The Effect of Realistic Mathematical Approaches and Learning Independence on Students' Mathematical Communication Ability SD Negeri Cot Meuraja

Mathematics learning objectives formulated by the National Council of Teachers of Mathematics Fahradina (2014: 55) are: (1) learning to communicate (mathematical comminication), (2) learning to reason (mathematical reasoning), (3) learning to solve problems (mathematical problem solving), (4) learning to link ideas (mathematical connections), (5) forming positive attitudes toward mathematics. Irhamna (2020) states that Mathematics is a universal science. Mathematics is also seen as the queen of science. From the objectives of learning mathematics above, it appears that one of the aspects emphasized in the 2013 curriculum is improving students' mathematical communication skills. Mathematical communication is very important for students. This is contained in the National Council of Teachers of Mathematics which explains that communication is an essential part of mathematics and mathematics education. This opinion suggests the importance of communication in mathematics learning. The facts in the field show that the students 'mathematical communication skills are still low, the teacher is not able to improve mathematical communication so that the students' mathematical communication skills are only in short verbal answers to various statements submitted by the teacher. According to research conducted by Sokoine (2015: Abstract

499), the communication skills program felt by most of the course respondents is very important for the acquisition of communication skills needed in their academic life. So in the process of learning mathematics it is very important for students to have the ability to communicate mathematically. Learning communication in mathematics fosters interaction and expresses ideas in the classroom because students learn in an active atmosphere. The same is the case with research conducted by (Hernawati, Mahmudi, Lestari, 2013) which states that in reality students' mathematical communication skills are still low. This is due to the lack of student activity where they have not carried out activities that are maximally meaningful such as discovering mathematical concepts and principles, exploring concepts, and analyzing mathematical problems.
One of the ways that teachers can use toimproving mathematical communication and student learning independence is to use the Realistic Mathematical Approach. The development of a realistic mathematics approach actually combines views of what mathematics is, how students learn mathematics and how mathematics should be taught (Johar, 2007: 176). Realistic mathematics learning emphasizes the relationship between mathematical concepts that children have in everyday life. Learning using a realistic mathematics approach begins with solving problems that exist around students and is based on the knowledge already possessed by students, so that it is expected to improve mathematical communication skills and student learning independence. In a realistic mathematical approach, mathematics is presented as a process, as a human activity, not as a finished product. Rahmawati (2015: 226) in her journal states that the theory of realistic mathematical approaches is in line with learning theories that are currently developing, such as constructivism and contextual learning. Realistic mathematics approach is a learning theory developed specifically for mathematics. Learning focuses more on learning activities on students and the environment and teaching materials that are arranged in such a way those students are more active in constructing or developing their own acquired knowledge, while teachers act more as facilitators. A realistic mathematics approach can arouse students' enthusiasm because the problems and examples given are real and exist in the real life of students (Ernalita, 2016: 231). 226) in his journal states that the theory of realistic mathematical approaches is in line with currently developing learning theories, such as constructivism and contextual learning. Realistic mathematics approach is a learning theory developed specifically for mathematics. Learning focuses more on learning activities on students and the environment and teaching materials that are arranged in such a way those students are more active in constructing or developing their own acquired knowledge, while teachers act more as facilitators. A realistic mathematics approach can arouse students' enthusiasm because the problems and examples given are real and exist in the real life of students (Ernalita, 2016: 231). 226) in his journal states that the theory of realistic mathematical approaches is in line with currently developing learning theories, such as constructivism and contextual learning. Realistic mathematics approach is a learning theory developed specifically for mathematics. Learning focuses more on learning activities on students and the environment and teaching materials that are arranged in such a way those students are more active in constructing or developing their own acquired knowledge, while teachers act more as facilitators. A realistic mathematics approach can arouse students' enthusiasm because the problems and examples given are real and exist in the real life of students (Ernalita, 2016: 231).
Learning mathematics using a realistic mathematics approach is one of the hopes to strive for the development of knowledge, increase understanding, mastery, mathematical communication and student learning independence. Because learning with a realistic approach is designed starting from solving problems around students and based on the knowledge that students already have. The realistic mathematics learning approach can be one of several learning alternatives that can involve students actively, collaborate, discuss, and argue with classmates so that they can find mathematical concepts themselves through presenting problems that are close to student life (Gumanambo, 2016: 144).

Definition of a Realistic Mathematical Approach
Realistic Mathematics Education (RME) is a mathematics learning approach developed by Freudenthal in the Netherlands in 1970. Realistic mathematics is school mathematics which is implemented by placing students' realities and experiences as a starting point for learning. Realistic problems are used as the source of the emergence of mathematical concepts or formal mathematical knowledge. To be able to teach mathematics, a teacher must be able to prepare lesson plans so mathematics lessonscan be received by students (Nasution, 2020).
The realistic mathematics approach is a theory in mathematics education which is based on the idea that mathematics is a human activity and mathematics must be connected significantly to the context of students' daily lives as a source of development and as an area of application of both horizontal and vertical mathematical processes. Realistic mathematics learning begins with real problems, so students can use previous experiences directly. With realistic mathematics learning, students can develop a more complete concept. Then students can also apply the concepts to new fields and the real world. Anwar (2010) states that the realistic mathematics approach is a mathematics learning approach that tries to use students' experiences and environments as primary teaching aids. In line with that Supinah (2008: 15-16) states that the realistic mathematics approach is a learning theory that has been developed specifically for mathematics.
Based on Some of these opinions can be concluded that the realistic mathematics approach is an approach to learning mathematics that is close to the real life of students as a means of increasing understanding and reasoning power.

Definition of Independent Learning
Independent study does not mean studying alone. Often people confuse self-study as learning alone. Chapter II of Law Number 20 of 2003 concerning the National Education System states that national education functions to develop capabilities and shape a dignified national character and civilization in order to educate the nation's life, aiming at developing the potential of students to become human beings who believe and have faith in God Who The One and Only, have a noble character, are healthy, knowledgeable, capable, creative, independent, and become democratic and responsible citizens. It is clear that the word independency has emerged as one of the goals of our national education. Therefore, the handling requires special attention from all teachers, especially since there is no special subject on independence. a. Each individual tries to increase the responsibility for making various decisions. b. Independent learning is seen as a trait that is inherent in every person and learning situation. c. Learning to be independent does not mean separating yourself from other people. d. With independent learning, students can transfer their learning outcomes in the form of knowledge and skills into other situations.
e. Students doing self-study can involve a variety of resources and activities, such as: self-reading, group study, exercises, electronic dialogue, and correspondence activities. f. The effective role of teachers in independent learning is still possible, such as dialogue with students, finding resources, evaluating results, and providing creative ideas. g. Several educational institutions are developing independent learning into a more open program (such as the Open University) as an alternative to individual learning and other innovative programs. From the definition of independent learning above, it can be concluded that independence is the behavior of students in realizing their will or desires in a real way without depending on others, in this case the student is able to do self-study, can determine effective learning methods, is able to carry out tasks -the task of learning well and able to carry out learning activities independently. .

Mathematical Communication
When there is a discussion between students, communication skills are very important, where students are expected to be able to state, explain, describe, listen, ask and work together so that they can bring students to a deep understanding of mathematics. Communication skills are seen as the ability of students to communicate the mathematics they have learned as the message content that must be conveyed. According to Wahyudin (2012: 529) communication can support students' learning of new mathematical concepts when they play a role in situations, take, use objects, provide reports and oral explanations, use diagrams, write, and use symbols. -mathematical symbols.
Another definition of mathematical communication put forward by Romberg and Chair (Sumarmo, 2005: 22) is: connecting real objects, pictures, and diagrams into mathematical ideas; explain ideas, situations and mathematical relations orally or in writing with real objects, pictures, graphics and algebra; express everyday events in language or mathematical symbols; listening, discussing, and writing about mathematics; reading with the understanding of a written mathematical presentation, making conjectures, constructing arguments, formulating definitions and generalizations; explain and make questions about mathematics that have been learned.
Indicators of students' ability in mathematics communication in mathematics learning according to NCTM (1989: 214) can be seen from: a. Ability to express mathematical ideas through oral, written, and demonstrate and describe them visually. b. Ability to understand, interprets, and evaluates mathematical ideas both orally, in writing, and in other visual forms. c. Ability to use terms, mathematical notations and structures to present ideas and describe relationships with situation models. Broadly speaking, according to Nurahman (2011: 107) it can be concluded that mathematical communication consists of oral and written communication. In this study, oral communication can occur in group discussion activities. Whereas written communication can be grouped into three, namely (1) explaining an idea or situation from a picture or graphic in your own words in writing (writing), (2) stating a situation with a picture or graphic (drawing) and (3) expressing a situation in the form of a mathematical model (mathematical expression).

III. Research Methods
This research was conducted at SD Negeri Cot Meuraja. This research was conducted in the even semester starting from January to February, 2019/2020 school year. The population in this study were all fourth grade students of SD Negeri Cot Meuraja for the 2019/2020 academic year, totaling 50 people who were distributed in two classes, namely class IV-A with 25 students and VI-B with 25 students. Using total sampling, that is, all populations are used as samples. Class IV-A is used as experimental class I which is taught using Realistic Mathemetics learning and class IV-B is used as control class which is taught using conventional learning approaches. This research uses a quantitative approach and this type of research is pseudo experiment. The research design used was a quasy experiment with 2 factorialsx 2 with 2-way analysis of variance (ANOVA) techniques. The tool used to collect data to determine the level of mathematical communication skills of students is a description test. Measuring students' learning independence in the form of a questionnaire.
The question instrument was validated by statistical tests, carried out by calculating the validity, reliability, difficulty level, and distinguishing power so that the instrument was valid. The data that has been collected will be analyzed descriptively and inferentially. Descriptive statistical data is needed to find the mean, median, standard deviation, variance, range, frequency of graph data and other information using the SPSS 20.0 program. Before the data is analyzed inferentially, the data must be subjected to a prerequisite test in terms of normality and homogeneity.
Furthermore, testing the hypothesis aims to prove the truth of the research hypothesis. To test the hypothesis, a two-way analysis of variance (ANOVA) test was used with the help of SPSS 20.0.The statistical model of this research experiment according to Syahputra (2016: 169). Testing criteria,If F0> Ftable at the selected significant level with the db of the numerator is the appropriate db, then H0 is rejected, so it can be concluded that there is a difference in the average increase between the groups tested, on the contrary for F0 ≤ Ftable, then H0 is accepted. For two-way ANOVA, the first step is to test the statistical hypothesis of the interaction effect, namely F0 (AB). If F0 (AB) ≤ Ftable or H0 is accepted which means there is no interaction effect, then the main effect hypothesis test is carried out, namely the F0 (A) test to study the average difference between A. Conversely, if F0 (AB)> Ftable or H0 is rejected, meaning that there is a significant interaction effect, so the consequence must be a simple effect test.

Description of Data a. Pre-Test of Students' Mathematical Communication Ability in Experiment Class
The lowest score obtained by the students was 45, while the highest score was 75 with an average value of 60.40; the median is 60.00; and mode namely 60; standard deviation of 7.06; and a variant of 49.83. Furthermore, the frequency distribution of the data can be described in the following histogram form.

Figure 1. Pre-Test Histograms of Students' Mathematical Communication Ability in Experiment Class
Based on the histogram, it can be seen that the pre-test results of the experimental class with the highest number of frequencies are in the 57-62 interval class, while the least number of frequencies is in the 75-80 interval class.

b. Pre-Test Mathematical Communication Skills for Control Class Students
The lowest score obtained by the students was 40, while the highest score was 80 with the average score obtained was 59.60; the median is 60.00; and mode namely 65; standard deviation of 10.20; and a variant of 104.00. Students' mathematical communication skills in the control class with the highest number of frequencies are in the 54-60 interval class, while the least frequency is in the 75-81 interval class.

c. Description of Independent Learning Experiment Class Students
In addition to giving pre-test to students before treatment, the study also looked at students' early learning independence which was then divided into 2 groups, namely a group of students who had high learning independence and a group of students who had low learning independence. The following shows the frequency distribution of students' independent learning in the experimental class.

d. Description of Independent Learning of Control Class Students
Furthermore, it can be seen the distribution of independent learning in the control class. The following is the frequency distribution of students' independent learning in the control class. Based on table It can be seen that of the 25 students in the control class there are 13 students who have high learning independence and 12 students who have low learning independence.

e. Post-test Students' Mathematical Communication Skills Taught with a Realistic Mathematical Approach
Based on the data obtained and the results of statistical calculations, it is known that the mathematical communication skills of students who are taught with a realistic mathematical approach get the lowest score, namely 73, and the highest score, namely 100, with an average of 86.17; the variant is 57.47 and the standard deviation is 7.58. The frequency distribution of students' mathematical communication skills scores taught using a realistic mathematical approach is presented in the following table. From the table, the frequency distribution of students 'mathematical communication skills taught with a realistic mathematical approach can be seen that 8 students or 32% of 25 students have mathematical communication skills below the average of 82 (post-test average results of students' mathematical communication skills). experimental class and control class). Meanwhile, 17 or 68% of other students have mathematical communication skills above average.

f. Post-test Students' Mathematical Communication Skills Taught with Conventional Approaches
From the data obtained and the results of statistical calculations, it is known that the mathematical communication skills of students who are taught using the conventional approach get the lowest score, namely 63, and the highest score, namely 93, with an average of 78.40; the variant is 60.30 and the standard deviation is 7.77. The frequency distribution of students' mathematical communication skills scores taught by conventional approaches is presented in the following table.

j. Students' Mathematical Communication Skills Taught with Realistic Mathematical Approaches and Have Low Learning Independence
From the data obtained and the results of statistical calculations it is known that the mathematical communication skills of students who are taught with a realistic mathematical approach and have low learning independence get the lowest score, namely 73, and the highest score, namely 93, with an average of 93.30; the variance is 16.83 and the standard deviation is 4.10. The frequency distribution of students' mathematical communication skills scores who are taught with a realistic mathematical approach and have low learning independence, it can be seen that 12 students or 80% of 15 students have mathematical communication skills below average. Meanwhile, 3 or 20% of other students have mathematical communication skills above average.

k. Students' Mathematical Communication Skills Taught with Conventional Approaches and Have High Learning Independence
From the data obtained and the results of statistical calculations it is known that the mathematical communication skills of students who are taught with conventional approaches and have high learning independence get the lowest score, namely 67, and the highest score, namely 87, with an average of 77.69; variants of 41.45 and standard deviation of 6.44. The frequency distribution of students' mathematical communication skills scores who are taught using a conventional approach and have high learning independence is presented in the following table. From the table, it is known that 13 of the 25 students in the control class have high learning independence. Based on statistical calculations it can also be seen that 8 students or 62% of the 13 students have mathematical communication skills below average. Meanwhile, 5 or 38% of other students have mathematical communication skills above average.

l. Students' Mathematical Communication Skills Taught with Conventional Approaches and Have Low Learning Independence
From the data obtained and the results of statistical calculations it is known that the mathematical communication skills of students who are taught with conventional approaches and have low learning independence get the lowest score, namely 63, and the highest score, namely 93, with an average of 79.17; the variant is 85.10 and the standard deviation is 9.23. The frequency distribution of students' mathematical communication skills scores who are taught using conventional approaches and have low learning independence, it can be seen that 12 out of 25 students in the control class have low learning independence. Based on statistical calculations it can also be seen that 9 students or 75% of the 12 students have mathematical communication skills below average. Meanwhile, 3 or 25% of other students have mathematical communication skills above average.

Data Analysis Testing a. Normality Test
The results of the pre-test data normality test of students' mathematical communication skills with the Shapiro-Wilk test obtained a probability value or a significant value of 0.622> 0.05, thus it can be concluded that the pre-test data were normally distributed. The results of the post-test data normality test of students' mathematical communication skills with the Shapiro-Wilk test obtained a probability value or a significant value of 0.378> 0.05, thus it can be concluded that the post-test datadistributed normally.

b. Homogeneity Test
Homogeneity testing of the pre-test data obtained a probability value or a significant value of 0.52> 0.05, thus it can concluded that the research data group is relatively the same or is homogeneous. Based on the table above shows that the post-test data homogeneity test obtained a probability value or a significant value of 0.996> 0.05, thus concluded that the research data group based on post-test scores is relatively the same or homogeneous.

Testing Hypothesis
Hypothesis testing requirements for parametric tests have been met, namely group data is normally distributed and has homogeneous variance. Hypothesis testing in this study uses two-way ANOVA with 2x2 factorial, hypothesis testing is calculated with the help of SPSS version 23. Hypothesis testing data can be seen in the following table:  Based on Table 8; Table 9; Table 10; and Table 11. it can be explained about the research hypothesis testing as follows:

First Hypothesis
Based on the SPSS output in Table 8 regarding mathematical communication skills based on the learning approach, it is found that the value of Fcount = 23,476 andthe probability value or significant value of the learning approach is 0.000 <0.05. Thus it can be said that there is a significant difference between the average mathematical communication skills of students who are taught with realistic mathematics learning approaches compared to conventional approaches. Furthermore, based on the SPSS output on the comparison of mathematical communication skills based on the learning approach in Table 9, it was found that the average mathematical communication skills of students who were taught using a realistic mathematics learning approach were 87.383. Meanwhile, students' mathematical communication skills taught by conventional approaches amounted to 78.391. This indicates that the average mathematical communication skills of students taught with a realistic mathematics learning approach are higher than the average mathematical communication skills of students taught using conventional approaches. So that the hypothesis testing rejects Ho and accepts Ha. With the conclusion that the mathematical communication skills of students who are taught with realistic mathematics learning approaches are higher than conventional approaches.

Second Hypothesis
Based on the SPSS output in Table 8 regarding mathematical communication skills based on independent learning, it is found that the value of Fcount = 7,673 and value probability or significant value of 0.008 <0.05. Thus it can be said that there is a significant difference between the mathematical communication skills of students who have high learning independence compared to the mathematical communication skills of students who have low learning independence. Furthermore, based on the SPSS output on the comparison of mathematical communication skills based on student learning independence in Table 10. it was found that the mathematical communication skills of students who had high learning independence were 85.458. While the mathematical communication skills of students who have low learning independence amounted to 80.317. This shows that the average mathematical communication ability of students who have high learning independence is higher than the mathematical communication skills of students who have low learning independence. So that the hypothesis testing rejects Ho and accepts Ha. With the conclusion that the mathematical communication skills of students who have high learning independence are higher than students who have low learning independence.

Third Hypothesis
Based on the SPSS output in Table 8, it is found that Fcount = 13.003 and a significant value of 0.001 with α = 0.05. Then it can be seen that the sig. 0.001 <0.05 so that the hypothesis testing rejects Ho and accepts Ha. With the conclusion that there is an interaction between the learning approach and student learning independence in influencing students' mathematical communication skills.

Test Tukey
The explanation is as follows: 1) Based on the results of Tukey's test, it can be concluded that there is a significant difference between the mathematical communication skills of students who are taught with a realistic mathematics approach and have high learning independence compared to the mathematical communication skills of students who are taught with a realistic mathematics approach and have low learning independence (Mean Diff = 11, 83; and significant value = 0.000). 2) Based on the results of Tukey's test, it can be concluded that there is a significant difference between the mathematical communication skills of students who are taught with realistic mathematics approaches and have high learning independence compared to the mathematical communication skills of students who are taught with conventional approaches and have high learning independence (Mean Diff = 15.68 ; and significant value = 0.000). 3) Based on the results of Tukey's test, it can be concluded that there is a significant difference between the mathematical communication skills of students who are taught with a realistic mathematics approach and have high learning independence compared to the mathematical communication skills of students who are taught with conventional approaches and have low learning independence (Mean Diff = 14,13 ; and the value sig = 0,000).

V. Conclusion
Based on the discussion previously described, several conclusions can be drawn among them as follows: 1. Students' mathematical communication skills taught with realistic mathematical approaches are higher than conventional approaches (Fcount = 23.476 and sig. 0.000> 0.05). 2. The mathematical communication skills of students who have high learning independence are higher than students who have low learning independence (Fcount = 7.673 and sig. 0.008> 0.05). 3. There is an interaction between the learning approach and learning independence in influencing students' mathematical communication skills (Fcount = 13.003 and sig. 0.001> 0.05).