JURNAL PENDIDIKAN MATEMATIKA DAN IPA
Vol.
No.
http://jurnal.
id/index.
php/PMP EFFECTIVENESS OF USING META AI IN IMPROVING
UNDERSTANDING OF ALGEBRA CONCEPTS BY GRADE
VII STUDENTS AHMAD YANI JUNIOR HIGH SCHOOL
Feri Prastyo1.
Wildan Hakim2 Universitas Al-Qolam Malang.
Jl.
Raya Ketawang.
Dusun Baron.
Putat Lor.
Kec.
Gondanglegi.
Kabupaten Malang.
Jawa Timur 65174 E-mail: wildan@alqolam.
DOI: http://dx.
org/10.
26418/jpmipa.
Abstract This study aims to analyze the effectiveness of using Meta AI in improving the understanding of algebra concepts of seventh-grade students at Ahmad Yani Pagelaran Junior High School.
The background of the problem shows that students have difficulty understanding basic algebra concepts, including the introduction of variables, coefficients, constants, and simplification of algebraic equations due to the use of conventional learning methods that are less interactive.
The study used a quasi-experimental design with 42 students divided into an experimental group .
sing Meta AI) and a control group .
onventional metho.
Data were collected through pretests, posttests, and learning observations, then analyzed using descriptive and inferential statistics with t-tests and N-Gain The results showed a significant difference between the two groups .
< 0.
The experimental group achieved an average posttest of 83 with an N-Gain of 73.
4% .
igh categor.
, while the control group only achieved 35 with an N-Gain of 17% .
ow The classical completeness of the experimental group reached 100%, exceeding the minimum standard of 85%.
Meta AI has been shown to increase student engagement by up to 90%, reduce procedural errors by 70%, and improve the accuracy of identifying algebraic components by up to 85%.
The study concluded that Meta AI is highly effective in improving understanding of algebraic concepts through adaptive, personalized, and interactive learning tailored to individual students' needs.
Keywords:Meta Ai.
Algebra Learning.
Artificial Intelligence.
Educational Technology.
Received Revised Accepted : 2025-08-13
: 2026-01-02
: 2026-01-03
This work is licensed under a Creative Commons Attribution 4.
0 International License Jurnal Pendidikan Matematika dan IPA Vol.
No.
INTRODUCTION
Mathematics is an essential core subject in education, where students are taught not only numerical and symbolic concepts but also logical, analytical, and methodical thinking Algebra is one of the fundamental topics taught in junior high school, with key components including variables, coefficients, constants, and terms in algebraic Understanding algebraic expressions is crucial because it forms the foundation for understanding more advanced mathematics and has broad applications in everyday life.
Through algebra, students can transform realworld problems into mathematical models that are easier to analyze and solve, while developing critical, creative, and abstract thinking .
states that algebra is a key subject in shaping students' character because it fosters important skills such as critical thinking, creativity, reasoning, and abstract thinking.
However, observations at Ahmad Yani Junior High School showed that seventh-grade students had difficulty concepts, including the introduction of variables, coefficients, constants, and simplification of algebraic equations.
This difficulty negatively impacted students' mathematical skills and their Previous research has shown that students' proficiency in algebraic reasoning is crucial for their mathematical problems, both in academic and real-world contexts, so new teaching methods are needed to improve students' understanding.
Based on direct observation findings, many students have difficulty understanding the basic concepts of algebra from the time they first learn the material until when they need to simplify algebraic expressions or solve related problems.
Jupri noted that students' challenges in algebra often stem from difficulty understanding basic concepts, particularly variables and coefficients.
This lack of understanding leads students to repeat the same mistakes when solving A common problem is the inability to distinguish variables from coefficients, which leads to errors in problem-solving because students don't realize that variables are unknown values or common symbols for numbers.
As a result, students often make mistakes that lead to poor academic performance.
One major contributing factor to this problem is the reliance on conventional teaching styles that use lecture formats.
This approach reduces student engagement in learning because it focuses on oneway communication, where students receive information without the opportunity to ask questions or participate in discussions.
This results mathematical concepts and makes students passive and less engaged in The lack of interaction in a lecture format means students do not have sufficient opportunities to explore and form their own understanding of mathematics.
As a mathematical concepts tends to be shallow, even though mastering in-depth knowledge and regular practice.
Students enthusiasm for learning mathematics due to limited interactive educational .
Those who are less Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
engaged in the learning experience struggle to develop the critical and inventive thinking skills essential for solving math problems.
This situation creates an urgent need for innovation in teaching methods to help students grasp algebraic concepts.
The application of technologies such as artificial intelligence can foster an active and personalized learning environment, tailored to each student's Integrating AI technology into an educational framework allows for the creation of content tailored to the student's learning pace, providing immediate feedback and interactive tasks to help students learn complex subjects like algebra.
MetaAI offers a more stimulating and individualized conventional methods.
This AI system provides adaptive learning features that can adapt to each student's pace and learning style while providing immediate feedback.
Students are motivated to engage more actively in learning because they can progress at their own pace.
The application of Meta AI in education is expected to traditional, less engaging lecture With its diverse features.
Meta AI helps educators present content in a more engaging manner, encourages conversation, and offers questions tailored to students' individual needs.
The application of AI technology in education has been shown to improve the overall engagement and Educators have the ability to track student progress in real .
and modifying materials and practice questions to better meet each student's unique needs, resulting in a more adaptive and responsive learning The AI system can generate practice questions tailored to the student's skill level with immediate feedback that allows students to immediately recognize their errors.
This understanding and helps students gradually develop algebra skills.
This personalized feature is key to addressing individual difficulties students experience in understanding basic algebra concepts.
The use of technology-based teaching methods such as Meta AI is crucial for improving the quality of mathematics education at Ahmad Yani Junior High School.
This strategy aims to help students better understand algebra, reduce errors when simplifying algebraic expressions, and improve overall learning outcomes.
This new approach not only benefits students but also supports teachers in making the learning experience more effective.
The implementation of Meta AI is expected to revolutionize mathematics learning, making it more interactive and Therefore, this study focuses on evaluating the effectiveness of using Meta AI in the context of algebra learning for seventh-grade .
Based on the background of the problem that has been described, this study aims to provide empirical evidence regarding the effectiveness of using Meta AI in improving the understanding of algebraic concepts of seventh-grade students at Ahmad Yani Middle School.
The focus of the study is directed at how Meta AI can overcome students' difficulties in understanding variables, coefficients.
Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
constants, and simplification of algebraic expressions.
This study is expected to contribute to the development of innovative and The results obtained will provide practical guidance for educators in integrating AI technology into mathematics learning, while also providing a theoretical basis for further research in the field of artificial technology-based This study is limited to the study of algebraic representations for seventh-grade students at Ahmad Yani Pagelaran Middle School, with Meta AI as the only artificial intelligence platform used.
Understanding algebraic concepts is defined as students' ability to recognize, explain, and solve problems related to basic algebraic concepts, as well as simplify algebraic expressions.
Learning using Meta AI is a mathematics learning process with the help of an AI system as the main medium, providing interactive and personalized materials, exercises, and feedback.
Learning effectiveness is measured based on the achievement of learning objectives, namely a significant increase in students' understanding of algebraic concepts after participating in Meta AI-based learning compared to conventional learning, so that it can make a real contribution to innovation in mathematics education.
METHOD
Research Design This study used a quantitative approach with a quasi-experimental design that aimed to examine the causal relationship between the implementation of Meta AI as an improvement in understanding of algebraic concepts as a dependent The quasi-experimental design was chosen because the study was conducted in a real classroom environment with naturally formed groups, so full randomization was not This methodology still allowed the researcher to control confounding variables as much as possible and make an accurate assessment of the effectiveness of Meta AI.
The study consisted of three main stages: .
a pre-experimental stage with a pretest administered to both groups to evaluate students' initial abilities, .
a treatment stage where the experimental group used Meta AI while the control group used conventional methods, and .
a postexperimental stage with a posttest administered to assess the effects of the Initial disparities in students' abilities before the treatment were addressed by adding a pre-test to both groups.
Location and Time of Research The research was conducted at Ahmad Yani Pagelaran Junior High School, located on Jl.
KH.
Syuhud Zayyadi.
Karangsuko Village.
Pagelaran District.
Malang Regency.
The research took place in the second semester of April 2025.
The location was selected based on the availability of technological facilities that support the implementation of MetaAI and the homogenous characteristics of the Population and Sample The study population consisted of all 7th grade students of Ahmad Yani Pagelaran Junior High School in the 2024/2025 academic year, totaling 42 students from two classes.
Students Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
backgrounds and academic abilities, allowing for a fair comparison between the control and experimental groups.
The purposive sampling to ensure that both groups had comparable characteristics before the experiment.
The sample consisted of two 7th grade classes with an estimated 21-33 students per class, with one class serving as the experimental group and the other as the control group.
Research Variables The independent variable in this study is the implementation of Meta AI as an interactive learning medium, with the indicator of student learning effectiveness measured by the increase in understanding of algebraic concepts after using the medium.
The dependent variable is the understanding of algebraic concepts of seventh-grade students at Ahmad Yani Middle School, which is the main focus for evaluation as a result of the implementation of Meta AI.
The main aspects assessed include the ability to constants, and terms in algebraic expressions, as well as the ability to simplify and operate algebraic forms.
Research Instruments The consisted of two main types:
observation tools and test tools.
Observation guidelines were used to observe students' learning activities while using Meta AI for 80 minutes .
wo lesson hour.
using a nonparticipant Observations focused on students' understanding of algebra topics, challenges faced, and the role of Meta AI in supporting learning.
The test instruments consisted of a pretest and posttest based on indicators of understanding algebraic concepts.
The test is designed to measure students' abilities in: .
identifying variables, coefficients, constants, and terms in algebraic expressions, .
simplifying algebraic expressions, .
performing algebraic operations, and .
applying algebraic concepts in problem solving.
Table 1.
Signs of understanding mathematical concepts Concept Understanding Indicators Re-explaining a concept Grouping objects Indicator Providing examples and nonexamples of a concept Rewrite algebraic forms.
Students are able to define algebraic terms .
ariables, coefficients, constant.
in their own words.
Mentioning variables, coefficients, constants and terms of several algebraic Students can identify and classify similar/unsimilar tribes.
Identify examples and give examples of algebraic forms.
Students are able to differentiate between algebraic and non-algebraic Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
Expressing concepts in various mathematical representation concepts Applying concepts or algorithms in problem Research Procedures The research was conducted through three systematic stages.
The initial stage included requesting research permission from the principal, discussions with mathematics teachers to identify materials and schedule instruments, and preparing the MetaAI learning tools.
The implementation stage included administering a pretest to both groups, implementing Meta-AI learning for the experimental group and conventional methods for the control group, observing learning activities, and administering a posttest and evaluation questionnaire.
The Changing algebraic forms to their simplest form and changing story problems into algebraic forms.
Students can convert algebraic forms into visual models .
raph diagram.
or story problems.
Using the concept of addition and subtraction of algebraic forms in solving everyday life problems.
Students use algebraic operations to solve contextual problems.
evaluation and reporting stage focused on analyzing student learning outcomes and comparing them between the two groups.
Data Analysis Techniques Descriptive Statistical Analysis Descriptive statistical analysis was used to describe and explain the especially in the experimental group.
Calculations statistics, highest and lowest scores using SPSS software, frequency distribution, mean, median, mode, quartiles, and measures of variability such as variance and standard Table 2 Categories of learning model implementation Average Score (G) 5OG 4.
5OG 3.
5OG 2.
1OG 1.
Learning Implementation Analysis Learning implementation data was collected through observation forms during learning sessions.
The quality of learning implementation was observation score divided by the number of aspects observed.
Learning Category Very well executed Well executed Quite well implemented Not implemented well classified based on score ranges.
Analysis of Student Learning Outcomes Classification outcome data uses the National Department of Education standards with categories of very high .
, high .
, medium .
, low Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
, and very low .
Student performance improvement is measured using the normalized gain equation.
Inferential Statistical Analysis Inferential analysis was used to estimate and extrapolate findings to a broader population.
Prior to analysis, prerequisite tests were performed, including a normality test using the Kolmogorov-Smirnov test and a homogeneity test using the Bartlett test with a significance level of 5% .
Table 3.
Explanation of mathematics learning achievement score categories Average Score (G) 5OG 4.
5OG 3.
5OG 2.
1OG 1.
Hypothesis Testing Hypothesis conducted using a paired sample t-test with three main hypotheses: .
the difference in pretest results between the experimental and control groups, .
the difference in posttest results between the two groups, and .
a comparison of improvements in learning outcomes.
The testing criteria used a significance level of 0.
05, where HCA was accepted if the significance value was > 0.
05 and HCA was accepted if the significance value was < 0.
Effectiveness Test (N-Gai.
The effectiveness of the treatment was evaluated using the normalized N-gain formula with high .
70 < N-Gain O 1.
, medium .
< N-Gain O 0.
, and low (N-Gain O .
Interpretation of effectiveness used percentages with effective (>76%), quite effective .
5%), less effective .
-55%), and ineffective (<40%) categories.
Effectiveness Criteria Learning is considered effective if it meets three criteria: .
the average posttest score exceeds the Category Very well executed Well executed Quite well implemented Not implemented well KKM, namely 71, .
the normalized gain is at least in the moderate category, and .
the classical completion level exceeds 85%.
These three indicators serve as a assessing the effectiveness of the teaching strategies used in improving students' understanding of algebraic RESULTS AND DISCUSSION General Description of Research Results The observation results showed a significant difference in student engagement between the experimental class using Meta AI and the control .
In the experimental class, 90% of students were actively engaged in Meta AI-based learning, such as asking questions directly through the chat feature and participating in interactive drag-and-drop exercises.
Student completing reward-based challenges such as interactive quizzes.
Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
Table 4.
Explanation of the effectiveness of n-gain values Normalized N-Gain Average.
70 < N-GainO1.
30 < N-GainO0.
N-GainO0.
Conceptually, the accuracy of identifying algebraic components .
ariables, coefficients, constants, and like term.
reached 85% thanks to Meta AI's dynamic visualizations.
Procedural errors were reduced by 70% due to the system's real-time For example, a student who initially struggled to differentiate 2ycuA and 3ycu successfully simplified 4ycuA 2ycu Oe ycuA 5 to 3ycuA 2ycu 5 after receiving adaptive guidance from Meta AI.
In contrast, in the control class, only 40% of students actively participated, and 60% of students experienced misconceptions such as assuming 5ycu 3yc = 8ycuyc or making repeated errors in sign operations.
Classification.
Tall.
Currently.
Low.
Student Learning Outcomes in Experimental and Control Classes Experimental Class (Meta AI) The results of the analysis of pretest and posttest data for class VII A using Meta AI showed a significant increase in understanding of algebraic Based on the data, the average student score on the pretest was 25, while the average posttest score reached 83.
The pretest score was below the Minimum Competency (KKM) .
, but the posttest score exceeded the established standard.
This improvement demonstrates Meta AI's effectiveness in helping students understand algebraic concepts.
Table 5.
Description of pretest and posttest scores for experimental class (VII A) Statistics Number of Students Maximum Value Minimum Value Average value Standard Deviation The frequency distribution shows that in the pretest, all students .
%) had not achieved the KKM, with the majority obtaining a score of After the implementation of Meta
Pretest
6,725
Posttest
4,031
AI, all students .
%) succeeded in achieving the KKM in the posttest, with a higher and more even distribution of scores.
Table 6.
Frequency distribution of posttest scores for the experimental class Valid Total Mark Frequency Percent Valid Percent Cumulative Percent Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
Control Class (Conventional Metho.
Class VII B as a control group using conventional learning methods showed limited improvement.
The average pretest score for the control class was 21.
57, while the average posttest only reached 35.
Both averages were still below the required KKM .
These results indicate that conventional methods are less effective in improving students' understanding of algebraic concepts.
Table 7.
Description of pretest and posttest values for control class VII B
Data
Number of Students Maximum Value Minimum Value Average value Standard Deviation Pretest
3,355
Effectiveness of Using Meta AI
N-Gain Analysis To measure the effectiveness of learning, an N-Gain analysis was Posttest
6,871
conducted which compared the increase in learning outcomes between the two groups.
Table 8.
Results of n-gain test calculation for experimental class Class Statistics Statistics Std.
Error NGain_Percent Experiment Mean Control 95% Confidence Interval for Mean - Lower Bound 95% Confidence Interval for Mean - Upper Bound 5% Trimmed Mean Median Variance Standard Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis Mean 95% Confidence Interval for Mean - Lower Bound 95% Confidence Interval for Mean - Upper Bound 5% Trimmed Mean Median 75,0000 42,726 Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
Class Statistics Statistics Variance Standard Deviation Minimum Maximum Range Interquartile Range Skewness Kurtosis 80,252 The N-Gain assessment results showed that the experimental group achieved an average score of 73.
which is considered highly effective.
In contrast, the control group only achieved a score of 17%, indicating low effectiveness.
This significant difference demonstrates the superiority Std.
Error
1,014
of the Meta-AI approach in improving student learning outcomes.
Classical Completion Based on the KKM set at Ahmad Yani Pagelaran Middle School, namely 70, the classical completion level of the experimental class showed optimal results.
Table 9.
Classical completion data Test Minimum Competency (KKM) Completed Not Completed Pretest Posttest Data shows that in the pretest, all students in the experimental class .
%) did not meet the Minimum Competency (KKM).
However, after implementing MetaAI, all students .
%) successfully achieved the KKM in the posttest.
This 100% completion rate exceeds the minimum effectiveness standard of 85%, thus declaring the learning process highly Prerequisite and Hypothesis Testing Normality and Homogeneity Test Before conducting inferential analysis, prerequisite tests were carried out to ensure that the data met the assumptions of parametric analysis.
Table10.
Normality test after experiment for experimental and control groups Class Learning Experimental Class Control Class KolmogorovSmirnov^a^ Statistics The results of the normality test using the Kolmogorov-Smirnov test showed a significance value of 0.
df Sig.
ShapiroWilk Statistics df Sig.
for the experimental class and 0.
for the control class.
Both values were Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
> 0.
05, indicating that the data were normally distributed.
Table 11.
Homogeneity of variance test Learning Levene Statistics Sig.
Based on Mean
3,629
Based on Median Based on Median and with adjusted df
Based on trimmed mean
3,663
3,663
4,011
The homogeneity test showed a significance value of 0.
064 > 0.
indicating that the variances of both groups were homogeneous.
With the 29,082 .
assumptions met, the analysis could proceed using parametric statistics.
Hypothesis Testing Comparison of Learning Outcomes Between Groups Table 14.
Results of the independent sample t test Posttest Equal variances 27,836 39 The independent sample t-test showed a significance value of 0.
05, with a calculated t value of 836 > t table of 0.
These results indicate the rejection of HCA, which means there is a significant difference between student learning outcomes using Meta AI compared to conventional methods.
The average posttest score of the experimental class .
was much higher than that of the control class .
Interpretation and Implications of Results Research results consistently show the effectiveness of Meta AI in learning algebraic concepts.
All effectiveness indicators were met: .
the posttest average .
exceeded the KKM .
, .
N-Gain reached the high category .
4%), and .
Sig.
Mean Difference 48,181 Standard Error Difference 1,7309 classical completeness reached 100% (>85%).
Meta AI plays a crucial role in creating personalized and efficient This platform is able to detect individual student error patterns and provide specific exercises as needed.
Time efficiency was also achieved because teachers saved 50% of their time correcting assignments thanks to the automation of AI assessments, allowing them to focus more on discussing complex problems such as the application of algebra in story The significant difference between the experimental and control groups indicates that AI technology can be an effective solution to overcome students' difficulties in understanding abstract mathematical These results support the theory of adaptive learning which Feri Prastyo.
Wildan Hakim Effectiveness of Using Meta AI in Improving Understanding of Algebra Concepts by Grade VII Students Ahmad Yani Junior High School Jurnal Pendidikan Matematika dan IPA Vol.
No.
emphasizes the importance of personalization in the learning process to achieve optimal results.
CONCLUSIONS
Based on the results of a quasiexperimental study conducted at Ahmad Yani Pagelaran Junior High School, it can be concluded that the implementation of Meta AI has proven highly effective in improving the understanding of algebra concepts in seventh-grade This effectiveness is demonstrated through three main indicators, all of which were optimally met.
First, the average posttest score of the experimental group reached 83, exceeding the minimum minimum completion rate of 70, while the control group only Second, the N-Gain analysis showed a high effectiveness category with a score of 73.
4% in the experimental group, significantly different from the control group which only achieved 17%.
Third, the classical completion rate reached 100%, far exceeding the minimum standard of Meta AI's strength lies in its ability to provide adaptive and personalized learning that can adjust to the individual pace and learning style of students.
This platform is capable of providing dynamic visualizations for real-time feedback that reduces procedural errors by 70%, and interactive features that increase student engagement by up to 90%.
The significant difference with conventional methods .
< 0.
proves that AI technology can be an innovative solution to overcome students' difficulties in understanding especially in identifying variables, simplifying algebraic expressions.
This finding provides an important contribution to the development of technology-based SUGGESTION Based on the research findings.
First, schools need to infrastructure, including a stable internet connection and devices that support optimal use of Meta AI.
Second, mathematics teachers should undergo special training on integrating AI technology into learning to maximize the platform's potential.
Third.
Meta AI implementation should be carried out in stages with intensive mentoring in the initial phase to ensure effective adaptation.
Fourth, further research is needed with a broader range of mathematics material and at different educational levels to test the consistency of Meta AI's effectiveness.
Fifth, the development of AI-based learning content should be adapted to local curricula and the characteristics of Indonesian students.
Sixth, regular
evaluation of the impact of AI
technology use on students' critical thinking and creativity skills is necessary to ensure a balance between technological efficiency and the development of holistic cognitive REFERENCES