Influence of the Guided Discovery Learning Model on Primary School Students ’ Mathematical Problem-solving Skills

Difficulties of primary school students in solving mathematical problems that require problem solving or reasoning abilities indicate the existence of problems in the process of learning mathematics, because one of the purposes of learning mathematics in primary school is to develop the students’ mathematical problem-solving skills. As a solution to overcome the problems, this research aims at understanding how the implementation of the guided discovery learning model influences mathematical problem-solving skills. This research employs a quasi-experimental design by involving all fifth-grade students at one of primary schools in Tebo Regency, Jambi. The simple random sample was conducted to obtain class VB as experimental class, and class VA as control class. Based on the results of the final score, experiential class obtained an average score of 86.00, and control class obtained an average score of 76.93. The results of hypothesis test using t-test obtained tcount = 2,203 > ttable = 2,020, which indicated that H0 was rejected, and H1 was accepted. Therefore, the guided discovery learning model influenced the fifth-grade students’ mathematical problem-solving skills, thus it is recommended for teachers to be able to use the guided discovery learning model in the mathematics learning process.

INTRODUCTION ~ Primary school is the most important part of learning that every student must pass successfully.The success at primary school level will determine success at a further stage.At this level, classroom teachers have the responsibility to teach five fields of study.Mathematics is one of five subject areas, which is taught by teachers as an effort to improve and maintain the quality of human resources (Şahin, 2018).Through the implementation of mathematics learning, every generation in Indonesia is expected to develop the ability of students to think logically in determining how to solve a problem, to train students' analytical thinking in solving a problem accurately and appropriately, and to train students in sorting a problem solving systematically, so that the students have the ability to solve problems experienced directly in daily life (Elfiadi, 2015).
The occurrence of a fundamental change in people's lives in the 21st century (Wijaya, Sudjimat, & Nyoto, 2016)  Mazurier , 2016).This causes the needs of learning that is able to produce educational needs to produce students with problem-solving skills in mathematics, has become something important to be considered.Changes that occur rapidly in the present will provide opportunities, if it is properly utilized, but it can be disastrous, if it is not systematically planned, structured, and measurable (Redhana, 2019).
Basically, the mission of the educational system is to produce students who have the ability and skills in accordance with the needs of every generation (Gürbüztürk, 2018), because in fact, mathematics is considered as a science education that always seeks to overcome problems of each generation (Condon & Wichowsky, 2018).
Indonesia, as a large developing country, must consider the current development, especially those related to basic education (Prearse, 2014), as predicted that in the year 2045, Indonesia will obtain demographic bonus (Kemenku, 2017), in which the number of working or productive age population (Indonesian golden generation) is greater than the old or unproductive population.If this generation is well prepared by the Indonesian educational institutions, they will be able to positively accelerate the progress of Indonesia in every aspect of life, especially in the aspects of technology, education, arts and culture (Prayitno, 2015).answer from a variety of different problems.
Based on these objectives, it is expected that the learning process of mathematics in primary schools' students regarding problem-solving, will produce students who are able to answer the challenges of dynamic world change.
However, in fact, the main problem in the  & Budiningsih, 2017).
The most important aspect to be recognized is that the learning conducted in the classroom by teachers determines the quality of students' mathematical problem-solving skills (Wasserberg, 2018).
Mathematics, as a subject related to counting activities, indeed determines how good the quality of students' problemsolving skills, because it is related to the arithmetical skills (Ulu & Özdemİr, 2018).
Therefore, Griffen (2015)  Based on the aforementioned general problems, it was found that in fifth-grade at one of primary schools in Tebo Regency, Jambi, the learning was still found implementing the one-way learning, i.e.
teachers are more active in delivering learning material to students (Astra & Sumiati, 2008), thus it resulted in some students could not focus on the learning, and they only passively listened to teachers, got sleepy, and played objects that are not related to its learning.In this learning process, the students were assigned to memorize some of the formulas learned in each meeting, without applying it correctly to complex questions.
In addition, the students have not been taught to find their own concepts or formulas to be mastered.In the learning activities conducted by the teachers, the students passively obtained learning material presented by the teachers.These problems are also found in other research in which the results revealed that the students still had low mathematical problem-solving skills, so that it raised the idea in the students' mind that mathematics is a difficult or boring subject to learn (Rahayu & Afriansyah, 2015).This understanding resulted in the lack of the implementation of learning conducted by the teachers, and some students had a low attendance when learning mathematics, which is very influential on students' understanding of the material and subsequent problem-solving skills (Kirksey & Gottfried, 2018).
The problem identification in this research is that the students had low problem-solving skills, due to conventional mathematics teaching performed by the teachers.
Hence, the students considered that mathematics was a difficult and boring subject to learn because it was related to calculations that are difficult to understand.
However, it should be realized that the learning process also has an impact on students' inactivity in classroom, because the learning activity is considered boring by students since they were only asked to memorize formulas without having opportunities to apply it.Thus, they were not used to find their own formula to solve the problems in mathematics.
If this kind of learning still remains, it can negatively affect the students' abilities to compete with other students in regions, national and international scopes.Thus, it is believed to be difficult to determine their development in Indonesia.. Hence, it requires the appropriate response to improve the learning process.with the implementation of KI-3 and KI-4 (Kemendikbud, 2016), which is as follows: Table 1.The Core Competence and Basic Competence.

KI (Kompetensi
Inti -Core Competence) KD (Kompetensi Dasar -Basic Competence) 3. Understand factual and conceptual knowledge by observing and questioning based on curiosity about himself, God's creatures and their activities, and objects found at home, at school, and at playgrounds.

Explain and determine simple building blocks (cubes and blocks).
4. Provide factual and conceptual knowledge in clear, systematic, logical and critical language, in aesthetic works, in movements reflecting healthy children, and in actions reflecting the behavior of children who have a faith and noble character.

Making simple building blocks (cubes and blocks).
In order to find out the findings of this

RESULTS
After conducting the learning process, and providing the final test to the sample classes, the obtained results are as follows.and Ltable with a real level = 0.05, which is presented in Table 2:

Homogeneity Test
The data homogeneity test was conducted to determine whether two sample classes had homogenous in variance or not.The homogeneity test was conducted using Ftest.
The results of determining homogeneity from the research data obtained that the calculation results are as follows:

DISCUSSION
The research results revealed that the guided discovery learning model with a scientific approach was able to achieve KI in mathematics subjects, because the model is considered suitable for learning in primary school.The characteristics of mathematics subjects in primary education units, starts from knowing the concepts, finding formulas, doing problem solving, to reasoning, so that this learning model can change the learning process in a good direction, it is expected to be able to create future generations that will be distinctive from current generations (Uusiautti, Paksuniemi, & Matta, 2013).

CONCLUSION
The conclusion of the implementation of this research showed that the average score of the students' final test obtained in the experimental class was 86.00 while in the control class, the students had an average value of 76.93.Then the students' final test data was analyzed using a statistical test to answer the hypothesis then tcount > ttable or 2.203 > 2.020, in meaning thatH0 was rejected and H1 was accepted.
Pasaribu (2017) defines mathematics as a determinant part of the development and students' self-realization of in primary school education.It is expected to realize the golden generation of Indonesia with the ability to implement their knowledge in every development situation (problemsolving) that develops rapidly(Prastiwi, 2018), thus, the students are used to solving problems that occur in complicated daily life(Nurhanifah, 2018), and being as provisions to answer future needs by being actively involved in determining change in the world.The problem-solving ability is a logical thinking process in determining the most appropriate way to solve a problem.According to Polyo (in Rahayu & Afriansyah, 2015), problem solving is defined as an intellectual activity in solving a problem encountered based on the provision of knowledge, so that every student needs to have it as evidence of their participation in learning mathematics taught by teachers in primary school.According to Charles and O'Daffer (in Harahap & Surya, 2017), the purpose of teaching mathematical problem-solving skills of students is to boost their thinking skills, to enhance the ability to choose and determine the right strategy in solving a problem, to develop attitudes and confidence in solving problems, to get used to using their knowledge in connecting theory with practice, to train them to monitor and assess their own thoughts and work in solving problems, to get used to solving problems in a cooperative condition, to get used to finding the correct Akhmad Badrul Lubis, Yelvema Miaz, & Indah Eka Putri, Influence of the Guided Discovery Learning… [255] states that learning in classroom will determine the quality of students produced.Homeroom teachers as an educator who are responsible in classroom become the main factor in determining the students' problem-solving skills.According toShabir (2015), since homeroom teachers directly face the students in classroom, they must be able to implement the learning in classroom that is interesting for students, so that they can students to find their own knowledge to be applied in solving their problems.Hence, they will get used to carry out certain problem-solving activities related to mathematics, and even other four subjects.However, this learning is even final test to the sample classes to making a research).The instrument used to determine the students' mathematical problem-solving skills is a list of questions that required problem-solving skills, which were administered to the students to solve by both of the classes.The results were checked using 4-point scale rubric, namely analytical rubric, so that teachers could make assessments based on certain criteria, so they were able to analyze the strengths and weaknesses of each student based on predetermined criteria.This research was conducted from 20 April 2018 to 5 May 2018 at one of the primary schools in Tebo Regency, Jambi, in which the 5B class was determined as an experimental class and the 5A class was determined as a control class.In each class, there were five meetings conducted Moreover, the experimental class with 21 students obtained the maximum score of 100, and the minimum score of 75, and the number of students who had a score above KKM was 18 students (85.71%), which was higher than the control class with 22 students that obtained the maximum score of 90, and the minimum score of 53, and the number of students who had a score above KKM was 15 students (68.18%).Data Analysis After obtaining the final test calculation results, several tests were conducted, namely normality tests, data homogeneity tests, and statistical tests conducted to answer the research hypotheses, with the explanation as follows: Normality Test In this research, the Lilliefors test was used for data normality testing.Based on the Lilliefors test on the experimental and control classes, the results obtained that L0 of the guided discovery learning model has developed the fifth-grade primary school students' mathematical problem-solving skills(Abalı & Kaan, 2018).According toNahdi (2018), the research results revealed that the guided discovery learning model had a good influence on the students' problem-solving skills.Similarly,Nupita's (2013) research revealed that the guided discovery learning model had a positive influence on the fifth-grade primary school students' problem-solving skills in natural sciences.These research results are in line with the results of this research, which revealed that the guided discovery learning model had an influence on the fifththe scientific approach in developing the fifth-grade primary school students' mathematical problemsolving skills, which is in the final stages of development of knowledge from concrete operations to formal operations.This learning had been conducted to change the students' knowledge(Winne, 1985).Through various series of activities, the results of the final test of the experimental class that was exposed toguided discovery learning model, was better than the control classes that was exposed toconventional learning.This was evidenced by the final test scores of the experimental class students obtaining an average score of 86.00, while the control class obtaining an average score of 76.93.This research can be used to differentiate the quality of learning (Park, 2017), because what is received (information) by the students at each meeting will be able to have a significant influence on the development of the students' problem-solving skills (Temur, 2012).If the experimental class and the control class are compared with the two data, the obtained conclusion would-be that the experimental class with the guided discovery learning method resulted in the better mathematical problem-solving skills, of tests, such as normality, homogeneity, and hypotheses tests, the results showed that the data were normally distributed, and homogeneous variance, which fulfilled the prerequisites to answer the hypothesis by using t-test.The results revealed that tcount of 2.203 > ttable of 2.020, which indicated that the guided discovery learning model, had an influence on the fifth-grade students' mathematical problem-solving skills at one of primary schools in Tebo Regency, Jambi.Based on the final test conducted by the researcher, the test given has an important role in the learning process (Şahin, 2018), In attempting to understand the students' problem-solving skills in the experimental and control classes, the percentage of indicators for each item was analyzed to see the students' mathematical problemscale of 3 with the indicators of A and C, and the average items of 100%, while the lowest percentage of questions were in question 7 on the scale of 3 with the indicators of A and B, and the average items of 44.44%.Meanwhile, in the control class, the highest percentage of questions were in question 2b on the scale of 3 with the indicators of A and B, and the average items of 75.45%, while the lowest percentage of questions were in question 7 on the scale of 3 with the indicators of A, B, and C, and the average items of 47.47%.The results indicated that the learning model could affect the students' problemsolving skills well, because the future education must be able to produce the students who can answer world changes occurred quickly, with the students'problem-solving skills(Kim & Lee, 2018).Based on the aforementioned explanation, in the process of problem-solving activities, the students in the experimental class had a better ability to solve problems provided by the teacher.In contrary, the students in the control class tended to have difficulty in solving problems, since many of the students answered the questions incorrectly , so that the results revealed that the guided discovery learning model had an influence on the fifth-grade primary school students' mathematical problem-solving skills, and the interrelation of each , mathematics is a field of study consisting of concepts, abstracts, algorithms, and symbols, which are directly related to the real world of students(Kara

Table 2 .
Result of Students' Final Tests.

Table 3 .
Result of Normality Test on the Sample Class.

Table 4 .
Result of Homogeneity Test on the Sample Class.

Table 5 .
Result of Hypothesis Test on the Sample Class.