Comparison of Favourable Pre-test and Post-test Proportions
Question 11 12 13 14 15 16 17 18 Mean
Pretest 0,03 0,17 0,07 0,00 0,18 0,17 0,18 0,07 0,11
Postest 0,47 0,50 0,43 0,40 0,53 0,48 0,57 0,45 0,48
Note: The values correspond to the proportion of favourable responses (A+AA) relative to the total number of students
(n = 60).
Prior to the comparative analysis between the pre-test and post-test results, it was
necessary to verify the statistical behaviour of the data obtained. To this end, a normality test was
applied in order to determine whether the distribution of values met the assumptions required for
the use of parametric tests (Sánchez et al., 2024). Given that 8 pairs of data were analyzed, the
Vol. 13/ Núm. 2 2026 pág. 1226
values of the Shapiro-Wilk test were considered, as this procedure is most recommended when
the sample size is small.
Table 11
Normality Tests
Kolmogorov-Smirnova Shapiro-Wilk
Statistic gl Sig. Statistic gl Sig.
Pretest 0,295 8 0,039 0,833 8 0,064
Postest 0,116 8 0,200* 0,990 8 0,995
Note: Data obtained using the SPSS programme
Upon reviewing the results recorded in Table 11, it can be observed that, in the case of
the pre-test, the significance value obtained was 0.064, whilst in the post-test it was 0.995. In both
cases, these values are above the established significance level of 0.05. This indicates that both
the pre-test and post-test data present a normal distribution, which allows the analysis to continue
using parametric statistical tests.
Based on the normal distribution of the data, a parametric statistical test was applied to
carry out the comparison between both evaluation stages. The choice of this type of test is
grounded in the fact that parametric techniques allow for the analysis of mean differences with
greater precision when the normality assumptions are met, offering more robust and reliable
results for determining whether the changes observed following the intervention are statistically
significant (Bautista et al., 2020). The data obtained from the application of this test are presented
below in Table 12.
Table 12
Paired Samples Test
Paired Differences
t gl
Sig.
(bilateral)Mean
Dev.
Deviation
Dev.
Standard
Error
95% Confidence interval of
the difference
Lower Upper
Pretest
Postest
-,370000 0,041404 0,014639 -,404615 -,335385 -25,276 7 0,000
Note: Data obtained using the SPSS programme.
The mean difference recorded between both stages was -0.370, indicating that the post-
test values were higher than those of the pre-test. This average increase reflects an improvement
in the proportion of favourable responses following the implementation of the didactic process
supported by artificial intelligence tools. Likewise, the internal consistency of the instrument was
high, with a Cronbach's alpha of 0.878, whilst the statistical significance obtained (p < 0.05)
confirms that the observed differences are not attributable to chance, but rather to the effect of the
intervention applied.
Furthermore, the 95% confidence interval for the mean difference ranged from -0.404 to
-0.335, demonstrating that the observed improvement is not the result of chance, but rather
values of the Shapiro-Wilk test were considered, as this procedure is most recommended when
the sample size is small.
Table 11
Normality Tests
Kolmogorov-Smirnova Shapiro-Wilk
Statistic gl Sig. Statistic gl Sig.
Pretest 0,295 8 0,039 0,833 8 0,064
Postest 0,116 8 0,200* 0,990 8 0,995
Note: Data obtained using the SPSS programme
Upon reviewing the results recorded in Table 11, it can be observed that, in the case of
the pre-test, the significance value obtained was 0.064, whilst in the post-test it was 0.995. In both
cases, these values are above the established significance level of 0.05. This indicates that both
the pre-test and post-test data present a normal distribution, which allows the analysis to continue
using parametric statistical tests.
Based on the normal distribution of the data, a parametric statistical test was applied to
carry out the comparison between both evaluation stages. The choice of this type of test is
grounded in the fact that parametric techniques allow for the analysis of mean differences with
greater precision when the normality assumptions are met, offering more robust and reliable
results for determining whether the changes observed following the intervention are statistically
significant (Bautista et al., 2020). The data obtained from the application of this test are presented
below in Table 12.
Table 12
Paired Samples Test
Paired Differences
t gl
Sig.
(bilateral)Mean
Dev.
Deviation
Dev.
Standard
Error
95% Confidence interval of
the difference
Lower Upper
Pretest
Postest
-,370000 0,041404 0,014639 -,404615 -,335385 -25,276 7 0,000
Note: Data obtained using the SPSS programme.
The mean difference recorded between both stages was -0.370, indicating that the post-
test values were higher than those of the pre-test. This average increase reflects an improvement
in the proportion of favourable responses following the implementation of the didactic process
supported by artificial intelligence tools. Likewise, the internal consistency of the instrument was
high, with a Cronbach's alpha of 0.878, whilst the statistical significance obtained (p < 0.05)
confirms that the observed differences are not attributable to chance, but rather to the effect of the
intervention applied.
Furthermore, the 95% confidence interval for the mean difference ranged from -0.404 to
-0.335, demonstrating that the observed improvement is not the result of chance, but rather
Vol. 13/ Núm. 2 2026 pág. 1227
remains within a consistent range of increase. The t-value obtained (-25.276) with 7 degrees of
freedom and a two-tailed significance of 0.000 demonstrates that the probability of this difference
being attributable to random variation is practically null (p < 0.05).
By considering the significance value of 0.000 as lower than the established threshold of
0.05, the null hypothesis is rejected and the alternative hypothesis (H₁) is accepted, which holds
that artificial intelligence educational applications do have an impact on the development of
integrated English language skills. These results allow us to affirm, from a statistical analysis
perspective, that the intervention applied contributed significantly to the improvement of students'
communicative performance.
DISCUSSION
The results showed that, on the one hand, the majority of students reported that the use
of artificial intelligence applications in the classroom was infrequent, and on the other, they
demonstrated a more favourable disposition when asked about the usefulness of AI in supporting
learning, particularly in writing and, to a lesser extent, in motivation and pronunciation. This
combination is consistent with what Li (2022) highlights, namely that AI can provide immediate
feedback and personalised activities, but also suggests that these benefits are realised only when
there is genuine and sustained use. In the same way, the results are related to the findings of Huang
et al. (2022), by demonstrating that AI can complement the development of speaking and writing,
although in this case the initial evidence indicated that such integration had not yet been
consolidated in everyday classroom practice.
The low results regarding immediate feedback in writing and pronunciation correction
suggest that students were not regularly receiving one of the most distinctive benefits of these
technologies. This partially contrasts with what Ayala and Alvarado (2023) and Kolegova and
Levina (2024) propose, highlighting the flexibility and personalization of AI; in the context
investigated, students' perceptions showed that these advantages were not yet being fully
expressed due to the limited presence of AI in the classroom. However, the fact that many students
considered AI to facilitate writing reinforces what Fitria (2021) maintains: these technologies tend
to generate more dynamic environments and raise motivation, although in this case that effect
appears more as a positive expectation than as a widespread prior experience. Likewise, the
intermediate values regarding motivation are consistent with Schmidt and Strassner (2022), who
associate AI tools with autonomy and self-regulation; even when the classroom did not integrate
them systematically, some students appear to recognize their value for learning in a more
independent manner.
The research data showed a low initial level in basic skills, particularly in pronunciation,
fluency, oral structuring and use of punctuation. This situation is consistent with what Syuhra et
al. (2025) indicate regarding the persistence of approaches centred on grammar and
remains within a consistent range of increase. The t-value obtained (-25.276) with 7 degrees of
freedom and a two-tailed significance of 0.000 demonstrates that the probability of this difference
being attributable to random variation is practically null (p < 0.05).
By considering the significance value of 0.000 as lower than the established threshold of
0.05, the null hypothesis is rejected and the alternative hypothesis (H₁) is accepted, which holds
that artificial intelligence educational applications do have an impact on the development of
integrated English language skills. These results allow us to affirm, from a statistical analysis
perspective, that the intervention applied contributed significantly to the improvement of students'
communicative performance.
DISCUSSION
The results showed that, on the one hand, the majority of students reported that the use
of artificial intelligence applications in the classroom was infrequent, and on the other, they
demonstrated a more favourable disposition when asked about the usefulness of AI in supporting
learning, particularly in writing and, to a lesser extent, in motivation and pronunciation. This
combination is consistent with what Li (2022) highlights, namely that AI can provide immediate
feedback and personalised activities, but also suggests that these benefits are realised only when
there is genuine and sustained use. In the same way, the results are related to the findings of Huang
et al. (2022), by demonstrating that AI can complement the development of speaking and writing,
although in this case the initial evidence indicated that such integration had not yet been
consolidated in everyday classroom practice.
The low results regarding immediate feedback in writing and pronunciation correction
suggest that students were not regularly receiving one of the most distinctive benefits of these
technologies. This partially contrasts with what Ayala and Alvarado (2023) and Kolegova and
Levina (2024) propose, highlighting the flexibility and personalization of AI; in the context
investigated, students' perceptions showed that these advantages were not yet being fully
expressed due to the limited presence of AI in the classroom. However, the fact that many students
considered AI to facilitate writing reinforces what Fitria (2021) maintains: these technologies tend
to generate more dynamic environments and raise motivation, although in this case that effect
appears more as a positive expectation than as a widespread prior experience. Likewise, the
intermediate values regarding motivation are consistent with Schmidt and Strassner (2022), who
associate AI tools with autonomy and self-regulation; even when the classroom did not integrate
them systematically, some students appear to recognize their value for learning in a more
independent manner.
The research data showed a low initial level in basic skills, particularly in pronunciation,
fluency, oral structuring and use of punctuation. This situation is consistent with what Syuhra et
al. (2025) indicate regarding the persistence of approaches centred on grammar and
Vol. 13/ Núm. 2 2026 pág. 1228
memorization, which ultimately weaken the development of real communicative skills. In the
same vein, the initial results support what Armendáriz et al. (2024) describe, noting that when
learning is concentrated on formal structures without meaningful practice, fluency and functional
communication are adversely affected. From this perspective, the initial data reflect precisely that
need for interactive scenarios and constant feedback, in line with what Mayorga and Tibán (2024)
propose regarding meaningful learning in speaking and writing.
When analyzing the effects of the didactic process and the pre-test/post-test comparison,
a substantial increase was observed across all research questions. This improvement is related to
findings reported by studies that highlight the potential of specific tools; for example, Anggraini
(2022) notes improvements in pronunciation associated with ELSA Speak, and in the post-test of
this research a notable increase in the perception of correct pronunciation and fluency is observed.
Similarly, the improvement in writing is related to what authors who have studied adaptive
platforms and automated feedback describe, such as Xiaofan and Annamalai (2025) and Sehlaoui
(2024), who link these environments with improvements in performance, motivation and progress
in writing.
Furthermore, the improvement in speaking and writing skills is also consistent with what
Alqaed (2024) and Sari (2023) propose, highlighting that the use of AI can reduce dependence on
rote learning, promote more consistent corrections and enhance meaningful interaction. In this
case, the increase in values from the pre-test to the post-test demonstrates that, when exercises
were applied in an organised manner, students had greater opportunities to practice, adjust and
improve. This is also related to what Marghany (2023) notes regarding the impact of
individualized feedback on oral fluency and accuracy, as the post-test shows a significant advance
in areas that were particularly low at the outset. In turn, the idea that AI can support autonomy
and fluency, as mentioned by Al-khresheh (2024), is reflected in the general pattern of
improvement following the implementation of the didactic process. The results allow us to
maintain that the effect of artificial intelligence tools depends not solely on their technological
availability, but on the manner in which they are integrated into a structured didactic sequence.
Authors such as Hipo et al. (2022) and Murillo et al. (2021) maintain that when skills are
worked on in an articulated manner, each one reinforces the other; in this research, speaking and
writing were addressed in parallel, which may explain why the improvements were not limited to
a single aspect, but were observed in both oral and written expression. Similarly, the idea of skills
integration and contextualized activities is related to what Mahapatra (2024) puts forward, whilst
the subsequent advances in writing and vocabulary are consistent with the results reported by
Sapan and Uzun (2024) in experiences where digital resources are incorporated as a support for
English language learning.
CONCLUSIONS
memorization, which ultimately weaken the development of real communicative skills. In the
same vein, the initial results support what Armendáriz et al. (2024) describe, noting that when
learning is concentrated on formal structures without meaningful practice, fluency and functional
communication are adversely affected. From this perspective, the initial data reflect precisely that
need for interactive scenarios and constant feedback, in line with what Mayorga and Tibán (2024)
propose regarding meaningful learning in speaking and writing.
When analyzing the effects of the didactic process and the pre-test/post-test comparison,
a substantial increase was observed across all research questions. This improvement is related to
findings reported by studies that highlight the potential of specific tools; for example, Anggraini
(2022) notes improvements in pronunciation associated with ELSA Speak, and in the post-test of
this research a notable increase in the perception of correct pronunciation and fluency is observed.
Similarly, the improvement in writing is related to what authors who have studied adaptive
platforms and automated feedback describe, such as Xiaofan and Annamalai (2025) and Sehlaoui
(2024), who link these environments with improvements in performance, motivation and progress
in writing.
Furthermore, the improvement in speaking and writing skills is also consistent with what
Alqaed (2024) and Sari (2023) propose, highlighting that the use of AI can reduce dependence on
rote learning, promote more consistent corrections and enhance meaningful interaction. In this
case, the increase in values from the pre-test to the post-test demonstrates that, when exercises
were applied in an organised manner, students had greater opportunities to practice, adjust and
improve. This is also related to what Marghany (2023) notes regarding the impact of
individualized feedback on oral fluency and accuracy, as the post-test shows a significant advance
in areas that were particularly low at the outset. In turn, the idea that AI can support autonomy
and fluency, as mentioned by Al-khresheh (2024), is reflected in the general pattern of
improvement following the implementation of the didactic process. The results allow us to
maintain that the effect of artificial intelligence tools depends not solely on their technological
availability, but on the manner in which they are integrated into a structured didactic sequence.
Authors such as Hipo et al. (2022) and Murillo et al. (2021) maintain that when skills are
worked on in an articulated manner, each one reinforces the other; in this research, speaking and
writing were addressed in parallel, which may explain why the improvements were not limited to
a single aspect, but were observed in both oral and written expression. Similarly, the idea of skills
integration and contextualized activities is related to what Mahapatra (2024) puts forward, whilst
the subsequent advances in writing and vocabulary are consistent with the results reported by
Sapan and Uzun (2024) in experiences where digital resources are incorporated as a support for
English language learning.
CONCLUSIONS
Vol. 13/ Núm. 2 2026 pág. 1229
Regarding the usefulness of the ELSA Speaking and Writing & Improve applications, it
is concluded that, at the initial stage of the research, these tools did not form a regular part of the
learning process in the classroom. The majority of students indicated that the use of artificial
intelligence applications was infrequent and that immediate feedback on aspects such as
pronunciation and writing was not an established practice. However, it was also evident that a
significant proportion recognized the potential of these technologies to facilitate learning,
particularly in written production. This allows us to affirm that, although implementation was
limited, a favourable perception existed regarding their pedagogical usefulness. The pre-test
results showed that students presented significant difficulties in both oral expression and written
production. The greatest weaknesses were identified in the structuring of sentences when
speaking, correct pronunciation and the appropriate use of punctuation marks. These data
reflected that, prior to the implementation of the didactic process, the group did not perceive itself
as having a solid command of the integrated speaking and writing skills, which justified the need
for a specific pedagogical strategy to strengthen these abilities.
A structured didactic process was designed and implemented that articulately integrated
oral and written production through the use of ELSA Speaking and Writing & Improve. The
progressive planning across four weeks, with activities oriented towards guided practice,
immediate feedback and reflection on learning, allowed for the work to be organised in a manner
consistent with the needs identified in the initial diagnosis. This process not only incorporated
technological tools, but integrated them within an intentional pedagogical sequence, centred on
the active and meaningful development of communicative skills.
The post-test results demonstrated a significant improvement in all assessed skills. The
proportions of favourable responses increased considerably in comparison with the pre-test, and
the statistical analysis using the paired samples t-test confirmed that the difference between both
stages was significant (p < 0.05). This allows us to conclude that the didactic intervention
supported by ELSA Speaking and Writing & Improve had a positive and verifiable impact on the
strengthening of integrated speaking and writing skills. As a future projection, it is recommended
that the research be extended to larger samples and diverse educational contexts, with the aim of
determining whether the observed effects are maintained at other levels of education and over
more extended periods.
Regarding the usefulness of the ELSA Speaking and Writing & Improve applications, it
is concluded that, at the initial stage of the research, these tools did not form a regular part of the
learning process in the classroom. The majority of students indicated that the use of artificial
intelligence applications was infrequent and that immediate feedback on aspects such as
pronunciation and writing was not an established practice. However, it was also evident that a
significant proportion recognized the potential of these technologies to facilitate learning,
particularly in written production. This allows us to affirm that, although implementation was
limited, a favourable perception existed regarding their pedagogical usefulness. The pre-test
results showed that students presented significant difficulties in both oral expression and written
production. The greatest weaknesses were identified in the structuring of sentences when
speaking, correct pronunciation and the appropriate use of punctuation marks. These data
reflected that, prior to the implementation of the didactic process, the group did not perceive itself
as having a solid command of the integrated speaking and writing skills, which justified the need
for a specific pedagogical strategy to strengthen these abilities.
A structured didactic process was designed and implemented that articulately integrated
oral and written production through the use of ELSA Speaking and Writing & Improve. The
progressive planning across four weeks, with activities oriented towards guided practice,
immediate feedback and reflection on learning, allowed for the work to be organised in a manner
consistent with the needs identified in the initial diagnosis. This process not only incorporated
technological tools, but integrated them within an intentional pedagogical sequence, centred on
the active and meaningful development of communicative skills.
The post-test results demonstrated a significant improvement in all assessed skills. The
proportions of favourable responses increased considerably in comparison with the pre-test, and
the statistical analysis using the paired samples t-test confirmed that the difference between both
stages was significant (p < 0.05). This allows us to conclude that the didactic intervention
supported by ELSA Speaking and Writing & Improve had a positive and verifiable impact on the
strengthening of integrated speaking and writing skills. As a future projection, it is recommended
that the research be extended to larger samples and diverse educational contexts, with the aim of
determining whether the observed effects are maintained at other levels of education and over
more extended periods.
Vol. 13/ Núm. 2 2026 pág. 1230
REFERENCES
Al-khresheh, M. H. (2024). The Future of Artificial Intelligence in English Language Teaching:
Pros and Cons of ChatGPT Implementation through a Systematic Review. Language
Teaching Research Quarterly, 43, 54-80. https://eric.ed.gov/?id=EJ1457241
Alqaed, M. A. (2024). AI in english language learning: Saudi Learners´ perspectives and usage.
Advanced Education, 125-142. https://doi.org/10.20535/2410-8286.318972
Anggraini, A. (2022). Improving Students' Pronunciation Skill Using Elsa Speak
Application. Journey, 5(1), 135-141. https://doi.org/10.33503/journey.v5i1.1840
Arias Gonzáles, J. L., & Covinos Gallardo, M. (2021). Diseño y metodología de la investigación.
Enfoques Consulting EIRL, 1(1), 66-78.
Armendáriz, I. A. U., Ramirez, P. L. V., & Alburqueque, C. R. M. (2024). Embracing
Communicative Language Teaching: Fostering Fluency, Accuracy, and Authentic
Communication: Adoptar la enseñanza comunicativa de idiomas: Fomentar la fluidez, la
precisión y la comunicación auténtica. Boletín Científico Ideas y Voces, 4(3), ág-117.
Ayala-Pazmiño, M., & Alvarado-Lucas, K. (2023). Integrating Artificial Intelligence into English
Language Education in Ecuador: A Pathway to Improved Learning Outcomes. Digital
Publisher CEIT, 8(3–1), 679-687.
Bautista-Díaz, M. L., Victoria-Rodríguez, E., Vargas-Estrella, L. B., & Hernández-Chamosa, C.
C. (2020). Pruebas estadísticas paramétricas y no paramétricas: Su clasificación,
objetivos y características. Educación y salud boletín científico Instituto de Ciencias de
la Salud Universidad Autónoma del estado de Hidalgo, 9(17), 78-81.
Castrillo, C. J. H. (2024). Paradigma Positivista. Boletín Científico de las Ciencias Económico
Administrativas del ICEA, 12(24), 29-32. https://doi.org/10.29057/icea.v12i24.12660
Chicaiza, V. A. P., Montaluisa, B. M. C., Cedeño, J. del C. M., Ayala, E. T. M., Teresa, S. V. E.,
& Parraga, A. P. B. (2025). Inteligencia Artificial y Aprendizaje de Idiomas:
Personalización del Aula de Inglés a Través de Plataformas Adaptativas. Revista Veritas
de Difusão Científica, 6(2), 477-506. https://doi.org/10.61616/rvdc.v6i2.643
Cisneros Caicedo, A. J., Guevara García, A. F., Urdánigo Cedeño, J. J., & Garcés Bravo, J. E.
(2022). Técnicas e Instrumentos para la Recolección de Datos que Apoyan a la
Investigación Científica en Tiempo de Pandemia. Dominio de Las Ciencias, 8(1), 1165-
1185. http://dx.doi.org/10.23857/dc.v8i41.2546
Fitria, T. N. (2021). The use Technology based on Artificial Intelligence in English Teaching and
Learning. ELT Echo : The Journal of English Language Teaching in Foreign Language
Context, 6(2). https://doi.org/10.24235/eltecho.v6i2.9299
Galarza, C. A. R. (2020). Los alcances de una investigación. CienciAmérica: Revista de
divulgación científica de la Universidad Tecnológica Indoamérica, 9(3), 1-6.
REFERENCES
Al-khresheh, M. H. (2024). The Future of Artificial Intelligence in English Language Teaching:
Pros and Cons of ChatGPT Implementation through a Systematic Review. Language
Teaching Research Quarterly, 43, 54-80. https://eric.ed.gov/?id=EJ1457241
Alqaed, M. A. (2024). AI in english language learning: Saudi Learners´ perspectives and usage.
Advanced Education, 125-142. https://doi.org/10.20535/2410-8286.318972
Anggraini, A. (2022). Improving Students' Pronunciation Skill Using Elsa Speak
Application. Journey, 5(1), 135-141. https://doi.org/10.33503/journey.v5i1.1840
Arias Gonzáles, J. L., & Covinos Gallardo, M. (2021). Diseño y metodología de la investigación.
Enfoques Consulting EIRL, 1(1), 66-78.
Armendáriz, I. A. U., Ramirez, P. L. V., & Alburqueque, C. R. M. (2024). Embracing
Communicative Language Teaching: Fostering Fluency, Accuracy, and Authentic
Communication: Adoptar la enseñanza comunicativa de idiomas: Fomentar la fluidez, la
precisión y la comunicación auténtica. Boletín Científico Ideas y Voces, 4(3), ág-117.
Ayala-Pazmiño, M., & Alvarado-Lucas, K. (2023). Integrating Artificial Intelligence into English
Language Education in Ecuador: A Pathway to Improved Learning Outcomes. Digital
Publisher CEIT, 8(3–1), 679-687.
Bautista-Díaz, M. L., Victoria-Rodríguez, E., Vargas-Estrella, L. B., & Hernández-Chamosa, C.
C. (2020). Pruebas estadísticas paramétricas y no paramétricas: Su clasificación,
objetivos y características. Educación y salud boletín científico Instituto de Ciencias de
la Salud Universidad Autónoma del estado de Hidalgo, 9(17), 78-81.
Castrillo, C. J. H. (2024). Paradigma Positivista. Boletín Científico de las Ciencias Económico
Administrativas del ICEA, 12(24), 29-32. https://doi.org/10.29057/icea.v12i24.12660
Chicaiza, V. A. P., Montaluisa, B. M. C., Cedeño, J. del C. M., Ayala, E. T. M., Teresa, S. V. E.,
& Parraga, A. P. B. (2025). Inteligencia Artificial y Aprendizaje de Idiomas:
Personalización del Aula de Inglés a Través de Plataformas Adaptativas. Revista Veritas
de Difusão Científica, 6(2), 477-506. https://doi.org/10.61616/rvdc.v6i2.643
Cisneros Caicedo, A. J., Guevara García, A. F., Urdánigo Cedeño, J. J., & Garcés Bravo, J. E.
(2022). Técnicas e Instrumentos para la Recolección de Datos que Apoyan a la
Investigación Científica en Tiempo de Pandemia. Dominio de Las Ciencias, 8(1), 1165-
1185. http://dx.doi.org/10.23857/dc.v8i41.2546
Fitria, T. N. (2021). The use Technology based on Artificial Intelligence in English Teaching and
Learning. ELT Echo : The Journal of English Language Teaching in Foreign Language
Context, 6(2). https://doi.org/10.24235/eltecho.v6i2.9299
Galarza, C. A. R. (2020). Los alcances de una investigación. CienciAmérica: Revista de
divulgación científica de la Universidad Tecnológica Indoamérica, 9(3), 1-6.
Vol. 13/ Núm. 2 2026 pág. 1231
González, O. H. (2021). Aproximación a los distintos tipos de muestreo no probabilístico que
existen. Revista Cubana de Medicina General Integral, 37(3).
Guzmán Sánchez, X., & Esquivel Rivero, Y. (2025). Uso de herramientas digitales innovadoras
para optimizar la enseñanza y el aprendizaje del inglés en bachillerato. 593 Digital
Publisher CEIT, 10(3), 196-203. https://doi.org/10.33386/593dp.2025.3.3133
Hernández-Sampieri, R., & Mendoza, C. (2020). Metodología de la investigación: Las rutas
cuantitativa, cualitativa y mixta.
Hipo, L. A. Q., Costales, S. N. C., Llerena, K. Y. C., & Guzmán, D. C. C. (2022). Integrated Skills
Approach improves the communicative competence in EFL students. Polo del
Conocimiento: Revista científico-profesional, 7(7), 1189-1213. http://10.23857/pc.v7i7
Huang, H.-L., Hwang, G.-J., & Chen, P.-Y. (2022). An integrated concept mapping and image
recognition approach to improving students’ scientific inquiry course performance.
British Journal of Educational Technology, 53(3), 706-727.
https://doi.org/10.1111/bjet.13177
Kolegova, I. A., & Levina, I. A. (2024). Using artificial intelligence as a digital tool in foreign
language teaching. https://10.14529/ped240110
Li, Y. (2022). Teaching mode of oral English in the age of artificial intelligence. Frontiers in
Psychology, 13. https://doi.org/10.3389/fpsyg.2022.953482
López Minotta, K. L., Chiappe, A., & Mella Norambuena, J. (2025). Implementation of artificial
intelligence to improve english oral expression.
Mahapatra, S. (2024). Impact of ChatGPT on ESL students’ academic writing skills: A mixed
methods intervention study. Smart Learning Environments, 11(1), 9.
https://doi.org/10.1186/s40561-024-00295-9
Marghany, M. M. (2023). Using artificial intelligence-based instruction to develop EFL higher
education students’ essay writing skills. CDELT Occasional Papers in the Development
of English Education, 82(1), 219-240. https://doi.org/10.21608/opde.2023.313623
Mayorga Román, M. G., & Tibán Huilca, S. F. (2024). Impacto de una estrategia contextualizada
en la enseñanza de unidades de masa y volumen. Educación química, 35(3), 86-101.
https://doi.org/10.22201/fq.18708404e.2024.3.87683
Murillo, M. G. E., Murillo, R. de los Á. B., Valle, V. V. Y., & Sangucho, N. F. T. (2021). La
tecnología como herramienta combinanda para la enseñanza del inglés. Polo del
Conocimiento: Revista científico-profesional, 6(9), 1270-1284.
http://10.23857/pc.v6i9.3109
Puente, R. T. (2020). El método de encuesta. Los métodos de investigación para la elaboración
de las tesis de maestría en educación, 51-60.
Ramírez, E. R. S. (2021). Lo cuantitativo y lo cualitativo en la investigación científica
educacional. Una visión desde la filosofía: The quantitative and the qualitative in
González, O. H. (2021). Aproximación a los distintos tipos de muestreo no probabilístico que
existen. Revista Cubana de Medicina General Integral, 37(3).
Guzmán Sánchez, X., & Esquivel Rivero, Y. (2025). Uso de herramientas digitales innovadoras
para optimizar la enseñanza y el aprendizaje del inglés en bachillerato. 593 Digital
Publisher CEIT, 10(3), 196-203. https://doi.org/10.33386/593dp.2025.3.3133
Hernández-Sampieri, R., & Mendoza, C. (2020). Metodología de la investigación: Las rutas
cuantitativa, cualitativa y mixta.
Hipo, L. A. Q., Costales, S. N. C., Llerena, K. Y. C., & Guzmán, D. C. C. (2022). Integrated Skills
Approach improves the communicative competence in EFL students. Polo del
Conocimiento: Revista científico-profesional, 7(7), 1189-1213. http://10.23857/pc.v7i7
Huang, H.-L., Hwang, G.-J., & Chen, P.-Y. (2022). An integrated concept mapping and image
recognition approach to improving students’ scientific inquiry course performance.
British Journal of Educational Technology, 53(3), 706-727.
https://doi.org/10.1111/bjet.13177
Kolegova, I. A., & Levina, I. A. (2024). Using artificial intelligence as a digital tool in foreign
language teaching. https://10.14529/ped240110
Li, Y. (2022). Teaching mode of oral English in the age of artificial intelligence. Frontiers in
Psychology, 13. https://doi.org/10.3389/fpsyg.2022.953482
López Minotta, K. L., Chiappe, A., & Mella Norambuena, J. (2025). Implementation of artificial
intelligence to improve english oral expression.
Mahapatra, S. (2024). Impact of ChatGPT on ESL students’ academic writing skills: A mixed
methods intervention study. Smart Learning Environments, 11(1), 9.
https://doi.org/10.1186/s40561-024-00295-9
Marghany, M. M. (2023). Using artificial intelligence-based instruction to develop EFL higher
education students’ essay writing skills. CDELT Occasional Papers in the Development
of English Education, 82(1), 219-240. https://doi.org/10.21608/opde.2023.313623
Mayorga Román, M. G., & Tibán Huilca, S. F. (2024). Impacto de una estrategia contextualizada
en la enseñanza de unidades de masa y volumen. Educación química, 35(3), 86-101.
https://doi.org/10.22201/fq.18708404e.2024.3.87683
Murillo, M. G. E., Murillo, R. de los Á. B., Valle, V. V. Y., & Sangucho, N. F. T. (2021). La
tecnología como herramienta combinanda para la enseñanza del inglés. Polo del
Conocimiento: Revista científico-profesional, 6(9), 1270-1284.
http://10.23857/pc.v6i9.3109
Puente, R. T. (2020). El método de encuesta. Los métodos de investigación para la elaboración
de las tesis de maestría en educación, 51-60.
Ramírez, E. R. S. (2021). Lo cuantitativo y lo cualitativo en la investigación científica
educacional. Una visión desde la filosofía: The quantitative and the qualitative in
Vol. 13/ Núm. 2 2026 pág. 1232
educational scientific research. A view from philosophy. Revista Iberoamericana de
Investigación en Educación, 1(1), 100-107.
Román, M. G. M., Paz, K. P. V., Bonifaz, C. A. C., & Román, M. A. U. (2025). Inclusión
educativa como garantía del derecho a una educación básica de calidad: Educational
inclusion as a guarantee of the right to quality basic education. LATAM Revista
Latinoamericana de Ciencias Sociales y Humanidades, 6(3), 864-879.
https://doi.org/10.56712/latam.v6i3.3995
Sanabria Navarro, J. R., Silveira Pérez, Y., Pérez Bravo, D. D., & Cortina Núñez, M. de J. (2023).
Incidencias de la inteligencia artificial en la educación contemporánea. Comunicar:
Revista Científica de Comunicación y Educación, (77), 97-107.
https://doi.org/10.3916/C77-2023-08
Sánchez-Solis, Y., Raqui-Ramirez, C. E., Huaroc-Ponce, E. J., & Huaroc-Ponce, N. M. (2024).
Importancia de Conocer la Normalidad de los Datos Utilizados en los Trabajos de
Investigación por Tesistas. Revista Tecnológica-Educativa Docentes 2.0, 17(2), 404-413.
https://doi.org/10.37843/rted.v17i2.554
Sapan, M., & Uzun, L. (2024). The Effect of ChatGPT-Integrated English Teaching on High
School EFL Learners’ Writing Skills and Vocabulary Development. International
Journal of Education in Mathematics, Science and Technology, 12(6), 1657-1677.
https://doi.org/10.46328/ijemst.4655
Sari, N. (2023). The role of artificial intelligence (AI) in developing English language learner’s
communication skills. Journal on Education, 6(01), 750-757.
http://jonedu.org/index.php/joe
Schmidt, T., & Strassner, T. (2022). Artificial Intelligence in Foreign Language Learning and
Teaching. Anglistik, 33(1), 165-184. https://doi.org/10.33675/ANGL/2022/1/14
Sehlaoui, F. Z. (2024). Integrating AI in Foreign Language Teaching and Learning: Learner
Autonomy and Tool Utilization in an Algerian University. Passerelle, 13(2), 116-139.
https://asjp.cerist.dz/en/article/259721
Syuhra, M. N., Chandra, N. E., & Rosalina, E. (2025). Artificial Intelligence in English Language
Teaching: A Systematic Literature Review of Tools, Impact, and Challenges. Voices of
English Language Education Society, 9(1), 193-205.
https://doi.org/10.29408/veles.v9i1.29745
Xiaofan, W., & Annamalai, N. (2025). Investigating the Use of AI Tools in English Language
Learning: A Phenomenological Approach. Contemporary Educational Technology,
17(2). https://eric.ed.gov/?id=EJ1470007
educational scientific research. A view from philosophy. Revista Iberoamericana de
Investigación en Educación, 1(1), 100-107.
Román, M. G. M., Paz, K. P. V., Bonifaz, C. A. C., & Román, M. A. U. (2025). Inclusión
educativa como garantía del derecho a una educación básica de calidad: Educational
inclusion as a guarantee of the right to quality basic education. LATAM Revista
Latinoamericana de Ciencias Sociales y Humanidades, 6(3), 864-879.
https://doi.org/10.56712/latam.v6i3.3995
Sanabria Navarro, J. R., Silveira Pérez, Y., Pérez Bravo, D. D., & Cortina Núñez, M. de J. (2023).
Incidencias de la inteligencia artificial en la educación contemporánea. Comunicar:
Revista Científica de Comunicación y Educación, (77), 97-107.
https://doi.org/10.3916/C77-2023-08
Sánchez-Solis, Y., Raqui-Ramirez, C. E., Huaroc-Ponce, E. J., & Huaroc-Ponce, N. M. (2024).
Importancia de Conocer la Normalidad de los Datos Utilizados en los Trabajos de
Investigación por Tesistas. Revista Tecnológica-Educativa Docentes 2.0, 17(2), 404-413.
https://doi.org/10.37843/rted.v17i2.554
Sapan, M., & Uzun, L. (2024). The Effect of ChatGPT-Integrated English Teaching on High
School EFL Learners’ Writing Skills and Vocabulary Development. International
Journal of Education in Mathematics, Science and Technology, 12(6), 1657-1677.
https://doi.org/10.46328/ijemst.4655
Sari, N. (2023). The role of artificial intelligence (AI) in developing English language learner’s
communication skills. Journal on Education, 6(01), 750-757.
http://jonedu.org/index.php/joe
Schmidt, T., & Strassner, T. (2022). Artificial Intelligence in Foreign Language Learning and
Teaching. Anglistik, 33(1), 165-184. https://doi.org/10.33675/ANGL/2022/1/14
Sehlaoui, F. Z. (2024). Integrating AI in Foreign Language Teaching and Learning: Learner
Autonomy and Tool Utilization in an Algerian University. Passerelle, 13(2), 116-139.
https://asjp.cerist.dz/en/article/259721
Syuhra, M. N., Chandra, N. E., & Rosalina, E. (2025). Artificial Intelligence in English Language
Teaching: A Systematic Literature Review of Tools, Impact, and Challenges. Voices of
English Language Education Society, 9(1), 193-205.
https://doi.org/10.29408/veles.v9i1.29745
Xiaofan, W., & Annamalai, N. (2025). Investigating the Use of AI Tools in English Language
Learning: A Phenomenological Approach. Contemporary Educational Technology,
17(2). https://eric.ed.gov/?id=EJ1470007