JOURNAL OF ENGLISH LANGUAGE PEDAGOGY
Vol.
No.
July 2025, pp.
INVESTIGATING THE EFFECTIVENESS OF QUESTIONGENERATION STRATEGY IN ENHANCING NARRATIVE
TEXT COMPREHENSION AMONG EFL LEARNERS
Jamilah Hatta Ira1.
Reflinda2.
Meylann Melani3.
Veni Roza4 Universitas Islam Negeri Sjech M.
Djamil Djambek Bukittingi Email : jamilahhattaira08@gmail.
Universitas Islam Negeri Sjech M.
Djamil Djambek Bukittinggi Universitas Islam Negeri Sjech M.
Djamil Djambek Bukittinggi Email : reflinda@uinbukittinggi.
Email : melyannmelani@gmail.
Universitas Islam Negeri Sjech M.
Djamil Djambek Bukittinggi Email : veniroz_501@yahoo.
ABSTRACT
Many students struggle with reading comprehension because they do not know how to identify key information, are less engaged in learning, and have limited exposure to interactive strategies.
This study examined the effect of the question-generation strategy on ninth-grade studentsAo comprehension of narrative texts at MTsN 2 Agam.
Employing a quasi-experimental non-equivalent control group design, the research involved 122 students across five classes, with two classes selected: IX.
as the experimental group and IX.
as the control group.
The experimental group received instruction through the question-generation strategy, while the control group was taught using conventional methods.
Reading comprehension was assessed using a multiple-choice test, and the data were analyzed through normality and homogeneity checks, followed by paired-sample and independent-sample t-tests.
The findings revealed a significant effect of the intervention, with the experimental group achieving greater gains from pre-test to post-test compared to the control group (Sig.
< 0.
These results suggest that the question-generation strategy effectively enhances reading comprehension and encourages more active and meaningful learning in EFL classrooms.
Keyword : Question-generation strategy, reading comprehension INTRODUCTION Reading comprehension plays a crucial role in the development of language proficiency, particularly for learners of English as a Foreign Language (EFL).
It enables readers to extract and construct meaning from written texts by integrating textual information with their background knowledge (Nunan, 2.
In academic contexts, effective reading comprehension not only supports studentsAo learning across subjects but also fosters critical thinking and problem-solving skills (Widayanti et al.
, 2.
However, in many EFL settings, 271 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 students still struggle to process information deeply, especially when texts are linguistically or culturally distant from their everyday experience (Brown, 2.
Reading comprehension is defined as the capacity to connect prior knowledge with textual information to construct meaning.
According to (Brassell.
D & Rasinski, 2.
, it is an active process requiring readers to apply existing knowledge and engage with the text to interpret its meaning .
Theoretically, this process can be explained through Schema Theory, which emphasizes the role of background knowledge in understanding texts, and the Construction-Integration Model, which explains how readers build coherent mental representations by integrating new textual information with prior knowledge (Islamiah et al.
Readers demonstrate comprehension when they can accurately interpret, recall, and explain a textAos ideas based on both explicit and implicit information.
One instructional approach that has shown promise in enhancing reading comprehension is the question-generation strategy.
Nunan .
describes reading strategies as the techniques readers employ to facilitate comprehension.
The question-generation strategy, as defined by Sharon Look et al.
, .
, encourages students to formulate and answer meaningful questions about a textAos main ideas while reading.
This active engagement promotes self-monitoring and deeper processing.
According to Langenberg .
, the question generation strategy includes 4 steps: .
Teachers ask children to generate questions during the reading of a passage.
The questions should integrate information across different parts of the passage.
Teachers ask children to evaluate their questions about whether the questions covered important material, were integrative and could be answered based on what is in the text.
Teachers provide feedback on the quality of the questions asked or assist students in answering the questions generated.
Teacher teach the students to evaluate whether their questions covered important information, whether the questions were integrative, and whether they themselves could answer the questions.
Similarly, (Al-Ahdal & Alolaywi, 2.
reported that incorporating metacognitive reading strategies, including question-generation, enhanced undergraduate EFL learnersAo ability to process and retain information.
There are some texts that will be learned by students based on the lesson plan.
Narrative text is one of them.
According to (Gerot-Wignell, 2.
Narrative text is a text to amuse, entertain and to deal with actual or vicarious experience in different ways.
Narrative text is a kind of text that retells the story in the past tense.
The purpose of the text is to entertain or to amuse the readers about the story (Han & Choi, 2.
272 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 Despite evidence supporting this approach, classroom observations and interviews conducted in October at MTsN 2 Agam revealed persistent challenges.
Among the eighthgrade classes (Vi.
1AeVi.
, many students read passively without integrating information, were unaware of whether they truly understood a text, and had difficulty recalling main ideas.
Instruction was largely limited to conventional teacher-led explanations, with little emphasis on active learning strategies.
These issues mirror broader trends in Indonesian EFL classrooms, where traditional methods often dominate and learner engagement remains low (Widayanti et al.
, 2.
However, few local studies have empirically tested the questiongeneration strategy in junior high school contexts, creating a gap this research seeks to METHOD The purpose of this study were as follows: to determine whether there is a significant effect of using the question generation strategy toward students reading comprehension at ninth grade in MTsN 2 Agam, to determine whether there is a significant difference between the class that uses the question generation strategy and the class that does not use the question generation strategy toward students reading comprehension at ninth grade in MtsN 2 Agam, and to determine whether the reading comprehension of students who use the question generation strategy is better than students who do not use the question generation strategy.
This study employed a quantitative approach with a quasi-experimental non-equivalent control group design (Creswell, 2.
This design was chosen because the research was conducted in intact classes where random assignment of individual students was not feasible.
Although this design may involve threats to internal validity such as selection bias, these were minimized by conducting pre-test equivalence checks between the groups.
The study was conducted at MTsN 2 Agam with ninth-grade students in the 2023/2024 academic year.
The population consisted of 122 students in five classes.
Using cluster sampling of intact classes, two were selected: Class IX.
1 as the experimental group .
and Class IX.
2 as the control group .
, with the selection considering comparability of prior academic performance based on teacher input.
The experimental group was taught using the question-generation strategy, implemented following the four steps described by (Langenberg, 2.
and (Sharon Look et al.
, 2.
First, students generated questions while reading passages, integrating information across parts of the text.
Second, they evaluated their questions for coverage of important ideas and 273 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 answerability.
Third, the teacher provided feedback and guided students in refining their Finally, students reflected on the quality and completeness of their questions.
The control group received reading instruction using conventional methods, which primarily involved teacher explanations, vocabulary discussion, and answering comprehension questions from the textbook.
To maintain consistency, lesson plans were prepared, and the implementation was monitored using observation notes.
The instrument was a reading comprehension test consisting of 20 multiple-choice items developed by the researcher.
The test was designed to assess different levels of comprehension, including literal and inferential understanding.
Content validity was reviewed by English teachers, and the items were based on curriculum-relevant texts.
Both groups completed a pre-test before the intervention and a post-test after the intervention, with the same test format used for both but containing different yet equivalent items to reduce test familiarity effects.
Data were analyzed using normality and homogeneity tests, followed by paired-sample t-tests to compare pre-test and post-test scores within groups and independentsample t-tests to compare post-test scores between groups.
Statistical significance was determined at Sig.
-taile.
< 0.
05 ( et al.
, 2.
FINDINGS AND DISCUSSION
Findings This section presents the results of the study in a systematic manner, beginning with the descriptive statistics of the pre-test and post-test, followed by the normality and homogeneity tests, and concluding with the hypothesis testing.
Pre-Test and Post-Test of the Experimental Class Before the treatment, the experimental group was given a pre-test to assess their initial reading comprehension ability.
After the implementation of the question generation strategy, the students were administered a post-test to evaluate their progress.
The detailed results of the experimental class are presented in Table 1.
Table 1: The Scores of Pre-Test and Post-Test of the Experimental Class Students Pre-test Post-test Gained score 274 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 To provide a clearer overview, the descriptive statistics are summarized in Table 2.
Table 2: Descriptive Statistics of Pre-Test and Post-Test of the Experimental Class Min Max Mean Median Standard Varience Pre-test Post-test As shown in Table 2, the mean score of the experimental class in the pre-test was 39.
with a minimum of 20 and a maximum of 60.
After the treatment, the mean score increased to 20, with a minimum of 45 and a maximum of 90.
This improvement of nearly 25 points indicates that the students in the experimental class benefited considerably from the strategy.
Pre-Test and Post-Test of the Control Class The same procedure was applied in the control class, which was taught using conventional instruction without the question generation strategy.
The pre-test and post-test scores of this class are presented in Table 3.
Table 3: The Scores of Pre-Test and Post-Test of the Control Class Students Pre-test Post-test Gained score 275 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 The descriptive statistics are provided in Table 4.
Table 4: Descriptive Statistics of Pre-Test and Post-Test of the Control Class Min Max Mean Median Standard deviation Varience Pre-test Post-test Based on Table 4, the mean pre-test score of the control class was 39.
79, which increased to 51.
88 in the post-test.
Although there was an improvement of about 12 points, the gain was notably smaller compared to the experimental class.
This suggests that while conventional instruction led to some progress, it was not as effective as the question generation strategy in improving studentsAo reading comprehension.
Normality Test Before conducting hypothesis testing, the normality of the data was examined using the Kolmogorov-Smirnov and Shapiro-Wilk tests.
The results are presented in Tables 5 and 6.
Table 5: Normality Test of Pre-Test Scores Tests of Normality Class Kolmogorov-Smirnova Shapiro-Wilk Statistic Sig.
Statistic Sig.
Reading Pre-test Experimenta l Class Pre-test Control Class Lilliefors Significance Correction 276 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 It could be seen that the significance of normality pre-test score in experimental class 085 > 0.
Meanwhile, the significance of normality pre-test score in control class was 111 > 0.
So, it could be concluded that the data of pre-test in experimental and control class were normally distributed.
Table 6: Normality Test of Post-Test Scores Class Reading Kolmogorov-Smirnova Statistic Sig.
Post-test Experime ntal Class Post-test Control Class Lilliefors Significance Correction Shapiro-Wilk Statistic Sig.
It could be seen that the significance of normality post-test score in experimental class 079 > 0.
meanwhile, the significance of normality post-test score in control class 70 > 0.
So, it could be concluded that the data of post-test in experimental and control class are normally distributed.
Homogeneity Test In addition to normality, the homogeneity of variance was tested to determine whether the two groups had equal variances.
The results of LeveneAos Test are displayed in Tables 7 Readi Table 7: Homogeneity Test of Pre-Test Scores Test of Homogeneity of Variance Levene Statistic Based on Mean Based on Median Based on Median and with adjusted df Based on trimmed Sig.
From table, it was found that the significance value .
based on mean is 0.
108 > 0.
it can be concluded that the variant of the pre-test of experimental class and control class is the same.
277 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 Table 8: Homogeneity Test of Post-Test Scores Levene Statistic Based on Mean Based on Median Based on Median and with adjusted df Based on trimmed mean Reading Sig.
The results show that the significance values for both pre-test .
and post-test .
were greater than 0.
Therefore, it can be concluded that the data of the experimental and control classes were homogeneous, allowing for meaningful comparisons between the two groups.
Hypothesis Testing This stage of analysis aimed to verify whether the application of the question generation strategy produced a significant effect on studentsAo reading comprehension.
Three hypotheses were tested: .
the effect of the strategy within the experimental class .
aired-sample tes.
, .
the comparison between the experimental and control classes .
ndependent-sample tes.
, and .
the superiority of the experimental class in terms of mean post-test scores.
The detailed results are presented below.
First Hypothesis (Paired Samples Test in the Experimental Clas.
The first hypothesis examined whether there was a statistically significant difference between the pre-test and post-test scores of the experimental class.
This test was conducted to determine the effectiveness of the question generation strategy in improving studentsAo reading comprehension within the same group.
The results are presented in Table 9.
Pair Pretest Post -test Table 9: Paired Samples Test for Experimental Class Paired Differences Mean Std.
Std.
95% Confidence Deviatio Error Interval of the Mean Difference Lower Upper Sig.
tailed As shown in Table 9, the mean difference between the pre-test and post-test was -24.
with a significance value of 0.
000 (< 0.
This indicates that the post-test scores were 278 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 significantly higher than the pre-test scores in the experimental class.
In other words, the application of the question generation strategy contributed meaningfully to the studentsAo The large negative t-value (-9.
further reflects a consistent increase in scores after the treatment.
These findings confirm that the strategy had a substantial positive effect on reading comprehension when implemented in the experimental group, thereby rejecting Ho and accepting Ha.
Second Hypothesis (Independent Samples Test of Post-Test Score.
The second hypothesis investigated whether there was a significant difference in posttest performance between the experimental class, which was taught using the question generation strategy, and the control class, which received conventional instruction.
This test compared the effectiveness of the two different teaching approaches.
The results are summarized in Table 10.
Table 10: Independent Samples Test of Post-Test Scores Levene's t-test for Equality of Means Test for Equality of Variances Sig.
Mean Std.
Differe Error Differe Pos Equal Equal ces not 95% Confidence Interval of the Difference Lowe Upper The results in Table 10 demonstrate that the significance value was 0.
003 (< 0.
confirming that there was a statistically significant difference in post-test scores between the two groups.
The mean difference of 12.
33 points shows that the experimental class outperformed the control class by a considerable margin.
These findings strongly support the conclusion that the use of the question generation strategy was more effective in enhancing 279 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 reading comprehension than conventional methods.
Therefore, the null hypothesis was rejected, and the alternative hypothesis was accepted.
Third Hypothesis (Comparison of Mean Score.
The third hypothesis focused on the comparison of mean post-test scores between the experimental and control classes.
This comparison provided further evidence of the effectiveness of the strategy in a broader sense.
The results of the independent samples test are presented below.
Levene's Test for Equality of Variances Pos Equal Equal ces not Table 11: Independent Samples Test t-test for Equality of Means Sig.
taile Mean Differe Std.
Error Differe 95% Confidence Interval of the Difference Lowe Upper The results reveal that the experimental class achieved a higher mean post-test score of 20 compared to the control class mean of 51.
This difference, which was statistically significant .
= 0.
003 < 0.
, highlights the superiority of the question generation strategy in fostering reading comprehension.
The higher average gain in the experimental class .
points from pre-test to post-tes.
compared to the control class .
09 point.
further strengthens this conclusion.
Thus, the null hypothesis was rejected and the alternative hypothesis was accepted, confirming that students who were taught using the question generation strategy achieved better reading comprehension outcomes than those who were 280 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 Discussion The findings of this study demonstrate that the use of the question-generation strategy significantly improved studentsAo reading comprehension.
In the experimental class, the mean score increased from 39.
60 in the pre-test to 64.
20 in the post-test, with a mean gain of 24.
The paired-samples t-test showed Sig.
-taile.
= 0.
000 < 0.
05, indicating a statistically significant improvement.
Furthermore, the independent-samples t-test comparing post-test scores between the experimental and control classes yielded Sig.
-taile.
= 0.
confirming that students taught using the question-generation strategy outperformed those taught through conventional methods.
These results suggest that incorporating questiongeneration into reading instruction can effectively enhance comprehension skills.
These results align with previous theoretical perspectives and empirical findings.
According to Sharon Look et al.
, the question-generation strategy involves readers actively asking and answering meaningful questions during reading, thereby fostering deeper understanding of the text.
Furthermore.
Khansir & Dashti .
, question generation strategy is designed to help students effectively understand complex reading demands.
Nurhailis .
found significant gains in reading comprehension among second-year students at MA Islamic Centre Al-Hidayah Kampar, while Khansir & Dashti .
reported similar results with Iranian EFL high school learners.
The present study corroborates these findings by demonstrating comparable or greater effects, possibly due to the targeted application of the strategy over the research period.
The effectiveness of the question-generation strategy in this study can be understood through several theoretical lenses.
First, it aligns with schema theory, which posits that comprehension improves when learners activate prior knowledge and connect it to new Generating questions prompts students to recall and apply relevant schemas while reading.
Second, from a metacognitive perspective, this strategy encourages learners to monitor their understanding, identify gaps, and seek clarification, thus promoting active Third, it may reduce extraneous cognitive load by directing attention to key ideas, enabling better integration of textual information into long-term memory (Ma et al.
, 2.
For educators, these findings highlight the practical value of integrating questiongeneration activities into reading lessons.
Teachers can guide students to create questions before, during, and after reading, encouraging deeper engagement with texts.
This approach not only enhances comprehension but also fosters critical thinking and learner autonomy.
line with Amanda & Dewi .
, reading becomes one of the language skills that must be 281 | Copyright A 2025.
ELP.
ISSN 2656-6451 EISSN 2502-2792 mastered by the learner to get information from books or texts that were used to enrich their While the improvements observed are promising, certain limitations should be The quasi-experimental design limits the ability to draw definitive causal The sample size was relatively small, and the research was conducted in a single school, which may affect generalizability.
Additionally, the study did not assess long-term retention of reading gains (Sun et al.
, 2.
Future research could address these issues by employing randomized controlled trials, involving larger and more diverse populations, and including follow-up assessments to examine sustained effects.
In conclusion, the results support the assertion that the question-generation strategy can significantly improve studentsAo reading comprehension, as evidenced by higher post-test scores in the experimental group compared to the control group.
This is consistent with the perspectives of Putri & Melani .
Nasir et al.
SyaAoban & Reflinda .
, and Wilda et al.
, who emphasize that reading comprehension involves not only decoding text but also processing and interpreting meaning.
By promoting active engagement and deeper understanding, question-generation serves as an effective instructional strategy in EFL
CONCLUSION AND SUGGESTION
This study concludes that the question-generation strategy has a significant positive effect on studentsAo reading comprehension, as it encourages learners to actively engage with texts, monitor their understanding, and construct deeper meaning.
The findings support schema theory, metacognitive approaches, and cognitive load perspectives, showing that generating questions fosters critical thinking and enhances comprehension more effectively than conventional instruction.
For teachers, integrating question-generation into reading lessons is highly recommended to promote learner autonomy and sustained engagement.
Future research should extend this work by involving larger and more diverse populations and examining long-term impacts to strengthen its applicability in broader educational contexts.
REFERENCES