Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026.
Page: 1-18 The Effectiveness of Using Artificial Intelligence in Advancing the Level of Biomechanical Analysis of Gymnastics Skills Ziadoon Salah Raheem Individual Games Physical Education and Sports Sciences collage.
University of Diyala.
Iraq DOI:
https://doi.
org/10.
47134/jpo.
*Correspondence: Ziedoon Salah Raheem Email: zaidoun.
salah@uodiyala.
Received: 28-01-2026 Accepted: 05-02-2026 Published: 17-03-2026 Copyright: A 2026 by the authors.
Submitted for open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) .
ttp://creativecommons.
org/licenses/ by/4.
0/).
Abstract: The research aims to identify the effectiveness of using artificial intelligence in advancing the level of biomechanical analysis of gymnastics skills by examining .
he analytical efficiency of artificial intelligence in biomechanical measurement, temporal effectiveness and speed of data processing, objectivity and support of training decisionmaking, and the contribution of intelligent analysis to improving skill performanc.
The researcher used the descriptive method due to its suitability to the nature of the research.
The researcher selected a random sample of .
gymnastics coaches and relied on a questionnaire form to collect data.
Among the most important conclusions was that the use of artificial intelligence in the biomechanical analysis of gymnastics skills demonstrates a high degree of efficiency and accuracy, contributes to reducing time and effort, enhances objectivity in evaluation, and supports the improvement of skill performance, which confirms its effectiveness as a modern scientific tool that can be employed to develop the training process in gymnastics.
Among the most important recommendations was the necessity of expanding the use of artificial intelligence technologies in analyzing gymnastics skills within clubs and sports institutions, and integrating intelligent biomechanical analysis systems into daily training programs due to their role in improving evaluation accuracy and accelerating the analysis process.
Keywords: Artificial Intelligence.
Biomechanical Analysis.
Gymnastics Introduction The current era is witnessing rapid development in digital technology applications, especially in the fields of artificial intelligence and big data analysis, which has clearly been reflected in various aspects of life, including the sports field.
Despite the relentless global competition over sports achievement, relying on modern scientific methods has been an absolute necessity of improving the athletic performance levels and increasing training program efficiency.
A sport high in motor complexity is likely to be characterized as gymnastics since it relies on the required fine technical performance and neuromuscular coordination, and associated with the constraints of biomechanical variables like joint angles, angular velocities, center of gravity path, and dynamic balance.
Thus, any small error in technique will impact the performance and judging assessment directly, while even increasing the risk of injury.
(Al-Laithi, 2022: .
Biomechanical analysis is one of the mainstays of motor performance in gymnastics because it allows movement to be studied from the mechanical laws by which they are governed, through kinematic variables .
isplacement, velocity, acceleration angle.
and https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 2 of 18 kinetic variables .
orces, torque.
Such analysis helps in understanding the strengths and weaknesses, providing explanation for tech errors and suggesting appropriate corrective (Abdelhamid.
Abdelwahab, 2015: .
Biomechanical analysis plays an important role, but traditional methods based on visual observation and manual video-based analysis may have some drawbacks such as being time- and effort-consuming, prone to personal bias, or how to obtain precise measurements at real-time.
Here, the role of artificial intelligence technologies forwarded, especially computer vision and deep learning techniques that can take body points tracking from point coordinates, extract joint angles in no time at all and analyze movement trajectories automatically and accurately.
(Hajji, 2023: .
Using machine learning in biomechanical analysis of gymnastics skills offers sophisticated capabilities like:
Immediate analysis of motor performance.
Providing precise feedback based on digital data.
Comparing performance with ideal standards.
Predicting potential errors or injury risks.
Supporting training decisions based on objective quantitative indicators.
Integrating these technologies within the training environment also contributes to developing teaching and training methods and enhances the coachAos ability to plan scientifically based on data, rather than relying entirely on personal experience or selfestimation.
(David, 2023: .
Based on this, the significance of the present study can be found in trying to investigate a machine learning approach to enhance the quality of individual biomechanical analysis in gymnastics skills and examine its extent towards raising levels of accuracy, objectivity and reducing the time needed for movement analysis while aiding motor performance development.
The study also aims to provide a scientific framework for the systematic use of such technologies within the sports field.
The researcher hopes that the results of this study will contribute to enriching the theoretical aspect related to artificial intelligence applications in the sports field, in addition to providing practical recommendations that can be utilized in developing training and motion analysis programs for gymnastics.
Research Problem Gymnastics is one of the sports that fundamentally depends on high technical precision, neuromuscular coordination, and control of body positions during performance in the different phases of movement .
reparatory phase Ae takeoff Ae flight Ae landin.
Accordingly, the success of skill performance does not depend solely on strength or flexibility, but is largely related to the degree of adherence to precise biomechanical requirements such as appropriate joint angles, movement sequencing, angular velocity, the trajectory of the center of gravity, and achieving balance during transitions between performance phases.
Despite the development witnessed in sports sciences, the process of evaluating motor performance in many training environments still largely depends on visual https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 3 of 18 observation and the coachAos experience, or on traditional manual video analysis, which may result in a number of issues, most notably:
A Difficulty in detecting precise technical errors that occur in fractions of a second, especially in flight and rotation skills.
A Time and effort consuming processes of manual measuring or analyzing A Differences in the way coaches or analysts makes things cooler makes it hard to be objective because they have different perspectives, angles and methods of A Limited ability to extract accurate quantitative indicators .
uch as angular velocity, phase time, and deviation of the center of gravity trajector.
continuously and regularly within the training environment.
In contrast, recent years have witnessed a clear expansion in the use of artificial intelligence techniques in the field of sports motion analysis, especially with regard to computer vision and deep learning techniques that allow tracking body points, extracting joint angles, and determining movement trajectories almost instantaneously, with the possibility of providing accurate and numerically supported feedback.
However, the reality of employing these technologies in the field of gymnastics within training institutions still faces scientific and practical questions related to:
A The accuracy and reliability of AI-based analysis in highly complex gymnastics A The ability of these technologies to improve the quality of biomechanical analysis compared to commonly used traditional practices.
A The applicability of these systems within the training environment and the possibility of integrating them into the daily programs of coaches and athletes.
A The extent to which AI-based analysis contributes to supporting training decisions and guiding the correction of skill errors more objectively.
Through the researcherAos review of the reality of training and field practice, and what some preliminary observations have shown regarding heavy reliance on personal estimation and a clear lack of the use of digital tools that provide accurate and immediate biomechanical data, the need has emerged for a scientific study that describes and analyzes the effectiveness of using artificial intelligence in advancing the level of biomechanical analysis of gymnastics skills, in a way that contributes to bridging the gap between modern technological development and practical application within sports Accordingly, the research problem is defined in the following main question: What is the effectiveness of using artificial intelligence techniques in advancing the level of biomechanical analysis of gymnastics skills in terms of accuracy, speed, objectivity, and applicability within the training environment? Research Objectives The research aims to identify the effectiveness of using artificial intelligence in advancing the level of biomechanical analysis of gymnastics skills by identifying:
A The analytical efficiency of artificial intelligence in biomechanical measurement.
https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 4 of 18 A Temporal effectiveness and speed of data processing.
A Objectivity and support of training decision-making.
A The contribution of intelligent analysis to improving skill performance.
Research Questions A What is the analytical efficiency of artificial intelligence in biomechanical A What is the temporal effectiveness and speed of data processing? A What is the level of objectivity and support for training decision-making? A What is the degree of contribution of intelligent analysis to improving skill Research Procedures Research Method:
The researcher used the descriptive method, due to its suitability to the nature of the research.
Research Population and Sample:
The researcher selected a random sample of .
gymnastics coaches.
The researcher divided the research sample as follows:
- Pilot study sample: Consisted of .
individuals for the purpose of standardizing the questionnaire form from the original research population and outside the basic study sample.
- Basic study sample: Consisted of .
individuals for the purpose of applying the questionnaire form for the research, as shown in Table .
Table .
Numerical distribution of the research population and sample Basic Sample )n( )n( Research Sample q.
Gymnastics coaches Total q.
Pilot Sample (.
Categories of Research Sample Data Collection Tools The researcher designed a questionnaire form for the individuals and categories of the research population shown in Table .
as a primary tool for data collection, with the aim of identifying the effectiveness of using artificial intelligence in advancing the level of biomechanical analysis of gymnastics skills, through a reference survey and the results of previous studies related to the research topic.
Through the previous steps, the researcher established four .
axes for the questionnaire form, represented as follows:
- The first axis: Analytical efficiency of artificial intelligence in biomechanical - The second axis: Temporal effectiveness and speed of data processing.
- The third axis: Objectivity and support for training decision-making.
https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 5 of 18 The fourth axis: The contribution of intelligent analysis to improving skill Scientific Coefficients of the Questionnaire Form:
Validity:
To establish the validity of the questionnaire form, the researcher relied on:
Content validity .
xpertsAo validit.
The researcher presented the axes to a group of specialized experts in the field of gymnastics training who meet the requirement of holding a doctoral degree, numbering .
Appendix No.
, to survey their opinions regarding the suitability of the proposed axes and to express their views on their appropriateness for achieving the research objectives.
Table .
Frequency and percentage of expertsAo opinions regarding the axes of the questionnaire form n = 7 experts Experts Agreeing Percentage Frequency 0 Axes First axis: Analytical efficiency of artificial intelligence in biomechanical Second axis: Temporal effectiveness and speed of data processing Third axis: Objectivity and support for training decision-making Fourth axis: The contribution of intelligent analysis to improving skill It is evident from Table .
that the opinions of the experts agreed on the suitability of the axes established by the researcher, with percentages ranging between .
100%).
The researcher then formulated the appropriate statements for each axis.
Subsequently, the axes and statements of the questionnaire form in its initial version.
Appendix .
, were presented to the experts.
The form was personally delivered to the experts by the researcher to obtain their opinions regarding:
A The extent to which each statement is appropriate to the axis to which it belongs in achieving its objective.
A The correctness of the linguistic formulation of each statement.
A The deletion, merging, transfer, or modification of what they deemed appropriate.
Table .
shows the frequencies and percentages of the expertsAo opinions regarding each statement of the questionnaire form.
Table .
Frequencies and percentages of the expertsAo opinions regarding each statement of the questionnaire form n = 7 experts Percen Frequ State Percen Frequ https://edu.
id/index.
php/jpo State Percen Frequ State Percen Frequ State Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 I.
6 of 18 It is evident from Table .
that the experts agreed on the initial version of the questionnaire form, as the percentage of agreement ranged between .
%: 100%), with a suggestion to delete some statements.
The researcher made the necessary modifications, as shown in the following Table No.
Table .
Modifications made to the questionnaire form Statement after ------------- Type of Deletion ------------- Deletion Statement before modification The artificial intelligence system is characterized by ease of use in the training environment The artificial intelligence system is characterized by ease of use in the training environment Statement Axis Third Fourth Based on the expertsAo modifications made to the questionnaire form, including the deletion of some statements, the questionnaire form became finalized in its final version before application.
Appendix No.
Internal Consistency Validity:
The validity of the questionnaire statements was verified by calculating the correlation coefficient between the score of each statement and the total score of the questionnaire form, as shown in Table .
Table .
Correlation coefficient between the score of each statement and the total score of the questionnaire form n = 17 Correlation Coefficient With Wit Question Axi *0.
*0.
*0.
*0.
*0.
State Correlation Coefficient With Wit Question Axi *0.
*0.
*0.
*0.
*0.
https://edu.
id/index.
php/jpo State Correlation Coefficient With Wit Question Axi *0.
*0.
*0.
*0.
*0.
State Correlation Coefficient With Wit Question Axi *0.
*0.
*0.
*0.
*0.
State Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 *0.
*0.
7 of 18 *0.
*0.
*0.
*Statistically significant at 0.
abulated Pearson correlation coefficient = )0.
It is evident from Table .
that the correlation coefficients between each statement and the total score of the questionnaire form are statistically significant at the significance level .
Table .
Internal consistency of the questionnaire axes .
= .
Pearson Correlation Coefficient with Total Questionnaire Score Axes *0.
First axis: Analytical efficiency of artificial intelligence in biomechanical Second axis: Temporal effectiveness and speed of data processing Third axis: Objectivity and support for training decision-making Fourth axis: The contribution of intelligent analysis to improving skill *0.
*0.
*0.
*Statistically significant at 0.
abulated Pearson correlation coefficient = )0.
It is evident from Table .
that the values of the correlation coefficients among the questionnaire axes and between them and the total score of the questionnaire are statistically significant at the significance level .
, which indicates the validity of the questionnaire form and that it measures what it was designed to measure.
Reliability:
Reliability was verified using CronbachAos Alpha coefficient to determine the reliability of the questionnaire statements and the reliability of the questionnaire axes, as shown in Table .
Table .
Reliability of the Questionnaire Using CronbachAos Alpha Method .
= .
Reliability Coefficient if Statement Deleted For For Question Ax State Reliability Coefficient if Statement Deleted For For Question Ax https://edu.
id/index.
php/jpo State Reliability Coefficient if Statement Deleted For For Question Ax State Reliability Coefficient if Statement Deleted For For Question Ax State Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 8 of 18 Table .
CronbachAos Alpha values for the questionnaire axes CronbachAos Alpha Coefficient For the For the axis as questionnaire as a whole a whole Questionnaire Axes First axis: Analytical efficiency of artificial intelligence in biomechanical measurement Second axis: Temporal effectiveness and speed of data processing Third axis: Objectivity and support for training decision-making Fourth axis: The contribution of intelligent analysis to improving skill performance It is evident from Table .
, which presents the reliability coefficients of the questionnaire axes using CronbachAos Alpha, that the reliability coefficients for the axes with the questionnaire ranged between .
873 Ae 0.
The reliability coefficients for the statements with the questionnaire as a whole reached .
All of them are statistically significant at the significance level .
, which indicates the reliability of the questionnaire form.
Application of the Questionnaire Form:
- Human Scope: The application was conducted on the members of the research sample, numbering .
gymnastics coaches.
- Spatial Scope: The research was conducted in the clubs of Diyala Governorate.
- Time Scope: The questionnaire form for the research was applied during the period from 17/11/2025 to 25/12/2025.
Method of Scoring the Questionnaire Form:
The researcher relied on a three-point scale in determining the responses to the questionnaire form, represented by (Agree Ae To some extent Ae Disagre.
The responses were assigned scores of .
, 2, .
Statistical Treatments Used in the Research:
The statistical treatments were conducted using SPSS version 2020 as follows:
n Percentage (%) to calculate the frequencies of expertsAo opinions on the statements and axes of the questionnaire.
n Pearson correlation coefficient to calculate the internal consistency validity between the questionnaire statements.
n CronbachAos Alpha reliability coefficient to calculate the reliability of the questionnaire form.
https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 n n n 9 of 18 Chi-square (NA) to calculate the differences between the responses of the research Predominant trend to determine the direction of the sample membersAo responses.
Approval percentage to determine the percentage of agreement of the research sampleAos opinions on the questionnaire statements.
Presentation and Discussion of Results Presentation and Discussion of the Results of the First Axis: Analytical Efficiency of Artificial Intelligence in Biomechanical Measurement Table .
Frequencies, percentages, and statistical significance of the statements of the first axis (Analytical efficiency of artificial intelligence in biomechanical measuremen.
= .
Appro % val a.
Overall Respon Likert Yes i.
Mean Significa nce Level ChiSquare Disagre To Some Extent Agree Yes u.
Yes w.
Yes O.
Yes https://edu.
id/index.
php/jpo Statement Content The artificial contributes to AI-based by consistency and stability in results.
The helps identify differences in the stages of Artificial indicators of the trajectory of the center of gravity.
The Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 10 of 18 C.
Yes I.
Yes and objective The error rate The results of be relied upon as a basis for Chi-square is significant at the 0.
05 level, where the significance level O 0.
05 at degree of freedom 2 = 5.
99, degree of freedom 1 = 3.
Likert Scale: Weighted mean ranges: 1.
00Ae1.
66 (Disagre.
, 1.
67Ae2.
33 (To some exten.
, 34Ae3.
00 (Agre.
It is evident from Table .
, which presents the frequencies, percentages, and statistical significance of the statements of the first axis (Analytical efficiency of artificial intelligence in biomechanical measuremen.
, that there are statistically significant differences between the responses.
For all statements, the chi-square values are significant 05 level .
pproval percent: 75.
58% - >94.
19%).
These results are consistent with the findings of the study by Dindorf .
, which confirmed that computer vision and deep learning techniques contribute to improving the accuracy of biomechanical measurements compared to traditional methods.
The study demonstrated that digital joint-tracking systems are capable of extracting movement angles and angular velocities with a high degree of consistency and stability, thereby reducing human error resulting from visual estimation.
The results of the study by Mohi El-Din Al-Laithi .
also indicated that the use of artificial intelligence in motion analysis provides accurate quantitative indicators of the trajectory of the center of gravity and the sequencing of movement phases, which enhances the objectivity of analysis and increases the efficiency of technical evaluation.
This is consistent with the high approval percentages for the statements related to measurement accuracy and the analysis of linear and angular velocities.
These results are also consistent with the findings of the study by Somaya Jamil AlSarayreh .
, which indicated that intelligent systems allow performance to be reanalyzed multiple times with the same level of accuracy without being affected by human factors, thereby enhancing the statistical reliability of measurements.
The study https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 11 of 18 further noted that reliance on digital analysis contributes to detecting subtle differences that are difficult to observe with the naked eye, especially in high-speed skills such as gymnastics skills.
This explains why the statements related to reducing error rates and the possibility of relying on results as a basis for biomechanical evaluation achieved high approval percentages exceeding .
%).
The researcher believes that the high approval percentages and mean scores in this axis reflect the sample membersAo awareness of the importance of shifting toward digital analysis supported by artificial intelligence, particularly in a sport characterized by precision and motor complexity such as gymnastics.
It furthermore reflects a professional belief that intelligent analysis is no longer just an ancillary technical assistance, but such that if it was applied as a laboratory formal method we could further promote better quality of biomechanical measurement parameters due to reduction of bias and also make evaluation more objective.
Presentation and Discussion of the Results of the Second Axis: Temporal Effectiveness and Speed of Data Processing Table .
Frequencies, percentages, and statistical significance of the statements of the second axis (Temporal effectiveness and speed of data processin.
= .
Appro % val c.
Overall Respon Likert Yes Mean Significa nce Level ChiSquare Disagre To Some Extent Agree Y.
Some Extent w.
Yes c.
Yes https://edu.
id/index.
php/jpo Statement
Content
AIA
The use of time required to analyze skill The feedback after skill execution.
It contributes to accelerating the process of organize data and present it in a fast and Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 12 of 18 .
Yes c.
Yes e manner.
It reduces the time and effort It shortens the Chi-square is significant at the 0.
05 level, where the significance level O 0.
05 at degree of freedom 2 = 5.
99, degree of freedom 1 = 3.
Likert Scale: Weighted mean ranges: 1.
00Ae1.
66 (Disagre.
, 1.
67Ae2.
33 (To some exten.
, 34Ae3.
00 (Agre.
The clear from Table .
that provides frequencies and percentages Statements of the second axis: Temporal effectiveness and speed of processing dat.
The Chi-square revealed statistically significant differences at the 0.
05 level for all judicial statements except statement number .
, and the approval percentages therefore were .
30% to 671%).
These results are in agreement with what Bakri, 2013 study found, confirming that AI systems have evidently shortened the time needed for data processing compared to traditional analysis dependent on manual frameworks or non-automated software.
Moreover, the result of this study revealed that automated motion-tracking techniques could extract multiple biomechanical indicators within a short timeframe while they can be stored, organized, and directly displayed in the data that shortens the associated technical procedures with the analysis process.
This is consistent with the high approval rates for statements regarding saving time and effort as well as data organization.
Also, the results of Odeh .
study showed that digital processing speed is one of AI systems' essential advantages in the sports domain compared to other attempts or athletes as intelligent systems allow to view instantaneous comparisons depending on accurate digital data.
However, some other works have reported that immediate feedback could be influenced by technical parameters as the speed of the device, connection quality and software performance used.
which may explain lower approval percentage for this statement regarding immediate feedback when compared with other statements, where high percentages were achieved.
This can be attributed to the perception and awareness of members of the sample on the time-saving aspect achieved by artificial intelligence in biomechanical analysis, particularly with regards to minimizing effort within a relatively short period that they https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 13 of 18 spend conducting manual analysis, which marks a qualitative orientation in the sports training environment.
As for immediate feedback, the differences in responses may indicate discrepancies between participantsAo nominal level of experience with intelligent systems and their actual level of experience with intelligent systems or inconsistencies in the technical solutions afforded to them.
Presentation and Discussion of the Results of the Third Axis: Objectivity and Support for Training Decision-Making Table .
Frequencies, percentages, and statistical significance of the statements of the third axis (Objectivity and support for training decision-makin.
= .
Appro % val A.
Overal Respo Likert Yes I.
Mean Significa nce Level ChiSquare Disagree To Some Extent Agree Yes p.
Yes C.
Yes `.
Some Extent Y.
Some Extent https://edu.
id/index.
php/jpo Statement
Content
AIA
Artificial enhances the objectivity of personal bias in judging the athleteAos level.
decisionmaking based digital data.
indicators that help identify strengths and It contributes It helps in training plans Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 14 of 18 Chi-square is significant at the 0.
05 level, where the significance level O 0.
05 at degree of freedom 2 = 5.
99, degree of freedom 1 = 3.
Likert Scale: Weighted mean ranges: 1.
00Ae1.
66 (Disagre.
, 1.
67Ae2.
33 (To some exten.
, 34Ae3.
00 (Agre.
The table .
, it is clear the frequencies, percentages, and significance of the statements for third axis .
equirements objectivity and support in decision trainin.
there are statistically significance differences between responses.
The Chi-square value at the 05 level is significant for all statements but statement No-18, with approval levels between .
30% to 86.
This comes in consonance with the results of a study that proved using AI helps to improve objectivity in performance evaluation because it is based on measurable quantitative data and minimizes personal influences and subjective opinions in judging Mohi El-Din Al-Laithi .
further suggested that backing training decisions with the correct digital indicators allows the coach to better identify strengths and weaknesses than through conventional visual appraisal.
This aligns with the high rates of approval for the statements regarding reducing bias and offering measurable Inversely.
Sanders's .
results prescribed that a full adaptation towards converging evaluation standards and constructing training plans from scratch using solely digital analysis might take time, practice adoption and coaching professional culture This could be the reason why statements about unifying standards and constructing training plans received fairly moderate approval rates.
This does not only depend on the accuracy of the system for decision-making process but this is more affected by how much user accepting it, technical experience of users and supportive The researcher believes that this data indicates a significant awareness of the impact of artificial intelligence on achieving objectivity in assessment and within evaluation, reducing human impact which is an important step towards creating methods of training decision-making in gymnastics.
From the effective role of digital indicators and databases as supportive source .
hich should continu.
in forming training plans, quantifying standards among coaches.
there should be some qualification, training and awareness of the importance of quantitative analysis.
The researcher maintains that in this light making objective and efficient decisions by training decision-makers, no matter how intelligent they may be, ultimately remains the best model.
Presentation and Discussion of the Results of the Fourth Axis: The Contribution of Intelligent Analysis to Improving Skill Performance Table .
Frequencies, percentages, and statistical significance of the statements of the fourth axis (The contribution of intelligent analysis to improving skill performanc.
= .
https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 Appro % val y.
Overall Respon Likert Yes v.
Mean Significa nce Level ChiSquare 15 of 18 Disagre To Some Extent Agree Yes X.
Some Extent p.
Yes e.
Some Extent E.
Yes
Statement
Content
AIA
AI-based contributes to improving the It helps correct motor errors It contributes to developing phases of skill It supports the motor learning process among It reduces the errors during It contributes Chi-square is significant at the 0.
05 level, where the significance level O 0.
05 at degree of freedom 2 = 5.
99, degree of freedom 1 = 3.
Likert Scale: Weighted mean ranges: 1.
00Ae1.
66 (Disagre.
, 1.
67Ae2.
33 (To some exten.
, 34Ae3.
00 (Agre.
It is evident from Table .
, which presents the frequencies, percentages, and statistical significance of the statements of the fourth axis (The contribution of intelligent analysis to improving skill performanc.
, that there are statistically significant differences between the responses.
The Chi-square values are significant at the 0.
05 level for all statements except statement No.
, with approval percentages ranging from .
14% to 72%).
These results are consistent with the findings of Dindorf .
, which indicated that AI-supported digital feedback contributes to improving technical performance by providing the athlete with accurate and immediate information about motor errors, https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 16 of 18 thereby helping to correct them more efficiently.
The study also clarified that the use of quantitative biomechanical indicators enhances motor control and reduces the gap between actual performance and the ideal model, which is positively reflected in achievement levels during official competitions.
This aligns with the high approval percentages for statements .
On the other hand, the findings of Souaifi .
indicated that the comprehensive development of all phases of skill performance may require a longer period and continuous repetition of feedback, in addition to the integration of physical and psychological factors alongside technical analysis.
This may explain the lack of statistical significance for statement .
and its classification within the AuTo Some ExtentAy category.
Furthermore, reducing the repetition of technical errors depends not only on the availability of digital analysis, but also on the athleteAos understanding of the feedback and their ability to apply it practically during training.
The researcher believes that the results indicate a clear positive effect of intelligent analysis in improving skill performance, especially in aspects related to technical precision and competitive achievement.
Nevertheless, such enhancements must be generalized overall performance phases by putting biomechanical analysis into practice, through planned training programs and systematic repeated use of intelligent systems.
This at the same time does not work separately from the rest of the parts of training process, but should be used within a complete system in order to gain as much benefit as possibly Conclusions In light of the research results, the following conclusions were reached:
Analytical Efficiency of Artificial Intelligence in Biomechanical Measurement shows AI is well trained as aligned to biomechanics manufactures, with an accuracy of obtaining biomechanical variables important for gymnastics skills.
High accuracy, consistency and stability in analysis results is achieved with the help of intelligent system.
Artificial intelligence lowers the measurement error rate relative to other methods.
Directly probes the data from a single AHRS sensor and produces precise numerical metrics for the center of gravity trajectory, joint angles, and angular velocities.
The results of intelligent analysis can be relied upon as a scientific basis for biomechanical evaluation.
The results confirmed a high level of analytical efficiency of artificial intelligence in biomechanical measurement, which enhances the objectivity and accuracy of motor performance analysis in gymnastics.
Temporal Effectiveness and Speed of Data Processing shows Artificial intelligence contributes to significantly reducing the time required to analyze skill performance,It reduces the time and effort expended in traditional manual analysis.
It helps organize and present data in a fast and easily understandable manner.
It shortens the technical procedures associated with the biomechanical analysis process.
The element of immediate feedback still requires further technical development to ensure full benefit from it.
Artificial intelligence achieved high temporal effectiveness in data processing and performance analysis, with a need to enhance the aspect of immediate feedback.
https://edu.
id/index.
php/jpo Pubmedia Jurnal Pendidikan Olahraga Vol: 3.
No 3, 2026 17 of 18 Objectivity and Support for Training Decision-Making shows Artificial intelligence enhances the objectivity of motor performance evaluation.
It reduces personal bias in judging playersAo levels.
It supports training decision-making through accurate and measurable digital data.
It helps identify strengths and weaknesses more clearly.
Unifying evaluation standards and building training plans based entirely on intelligent analysis still require further activation and training.
Artificial intelligence effectively contributes to supporting objectivity in evaluation and training decision-making.
however, its full integration into training planning requires further professional qualification.
Contribution of Intelligent Analysis to Improving Skill Performance.
Intelligent analysis contributes to improving the technical performance level of gymnastics players.
It helps correct motor errors more accurately and efficiently.
It supports the motor learning process and enhances training quality.
It contributes to raising performance efficiency in official competitions,Comprehensive development of all phases of skill performance requires integrating intelligent analysis with carefully planned training programs.
AIbased analysis is an efficient resource for skill performance enhancement.
nonetheless, to leverage its full potential it must be embedded in a holistic training system.
Artificial intelligence in the biomechanical analysis of gymnastics skills are more efficient and accurate, save time and effort, increase evaluation objectivity, and promote skill It proves this tool effectiveness as a modern scientific device to be used for formation of the training process in gymnastics.
Recommendations In light of the research findings, it is recommended to integrate artificial intelligenceAebased biomechanical analysis systems into regular gymnastics training programs to enhance the accuracy and speed of skill evaluation.
Sports clubs and associations should utilize digital performance indicators generated through intelligent analysis to design effective training plans and systematically correct technical errors.
It is also important to provide the necessary technical infrastructure, including high-resolution cameras, specialized motion analysis software, and high-performance computers, to maximize the benefits of these technologies.
Furthermore, implementing real-time feedback systems during training can significantly support the motor learning process.
Finally, organizing practical training workshops for coaches and promoting collaboration between sports scientists and artificial intelligence specialists will contribute to developing more accurate and efficient analysis systems for gymnastics performance.
References