THE VARK-FLEMING MODEL AND SELF-CONCEPT: DOES IT AFFECT MATHEMATICAL CONCEPTS UNDERSTANDING

Concept understanding is required for the learning process to be successful. This quantitative study investigates students' mathematical concept understanding abilities using the VARK-Fleming model and self-concept as a covariate variable. The purpose of this research is to determine the effect of the VARK-Fleming model on students' ability to understand mathematical concepts by controlling for students' self-concept and the effect of the covariate variable of self-concept on students' ability to understand mathematical concepts, as well as the simultaneous concept of the VARK-Fleming model and self-concept on students' ability to understand mathematical concepts. Sixty-four junior high school students participated in this research, with 32 learning the VARK-Fleming model and 32 learning the expository model. Questionnaires and descriptive tests were employed to collect information. The analysis of covariance was utilized for hypothesis testing (one-way ANCOVA). The findings of this research reveal that the VARK-Fleming learning model has a better effect on students' mathematical concept understanding capacity than the expository model and self-concept, both partially and concurrently.

The effectiveness of selecting a learning model will improve teachers' concept understanding ability (Pranata, 2016). The Visual, Auditory, Read, Kinesthetic (VARK) learning model is a learning model that focuses on giving direct and engaging learning experiences by utilizing the four learning modalities that students have: seeing (visual), hearing (auditory), reading (read), and moving (kinesthetic) (Hussain, 2017;Mirza & Khurshid, 2020;Nurhidayah, 2021). VARK-Fleming is a learning model that optimizes the four learning modes to make learning effective and engaging (Hariyani & Sejati, 2019). The VARK-Fleming model is learning that involves physical movement and student participation by employing the five senses they have in the learning process (Monariska & Komala, 2020).
Several researchers have researched the VARK-Fleming model, including the findings of Erawati et al. (Erawati et al., 2020;Monariska & Komala, 2020;Prihaswati & Purnomo, 2021), which discovered that employing the VARK-Fleming model can improve connection skills and mathematical proof abilities. However, the sample was at the university level, so whether this model improved mathematical abilities at the junior high school level was unknown. Also, VARK-Fleming can increase communication skills and mathematical thinking (Didit Chayono & Nuriyatin, 2019;K et al., 2021;Tahir, 2021). However, the sample utilized was at the high school level; thus, whether this model affected junior high school students' mathematical abilities was unknown.
The VARK-Fleming model used in this research adheres to Ngalimun's syntax (Ngalimun, 2012), namely stage 1, preparation. This first stage is concerned with preparing students to participate in the learning process by providing motivation and placing. Stage 2 is delivery, which refers to the range of methods and media in the learning process that maximizes the utilization of students' five senses in the form of visual, aural, read/write, and kinesthetic modalities so that the students can be actively engaged with the educational process. Stage 3 is training, which is concerned with deepening the material taught by the teacher using various methods and media in the form of material or questions connected to the material. Students receive training, which can be completed individually or in groups. Stage 4 is presentation, which is concerned with the disclosure of the findings of the students' thinking about the material learned. The results are disclosed through student presentations, which the teacher then checks. Stage 5 is the conclusion concerned with consolidating the material taught. The learning outcomes conclusion is provided as a summary created by students based on their learning styles (Hariyani & Sejati, 2019). ISSN 2089-8703 (Print) Volume 12, No. 2, 2023, 2009-2023ISSN 2442 Figure 1 showa depicts the syntax of VARK-Fleming. Concept understanding ability has a strong association with self-concept. This statement is consistent with research demonstrating a favorable and substantial link between self-concept and concept understanding ability (Satriani et al., 2021;Lestari & Rukmigarsari, 2021). The self-concept is a person's views, attitudes, feelings, and expectations when confronted with and addressing problems in general, scientific, and social contexts (Andinny, 2015;Hattie, 2014;Markus & Wurf, 1986;Juniarti & Margunayasa, 2020;Pamungkas et al., 2015). A person's self-concept is established first concerning the person closest to them (Kartika, 2017). Based on past research, no researcher has investigated the influence of the VARK-Fleming model and self-concept on students' mathematical concept understanding ability. As a result, this research will disclose the impact of the VARK-Fleming model and self-concept on junior high school student's ability to understand mathematical concepts.

RESEARCH METHODS a. Research Design
This research employs a quantitative approach and a quasiexperimental design. This research has one independent variable, the VARK-Fleming model, one covariate variable, self-concept, and one dependent variable, concept understanding ability. The sample was chosen randomly using the cluster random sampling technique, yielding two samples: the experimental class, which received treatment with the VARK-Fleming model, and the control class, which received treatment with the expository model. This research employs a 1 X 2 factorial design. The research design can be seen in Table 1.

2012|
questionnaire was adapted from (Widyastuti et al., 2020). It included subscales such as seriousness, distinguishing between qualities and deficiencies, optimism, collaboration, respecting opinions, communication skills, and understanding the function of learning mathematics. The questionnaire consisted of 30 statements with four possible responses, each with a score depending on its degree on a Likert scale, such as strongly disagree (score 1), disagree (score 2), agree (score 3), and strongly agree (score 4) (Satriani et al., 2021). The responses are chosen range from very positive to very negative. Positive and negative statement scores are inversely proportional to one another. After the validation, the questionnaire was administered to 32 students, yielding 28 valid and reliable items with a Cronbach's Alpha value of 0.850. The concept understanding ability test comprised 14 essay questions with the subject matter of exponentiation and radical forms representing 7 indicators of concept understanding ability. Additionally, 32 students were tested, and 8 questions were confirmed valid and reliable, with a Cronbach's Alpha value of 0.733. Thus, 8 questions were utilized to assess concept understanding ability.

c. Research Participants
The demographics of the participants in this study are presented in Table 2. Based on Table 2, it can be observed that This research involved 64 students between the ages of 12 and 15, with 32 students learning with the VARK-Fleming model and 32 with the expository model. The 64 were ninth-grade junior high school students in South Lampung Regency, Lampung Province. The problems in that area were particularly pertinent to our research; hence students from South Lampung Regency were selected. It turned out that many students had difficulty understanding mathematical concepts. Note. N = 64; the average age of 15 years ( = 0,056 . = 0,445).

a. Data Collecting Technique
In this research, a questionnaire and test were utilized to collect data, with the questionnaire used to examine self-concept and the test used to measure concept understanding ability.
The tests employed in this research were the first (pretest) and the final (posttest). The pretest was conducted before using the VARK-Fleming model to compare the starting ability of the experimental and control classes. Following the implementation of the VARK-Fleming model, a posttest was administered to determine whether there was a significant difference between the experimental and control classes. Table  3 shows the self-concept indicators employed in this research. ISSN 2089-8703 (Print) Volume 12, No. 2, 2023, 2009-2023ISSN 2442  According to Table 3, there are seven indicators of self-concept used in constructing the questionnaire, namely seriousness (interest, motivation, courage, and seriousness), distinguishing the self-qualities and shortcomings, optimism, collaborating and tolerant, respecting opinions, communication skills and being able to place themselves, and understanding the function of learning mathematics (Afri, 2019). Indicators of concept understanding ability used in this research are shown in Table 4.  forms of mathematical representation, explaining the necessary conditions for a concept, determining and using and referring to certain procedures or operations, and applying concepts or algorithms to problem-solving (Baiduri et al., 2021).

d. Data Analysis
The analysis of the covariance (one-way ANCOVA) test was utilized to analyze the data in this research. The one-way ANCOVA test is a hypothesis test performed onCE the prerequisite tests have been completed. The four prerequisite tests are normality, data variation homogeneity, regression linearity, and regression coefficient homogeneity tests (Rinaldi & Novalia, 2020;Sudjana, 2003;Zunita, 2018 Table 6 shows the results of the normality tests. The results of Table 6 show the results of the normality test of self-concept and students' concept understanding ability at the α = 0.05 level. Therefore, it can be concluded that the data obtained in the control and experimental classes are normally distributed because -is greater than α. The next prerequisite test is the homogeneity test. The results of the data variation homogeneity test can be seen in Table 7. Note. Normality test with = 64  Table 7 reveals that the homogeneity test of self-concept and concept understanding ability results came from the same variance or were homoge-neous because the p-value (0.523) is greater than α (0.05). The regression linearity test is the next required test. The regression linearity test is satisfied if there is a linear relationship between the covariates and the dependent variable. Table 8 displays the results of the regression linearity test.   Table 9 displays the results of the regression coefficient homogeneity test.

Hypothetical Test of the Analysis of Covariance (One-Way ANCOVA)
Analysis of covariance has been employed to test hypotheses (one-way ANCOVA). This test is a distinct or comparative test in which the independent variables combine factorial and numerical data, and the dependent variable is interval or ratio (quantitative) data. By removing treatment effect bias, the ANCOVA technique adjusts the dependent variable score (Montgomery, 2013). Eliminating treatment effect bias aims to reduce error variance by controlling for the influence of covariate factors that are thought to affect analysis outcomes. Statistical covariance analysis can equalize groups based on the influence of variables other than the treatment variable (Kadir, 2019). The one-way ANCOVA test was performed using SPSS 26 program in this research. Table  10 displays the outcomes of the oneway ANCOVA test.

2016|
In Table 10, row X1, the value of F observed is 5.860, with a p-value of 0.018 at a significance level of 0.05. This result suggests that the p-value is lower than 0.05, indicating that H 0 is rejected and H 1 is accepted. In conclusion, the VARK-Fleming model affects students' mathematical concept understanding ability by influencing self-concept. Table 10 shows that the value of F observed is 13,700 with a p-value of 0.000 at a significance level of 0.05. This finding demonstrates that the pvalue is lower than 0.05. As a result, H 0 is rejected, and H 1 is accepted. The conclusion is that self-concept covariate variables affect students' mathematical concept understanding ability.
The corrected model findings in Table 10 indicate a F observed value of 19,609 with a p-value of 0.000 at a significance level of 0.05. This number demonstrates that the p-value is lower than 0.05. Therefore, H 0 is rejected, and H 1 is accepted. In conclusion, students' mathematical concept understanding ability is simultaneously influenced by the VARK-Fleming model and selfconcept.
The results of the one-way ANCOVA test reveal that the VARK-Fleming model has a greater influence than the expository model on concept understanding ability. Students taught using the VARK-Fleming model had a higher posttest score of concept understanding ability than those taught with the expository model, particularly on presenting concepts in diverse forms of mathematical representation. This is because the Fleming VARK model involves actively engaging students in discovering and understanding concepts, allowing them to develop their individual potential to the fullest. The benefits of VARK-Fleming can make the learning environment energetic and interesting, causing students to appear eager during the learning process. According to its benefits, learning activities are more effective because it integrates the four learning styles and provides direct experience by integrating students as much as possible in understanding mathematical concepts. This model makes it easier for students to understand mathematical concepts (Monariska & Komala, 2020;Nurhidayah, 2021). However, some students are not yet accustomed to combining the four learning styles. This is consistent with previous research conducted by (Didit Chayono & Nuriyatin, 2019;Hidayati & Jahring, 2021;K et al., 2021), which found a positive and significant impact of the Fleming VARK learning model on students' mathematical reasoning and connections.
Similarly, students' self-concept influences their ability to understand concepts. Students with positive selfconcept answer mathematics problems better than students with negative selfconcepts. This finding can be seen in the results of the posttest of concept understanding ability, where students with positive self-concept tend to solve concept understanding ability problems correctly, particularly on indicators of determining, using, and referring to certain operating procedures. This is because students with a positive selfconcept have greater confidence in their abilities and put in more effort to complete their tasks, even when they find them challenging. As a result, they ISSN 2089-8703 (Print) Volume 12, No. 2, 2023, 2009-2023ISSN 2442 Sari & Pujiastuti, 2020) found that an individual's self-concept influences their mathematical abilities, whereby a more positive self-concept is associated with better mathematical skills in terms of formulating concepts and expressing ideas. The Fleming VARK model and self-concept simultaneously influence students' conceptual understanding abilities, as indicated by a p-value < α, suggesting that H1 is accepted. Therefore, when the Fleming VARK model and self-concept are combined, they have an influence on students' mathematical conceptual understanding. Further analysis will be conducted using the t-statistic in Table 11.  Table 11 in the row [X1 = 1.00] shows that the value of t0 is 2.421 with a p-value of 0.018 at a significance level of 0.05. This number demonstrates that the p-value is lower than 0.05, indicating that H 0 is rejected and H 1 is accepted. In conclusion, after adjusting for student self-concept, the mathematical concept understanding ability taught to students using the VARK-Fleming model is superior to that taught using the expository model. Because the VARK-Fleming model combines the four learning styles of visual, auditory, read/write, and kinesthetic, it not only provides direct experience by involving students as much as possible in understanding mathematical concepts (Mulabbiyah, Ismiati, 2018), but it also makes the learning environment active and fun to get students excited so that they do not experience difficulties understanding mathematical concepts. According to this statement, other research (Almola, 2022;Jumrah et al., 2022) discovered that the VARK-Fleming model is more effective in boosting concept understanding and mathematics learning achievement because it integrates the four learning styles. Based on the previous description, it is possible to conclude that the VARK-Fleming model is superior to the expository model.

CONCLUSION AND SUGGESTION
Based on the analysis and discussion of the research data, it can be concluded that: Firstly, there is an influence of the Fleming VARK learning model on students' mathematical conceptual understanding, while controlling for self-concept. Secondly, there is an influence of the covariate ISSN 2089-8703 (Print) Volume 12, No. 2, 2023, 2009-2023ISSN 2442 Based on the research findings and observations in the field, the following suggestions are proposed: The Fleming VARK learning model can be implemented in teaching and learning activities, especially in mathematics, to make students more active, independent, and creative, and to facilitate their understanding of the learning material. Educators should foster positive self-concepts towards mathematics in students to develop internal motivation to strive for excellence in mathematics. Therefore, it is recommended for educators to implement the Fleming VARK learning model as an alternative and varied learning environment that can enhance students' mathematical conceptual understanding. For future research, it is suggested to explore other abilities that can be applied through the Fleming VARK model. Hopefully, this research can provide benefits and contribute valuable insights for educators in general.