How may the ChatGPT usage of the respondents be described in terms of:
1.1 Frequency;
INDICATOR Mean Sd Descr
iptio
n
1. I use ChatGPT to complete English-related tasks. 2.98 2.43 Agree
2. I do not use ChatGPT as a regular study tool. 2.68 2.12 Agree
3. Without ChatGPT, I find it hard to complete English-related tasks. 2.45 1.89 Disag
ree
4. I depend on ChatGPT as a part of my study routine. 2.42 1.85 Disag
ree
5. I find myself relying on ChatGPT more than any English-related learning resources. 2.44 1.87 Disag
ree
6. I depend on ChatGPT when I find English tasks challenging. 2.73 2.17 Agree
7. I use ChatGPT 1-3 hours a week. 2.91 2.35 Agree
8. I use ChatGPT 4-6 hours a week. 1.62 1.00 Stron
gly
Disag
ree
9. I use ChatGPT 7-12 hours a week. 1.52 0.89 Stron
gly
Disag
ree
2.42 1.85 Agre
e
(Scale of interval: STRONGLY DISAGREE (1-1.75), DISAGREE (1.76 – 2.51), AGREE (2.52 – 3.27), STRONGLY AGREE (3.28 – 4.00)
The results show that students generally agree that they use ChatGPT to complete English-related tasks
(M=2.98, SD=2.43) and depend on it when faced with challenging activities (M=2.73, SD=2.17). This indicates that the
tool is primarily utilized as a form of academic support when difficulties in English tasks arise.
On the other hand, respondents disagreed with statements suggesting a strong dependence on ChatGPT. They reported
that they do not necessarily find it hard to complete English-related tasks without ChatGPT (M=2.45, SD=1.89), nor do
they consider it as part of their regular study routine (M=2.42, SD=1.85). Likewise, they disagreed with relying on
ChatGPT more than other learning resources (M=2.44, SD=1.87). This shows that while ChatGPT is used, it is not
viewed as the primary or indispensable study tool.
In terms of the duration of usage, students agreed that they spend around 1–3 hours per week using ChatGPT (M=2.91,
SD=2.35). However, they strongly disagreed with using the tool for longer periods, such as 4–6 hours (M=1.62,
SD=1.00) or 7–12 hours per week (M=1.52, SD=0.89).
Purpose
INDICATOR Mean Sd Descr
iptio
n
1. I rarely use ChatGPT because of my limited knowledge and familiarity with it. 2.68 2.12 Agree
2. I only use ChatGPT to simplify Englishrelated tasks that are hard to understand. 2.95 2.40 Agree
3. I involve ChatGPT in the process of completing my tasks, but it is not the central part of 2.84 2.29 Agree
it..
4. I think ChatGPT can enhance, but not replace, my own effort in completing English- 3.08 2.53 Agree
related tasks.
5. I see ChatGPT as a supporting tool rather than a necessity for English tasks. 2.81 2.26 Agree
6. I regularly use ChatGPT to generate ideas or outlines for my English writing tasks. 2.73 2.17 Agree
7. I often use ChatGPT to find answers 2.58 2.02 Agree
8. ChatGPT explains word meanings in a simple way that I can understand 2.97 2.42 Agree
9. ChatGPT helps me organize my thoughts and ideas when writing essays in English. 2.95 2.40 Agree
Using ChatGPT helps me practice giving reasons and evidence to support my answers in 2.81 2.26 Agree
English activities.
10. I use ChatGPT to check the grammar of my English-related works. 2.35 2.34 Disag
ree
2.85 2.29 Agre
e
(Scale of interval: STRONGLY DISAGREE (1-1.75), DISAGREE (1.76 – 2.51), AGREE (2.52 – 3.27), STRONGLY AGREE (3.28 – 4.00)
The results show that Grade 11 students generally use ChatGPT with a supportive purpose in their English-related tasks.
The composite mean of 2.85 (SD=2.29), interpreted as Agree, indicates that respondents acknowledge ChatGPT as a helpful tool
but not as the central element of their learning.
Students reported that they use ChatGPT mainly to simplify difficult English tasks (M=2.95, SD=2.40), to organize ideas
when writing essays (M=2.95, SD=2.40), and to understand word meanings more clearly (M=2.97, SD=2.42). They also agreed
that ChatGPT is helpful for generating ideas or outlines (M=2.73, SD=2.17) and for practicing reasons and evidence in English
activities (M=2.81, SD=2.26).
Moreover, students see ChatGPT as an enhancement rather than a replacement for their own effort (M=3.08, SD=2.53)
and as a supporting tool rather than a necessity (M=2.81, SD=2.26). This suggests that they recognize its value while still
acknowledging the importance of their own learning process.
However, respondents disagreed with the statement that they use ChatGPT to check grammar (M=2.35, SD=2.34). This
shows that while ChatGPT is used for ideas, explanations, and simplification, grammar checking is not a primary purpose for most
students.
How may the English performance of the respondents be described in terms of:
Grammar
Descriptive Statistics
N Sum Mean Std. Deviation
Statistic Statistic Statistic Std. Error Statistic
STRAND(STEM101) 16 16.00 1.0000 0.00000 0.00000
GRAMMAR 16 120.00 7.5000 0.52440 2.09762
STRAND(STEM102) 21 21.00 1.0000 0.00000 0.00000
GRAMMAR 21 161.00 7.6667 0.52705 2.41523
STRAND(HUMSS101) 12 12.00 1.0000 0.00000 0.00000
GRAMMAR 12 71.00 5.9167 0.78295 2.71221
STRAND(HUMSS102) 13 13.00 1.0000 0.00000 0.00000
GRAMMAR 13 70.00 5.3846 0.66543 2.39925
STRAND(AABM101 23 23.00 1.0000 0.00000 0.00000
GRAMMAR 23 175.00 7.6087 0.45222 2.16877
STRAND(TVL101) 23 23.00 1.0000 0.00000 0.00000
GRAMMAR 23 133.00 5.7826 0.43538 2.08799
N Sum Mean Std. Deviation
OVERALL Statistic Statistic Statistic Std. Error Statistic
GRAMMAR
121.67 6.64 0.56 2.31
The English performance of the respondents in terms of grammar was described using the computed descriptive statistics
across the different academic strands. Results show that students from the STEM strand obtained the highest grammar
performance. Specifically, STEM102 recorded the highest mean score of 7.67 (SD = 2.42), followed closely by STEM101 with a
mean of 7.50 (SD = 2.10). Similarly, students from the ABM strand also exhibited strong grammar performance with a mean score
of 7.61 (SD = 2.17), indicating a comparable proficiency level with their STEM counterparts.
On the other hand, students from the HUMSS strand obtained lower mean scores in grammar. HUMSS101 posted a mean
of 5.92 (SD = 2.71) while HUMSS102 registered the lowest grammar performance with a mean of 5.38 (SD = 2.40). Likewise, the
TVL strand students also recorded a relatively low grammar mean of 5.78 (SD = 2.09). These results indicate that HUMSS and
TVL students generally demonstrated weaker performance in grammar compared to STEM and ABM students. Taken collectively,
the overall mean score across strands was 6.64 (SD = 2.31).
Vocabulary
Descriptive Statistics
N Sum Mean Std. Deviation
Statistic Statistic Statistic Std. Error Statistic
STRAND(STEM101) 16 16.00 1.0000 0.00000 0.00000
VOCABULARY 16 100.00 6.2500 0.57373 2.29492
STRAND(STEM102) 21 21.00 1.0000 0.00000 0.00000
VOCABULARY 21 143.00 6.8095 0.38184 1.74983
STRAND(HUMSS101) 12 12.00 1.0000 0.00000 0.00000
VOCABULARY 12 72.00 6.0000 0.57735 2.00000
STRAND(HUMSS102) 13 13.00 1.0000 0.00000 0.00000
VOCABULARY 13 61.00 4.6923 0.57048 2.05688
STRAND(AABM101 23 23.00 1.0000 0.00000 0.00000
VOCABULARY 23 152.00 6.6087 0.38622 1.85225
STRAND(TVL101) 23 23.00 1.0000 0.00000 0.00000
VOCABULARY 23 110.00 4.7826 0.45317 2.17332
N Sum Mean Std. Deviation
OVERALL Statistic Statistic Statistic Std. Error Statistic
VOCABULARY
106.33 5.86 0.49 2.02
The English performance of the respondents in terms of vocabulary reveals varying levels of proficiency across strands.
Students from the STEM strand consistently performed well, with STEM102 obtaining the highest mean score of 6.81 (SD =
1.75), followed by STEM101 with a mean of 6.25 (SD = 2.29). Similarly, the ABM strand also showed strong vocabulary
performance, recording a mean of 6.61 (SD = 1.85), which is comparable to the performance of STEM students.
In contrast, the HUMSS strand reflected lower results. HUMSS101 posted a mean of 6.00 (SD = 2.00), while HUMSS102 had a
noticeably lower mean score of 4.69 (SD = 2.06). Likewise, students from the TVL strand also registered a lower vocabulary
performance, with a mean of 4.78 (SD = 2.17). These findings suggest that HUMSS102 and TVL students demonstrated the
weakest performance in vocabulary compared to their peers in other strands. Taken as a whole, the respondents’ overall mean
score in vocabulary was 5.86 (SD = 2.02)
Critical thinking skills
Descriptive Statistics
N Sum Mean Std. Deviation
Statistic Statistic Statistic Std. Error Statistic
STRAND(STEM101) 16 16.00 1.0000 0.00000 0.00000
CRITICAL_THINKING_SKILLS 16 48.00 3.0000 0.42817 1.71270
STRAND(STEM102) 21 21.00 1.0000 0.00000 0.00000
CRITICAL_THINKING_SKILLS 21 84.00 4.0000 0.27603 1.26491
STRAND(HUMSS101) 12 12.00 1.0000 0.00000 0.00000
CRITICAL_THINKING_SKILLS 12 41.00 3.4167 0.45157 1.56428
STRAND(HUMSS102) 13 13.00 1.0000 0.00000 0.00000
CRITICAL_THINKING_SKILLS 13 28.00 2.1538 0.35529 1.28103
STRAND(AABM101 23 23.00 1.0000 0.00000 0.00000
CRITICAL_THINKING_SKILLS 23 80.00 3.4783 0.28024 1.34400
STRAND(TVL101) 23 23.00 1.0000 0.00000 0.00000
CRITICAL_THINKING_SKILLS 23 57.00 2.4783 0.20727 0.99405
N Sum Mean Std. Deviation
Statistic Statistic Statistic Std. Error Statistic
CRITICAL_THINKING_SKILLS
56.33 3.09 0.33 1.36
The respondents’ English performance in terms of critical thinking skills revealed varying outcomes across strands.
Among the groups, STEM102 students recorded the highest mean score of 4.00 (SD = 1.26), indicating a stronger capacity for
analysis and reasoning compared to other strands. This was followed by students from ABM101 with a mean of 3.48 (SD = 1.34),
and HUMSS101 with a mean of 3.42 (SD = 1.56). Meanwhile, STEM101 obtained a moderate mean of 3.00 (SD = 1.71),
suggesting average performance in this domain.
On the other hand, students from HUMSS102 and TVL101 demonstrated lower levels of performance in critical thinking.
HUMSS102 posted a mean of 2.15 (SD = 1.28), the lowest among all groups, while TVL101 recorded a mean of 2.48 (SD =
0.99). These findings imply that students from HUMSS102 and TVL101 encountered greater challenges in applying logical
reasoning and problem-solving skills.
In general, the respondents achieved an overall mean score of 3.09 (SD = 1.36), which indicates that the level of critical
thinking skills was fair across the entire group.
Significant relationship between ChatGPT usage and the English performance of the respondents
Correlations
ChatGPT_USAGE GRAMMAR VOCABULARY CRITICAL_THINKING_SKILLS
ChatGPT_USAGE Pearson 1 0.045** -0.071* -0.096*
Correlation
Sig. (2- 0.000 0.037 0.014
tailed)
Sum of 6.956 2.962 -4.183 -8.273
Squares
and Cross-
products
Covariance 0.065 0.028 -0.039 -0.077
N 108 108 108 108
**. Correlation is significant at the 0.01 level (2-tailed).
The correlation analysis revealed a statistically significant relationship between ChatGPT usage and the English
performance of the respondents across the three domains: grammar, vocabulary, and critical thinking skills.
For grammar, the computed Pearson correlation coefficient was 0.045 with a significance level of p = 0.000.
This result indicates a positive but very weak correlation, suggesting that as students use ChatGPT, their grammar
performance tends to improve slightly.
In contrast, vocabulary showed a correlation coefficient of -0.071 (p = 0.037), which signifies a weak negative
correlation. This implies that higher ChatGPT usage was associated with slightly lower vocabulary performance among
the respondents. A similar trend was found in critical thinking skills, where the correlation coefficient was -0.096 (p =
0.014), also indicating a weak negative correlation. This suggests that increased reliance on ChatGPT may be
associated with reduced critical thinking performance.