ISSN 1798-4769

Journal of Language Teaching and Research, Vol. 15, No. 2, pp. 568-578, March 2024
DOI: https://doi.org/10.17507/jltr.1502.25

Defining the Role of Artificial Intelligence in

Improving English Writing Skills Among
Indonesian Students
Kaharuddin
English Education Department, Universitas Islam Negeri Alauddin Makassar, Indonesia

Djuwairiah Ahmad
English Education Department, Universitas Islam Negeri Alauddin Makassar, Indonesia

Mardiana
English Education Department, Universitas Islam Negeri Alauddin Makassar, Indonesia

Ismail Latif
English Education Program, Institut Agama Islam Negeri Parepare, Indonesia

Burhanuddin Arafah
English Department, Faculty of Cultural Sciences, Hasanuddin University, Makassar, Indonesia

Ray Suryadi
English Education Study Program, Faculty of Teacher Training and Education, Sembilanbelas November University of
Kolaka, Indonesia

Abstract—Artificial intelligence (AI) is a technological product in the form of an application usable in

education. The implementation of AI, specifically in the teaching and learning process in Higher Education,
not only broadens students' scientific knowledge academically but also facilitates awareness about current
technological developments. Therefore, this study aimed to explore the improvement of English writing skills
through the integration of AI into reading practices and provision of adequate feedback for skills enhancement.
Also, students' attitude toward the use of AI in the teaching and learning process served as the intervening
variables. This study was conducted at Higher Schools in Makassar City, Indonesia, and data were collected
from 80 students through questionnaires and subsequently analyzed using path analysis. The results showed
that (1) Students' attitude toward the use of AI played a mediating role in explaining the effect of reading and
feedback on writing skills, (2) Reading had an indirect effect through attitude on writing skills improvement,
and (3) Lecturers feedback had an indirect effect through attitude on writing skills.

Index Terms—artificial intelligence, reading skills, feedback, attitude, writing skills

I. INTRODUCTION
The Covid-19 pandemic impacted economic, social, and educational activities, where the teaching and learning
process could not be carried out in the classroom. Government policy required that the teaching and learning process be
carried out online through electronic media, such as computers, laptops, or cell phones. The face-to-face learning is
suddenly changed into an online learning method (Anggrawan et al., 2019). Even though the concept of online learning
application might not be considered optimal according to Kaharuddin (2022), it presented the initial effort to encourage
the use of technological instruments in the educational field. The expansion of internet had a significant impact at this
point (Arafah & Hasyim, 2023a). Information and material learning provided online worldwide entail students’
cognitive skills and abilities (Arafah & Hasyim, 2023b; Arafah et al., 2023). This allowed students to gain a practical
understanding of the relationship between technological developments and education.
Technology facilitated pedagogical approaches for lecturers (Tsui & Tavares, 2021, p. 110) and served as a medium
for creating a new way of teaching and learning (Keengwe et al., 2008, p. 83). Therefore, it is expected that the use of
electronic devices through artificial intelligence (AI) can provide expected learning outcomes. Depending on how it is
used, technology is very beneficial within digital media and digital literacy (Suhadi et al., 2022; Hasyim & Arafah,
2023a). Additionally, the integration of AI facilitated the teaching and learning process (Hasjim et al., 2020;
Kaharuddin et al., 2022). Through the internet, students can seek knowledge from any platform (Hasyim & Arafah,
2023b). For example, Google Translate provides translation facilities to translate teaching materials from English to

© 2024 ACADEMY PUBLICATION

JOURNAL OF LANGUAGE TEACHING AND RESEARCH 569

Indonesian. Caplar et al. (2017) stated that Artificial Intelligence (AI) can process language and translation
automatically. In 2015 a new computer model was discovered; it is called the D-Wave 2X. This computer model can
perform complex AI operations.
This study discussed the use of AI in English teaching and learning, as well as its effect on improving students'
writing skills. Fahimirad and Kotamjani (2018) stated that the quality of students' learning process was improved
through AI (p. 108). This study emphasized the improvement of students' writing skills on grammar mastery and
vocabulary through AI.
English writing skills have been taught with traditional learning models, but the results were unsatisfactory because
writing skills were low (Arafah & Kaharuddin, 2019). Furthermore, students’ perceptions of the use of AI were
descriptively discussed based on the results of quantitative analysis. The use of electronic devices was intended to
facilitate writing skills supported by good grammar and vocabulary mastery, which can be developed by reading
activities. Therefore, this analysis focused on how reading comprehension, as an independent variable, impacted writing
skills. Reading lessons were delivered using AI applications, and each lesson required feedback from lecturers.
Technology has basic features that enable virtual communicative interactions between lecturers and students. Feedback
from lecturers can help students improve their mistakes in grammar and vocabulary. In this study, feedback was the
second independent variable which was explored for its effect on writing skills. Therefore, "effective application of AI
methods can improve the quality of teaching and learning" (Arafah & Kaharuddin, 2019, p. 108).
Students' attitude to English lessons promoted learning activities and accelerated mastery. Research by Kaharuddin et
al. (2021) showed that there was a significant effect of students' attitude on motivation. The significance of character
education is in need due to the current situation of education in Indonesia (Mokoginta & Arafah, 2022). This study
discussed attitude as an intervening variable to explain the effect of reading and feedback on writing skills with the
integration of AI. The perceptions of digital-based learning were also analysed, where English writing skills were the
dependent variable while reading and feedback were the independent. This study directly and indirectly examined the
effect of the two independent variables on students' attitude.

II. LITERATURE REVIEW

A. Artificial Intelligence for Learning-Teaching

During the Covid-19 pandemic, the English teaching process was not merely carried out by lecturers in class but
included the use of AI online. This application was helpful in improving English skills, specifically writing. This is
because students could check the grammatical errors using AI in addition to lecturers' explanation. As a "simulation of
human knowledge on a machine programmed to think and imitate human actions" (Vasiljeva et al., 2021, p. 8), AI can
replace lecturers to carry out the learning tasks. For example, students learn English grammar, how to spell words,
match words, as well as make sentences through AI. The mistakes made can be detected by this application and
corrected as feedback. However, in the context of writing, the mastery of English grammar is needed as expressed by
Mozgovoy (2011) that “grammar check is important in text writing and language learning” (p. 209). Even a slight error
can drastically change the whole meaning (Iksora et al., 2022). Correction of grammar, diction, or sentence construction
was provided by AI through feedback-motivated learning activities. Cotos E. (2011) stated that "students obtain
feedback, then re-read and improve their writing, as well as practice being independent learners" (p. 107). At the end,
technology advancement in a form of online learning increases students’ writing skills (Arnawa & Arafah, 2023). Other
than that, grammar correction leads students to use meaningful and informative words to improve communication skills
(Kuswanty et al., 2023). Therefore, in communication, people should follow the exact patterns and procedures to avoid
any kind of mistake (Yulianti et al., 2022).
B. Reading on Writing Improvement
Reading is an important language skill for students to be carried out regularly. In addition to understanding the
message of the reading material, students can improve vocabulary mastery as a necessary aspect of English
communication. According to Ismail et al. (2020), vocabulary is a language element through which messages are
conveyed orally and in writing.
Students developed vocabulary through reading (Renandya, 2007; Arafah B. et al., 2023; Kaharuddin et al., 2023),
and reading material was taught separately by lecturers (Kroll J.F., 1993). In writing classes, students rarely engaged in
reading activities as expected (Jolliffe, 2007). Even lecturers and students neglected reading activities in composition
classes (Hirvela, 2004). On the other hand, vocabulary and grammar mastery is needed in writing. Horning and
Kraemer (2013) stated that one of the effective methods for developing writing skills in higher schools and colleges was
to engage in reading activities. Among the four language skills, reading and writing were the first to be learned,
according to Durukan (2011). This was in line with the opinion of Esmaeili (2002) that reading and writing had a
positive influence on students' academic success, especially in English lessons. Reading activity can also in a form of
reading literary work since literary work has escalated in academic studies (Takwa et al., 2022; Arafah et al., 2023).
Through literary work, students are willing to improve knowledge in a scope of culture and locality (Arifin et al., 2022),
nature and environmental situation (Manugeren et al., 2023; Siwi et al., 2022), and human relationship with other

© 2024 ACADEMY PUBLICATION

570 JOURNAL OF LANGUAGE TEACHING AND RESEARCH

creatures (Yudith et al., 2023). Therefore in learning English by reading literary work, students as readers are capable to
discover language styles that an author used that later will be useful for them to write (Asri et al., 2023).
Related to the effect of reading and writing on improving English competence, this study was conducted using AI
technology in a virtual teaching-learning process. The influence of reading on writing was directly and indirectly
discussed through attitude toward the process of using AI.
C. Feedback on Writing Improvement
Students may have a few questions, such as “Has the lecturer checked my assignment? Is my assignment correct? Is
there anything I need to fix?" The answers to these questions can be obtained through feedback from lecturers. However,
many lecturers did not provide feedback, leading to a situation where the questions were not answered due to
insufficient training (Corwin, 1976). In addition, there were many misconceptions only understood as lecturers'
comments (Boud & Molloy, 2013). Feedback can be understood as lecturers' comments regarding students' work. Since
the comments are related to the level of understanding and ability to carry out assignments (Hattie & Timperley, 2007),
feedback is students' understanding of lecturers' comments (Henderson et al., 2019).
Feedback is not the end of a learning process, but information related to the ability to complete English assignments
(Boud & Molloy, 2013). Therefore, it needs to comply with the principles of the right time, clear, educative,
proportional to the assessment rubric, communicate, and motivate students (Henderson & Phillips, 2014). Corrective
feedback improved writing skills (Boggs, 2019), reduced grammatical and lexical errors (Chandler, 2003), and
significantly improved writing (Khadawardi, 2021). Avoiding any lexical error in choosing words or phrases enhances
students’ vocabulary as well (Baa et al., 2023; Takwa et al., 2022; Arafah B. et al., 2024).
During the Covid-19 pandemic, the learning process took place virtually, and the comments feature in Microsoft
Word or Google Docs was used to send feedback to students. Furthermore, audio feedback was used to improve
communication (Cavanaugh & Song, 2014). A website-based application called Vocaroo (https://vocaroo.com) and
Mote, an integrated extension of Google with some features were applied to make a record of activities and sounds.
D. Learning Attitude on Writing Improvement
Before the Covid-19 pandemic, students studied English in class through traditional methods. Lecturers sometimes
used electronic devices and AI as teaching aids. However, during the pandemic, the learning-teaching process took
place virtually through electronic devices. There are 3 main areas in virtual learning, namely developing access to
education and training, developing the learning quality, and maintaining competitiveness in universities (Newton, 2003).
Face-to-face learning in class can enhance the social aspects between students, as well as between students and lecturers.
With virtual learning, there was a tendency for students to get bored studying from home (Male et al., 2020). Students
were aware of the importance of technology even though the usage was not encouraged (Farooq & Javid, 2012;
Mardiana et al., 2023), and they had a negative attitude to the virtual learning (Govindasamy, 2001). Meanwhile,
students of nursing department performed positive attitude to the E-learning. Learning attitudes of students and their
effect on writing skills are examined in this study.

III. METHODOLOGY
This was a descriptive quantitative study that explored students' perceptions on the use of AI as a tool in virtual
English learning. Reading comprehension and teachers feedback were the independent variables. The writing skill was
the dependent variable. The intervening variable was learning attitude. The effect of the independent variables (X 1,2) on
the dependent variable (Y2) through the intervening variable (Y1) was the focus of this study. The research was
conducted on 100 higher school students in Makassar City, as the research respondents. The questionnaires distributed
to the respondents were valued based on the 5 points of Likert scale which covered strongly agree, agree, neutral,
disagree, and strongly disagree (Arafah B., 2023; Lebba & Kaharuddin, 2023).
A. Test of Research Instrument
Validity and reliability tests were conducted to meet the criteria for validity and reliability (Abidin & Kaharuddin,
2021; Lebba et al., 2023).
a. Validity Test: This was conducted by correlating the value of each item and total variables. Pearson Correlation
Product Moment (r) was used with a significance level 5% or 0.05, the result was called r-calculation. If the rcalculation
on every item of questionnaire was significantly more than the value of r-table, the instrument was stated to be
valid.
b. Reliability Test: This unit was intended to test the consistency of students in answering the questionnaire. The
reliability of students' answers was determined using Cronbach Alpha. The answers were stated to be reliable
when the Cronbach Coefficient Alpha value was above 0.6.
B. Classical Assumption Test
a. Normality Test: This test was conducted to analyse the distribution of the residual value of the model using the
Kolmogorov-Smirnov test. A regression model is stated to be good when it has a normal distribution of residual
values. The basis for decision making is the asymp value, and when the significance value is above 0.05, then it is

© 2024 ACADEMY PUBLICATION

JOURNAL OF LANGUAGE TEACHING AND RESEARCH 571

called normally distributed.

b. Multicollinearity test: It was conducted by analysing the correlation between reading comprehension and teachers
feedback, both are independent variables. A good model does not show any multicollinearity. Multicollinearity is
not shown if multicollinearity test gives value lower than 10 or the tolerance value is greater than 0.10.
c. Heteroscedasticity test: It was conducted to analyse the residual variance from one observation to another
observation. The decision is based on the significance value. If the Significance value is greater than 0.05,
heteroscedasticity is not shown.
d. Linearity Test: This test was conducted to analyse the level of significant linear relationship. The basis for the
decision is that when the Sig value is > 0.05, the correlation between the independent variables and dependent
variable is significant.
C. Inferential Statistical Analysis
The Inferential statistical analysis was carried out by using path analysis with Windows SPSS version 26.
D. Hypothesis Test
In the hypothesis test, direct and indirect effect of independent variables on dependent variable was analysed. It
includes:
H0: Independent variables (X1 and X2) have no direct effect on dependent variable (Y2).
H1: Independent variables (X1 and X2) have direct effect on dependent variable (Y2).

IV. RESEARCH RESULT

Questionnaire was used to collect data, distributed to 80 students at Higher Schools in Makassar City, Indonesia.
Among them was only 70 returned of which 5 were flawed. Therefore, only 65 were processed, and the data were
analyzed using SPSS software version 26. The results were studied for the validity and reliability of the instrument.
Concerning the criteria used to determine validity, when the significance value was below 0.05, the items were
considered valid. Otherwise, when the value was above 0.05, the items were not considered valid. The validity test of
the instrument showed that the significance value of all the questionnaire items was below 0.05, which means the
instrument was valid.
Regarding the reliability test, the criterion was that when the Cronbach alpha value was above 0.70, the instrument
was considered reliable. The reliability test of this instrument denoted that the output of SPSS version 26 showed
Cronbach alpha of 0.823, and was above 0.70. Therefore, the questionnaire satisfied the reliability criteria.
TABLE 1
THE OUTPUT OF RELIABILITY TEST
RELIABILITY STATISTICS
N of
Cronbach’s Alpha Items
.823 16

The validity and reliability tests were conducted, and the classical assumption test was carried out, including
normality, multicollinearity, linearity, and heteroscedasticity tests. After the normality test, the output showed that the
Asymp.Sig, (2-tailed) showed the value of 0.200, greater than 0.05 (0.200 > 0.05). This indicated a normal distribution
of the residual variance.
TABLE 2
THE RESULT OF NORMALITY TEST
Unstandardized Residual
N 65
Normal Parameters,b Mean .0000000
Std. Deviation 1.30862592
Most Extreme Differences Absolute .062
Positive .059
Negative -.062
Test Statistic .080
Asymp. Sig. (2-tailed) .200c,d
a. The test distribution is normal
b. Calculated from data
c. Lilliefors Significance Correction.
d. This is a lower bound of the true significance.

The multicollinearity test was carried out to reveal the linear correlation of the independent variables (reading and
feedback). Furthermore, tolerance value of the variables and the value of VIF were compared with 0.10 and 10.00. The
output of SPSS version 26 showed that the tolerance value of each variable (X1: 0.904, X2: 0.722) was more significant
than 0.10, and the value of VIF (X1: 1.106, X2: 1.386) was less than 10.00. Therefore, there was no linear correlation

© 2024 ACADEMY PUBLICATION

572 JOURNAL OF LANGUAGE TEACHING AND RESEARCH

between one independent variable and the other one. It means that there was no multicollinearity problem.
TABLE 3
THE RESULT OF MULTICOLLINEARITY TEST
COEFFICIENTS
Unstandardized Coefficients Standardized Coefficients Collinearity Statistics
Model B Std. Error Beta T Sig. Tolerance VIF
1 (Constant) 6.133 1.755 3.494 .001
Rd -.119 .104 -.111 -1.150 .255 .904 1.106
Fb . -.017 .069 -.027 -.251 .803 .722 1.386
Att .673 .097 .724 6.948 .000 .769 1.300
a. Dependent Variable: Wr

The heteroscedasticity test was conducted to determine the regression model variance in similarity. A good model
does not show a heteroscedasticity. The conclusion was made based on the followings:
 Value of Sig> 0.05 = there was no heteroscedasticity
 Value of Sig<0.05 = there was heteroscedasticity
The table of coefficient in column sig. shows the sig. value of 0.874 for X1, 0.712 for X2 and 0.927 for Y1. The
regression model was bigger than 0.05 which means that there was no heteroscedasticity.
TABLE 4
THE OUTPUT OF HETEROSCEDASTICITY TEST
COEFFICIENTS
Unstandardized Standardized
Coefficients Coefficients Collinearity Statistics
Model B Std. Error Beta t Sig. Tolerance VIF
1 (Constant) 1.420 1.305 1.088 .281
Rd .012 .077 .021 .160 .874 .904 1.106
Fb -.019 .051 -.056 -.370 .712 .722 1.386
Att .007 .072 .013 .092 .927 .769 1.300
a. Dependent Variable: RES2

The linearity test was carried out to determine the significant linear correlation between the independent and
dependent variables. The criteria to determine the correlation was sign. Value greater than 0.05 which was indicating a
linearity in the correlation between the independent and dependent variables. The Anova analysis denoted that the
deviation from linearity showed the value of sign. 0.999 which was greater than 0.05. It means that there was lineaity in
the correlation between the independent variables and dependent variable.
TABLE 5
THE OUTPUT OF THE LINEARITY TEST
ANOVA TABLE
Sum of Squares Df Mean Square F Sig.
WR * FB Between Groups (Combined) 1.441 7 .206 .055 1.000
Linearity .046 1 .046 .012 .912
Deviation from Linearity 1.395 6 .232 .062 .999
Within Groups 213.698 57 3.749
Total 215.138 64

The independent variables in this research were Reading (X1) and Feedback (X2), the dependent variable was Writing
(Y2), and the intervening variable was Attitude (Y1). Path analysis was conducted twice, and produced two models:
Model 1 and Model 2. Those two models were discussed below.
Model 1
In this Model 1 construction, Reading (X1) and Feedback (X2) were the independent variables and Attitude (Y1) was
the dependent variable. The result of the analysis using SPSS version 26 was shown in the Table 6 below.
TABLE 6
MODEL 1
COEFFICIENTS
Unstandardized Coefficients Standardized Coefficients
Model B Std. Error Beta T Sig.
1 (Constant) 11.865 1.740 6.819 .000
Rd .048 .136 .041 .354 .724
Fb .321 .081 .466 3.980 .000
a. Dependent Variable: ATT

© 2024 ACADEMY PUBLICATION

JOURNAL OF LANGUAGE TEACHING AND RESEARCH 573

The value of standardized coefficient of the independent variables as the path coefficient of X1 and X2 is presented in
Table 6 above. The part coefficient matrix was as follows:

The value of determinant coefficient or the R square value was indicated by the model summary in the table below:
TABLE 7
DETERMINANT COEFFICIENT (R2) OF MODEL 1
MODEL SUMMARY
Model R R Square Adjusted R Square Std. Error of the Estimate
1 .480a .231 .206 1.758
a. Predictors: (Constant), Fb, Rd
b. dependent variable: Attitude

The R square value was 0.231. The path coefficient of X1 and X2 was manually calculated by a part matrix,
multiplied by Y1. The R square, pY1ɛ, calculates the coefficient of variables outside the model.

1) Path Coefficient of ρY1X1

For the path coefficient of pY1X1, namely from X1 to Y1, the column sig. in table 6 was 0.724, which was greater than
0.05. The value of tcalculation 0.354 was smaller than the value of ttable 1.669. It means that H0 was accepted and H1 was
rejected which implied that the path coefficient of X1 to Y1 was insignificant.
2) Path Coefficient of ρY1X2
Related to the path coefficient from X2 to Y1 (pY1X2), the value of sig. was 0.000 which was less than the value of
0.05. Meanwhile, the value of tcalculation 3,980 was greater than the value of ttable 1.669. Therefore, H0 was rejected, and
H1 accepted. It indicated that the path coefficient from X2 to Y1 was significant.
Model 2
In this Model 2, the dependent variable was writing skills (Y2) and the independent variables were Reading skill (X1),
Feedback (X2), and students’ attitude (Y1). The result of the analysis by SPSS in this Model 2 is presented below.
TABLE 8
MODEL 2
COEFFICIENTS
Unstandardized Coefficients Standardized Coefficients
Model B Std. Error Beta T Sig.
1 (Constant) 6.133 1.755 3.494 .001
Rd -.119 .104 -.111 -1.150 .255
Fb -.017 .069 -.027 -.251 .803
Att .673 .097 .724 6.948 .000
a. Dependent Variable: Wr

Standardized coefficient values as shown in the Table 8 above indicated that coefficient value for Reading (X1) was
0.111, Feedback (X2) was 0.027, and Attitude (Y1) was 0.724. The path coefficient matrix was:

TABLE 9
DETERMINANT COEFFICIENT (R2) OF MODEL 2
MODEL SUMMARY
Model R R Square Adjusted R Square Std. Error of the Estimate
1 .700a .491 .466 1.340
a. Predictors: (Constant), Att, Rd, Fb
b. Dependent Variable: Wr

The R Square value of 0.491 was manually calculated by changing the path coefficients of the independent variables
(X1, X2, and Y1) to the path matrix and multiplied by the column matrix Y2. Path coefficient of other variables outside
the model (ρY2Ɛ) was calculated by:

© 2024 ACADEMY PUBLICATION

574 JOURNAL OF LANGUAGE TEACHING AND RESEARCH

1) Path Coefficient of ρY2X1

The path coefficient from path X1 to Y2 indicated that the column sig. in table 8 was 0.255, which was greater than
0.05, and the value of tcalculation 1,150 was less than the ttable 1.663. In this case, H0 was accepted and H1 rejected,
meaning the path coefficient of X1 to Y2 was statistically insignificant.
2) Path Coefficient of ρY2X2
The path coefficient from X2 to Y2 indicated that the value of sig., as shown in table 8, was 0.803, which was greater
than 0.05, and the value of tcalculation 0,251 was less than the ttable 1.663. Hence, H1 was rejected and H0 accepted,
meaning the path coefficient of X2 to Y2 was insignificant.
3) Path Coefficient of ρY2Y1
In path coefficient from the path Y1 to Y2, the value of sig. was 0.000, which was greater than 0.05, and the value of
tcalculation 6,948 was bigger than the value of ttable 1.663. Hence, H0 was rejected and H1 was accepted, meaning the path
coefficient of Y1 to Y2 was significant.
Test of Hypothesis
The effect of reading (X1) on students' learning attitude (Y1) was tested. The analysis of the path coefficient of ρY1X1
indicated that the significant value of X1 was 0.724, and was greater than 0.05, and the value of tcalculation (1,150) was less
than the value of ttable (1.663) or tcalculation<ttable. This showed reading had no effect on students' learning attitude, hence,
Hypothesis 1: “Reading has a significant effect on Attitude” was rejected.
To reveal whether the Reading skill (X1) could affect the students’ writing competence ((Y2), the result of the data
analysis indicated that the value of X1 was 0.255, which was greater than 0.05, and tcalculation (1.150) was less than the
value of ttable (1.663) or tcalculation < ttable. It was concluded that X1 had no direct effect on Y2. Therefore, the Hypothesis 2,
Reading has significant effect on Writing, was rejected which proved that the students’ writing competence could not be
improved by the reading skill.
With regard to the contribution of Reading skill (X1) to the improvement of students Writing skill (Y2), the data
analysis showed that Reading (X1) has significant value 0.255, which was greater than 0.05, and tcalculation (1.150) was
less than the value of ttable (1.663) or tcalculation < ttable. It was concluded that X1 had no direct effect on Y2. Therefore, the
Hypothesis 2, Reading has significant effect on Writing, was rejected. It means that good competence on Reading could
not boost directly the improvement of students’ writing competence.
Indirect effect of Reading (X1) on Writing (Y2) mediated by Attitude (Y1) was analyzed by multiplying the β value of
Reading (X1) on Attitude (Y1) and the β value of Attitude (Y1) on Writing (Y2). It was noted in table 6 that the β value
of Reading (X1) on Attitude (Y1) was 0.111 and the β value of Attitude (Y1) on Writing (Y2) was 0.724. The
multiplication was then made, 0.111 x 0.724 = 0.027. The total effect of Reading (X1) on Writing (Y2) was the sum of
the direct and indirect effects, 0.041+0.027= 0.068. The result indicated that the indirect effect was greater than the
direct effect, which means that the Hypothesis 3, Reading has an indirect effect through Attitude on Writing, was
accepted. In this case, students’ attitude in learning English can bolster the Reading skill to improve the writing
competence.
Regarding the analysis of the effect of feedback (X2) on Attitude (Y1), the result showed that the significant value of
X2 was 0.000, which was less than 0.05, and the value of tcalculation (3.980) was greater than the value of ttable (1.663) or
tcalculation > ttable. Therefore, Hypothesis 4: “Feedback has effect on Attitude” was accepted. Furthermore, the direct effect
of feedback X2 on Writing (Y2) showed the significant value of X2 (0.803) which was greater than 0.05, and the value of
tcalculation (0.251) was less than the value of ttable (1.663) or tcalculation < ttable. In this case, it did not fulfil the criteria that
feedback directly affected students’ writing improvement. Therefore, Hypothesis 5: “Feedback has a significant direct
effect on Writing” was rejected.
The Indirect effect of Feedback (X2) on Writing (Y2) mediated by Attitude (Y1) was analyzed by multiplying the β
value of Feedback (X2) on Writing (Y2) and the β value of Attitude (Y1) on Writing (Y2). The result was 0.466 x 0.724
= 0.337, and the total effect of X2 on Y2 was 0.803+0.337= 1.140. The result showed that the indirect effect value
(1,140) was greater than the direct (0.803). This implied that Attitude (Y1) could strengthen the indirect effect of
Feedback (X2) on students’ Writing competence (Y2). It means that Hypothesis 6, Feedback has an indirect effect
through Attitude on Writing, was accepted.
The effect of Attitude (Y1) on Writing (Y2) was tested, and when the significance value of Y1 was bigger than 0.05 or
the value of tcalculation was bigger than the value of ttable, Attitude (Y1) was considered to affect the Writing Competence
(Y2). The analysis result showed that the significant value of Y1 was 0.000, greater than 0.05, and the value of tcalculation
(6.948) was bigger than ttable (1.663) or tcalculation > ttable. This showed attitude affected writing, hence, Hypothesis 7:
“Attitude affects Writing” was accepted.

V. DISCUSSION
Technological developments, specifically AI, have penetrated all sectors and tend to affect human civilization. The
education sector is one of the sectors most affected, mainly related to the use of AI in teaching and learning. During the
Covid-19 pandemic, the learning process was not carried out by face-to-face interaction in class, but virtually using AI.
In the context of learning English in Higher Schools, this study was conducted to analyze the use of AI in teaching
reading and providing feedback to improve writing skills. Furthermore, students' attitude toward the use of AI was

© 2024 ACADEMY PUBLICATION

JOURNAL OF LANGUAGE TEACHING AND RESEARCH 575

investigated. AI can perform automatic tasks, and the possibilities of performing complex tasks in higher education are
unimaginable (Soto et al., 2012). Technological advancements will become a standard solution (Bengio et al., 2013) for
teaching and learning problems.
The reading course was virtually disseminated by uploading the material on Google Classroom in the form of
PowerPoint. According to Pence (2019), classes were held virtually, presented by a slide show program like PowerPoint.
Students were assigned task through Google Classroom. Lecturers provided feedback to notify students about their
mistakes in using appropriate words and the grammatical aspect. Ability to understand the reading materials people
send virtually could boostle the students to improve their writing skill. This should be boosted by the mastery of English
grammar and vocabulary, especially the vocabulary used in those reading materials. Furthermore, students’ positive
attitude toward using AI in the learning-teaching process should support their learning English progress.
This study found that virtual teaching through AI with a positive attitude had an indirect significant effect on writing
skills improvement. The total indirect effect of reading on writing was 1.140, which showed reading contributed
11.40% indirect effect to the writing skills improvement. In this case, the teaching of reading did not significantly affect
writing skills without a positive attitude of students. The improvement should be boosted by a positive attitude toward
the process. Feedback had indirect effects through attitude on writing. In this case, it cannot significantly affect writing
skills without positive attitude. This study proved that virtual teaching of reading and feedback with a positive attitude
had a significant effect on writing skills improvement. Improving writing skills requires good mastery of grammar and
vocabulary, which can be developed by reading and feedback from lecturers. Furthermore, electronic devices can
facilitate this process with the application of AI.

VI. CONCLUSION
In conclusion, students' skills in learning English using AI needed to be improved. This was because students'
attitude was shown to have a mediating role in explaining the effect of reading and teacher feedback on writing skills
development. In this case, feedback had no direct influence on the development of writing skills. Therefore, building a
positive attitude towards the use of AI in learning needs to be prioritized. Vocabulary development from reading and
remedial exercises based on lecturers' feedback improved students' writing skills.

REFERENCES
[1] Abidin & Kaharuddin. (2021). Analysis of Syariah Credit Card from Islamic Economic Perspective: Evidence from BNI Bank
in Indonesia. Academy of Strategic Management Journal, 20(4S), 1–11.
[2] Anggrawan, A., Yassi, A.H., Satria, C., Arafah, B., & Makka, H.M. (2019). Comparison of Online Learning versus Face-to-
Face Learning in English Grammar Learning. 5th International Conference on Computing Engineering and Design (ICCED),
1-4.
[3] Arafah, B. & Hasyim, M. (2023a). Digital Literacy: The Right Solution to Overcome the Various Problems of Meaning and
Communication on Social Media. Studies in Media and Communication, 11(4), 19-30.
[4] Arafah, B. & Hasyim, M. (2023b). Digital Literacy on Current Issues in Social Media: Social Media as a Source of Information.
Journal of Theoretical and Applied Information Technology, 101(10), 3943-3951.
[5] Arafah, B., Hasyim, M., Khaerana, A.S.A., Soraya, A.I., Ramadhani, R., Ismail, A., & Takwa. (2023). The Digital Culture
Literacy of Generation Z Netizens as Readers, Producers and Publishers of Text on Social Media. International Journal of
Intelligent Systems and Applications in Engineering, 11(3), 112-123.
[6] Arafah, B., & Kaharuddin. (2019). The Representation of Complaints in English and Indonesian Discourses. Opción, 35, 501-
517. https://doi.org/10.18502/kss.v3i19.4829.
[7] Arafah, B., Kaharuddin, Hasjim, M., Arafah, A. N. B., Takwa, & Karimuddin. (2023). Cultural Relations Among Speakers of
South Halmahera Languages. Theory and Practice in Language Studies, 13(1), 168-174.
[8] Arafah, B., Rofikah, U., Kaharuddin, A., Room, F., Takwa, & Purwarno, P. (2023). Evaluating the Suitability of Printed
Materials in the Context of Online Learning. XLinguae, 16(2), 16-32.
[9] Arafah, B., Sirajuddin, J., Thayyib, M., Room, F., Takwa, & Anayati, W. (2023). Emotional Management of Defoe's Robinson
Crusoe's Main Character. Journal of Language Teaching and Research, 14(5), 1414-1421.
[10] Arafah, B., Room, F., Suryadi, R., La Ode Muhammad Idrus Hamid, B., & Juniardi, Y. (2024). Character Education Values in
Pullman's the Golden Compass. Journal of Language Teaching and Research, 15(1), 246-254.
[11] Arifin, M. B., Arafah, B., & Kuncara, S.D. (2022). Dayak’s Sociocultural Situation through Locality in Lumholtz’s Through
Central Borneo Travel Writing. Theory and Practice in Language Studies, 12(12), 2695-2703.
[12] Arnawa, I. G. N. E. V. & Arafah, B. (2023). Students’ Self-Regulated Strategies in Approaching Second Language Writing.
Theory and Practice in Language Studies, 13(3), 690-696.
[13] Asri, D., Arafah, B., Sahib, H., & Abbas, H. (2023). Male Domination in Helen Garner’s Monkey Grip. Theory and Practice in
Language Studies, 13(7), 1651-1658.
[14] Baa, S., Wardani, S. B., Iskandar, Weda, S., & Arafah, B. (2023). Lexical Metaphors in Westlife's Selected Song Lyrics.
XLinguae, 16(1), 132-154.
[15] Bengio, Y., Courville, A., & Vincent, P. (2013). Representation learning: A review and new perspectives. IEEE transactions on
pattern analysis and machine intelligence, 35(8), 1798-1828.
[16] Boggs, J. A. (2019). Effects of Lecturer-Scaffolded and Self-Scaffolded Corrective Feedback Compared to Direct Corrective
Feedback on Grammatical Accuracy in English L2 Writing. Journal of Second Language Writing, 46, 1–34.
[17] Boud, D., & Molloy, E. (2013). Rethinking models of feedback for learning: the challenge of design. Assessment & Evaluation

© 2024 ACADEMY PUBLICATION

576 JOURNAL OF LANGUAGE TEACHING AND RESEARCH

in Higher education, 38(6), 698–712.

[18] Caplar, N., Tacchella, S., & Birrer, S. (2017). Quantitative Evaluation of Gender Bias in Astronomical Publications from
Citation Counts. Nature Astronomy, 1(6), 1-5.
[19] Cavanaugh, A. J., & Song, L. (2014). Audio Feedback versus Written Feedback: Instructors’ and Students’ Perspectives.
Journal of Online Learning and Teaching, 10(1), 122–138.
[20] Chandler, J. (2003). The efficacy of various Kinds of Error Feedback for Improvement in the Accuracy and Fluency of L2
Student Writing. Journal of Second Language Writing, 12(3), 267–296.
[21] Corwin, R. G. (1976). “Dan C. Lortie”: School Lecturer (Book Review). Social Forces, 54(4), 29-48.
[22] Cotos, E. (2011). Potential of Automated Writing Evaluation Feedback. Calico Journal, 28(2), 420–459.
[23] Durukan, E. (2011). Effects of cooperative integrated reading and composition (CIRC) technique on reading-writing skills.
Educational Research and Reviews, 6(1), 102–109.
[24] Esmaeili, H. (2002). Integrated reading and writing tasks and ESL students' reading and writing performance in an English
language test. Canadian Modern Language Review, 58(4), 599–620.
[25] Fahimirad, M., & Kotamjani, S. S. (2018). A review of the application of artificial intelligence in teaching and learning in
educational contexts. International Journal of Learning and Development, 8(4), 106-118.
[26] Farooq, M. U., & Javid, C. Z. (2012). Attitude of students towards E-learning: A study of English language learners at Taif
University English Language Centre. NUML Journal of Critical Inquiry, 10(2), 17-28.
[27] Govindasamy, T. (2001). Successful implementation of e-learning: Pedagogical considerations. The internet and higher
education, 4(3-4), 287-299.
[28] Guven Ozdemir, N., & Sonmez, M. (2021). The relationship between nursing students’ technology addiction levels and
attitudes toward e‐learning during the COVID‐19 pandemic: A cross‐sectional study. Perspectives in Psychiatric Care, 57(3),
1442-1448.
[29] Hasjim, M., Arafah, B., Kaharuddin, S. V., & Genisa, R. A. A. (2020). Principles behind Semantic Relation between Common
Abbreviations and their Expansions on Instagram. International Journal of Criminology and Sociology, 9, 2270-2276.
[30] Hasyim, M. & Arafah, B. (2023a). Semiotic Multimodality Communication in the Age of New Media. Studies in Media and
Communication, 11(1), 96-103.
[31] Hasyim M. & Arafah, B. (2023b). Social Media Text Meaning: Cultural Information Consumption. Wseas Transactions on
Information Science and Applications, 20(2023), 220-227.
[32] Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112.
[33] Henderson, M. et al. (2019). Conditions that enable effective feedback. Higher Education Research & Development, 38(7),
1401-1416.
[34] Henderson, M., & Phillips, M. (2014, September). Technology enhanced feedback on assessment. In Australian Computers in
Education Conference.
[35] Hirvela, A. (2004). Connecting reading & writing in second language writing instruction. University of Michigan Press.
[36] Horning, A. S., & Kraemer, E. W. (Eds.). (2013). Reconnecting reading and writing. Parlor Press LLC.
[37] Iksora, Arafah, B., Syafruddin, S., Muchtar, J., & Lestari, P.A. (2022). Typos’ Effects on Web-Based Programming Code
Output: A Computational Linguistics Study. Theory and Practice in Language Studies, 12(11), 2460-2469.
[38] Ismail., Sunubi, A. H., Halidin, A., Nanning & Kaharuddin. (2020). Paraphrasing Technique to Develop Skill for English
Writing Among Indonesian College Students of English. Systematic Review in Pharmacy, 11(11), 291–297.
[39] Jolliffe, W. (2007). Cooperative learning in the classroom: Putting it into practice. Sage.
[40] Kaharuddin, K., Ahmad, D., Mardiana, M., & Rusni, R. (2021). Contributions of Technology, Culture, and Attitude to English
Learning Motivation during Covid-19 Outbreaks. Systematic Reviews in Pharmacy, 11(11). 76-84.
[41] Kaharuddin, Mardiana, N., Ahmad, D., Sari, A.A.I. (2022). Examining the Skill in Writing Descriptive Text among Indonesian
Learners of English: The Effects of Task-Based Language Teaching (TBLT). Journal of Language Teaching and Research,
13(1), 46-57.
[42] Kaharuddin (2022). E-Learning During the COVID-19 Outbreak: The Effect of the Grammar Translation Method and the
Direct Method on Students’ English Competence, Journal of Language Teaching and Research, 3(2), 271–278.
[43] Kaharuddin, Arafah, B., Nurpahmi, S., Sukmawaty, S., Rahman, I. F., & Juniardi, Y. (2023). Exploring How Reading Aloud
and Vocabulary Enrichment Shape English Speaking Skills Among Indonesian Learners of English. World Journal of English
Language, 13(8), 436-445.
[44] Khadawardi, H. A. (2021). The Effect of Implicit Corrective Feedback on English Writing of International Second Language
Learners. English Language Teaching, 14(1), 123-139.
[45] Keengwe, J., Onchwari, G., & Wachira, P. (2008). The use of computer tools to support meaningful learning. AACE Review
(formerly AACE Journal), 16(1), 77-92.
[46] Kuswanty, W.H., Arafah, B., Budiman, A.N.A., Ali, T., Fatsah, H., & Room, F. (2023). Students’ Perception of Explicit and
Implicit Methods in Learning Tenses in SMP DDI Mangkoso. Theory and Practice in Language Studies, 13(6), 1473-1482.
[47] Lebba & Kaharuddin. (2023). The Meaning of Death Ceremony ‘Rambu Solo’ in Tana Toraja, Kurdish Studies, 11(2), 3958-
3967.
[48] Lebba, Kaharuddin, Arafah, B., Nur, S., & Mashuri, S. (2003). The Role of Santri Maghrib Recitation Movement in Islamic
Axiology, Kurdish Studies, 11(2), 5624–5633.
[49] Male, H. et.al. (2020). Attitude of Undergraduate Students Towards Online Learning During Covid-19 Pandemic. PalArch’s
Journal of Archaeology of Egypt/Egyptology, 17(4), 1628–1637.
[50] Manugeren, M., Arafah, B., Purwarno, P., Siwi, P., Ekalestari, S., & Wulan, S. (2023). An Ecoliterature Approach to
Environmental Conservation: Take Four Selected Literary Works as Examples. Theory and Practice in Language Studies,
13(5), 1318-1327.
[51] Mardiana, Arafah, B., Ahmad, D., Kaharuddin, Room, F., & Barus, E. (2023). Time Allocation Effect on Test Scores for
Academic Writing of Indonesian English Learners. Journal of Language Teaching and Research, 14(6), 1628-1636.

© 2024 ACADEMY PUBLICATION

JOURNAL OF LANGUAGE TEACHING AND RESEARCH 577

[52] Mokoginta, K. & Arafah, B. (2022). Negotiation in Indonesian Culture: A Cultural Linguistic Analysis of Bahasa Indonesia
Textbooks. Theory and Practice in Language Studies, 12(4), 691-701.
[53] Mozgovoy, M. (2011, September). Dependency-based rules for grammar checking with LanguageTool. In 2011 Federated
Conference on Computer Science and Information Systems (FedCSIS) (pp. 209-212). IEEE.
[54] Newton, R. (2003). Staff attitudes to the development and delivery of e‐learning. New library world.
[55] Pence, H. E. (2019). Artificial Intelligence in Higher Education: New Wine in Old Wineskins?. Journal of Educational
Technology Systems, 48(1), 5–13.
[56] Renandya, W. A. (2007). The power of extensive reading. RELC Journal, 38(2), 133-149.
[57] Siwi, P., Arafah, B., Wulan, S., Purwarno, P., Ekalestari, S., & Arafah, A.N.B. (2022). Treatment of Nature: An Ecocriticism
Approach in ‘Komat Kamit’ of Tejo and Kamba’s Tuhan Maha Asik. Theory and Practice in Language Studies, 12(7), 1278-
1285.
[58] Soto, J. A., et al. (2012). Strength in numbers? Cognitive reappraisal tendencies and psychological functioning among Latinos
in the context of oppression. Cultural Diversity and Ethnic Minority Psychology, 18(4), 384-394.
[59] Suhadi, J., Arafah, B., Makatita, F.P., Abbas, H., & Arafah, A.N.B. (2022). Science and Society: The Impact of Science Abuse
on Social Life in Well’s The Invisible Man. Theory and Practice in Language Studies, 12(6), 1214-1219.
[60] Takwa, Arafah, B., Kaharuddin, Putra, E., Masrur, & Arafah, A.N.B. (2022). The Shift of Lexicon in Traditional Technology
System in Tolaki Community at Konawe District of Southeast Sulawesi. Theory and Practice in Language Studies, 12(5), 980-
989.
[61] Takwa, Arafah, B., Sopiandy, D., Taqfiah, S. J. A., & Arafah, A. N. B. (2022). Humanistic Values in Metaphoric Expressions
of Traditional Marriage in Tolaki Mekongga Kolaka. Theory and Practice in Language Studies, 12(8), 1602-1608.
[62] Tsui, A. B., & Tavares, N. J. (2021). The Technology Cart and the Pedagogy Horse in Online Teaching. English Teaching &
Learning, 45(1), 109-118.
[63] Vasiljeva, T., Kreituss, I., & Lulle, I. (2021). Artificial Intelligence: The Attitude of the Public and Representatives of Various
Industries. Journal of Risk and Financial Management, 14(8), 1-17.
[64] Yudith, M., Arafah, B., Sunyoto, F.G., Fitriani, Rostan, R.B., & Nurdin, F.E. (2023). The Representation of Animalism Issue in
Sewell’s Black Beauty. Theory and Practice in Language Studies, 13(1), 108-116.
[65] Yulianti, S., Arafah, B., Rofikah, U., Idris, A.M.S., Samsur, N., & Arafah, A.N.B. (2022). Conversational Implicatures on
Saturday Night Live Talk Show. Journal of Language Teaching and Research, 13(1), 189-197.

Kaharuddin, an Associate Professor in English Language Teaching, was born in Makassar, Indonesia, on
August 14, 1975. He got his Master's degree in English Language Studies in 2004 and his doctoral degree in
Applied Linguistics in 2014, both from Hasanuddin University of Makassar, Indonesia. He is currently
teaching English Language and Linguistics at the English Education Department, Faculty of Education and
Lecturer Training, Universitas Islam Negeri (UIN) Alauddin Makassar, Indonesia. He is the author of a
number of books, including: Syllabus Design for English Language Teaching, Prenada Media (2018); It’s All
About Transactional Speaking Skills, Eliva Press, Moldova Europe (2021); and Linguistics: Basic theory and
Roles in ELT, Prenada Media (2021). His research focuses on English education and Linguistics.
Dr. Kaharuddin is currently a member of ASIA TEFL Association, Busan, South Korea.

Djuwairiah Ahmad, a Professor in English Education, was born in Makassar on June 19, 1970. She
completed her undergraduate study in the English Education Department of IKIP Ujung Pandang, Indonesia,
in 1992. She got her Master of Education majoring in English Education at State University of Makassar,
Indonesia, in 2000. Then, she achieved her second master degree in TESOL at Victoria University,
Melbourne-Australia in 2006. Lastly, she got her doctoral degree in Linguistics at the Faculty of Letters,
Postgraduate Program of Hasanuddin University, Indonesia, in 2014.
Currently, she is the head of Language Center of Alauddin State Islamic University of Makassar, Indonesia.
She was previously the head of English Education Department for two consecutive periods (years of 2008 to
2015) and the secretary of the same department from 2000 to 2004. A number of books, written works, and
published articles that have been produced by the writer include: Translation (Book, 2013), Effect of Time Allotment on Test Scores
for Academic Writing of Indonesian Learners of English (Journal of Multicultural Education, 2021), Exploring Student Achievement
and Perceptions in an Online Flipped Grammar Course (Indonesian Journal of Applied Linguistics, 2021). Her previous research
interests were in English teaching and curriculum, materials development, and language assessment.
Prof. Ahmad is a member of Indonesian English Lecturers Association (TEFLIN), Asia TEFL, Indonesian Linguistic Society
(MLI), and Association of America-Indonesia Scholarship Awardee Alumni (ALPHA-I).

Mardiana is a lecturer in English Language Teaching at the English Education Department, Faculty of
Education and Teacher Training, Universitas Islam Negeri (UIN) Alauddin Makassar. She was born on August
15, 1969, in Pinrang, South Sulawesi. She studied at the Institute of Teacher Training and Education (IKIP) in
Ujung Pandang, earning a bachelor's degree in English education in 1993. She earned a Master's Degree in
English Education Department after completing her studies at Hasanuddin University (UNHAS) in 1996 and
working as an English Lecturer at UIN Alauddin Makassar since 1999. She then obtained her doctorate at
Universitas Negeri Jakarta (UNJ) in 2012. Therefore, her strong interest in English education has provided her
with numerous opportunities to participate in professional training programs at foreign universities, such as

© 2024 ACADEMY PUBLICATION

578 JOURNAL OF LANGUAGE TEACHING AND RESEARCH

four months. There are short courses at Ohio State University in Columbus and two-month courses at Gottingen University in
Germany. She also spent two months in Canada (Montreal, Ottawa, and Kitchener) for a community outreach program and published
inrespectable publications.

Ismail Latif is a lecturer at the English education study program of Institut Agama Islam Negeri Parepare. He
was born in Bottoe Barru South Sulawesi on December 1963. After graduating from high school in 1981, he
attended IAIN Alauddin Ujung Pandang from which he received a Drs degree in English education, and
then he received a Magister degree in Management of Human Resources at STIE PATRIA ARTHA
Makassar in 2003. He has written some academic journals such as, to name only the recent ones, The
Effectiveness of Using Modifying Reading Text to Develop the First Year Students’ Reading Skills at MAN 1
Parepare (2014), Using Pictures to Increase the Second Year Students’ vocabulary Mastery at MTs
Addariyah DDI Kaballangang Pinrang (2014). His research interest centers on English education such as
Analysing Students’ Needs in English Course for Students of Non-English Study Program at STAIN Parepare
(2010). Integrating Moral Characters to Develop ELT Materials for Transactional Speaking Skills (2015).
Besides, he has participated in some international seminars, symposiums, and conferences.

Burhanuddin Arafah obtained his PhD in English literature at the University of Newcastle, Australia, in
2003. He earned his Master's degree in American literature at Gadjah Mada University (UGM) Yogyakarta of
Indonesia in 1995 and his Bachelor's degree in English literature at Hasanuddin University (UNHAS) in 1988.
He is currently a full Professor in English literature at the English Literature Study Program, Faculty of
Cultural Sciences of UNHAS, Indonesia. He has published eight books in English and literature and more
than 70 research articles ranked in international journals published in English. He also has received 26
Intellectual Property Right Certificates from the Indonesian government.
His areas of interest are English literature, language education, and cultural studies. He was the Dean of
the Faculty of Cultural Sciences of UNHAS from 2009-2017. He is actively involved at the National
Accreditation Board-Higher Education, Ministry of Education and Culture of the Republic of Indonesia in his position as Assessor.

Ray Suryadi is a lecturer at the English Department of the Faculty of Teacher Training and Education,
Sembilanbelas November University Kolaka (USN). He completed his Bachelor's degree in English Education
in 2008 and his Master's in 2013 at Makassar State University. In 2023, he published two articles in national
journals, namely The Impact of Kadatua Dialect on Students' English Pronunciation (Lingual, Vol. 15, No. 1)
and Student's Perception on Learning Media Augmented Reality based at English for Specific Purpose Subject
(Epigram, Vol. 20, No. 1). He has also written some books such as English for Physical Education and Sport,
Skill of Speaking, Travel Around the World with English, Learning Physics and Chemistry with English.

© 2024 ACADEMY PUBLICATION

Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.