ISSN 2953-6367  
http://revistainvestigo.com  
Octubre 2026  
Vol. 7 No. 21, PP. 31-49  
https://doi.org/10.56519/y1rf8q35  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL  
INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO RURAL  
FRONTERIZO DEL CARCHI, ECUADOR  
DIGITAL PLATFORMS AND INTEGRATED ENGLISH SKILLS: A  
CENSUS DIAGNOSIS IN A RURAL BORDER SETTING IN CARCHI,  
ECUADOR  
Karen Lizbeth Ayala Caratar1, Elena Valeria Flores Borja2  
{karen.ayala@upec.edu.ec¹ elena.flores@upec.edu.ec2}  
Fecha de recepción: 01/07/2026  
/ Fecha de aceptación: 21/08/2026  
/ Fecha de publicación: 14/10/2026  
RESUMEN: La integración de plataformas digitales en la enseñanza del inglés se ha  
expandido en América Latina, pero la evidencia procede sobre todo de intervenciones urbanas  
y describe poco el punto de partida de las instituciones rurales. El estudio caracteriza la  
frecuencia de uso pedagógico de plataformas y el desempeño auto percibido en las cuatro  
destrezas integradas del inglés en una unidad educativa ubicada en la parroquia Chical del  
cantón Tulcán, Carchi, Ecuador. Se adoptó un enfoque cuantitativo, no experimental,  
transversal y descriptivo. Se censó a los 152 estudiantes de octavo, noveno y décimo año de  
Educación General Básica Superior durante el año lectivo 2025-2026, con un cuestionario de  
veintidós ítems en escala ordinal de cuatro puntos validado por tres expertos. Las distribuciones  
se resumieron mediante media ponderada, mediana, moda e índice normalizado de intensidad  
de cero a cien puntos, agregado por destreza. La mediana fue a veces en los nueve ítems de uso  
y la moda coincidió en ocho; el índice global de uso alcanzó 43,9 puntos, bajo el punto medio  
de la escala, y los dos ítems de utilidad percibida promediaron 66,8 puntos. Ambos valores  
proceden de tipos de pregunta distintos y se presentan como contraste descriptivo, sin calcular  
diferencias. La reproducción de audio mediante plataformas registró el valor más bajo del  
estudio y la fluidez oral, el desempeño auto percibido más bajo. La preferencia se concentró en  
una única aplicación generalista en el 69,1 % de los casos. Los hallazgos sitúan el problema en  
la mediación didáctica y no en la disponibilidad tecnológica.  
Palabras clave: Tecnología educativa; enseñanza de lenguas; inglés; educación rural;  
aprendizaje en línea.  
1Unidad Educativa Ecuador, Ministerio de Educación del Ecuador; https://orcid.org/0009-0006-1948-4705; +593 996455295.  
2Universidad Politécnica Estatal del Carchi, Tulcán - Ecuador; https://orcid.org/0009-0005-4800-8938; +593 992461397.  
31  
Revista Científica Multidisciplinaria InvestiGo  
Riobamba – Ecuador  
Cel: +593 97 911 9620  
revisinvestigo@gmail.com  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
ABSTRACT: The integration of digital platforms into English language teaching has expanded  
rapidly in Latin America, yet the available evidence comes mostly from urban interventions and  
says little about the starting point of rural institutions. This study characterizes the frequency  
of pedagogical use of digital platforms and self-perceived performance in the four integrated  
English skills at a rural border school located in the Chical parish in the canton of Tulcán,  
province of Carchi, Ecuador. A quantitative, non-experimental, cross-sectional and descriptive  
approach was adopted. A census was administered to all 152 students of the eighth, ninth and  
tenth years of Upper Basic General Education during the 2025-2026 school year, using a twenty-  
two-item questionnaire with a four-point ordinal scale validated by three expert judges.  
Distributions were summarized through a weighted mean, the median, the mode and a  
normalized intensity index from zero to one hundred points, aggregated by skill. The median  
fell in the category sometimes for all nine use items and the mode coincided in eight of them;  
the global use index reached 43.9 points, below the midpoint of the scale, whereas the two  
perceived-usefulness items averaged 66.8 points. Both figures come from different question  
types and are presented as a descriptive contrast, without computing differences. Audio  
playback through platforms recorded the lowest value in the study and oral fluency with the  
lowest self-perceived performance. Preference concentrated on a single generalist application  
in 69.1 % of cases. The findings locate the problem in didactic mediation rather than in  
technological availability.  
Keywords: Educational technology; language teaching; English; rural education; online learning  
INTRODUCTION  
English has become an enabling competence for accessing scientific information, academic  
mobility and entering the labor market. In Ecuador, that competence is shaped in a system of  
education that has significant territorial disparities, and whose learning outcomes are more  
related to the content and pedagogical practices of the school where learning is carried out than  
to the curriculum itself. The most limiting constraints are faced by rural and border institutions:  
limited equipment, poor connectivity, and teacher training on pedagogical technology use that is  
typically limited to one-off workshops. Before proposing any technological innovation, it is crucial  
to understand that point because a strategy developed based on urban assumptions seldom  
survives the first encounter with a rural classroom.  
The size of the structural constraint can be measured. The latest official data on information and  
communication technologies (ICTs) indicates that 66.0 % of Ecuadorian households have access  
to the Internet, 73.6 % of which are in urban areas and 48.1 % in rural areas (1). The gap of 25.5  
percentage points between the two territories (Figure 1) is not contextual detail, but a boundary  
condition that will affect the viability of any proposal that is supported by available online  
resources, and especially autonomous practice outside school hours. Studies in rural areas across  
the country have found that the restriction doesn't stop connectivity, and that no specific teacher  
training is a second constraint when devices are present (2,3).  
32  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
Figure 1. Internet access in Ecuadorian households by area, 2024.  
Source: (1).  
Note: the gap between urban and rural areas amounts to 25.5 percentage points.  
There is a consistent body of specialized literature that provides evidence of the potential of  
digital platforms for the development of specific skills. Research on language-learning  
applications based on spaced repetition and immediate feedback have found improvements in  
receptive knowledge and in productive skills in the case of continued practice (4,5) but also note  
that improvements in oral production are far less than in comprehension (6) and that the gamer  
aspect alone is not enough to ensure learning (7). For written production, systematic reviews of  
the use of automated written feedback indicate that formative use, with teacher supervision, of  
the system results in gains in grammatical and spelling accuracy, but that summative use, with or  
without teacher supervision, does not (8-10). In the field of oral production, the use of automatic  
speech recognition has led to positive results in terms of pronunciation, intonation and fluency,  
as well as high levels of user satisfaction (11-15); however, recent studies have emphasized that  
the success of these applications relies on teacher mediation, as learners require assistance in  
understanding the feedback and in avoiding instrumental dependence (16-19).  
That evidence is all linked to the curricular integration of the tool, and that is the common  
denominator. The findings of the studies on Latin American platforms are similar: platforms boost  
motivation and participation, but only when they are used regularly, and with clear learning goals  
(20-22). Studies of learner perceptions of mobile-assisted language learning converge on the  
same point: a favorable view of the resource does not by itself translate into systematic classroom  
use (23). In the Ecuadorian context, this qualification is critical because the use of technology is  
often reported as availability of resources, and not as systematic pedagogical practice (24). This  
collection of work provides a working hypothesis that skills that are practiced more on the  
platform will be perceived as being performed better, and that the ordering of skills by intensity  
of use will be parallel to the ordering of skills by perceived performance. The expectation is  
descriptive, linear and monotonic, but not statistically tested here; it is simply stated as the  
pattern of ordering against which the observed ordering is read. Explicitness is important because  
any deviation from this parallel ordering (especially a reversal at the extremes) would suggest  
33  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
that exposure is not sufficient to explain perceived competence, and would shift focus from the  
amount of technological mediation to the quality of it. However, there is a lack of evidence. Most  
published studies are based on an intervention design, and the results are reported for one  
isolated skill in specific groups, with few reporting the baseline of an entire institution, and almost  
none in rural border schools in northern Ecuador. This lack has an impact on practice: When  
developing improvement suggestions, the actual usage of platforms is not considered, as is the  
lack of skills and the relationship between these and what students think they perform. If no such  
diagnosis is made, the digital divide is perceived as an access issue as opposed to a mediation  
issue. The general aim of this study is to describe the frequency of pedagogical use of digital  
platforms and the self-perceived performance in the integrated English skills of Upper Basic  
General Education students at a rural border school in the canton of Tulcán, and to compare the  
ordering of skills that results from each description. Three specific aims follow: (i) to establish how  
often digital platforms are used for each skill; (ii) to describe self-perceived performance in  
listening comprehension, oral production, reading comprehension and written production; and  
(iii) to identify declared platform preferences and their degree of concentration. The question of  
organizing the analysis is whether the ordering of skills by declared intensity of use coincides with  
the ordering by self-perceived performance; the answer indicates whether the intervention  
priority is technological provision or didactic mediation.  
MATERIALS AND METHODS  
Research design  
The study used a quantitative, non-experimental, cross-sectional and descriptive design: no  
variables were manipulated, and the aim was to describe one educational situation at a single  
point in the 2025–2026 school year, establishing a baseline for later interventions.  
Setting and study area  
The study was carried out in the public school Unidad Educativa Ecuador, located in the parish of  
El Chical, in the canton of Tulcán, province of Carchi (Zone 1, Sierra calendar, morning shift). The  
parish is rural and is located next to Colombia, and can only be accessed by road, which is  
approximately 110 km from the cantonal capital (Figure 2). That location affects the continuity of  
connectivity and replacement of equipment and is the key to making the case different from the  
urban settings typically studied; it also makes the school community small enough that individual  
response records would be at risk of re-identification.  
34  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
Figure 2. Location of the study area.  
Source: prepared by the author.  
Note: schematic representation; distances are not to scale.  
Population  
The target population consisted of all students in the 8th, 9th and 10th year of Upper Basic  
General Education. A census (as opposed to a sample) was conducted because the number of  
cases was manageable; all 152 students at those three levels participated. The students in this  
stage were chosen because it is where the shift towards functional uses of the language and the  
four skills consolidation are expected.  
Instrument  
A structured 22-item questionnaire was used to collect data, which was administered face-to-  
face and anonymously. Twenty-one items were rated on a four-point ordinal scale (always, almost  
always, sometimes, never) and the last item was a single-choice question with four alternatives  
on platform preference. The instrument consists of four blocks of items, which are presented with  
the skills in Table 1, with the two perceived-usefulness items in between the use items.  
The  
content validity is based on the judgment of three experts in the pedagogy of national and foreign  
languages who gave ratings on the relevance, clarity and sufficiency of each item before  
administration. Therefore, the only validity evidence for the instrument is expert judgement.  
There was no individual record kept and questionnaires were counted directly into a frequency  
matrix by item and response category, so that the anonymity of the participants is guaranteed in  
this school of a small size, but no participant-level data were retained. Because reliability  
coefficients that depend on a case-by-item matrix are not estimable (Cronbach's alpha,  
McDonald's omega, or an ordinal alpha on polychoric correlations), no reliability coefficient is  
reported anywhere in this article. The block indices are therefore interpreted as descriptive  
summaries of the observed distributions and the Limitations section is presented as an explicit  
opportunity for future applications. Block A is a student-reported proxy for classroom practice,  
35  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
not an actual observation of classroom practice. Triangulation with teacher reports and classroom  
observation was not done.  
Table 1. Composition of the instrument and correspondence with the integrated skills.  
Block  
A
Items  
9
Construct measured  
Frequency of pedagogical use of  
digital platforms  
Perceived usefulness of digital  
resources  
Associated skills  
All four skills plus two  
cross-cutting items  
Crosscutting  
Scale  
4-point ordinal  
B
C
2
4-point ordinal  
4-point ordinal  
10  
Self-perceived language  
performance  
Written production (2),  
oral production (4),  
reading comprehension  
(2), listening  
comprehension (2)  
D
1
Declared platform preference  
Crosscutting  
Single choice (4  
options)  
Source: prepared by the author.  
Data processing and analysis  
There were two levels of processing. Absolute and relative frequencies were first calculated for  
each response category and the median and mode of each item were also calculated. The ordinal  
distributions were summarized using a weighted mean at the second, with weights of four, three,  
two and one for the categories always, almost always, sometimes and never, respectively, as per  
equation (1). This transformation assumes equal spacing between adjacent response categories,  
which is a common assumption that cannot be tested with ordinal data; therefore, the media and  
the mode are also reported with each index and are used to interpret the index whenever the  
two values differ:  
M = (Σ wᵢ · nᵢ) / N  
with wᵢ ∈ {4, 3, 2, 1}  
(1)  
where nᵢ is the frequency observed in category i, wᵢ the weight assigned to that category and N  
the total number of cases. To allow comparison across items and across skills, the resulting  
meaning was transformed into a normalized intensity index (NII) ranging from zero to one  
hundred points, following equation (2):  
NII = [ (M − 1) / 3] × 100  
(2)  
Fifty points is the same as the middle of the scale. The individual indices were then averaged by  
skill, one for the pedagogical use index (based on the items in block A) and the other for the self-  
perceived performance index (based on the items in block C). The two indices reflect different  
constructs, one a report of the frequency of a classroom event and the other a self-assessment  
36  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
of ability and therefore are reported together on a common normalized metric to be able to order  
the skills, but without computing arithmetic differences between them. The number of items  
averaged (k) is reported in each table and varies by skill. Two use indices are based on one item  
each: listening comprehension (audio playback through platforms) and reading comprehension  
(readings on digital platforms). A single-item index has no internal redundancy, has no  
consistency check, and is completely specified by the wording of that one item; both are therefore  
treated as indicative values, and every claim about these two skills is qualified accordingly in the  
Results and the Discussion. No confidence intervals were calculated, and no significance tests  
were used throughout the analysis, since the study is based on the entire population, and not a  
sample. Spreadsheets were used for computations, as well as open-source statistical analysis  
routines.  
Ethical considerations  
The principles of autonomy, beneficence, non-maleficence and justice were observed during the  
research. The school authorities and the legal guardians of the participants gave their institutional  
authorization and their informed consent, and the students themselves gave verbal assent, in  
which the purpose of the study, the voluntary nature of participation and the right to withdraw  
at any time without any consequences were explained. The names and identifying data were not  
recorded: the information was coded and is presented in an aggregate form. The information  
gathered was not shared with anyone for any purpose other than what stated.  
RESULTS  
There were no missing values, and no imputation procedures were required as all 152 students  
who formed the population of interest responded to all items. The results are presented in four  
parts, following the specific objectives stated above: Pedagogical use of platforms (frequency),  
Self-perceived performance in the integrated skills (frequency), Overall comparison of the two  
dimensions (frequency), and stated tool preference (frequency).  
Frequency of pedagogical use of digital platforms  
The first block of results reflects the frequency with which students feel digital platforms are used  
in classroom activities. The distribution of responses, summary statistics and the normalized index  
for each item are shown in Table 2, which also reports, in its final section, the two perceived-  
usefulness items of block B.  
37  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
Table 2. Frequency of pedagogical use of digital platforms and perceived usefulness of digital resources (N = 152).  
Item  
Always  
Almost  
always  
27  
Sometimes  
Never  
M
SD  
NII  
Mdn / Mo  
General use of  
digital platforms  
in class  
Readings on  
digital platforms  
17  
(11.2)  
67 (44.1)  
41  
(27.0)  
2.13  
0.94  
37.7  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/ Never  
(17.8)  
21  
(13.8)  
52  
(34.2)  
54 (35.5)  
66 (43.4)  
47 (30.9)  
58 (38.2)  
51 (33.6)  
53 (34.9)  
49 (32.2)  
53 (34.9)  
25  
(16.4)  
2.45  
2.24  
2.01  
2.50  
2.38  
2.30  
2.33  
2.51  
0.92  
0.82  
1.01  
0.98  
1.03  
1.05  
1.10  
1.04  
48.5  
41.2  
33.8  
50.0  
46.1  
43.4  
44.3  
50.4  
Digital self-  
assessment  
quizzes  
Audio playback  
through  
platforms  
Writing  
exercises on  
platforms  
Assisted  
8 (5.3)  
49  
(32.2)  
29  
(19.1)  
17  
(11.2)  
28  
(18.4)  
60  
(39.5)  
30  
(19.7)  
40  
(26.3)  
24  
(15.8)  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/
27  
(17.8)  
39  
(25.7)  
35  
(23.0)  
grammar  
correction  
Assisted spelling  
correction  
27  
(17.8)  
32  
(21.1)  
40  
(26.3)  
Pronunciation  
exercises on  
platforms  
Assisted  
pronunciation  
correction  
33  
(21.7)  
27  
(17.8)  
43  
(28.3)  
Sometimes  
Sometimes  
/
35  
(23.0)  
36  
(23.7)  
28  
(18.4)  
Sometimes  
Block B. Perceived usefulness of digital resources  
Listening  
comprehension  
exercises  
facilitate  
learning  
Educational  
games facilitate  
learning  
60  
(39.5)  
49  
(32.2)  
40 (26.3)  
48 (31.6)  
3
(2.0)  
3.09  
2.91  
0.85  
0.97  
69.7  
63.8  
Almost  
always /  
Always  
55  
(36.2)  
39  
(25.7)  
10  
(6.6)  
Almost  
always /  
Always  
Source: prepared by the authors.  
Note: each cell shows the absolute frequency and, in brackets, the percentage. The first nine items belong to block  
A and the last two to block B (positions 4 and 5 of the questionnaire). M = weighted mean (range 1–4); SD =  
standard deviation; NII = normalized intensity index (range 0–100); Mdn = median; Mo = mode.  
38  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
Figure 3. Distribution of responses and normalized index by pedagogical use item.  
Source: prepared by the author.  
Note: the dashed line in the right-hand panel marks the midpoint of the scale (50 points).  
The most consistent pattern is that the median falls in the category sometimes for all nine items  
of the block, and the mode coincides with it in eight of the nine; the single exception is audio  
playback through platforms, whose mode falls in the category never. For none of the items the  
percentage of respondents who responded always or almost always (favorable responses) equal  
or surpassed 50 % of the total. The values closest to the midpoint of the scale were assisted  
pronunciation correction (NII = 50.4) and writing exercises on platforms (NII = 50.0); every other  
item fell below it. On the other side of the scale, audio playback via platforms was the least used  
resource in the entire study (NII = 33.8): 39.5 % of students said they never use this resource and  
11.2 % said that this is permanent practice. The general use of digital platforms in class also had  
a low value (NII = 37.7), with 71.1 % of the responses being in the categories sometimes and  
never. It is obvious from Figure 3 that the band of those two categories is dominant for each item.  
The two relevant contrast items of usefulness in block B are different. The listening  
comprehension exercises were considered to always or almost always facilitate learning by 71.7  
% of the students (NII = 69.7), while 61.8 % of the students gave the same value to educational  
games (NII = 63.8). In both items the median falls in almost always and the mode always, in  
contrast with the median of sometimes that characterizes every use item. The perceived  
usefulness of both items is above the mid-point of the scale (66.8 points on average), while the  
nine use items are below the mid-point of the scale (43.9 points on average). The two figures are  
from two different types of questions (agreement with a statement in one, frequency of a  
classroom event in the other) and are not subtracted here. What they describe together is a  
school where the resource is valued but rarely seen, and that contrast is the main outcome of this  
first block.  
39  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
Self-perceived performance in the integrated skills  
The second block collects students' self-assessment of their performance in the four skills. Table  
3 details the distributions and Figure 4 represent them graphically. Again, as in the previous block,  
the distribution of the full responses is reported along with the weighted mean, the median, the  
mode and the normalized index, allowing items to be ranked on a common measure regardless  
of the skill level they are associated with.  
Table 3. Self-perceived performance in the integrated English skills (N = 152).  
Item  
Always  
Almost  
always  
32  
Sometimes  
Never  
M
SD  
NII  
Mdn / Mo  
Correct spelling  
when writing  
28  
(18.4)  
67 (44.1)  
25  
(16.4)  
2.41  
0.97  
47.1  
Sometimes  
/
(21.1)  
Sometimes  
Sometimes  
/
Correct  
24  
(15.8)  
46  
(30.3)  
60 (39.5)  
22  
(14.5)  
2.47  
0.92  
49.1  
sentence  
structure in  
writing  
Sometimes  
Correct  
pronunciation  
20  
(13.2)  
38  
(25.0)  
71 (46.7)  
65 (42.8)  
67 (44.1)  
70 (46.1)  
63 (41.4)  
61 (40.1)  
52 (34.2)  
23  
(15.1)  
2.36  
2.03  
2.39  
2.41  
2.43  
2.42  
2.53  
0.89  
0.90  
0.90  
0.92  
0.94  
0.94  
0.92  
45.4  
34.4  
46.3  
46.9  
47.6  
47.4  
50.9  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Sometimes  
/
Sometimes  
Almost  
always /  
Almost  
always  
Oral fluency  
12  
(7.9)  
28  
(18.4)  
47  
(30.9)  
Sentence  
structure when  
speaking  
20  
(13.2)  
42  
(27.6)  
23  
(15.1)  
Appropriate  
vocabulary  
when speaking  
Understanding  
main ideas in  
reading  
Recognising  
vocabulary in  
reading  
24  
(15.8)  
36  
(23.7)  
22  
(14.5)  
24  
(15.8)  
41  
(27.0)  
24  
(15.8)  
23  
(15.1)  
43  
(28.3)  
25  
(16.4)  
Understanding  
main ideas in  
audio  
24  
(15.8)  
54  
(35.5)  
22  
(14.5)  
Understanding  
vocabulary in  
audio  
26  
(17.1)  
48  
(31.6)  
57 (37.5)  
21  
(13.8)  
2.52  
0.93  
50.7  
Sometimes  
/
Sometimes  
Source: prepared by the author.  
Note: the first two items correspond to written production; the next four, to oral production; the following two,  
to reading comprehension; and the last two, to listening comprehension. M = weighted mean; SD = standard  
deviation; NII = normalized intensity index; Mdn = median; Mo = mode.  
40  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
Figure 4. Distribution of responses and normalized index by self-perceived performance item.  
Source: prepared by the author.  
Note: the dashed line in the right-hand panel marks the midpoint of the scale (50 points).  
The mean of the ten items was 46.6 points, which is near the mid-point of the scale. The category  
sometimes was modal in nine of the ten items, with percentages ranging from 34.2% to 46.7%; in  
the remaining item the mode fell almost always. The lowest self-perceived performance was for  
oral fluency (NII = 34.4), with nearly one third of students (30.9 %) not meeting the criteria and  
only 7.9 % always meeting the criterion. Correct pronunciation (NII = 45.4), sentence structure  
when speaking (NII = 46.3) and the use of appropriate vocabulary in oral expression (NII = 46.9)  
were followed. The four items with the lowest scores are all related to oral production.  
The listening comprehension was measured by the highest values: understanding the main ideas  
in audio material reached a maximum index (NII = 50.9) and was the only item in the whole  
instrument whose median fell in almost always. The indices of reading comprehension and  
written production were intermediate (47.1-49.1 points) and were quite stable, with narrow  
dispersion across the items.  
The other two skills, however, have use indices in block A that are based on only one item (audio  
playback through platforms and readings on digital platforms), whereas the performance side of  
the comparison has two items per skill, and thus more redundancy in the measurement of  
exposure.  
Aggregate comparison between use and performance  
Presenting the two indices by skill allows the profile of platform use to be read against the profile  
of self-assessment. Both are shown in Table 4, along with the average number of items in each  
case, and both are shown in Figure 5. It is important to note that the two indices are not  
subtracted from each other, but are both normalized to a common scale to enable the ordering  
of the four skills.  
41  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
It is also not a symmetrical comparison in terms of the number of items measured for each skill;  
the use indices for listening and reading comprehension are each based on one item of block A,  
while the other use indices and all performance indices are based on two or more items. The  
position on the use axis for a single-item index is not a stable estimate as there is no internal  
consistency check and it is only a reflection of the wording of the item. Table 4 shows k for each  
index for which it is desired to weigh.  
Table 4. Pedagogical use index and self-perceived performance index by skill.  
Skill  
Listening  
comprehension  
Oral production  
k
use  
1
UI  
Basis of the use index  
k perf.  
PI  
33.  
8
47.  
4
48.  
5
46.  
5
single item: audio playback through  
platforms  
2
4
2
2
50.  
8
43.  
3
47.  
5
48.  
1
2
1
3
2 items averaged  
Reading  
comprehension  
Written  
single item: readings on digital  
platforms  
3 items on averaged  
production  
Source: prepared by the author.  
Note: k = number of items averaged; UI = pedagogical use index; PI = self-perceived performance index. Both range  
from 0 to 100 points. The two indices measure different constructs and are not subtracted from one another; they  
share a normalized metric only so that the ordering of skills can be compared. The use indices for listening and  
reading rest on a single item each.  
Figure 5. Pedagogical use index and self-perceived performance index, by skill.  
Source: prepared by the author.  
Note: the two indices measure different constructs and share a normalized metric only to allow the ordering of  
skills to be compared.  
The two orderings do not coincide. Listening comprehension is the least used and the most self-  
perceived in terms of performance while oral production is the second most used and the least  
42  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
self-perceived in terms of performance. The values of reading comprehension and written  
production fall in the middle of both orders and the difference between them is small. The  
difference is thus skewed to the extremes of the ranking and not distributed evenly.  
There are two caveats to this comparison. First, the listening comprehension use index is based  
on a single item (playback via platforms) and the reading comprehension use index is based on a  
single item (readings on digital platforms), which means the two use indices have no internal  
redundancy; if the items were reworded or replaced, both use indices would change significantly.  
Second, the assignment of items to skills was done conceptually and was not tested against the  
data of the responses. It is important to note that the reversal at the extremes of the two  
orderings should be interpreted as a pattern that should be investigated rather than a  
relationship.  
Declared platform preference  
The final item asked which platform students consider most appropriate for learning English.  
Table 5 presents the distribution, together with the frequency that an even split across the four  
options would produce; Figure 6 represents it graphically.  
Table 5. Declared platform preference (N = 152).  
Platform  
f
%
f under an  
even split  
Tool orientation  
Duolingo  
105  
17  
69.1  
11.2  
10.5  
9.2  
38  
38  
Generalist (four skills)  
Reading comprehension  
Oral production  
Written production  
—
ReadTheory  
ELSA Speak  
Write & Improve  
Total  
16  
38  
14  
38  
152  
100.0  
152  
Source: prepared by the author.  
Note: the fourth column gives the frequency that an even split across the four alternatives would produce, reported  
as a descriptive reference. Since the study covers the whole population, no significance test is applied.  
43  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
Figure 6. Declared preference of digital platforms for learning English.  
Source: prepared by the author.  
Note: the dashed line marks the 25 % corresponding to an equiprobable distribution.  
Preference focused mainly on one application: 105 students (69.1 % of total) chose Duolingo. The  
remaining three tools shared the remaining 30.9 % in similar amounts: ReadTheory (11.2 %), ELSA  
Speak (10.5 %) and Write & Improve (9.2 %). If the four options were split evenly, there would be  
38 students on each; the highest option has nearly three times this reference value, and none of  
the other three options has more than half of this value. This distribution is the population  
distribution as it is a study of the entire population and there is no test of significance to be  
performed or reported.  
It is important to note that the three minority options are specialist applications, geared towards  
reading comprehension, oral production and written production, respectively, while the majority  
option is a generalist application. Declared preference is more toward the tool that is most visible  
to the public and not toward the tools that are specifically designed to address the skills that  
students say they struggle with most.  
Overall, the four sections tell a unified story. The pedagogical use of platforms is lower than the  
mid-point of the scale in all items assessed, the usefulness is higher than the mid-point of the  
scale in all items assessed, and the self-perceived performance is around the mid-point of the  
scale, with oral production being clearly lower. Preference is for a generalist platform rather than  
specialized ones. The four results are not dependent on any single item as the pattern recycles  
regularly for each section of the instrument.  
DISCUSSION  
These data suggest that the problem described in this school is not a technological availability  
problem but a problem of didactic systematicity. The most convincing evidence in that regard is  
that the median falls in the category sometimes across all nine use items and the mode does so  
in eight of them: it is not that some use items are used and others are not, but that none of the  
44  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
pedagogical functions of the platforms reaches a frequency that would allow them to build up  
routines of practice. This reading is consistent with the findings of research in Latin America,  
which has found that the sporadic incorporation of technology is not related to explicit learning  
goals (20-22) and with studies of rural Ecuadorian teachers, where pedagogical training in  
technology is found as a more severe constraint than equipment (2,3).  
It works the same with perceived usefulness. Students do not oppose the resource, rate it above  
the midpoint and listening comprehension exercises score the highest in the whole instrument  
and are reported as infrequent. There is no claim of an arithmetic distance between the two sets  
of items, as the questions are different in kind. What can be said is that a positive attitude coexists  
with low exposure, and that closing that gap is a planning decision, not an equipment purchase.  
Theoretically the most interesting finding is the reversal at the extremes of the two orderings.  
Listening comprehension is the lowest platform-mediated use with the highest self-perceived  
use, while oral production is the most with the lowest self-perceived use. This is in opposition to  
the parallel ordering that was expected in the introduction, where more exposure to platforms  
should be related to higher self-perceived performance. There are two non-exclusive  
explanations. The first is the inherent imbalance between receptive and productive skills:  
research on language-learning applications has consistently found that comprehension skills are  
more likely to improve than oral skills, even when the amount of practice is the same (4-6). The  
second is the contact with English in an extramural context. In the out-of-school environment,  
students encounter the language primarily in the form of listening, such as in song lyrics, audio  
and voice communication in video games, streamed series, and short-form video on social media.  
That contact is dense, self-selected and comprehension-driven and there is no production and no  
corrective feedback, which is the profile of the skill that scored highest here. It would also account  
for the fact that a skill such as listening can be scarce in a classroom and high in self-perceived  
competence, since the exposure that maintains listening is not administered by the teacher, and  
thus has no effect on block A, which records only the activity of the classroom mediated by the  
platform. Oral production, on the other hand, requires continuous corrective feedback,  
something that is not provided by sporadic app use or informal exposure (16,18). The instrument  
did not include any out-of-school contact with the language, and therefore cannot be tested here;  
and it is presented as a hypothesis for future research; an extramural-exposure module would be  
the single most informative addition to a replication. The use index for listening comprehension  
is also based on one item in both readings, so the magnitude of the reversal (not its direction) is  
not determined.  
A single generalist application is a risk because of the concentration of preference. The most  
publicly visible tool is the most-mentioned, not the one that matches to students' reported skill  
areas where they feel they have difficulties: the pronunciation and fluency tool receives the least.  
When the choice of resources is left to spontaneous preference, it is a foregone conclusion that  
the skill that is currently best will be strengthened and the one that is poorest will be neglected.  
The literature on automatic speech recognition (ASR) and on automated written feedback (AWF)  
both concur that the formative value of these tools relies on teachers' choice, sequence, and  
45  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
support of their use (8,10,17); allowing students to make their own choices is a way of losing the  
mediating function.  
The implication is that the lever is planning and not provision. It is more cost effective to have a  
specific tool for each skill, at a specific frequency, and monitored by the teacher, than to increase  
the technological stock and to directly address the oral production deficit in the context of limited  
connectivity.  
Study limitations  
The results should be interpreted with five caveats. The first is that the performance measures  
are self-perceived, rather than derived from a standardized language proficiency test, so they  
reflect the students' own assessment of their level of language proficiency rather than their  
objective level. The second is that the design is cross-sectional, meaning that it cannot be used to  
infer causal relationships between use and performance. Third, the processing was based on  
aggregate distributions, not on individual records a conscious protection of anonymity in a small  
community, but one that precludes participant-level correlations and internal consistency indices  
for an instrument validated by expert judgement alone. The individual data matrix should be kept  
and McDonald's omega should be reported in future applications of larger or multi-site samples.  
The fourth is that the use indices for listening and reading comprehension rest on a single item  
each, so the claims concerning those two skills carry the weakest empirical support in the study.  
The fifth is that the study is limited to one institution, meaning that the results only reflect one  
rural border case and cannot be generalized to the education system. These restrictions are not  
intended to detract from the diagnostic value of the work.  
CONCLUSIONS  
The use of digital platforms in the English language teaching in the school studied is systematically  
sporadic. The median falls in the category sometimes across the nine use items, and the mode  
coincides in eight of them, along with the global index of 43.9 points out of one hundred, suggests  
that no pedagogical function of these tools is regular enough to establish routines of practice. The  
two perceived-usefulness items, on the other hand, average 66.8 points and are well above the  
scale center. The two sets of items ask different types of questions, and the distance between  
them is not measured in this study, but the direction of the contrast is clear: that which hinders  
the use of digital platforms in this school is not the disposition of the students but teaching  
planning.  
The intensity of ordering of skills does not match the self-perceived performance ordering of  
skills. Listening comprehension is the lowest platform-mediated and the highest declared  
performance, whereas oral production is the opposite, with the lowest four items of the  
instrument being found under the oral production section, led by oral fluency at 34.4 points. This  
trend at the extremes of the two orderings runs contrary to the parallel ordering that greater  
exposure would predict, and it points to didactic mediation rather than resource availability as  
the variable to be investigated. The observation should be interpreted with great care as the use  
46  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
indices for listening and reading are based on one item for each skill, and the mapping of items  
to skills has not been tested empirically.  
Preference focuses on a generalist application (69.1 %), and the three specialised tools do not add  
up to more than 30.9 % (with the one focused on pronunciation and fluency). The skill with the  
lowest score is exactly the one for which there is a dedicated tool and is the least selected;  
resource selection cannot be spontaneous student choice. The following practical  
recommendation is one of sequence. If budgets are constrained, the focus is not on increasing  
the technological stock, but on planning pedagogical mediation guided by the stock already  
available: assign a specific platform to each skill, establish a frequency of use, and put it under  
teacher control, starting from the skill of oral production. If additional hardware is added prior to  
this mediation, the pattern described here will be repeated at a higher cost. The investment in  
equipment is warranted after the pedagogical sequence is in place and the existing equipment is  
not sufficient to support the sequence, rather than before. Further research is needed to compare  
these self-perceived measures with standardised measures of proficiency, to record out-of-school  
contact with the language, and to expand the diagnosis to other rural border institutions to  
determine if the pattern is situational or specific to this case.  
REFERENCES  
1.  
Instituto Nacional de Estadística y Censos. Tecnologías de la Información y Comunicación  
(TIC) 2024 [Internet]. Quito: INEC; 2024 [cited 20 July 2026]. Available from:  
https://www.ecuadorencifras.gob.ec/documentos/web-  
inec/Estadisticas_Sociales/TIC/2024/202407_Tecnologia_de_la_Informacion_y_Comunicaci  
on-TICs.pdf  
2.  
3.  
4.  
5.  
6.  
7.  
Almache Granda G, Ramírez Romero E, Ley Leyva N, Gortaire Díaz D. La enseñanza del inglés  
en entornos virtuales: desafíos y estrategias para el éxito educativo de estudiantes rurales.  
Revista InveCom. 2024;4(2):1-20. doi: https://doi.org/10.5281/zenodo.10558687  
Cárdenas-Ramírez NJ, Calupiña-Rivera CA. Integración de la tecnología en la enseñanza del  
idioma inglés. Revista Multidisciplinaria Perspectivas Investigativas. 2025;5(1):1-12. doi:  
https://doi.org/10.62574/rmpi.v5i1.253  
Loewen S, Isbell DR, Sporn Z. The effectiveness of app-based language instruction for  
developing receptive linguistic knowledge and oral communicative ability. Foreign Language  
Annals. 2020;53(2):209-33. doi: https://doi.org/10.1111/flan.12454  
Jiang X, Rollinson J, Plonsky L, Gustafson E, Pajak B. Evaluating the reading and listening  
outcomes of beginning-level Duolingo courses. Foreign Language Annals. 2021;54(4):974-  
1002. doi: https://doi.org/10.1111/flan.12600  
Loewen S, Crowther D, Isbell DR, Kim KM, Maloney J, Miller ZF, et al. Mobile-assisted  
language  
learning:  
a
Duolingo  
case  
study.  
ReCALL.  
2019;31(3):293-311.  
doi:  
https://doi.org/10.1017/S0958344019000065  
Shortt M, Tilak S, Kuznetcova I, Martens B, Akinkuolie B. Gamification in mobile-assisted  
language learning: a systematic review of Duolingo literature from public release of 2012 to  
47  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
early  
2020.  
Computer  
Assisted  
Language  
Learning.  
2023;36(3):517-54.  
doi:  
https://doi.org/10.1080/09588221.2021.1933540  
8.  
9.  
Shi H, Aryadoust V. A systematic review of AI-based automated written feedback research.  
ReCALL. 2024;36(2):187-209. doi: https://doi.org/10.1017/S0958344023000265  
Shi H, Aryadoust V. A systematic review of automated writing evaluation systems. Education  
and Information Technologies. 2023;28(1):771-95. doi: https://doi.org/10.1007/s10639-  
022-11200-7  
10. Li Z. Teachers in automated writing evaluation (AWE) system-supported ESL writing classes:  
perception, implementation, and influence. System. 2021;99:102505. doi:  
https://doi.org/10.1016/j.system.2021.102505  
11. Pham VTT, Pham AT. English major students' satisfaction with ELSA Speak in English  
pronunciation courses. PLOS ONE. 2025;20(1):e0317378. doi:  
https://doi.org/10.1371/journal.pone.0317378  
12. Dhivya DS, Hariharasudan A, Ragmoun W, Alfalih AA. ELSA as an Education 4.0 tool for  
learning Business English communication. Sustainability. 2023;15(4):3809. doi:  
https://doi.org/10.3390/su15043809  
13. Kholis A. ELSA Speak app: automatic speech recognition (ASR) for supplementing English  
pronunciation skills. Pedagogy: Journal of English Language Teaching. 2021;9(1):1-14. doi:  
https://doi.org/10.32332/joelt.v9i1.2723  
14. Anggraini A. Improving students' pronunciation skill using ELSA Speak application. Journey:  
Journal  
of  
English  
Language  
and  
Pedagogy.  
2022;5(1):135-41.  
doi:  
https://doi.org/10.33503/journey.v5i1.1840  
15. Akhmad NW, Munawir A. Improving the students' pronunciation ability by using ELSA Speak  
app. IDEAS: Journal on English Language Teaching and Learning, Linguistics and Literature.  
2022;10(1):846-57. doi: https://doi.org/10.24256/ideas.v10i1.2868  
16. Nguyen KN, Nguyen TMT. Tertiary students' perceptions on ELSA Speak application for  
pronunciation learning. European Journal of Applied Linguistics Studies. 2024;7(1):1-13. doi:  
https://doi.org/10.46827/ejals.v7i1.483  
17. Yosintha R, Rekha A. "ELSA Speak" in an online pronunciation class: students' voices. IJEE:  
Indonesian  
Journal  
of  
English  
Education.  
2022;9(1):160-75.  
doi:  
https://doi.org/10.15408/ijee.v9i1.23033  
18. Pennington MC. Teaching pronunciation: the state of the art 2021. RELC Journal.  
2021;52(1):3-21. doi: https://doi.org/10.1177/00336882211002283  
19. Al-khresheh MH. Bridging technology and pedagogy from a global lens: teachers'  
perspectives on integrating ChatGPT in English language teaching. Computers and Education:  
Artificial Intelligence. 2024;6:100218. doi: https://doi.org/10.1016/j.caeai.2024.100218  
20. García García EK. Las plataformas digitales como recurso didáctico para reforzar el  
aprendizaje del inglés en educación primaria. Ciencia Latina Revista Científica  
Multidisciplinar. 2024;8(1):3006-22. doi: https://doi.org/10.37811/cl_rcm.v8i1.9633  
48  
PLATAFORMAS DIGITALES Y DESTREZAS INTEGRADAS DEL INGLÉS: DIAGNÓSTICO CENSAL EN UN CONTEXTO  
RURAL FRONTERIZO DEL CARCHI, ECUADOR  
21. Velázquez Concepción Y, Solis Maldonado LA. Plataformas digitales en la enseñanza del  
inglés: impacto en la motivación y el rendimiento estudiantil. Journal TechInnovation.  
2025;4(1):20-7. doi: https://doi.org/10.47230/Journal.TechInnovation.v4.n1.2025.20-27  
22. Rodríguez-Cajamarca LP, García-Herrera DG, Guevara-Vizcaíno CF, Erazo-Álvarez JC. Alianza  
entre aprendizaje y juego: gamificación como estrategia metodológica que motiva el  
aprendizaje del inglés. Revista Arbitrada Interdisciplinaria Koinonía. 2020;5(1):370-91. doi:  
https://doi.org/10.35381/r.k.v5i1.788  
23. Darsih E, Asikin NA. Mobile assisted language learning: EFL learners' perceptions toward the  
use of mobile applications in learning English. English Review: Journal of English Education.  
2020;8(2):183-94. doi: https://doi.org/10.25134/erjee.v8i2.2999  
24.  
Rebolledo Font De La Vall R, González Araya F. Exploring the benefits and challenges of AI-  
language learning tools. International Journal of Social Sciences and Humanities Invention.  
2023;10(1):7569-76. doi: https://doi.org/10.18535/ijsshi/v10i01.02  
49