







































 

 

 

 

 

 
 

LEARNING PROCESS ANALYSIS MIS GUPPI PADANG HILIR 

(Using a Multivariate Analysis Approach) 

1Rizki Maulida,2 Riska Fadhilah Hutasuhut,3 Juwita Tindaon,4 Rika Restella,5 Hasratuddin 

Universitas Potential Utama, Universitas Muhammadiyah Sumatera Utara, Universitas 

Quality Berastagi, Institut Agama Islam Negeri Langsa, Universitas Negeri Medan 

 

Email : rizkimaulida24@gmail.com 

Email : riskafadhilahhutasuhut@gmail.com 

Email : wieta.niez@gmail.com 

Email : restelastela@gmail.com 
 

Abstract. This study aims to look at student learning outcomes through National Exam scores 

conducted in the last ten years. This research also looks at the correlation between the subjects 

tested in the national exam. This research examines the suitability of lesson objectives with the 

learning outcomes obtained by students. The data in this control process used data on student UN 

scores in 2012 - 2023 in three subjects, namely: B Indonesia, Science, and Mathematics. This 

process analysis uses data analyzed in 2017 - 2018 with the results showing the highest 

correlation value is 59.8121, namely between science and Indonesian subjects. While the lowest 

correlation is 33.13578, namely between math and science subjects. Process data is seen in 2019 

-2020 with the results of the highest correlation value is 8.265306, namely in Indonesian and 

mathematics subjects. While the lowest correlation value is 7.281746, namely in math and 

science subjects. 

 

 
Keywords: Control, Process, Learning, UN, MIS, Tebing Tinggi 

 
 

A. Introduction 

The function of education according to Law No. 20 of 2003 is to develop knowledge and 

form a dignified character in order to educate the nation's life (Law No. 20 of 2003.Pdf, n.d.). 

based on the function of education listed in Law no. 20, we can see that education is a place to 

build human character in order to build an understanding of critical thinking, independent and 

structured. The implementation of education should have control over the implementation 

process. control activities on the implementation of learning can be seen by analyzing student 

test scores for severalyears. Control activitieson the results of this learning aim to see the 

suitability of learning implementation standards and learning outcomes. 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 1

mailto:rizkimaulida24@gmail.com
mailto:riskafadhilahhutasuhut@gmail.com
mailto:wieta.niez@gmail.com
mailto:restelastela@gmail.com


Educational facilities are an important aspect in learning activities (Antoro, 2023). 

Education management is a stimulus to advance education in Indonesia. Activities in education 

implementation management are the beginning of learning activities that are appropriate and in 

accordance with learning objectives. one of the learning management activities is evaluation in 

the learning process. according to (Darodjat & Wahyudhiana, 2015) evaluation has three terms, 

namely measurement, assessment, evaluation. Evaluation activity is a process to collect, analyze 

and interpret information with the aim of knowing the level of achievement of learning 

objectives (Ratnawulan & Rusdiana, 2014). Research conducted with the aim of obtaining 

information on the learning process and as material for evaluating the learning process has also 

been conducted by (Antoro, 2023). The research conducted by Antoro found that a larger portion 

of student learning was experienced by students who were in fullday class learning, students who 

studied with a larger portion had a high average score. 

According to (Hasratuddin, n.d.) multivariate statistical methods are data analysis 

techniques to see the correlation between variables as a system by taking into account the 

correlation between these variables. The purpose of multivariate analysis is to find the structure 

of the data characteristics. 

This paper is to analyze a learning process that takes place at MIS Teladan GUPPI 

Tebing Tinggi. the purpose of this study is to see the correlation between students' national exam 

results and the learning process. whether the students' national exam results are in accordance 

with the assessment standards enforced in Indonesia. 

 
B. Discussion 

 

This research was conducted at MIS Teladan GUPPI Tebing Tinggi which is located at Jl. 

Bhakti Gg. Karya Satria Village, Padang Hilir Subdistrict, District / City Tebing Tinggi, North 

Sumatra Province. The sample in this study were all students' National Exam scores in 2014 - 

2023 in the subjects of Mathematics, Science, and Indonesian. The amount of data for all 

students for 10 years is 258 students. Each year 2012 - 2013 is 17 

students, 2013 - 2014 as many as 14 s t u d e n t s , 2014 - 2015 as many as 12 students, 2015 - 2016 

17 students, 2016 - 2017 as many as 29 students, 2017 - 2018 as many as 39 students, 2018 -2019 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 2



as many as 28 students, 2019 - 2020 as many as 42 s tudents,  2022 - 2021 as many as 18 students, 

2021 - 2022 with 39 students, 2022 - 2023 with 19 students. 

 

The focus of data analysis in this study is; 1) the relationship between math, science, and 

Indonesian subjects. The data used is data from the national exam results (UN) of students in 

2012 - 2023, namely the data used is data for the last 11 years. The use of this data aims to see 

the correlation between subjects. It is used data in 2017 - 2018. The latest data to be seen is in 

2019-2020 as many as 42 students. With the subjects of math, science, and Indonesian language. 

see the control of the learning process for eleven years of UN data using the control model. 

𝑈𝑝𝑒𝑟𝐶𝑜𝑛𝑡𝑟𝑜𝑙 𝐿𝑖𝑚𝑖𝑡 (𝑈𝐶𝐿), 𝑇2 = 𝑚 ( 𝑋 ̅𝑦   -̿𝑋)̅�̅̅�-̅1(̅ �̅�𝑦 -̿𝑋) 𝑡 

C. Data Analysis 2017 - 2018 

 
Data analysis in this study used national exam data (UN) in 2003 - 2004 at MIS Teladan 

GUPPI Tebing Tinggi, namely three subjects of Indonesian Language (X1) Science (X2), and 

Mathematics (X3). The amount of data for that year is 39 data. The results of the average 

calculation can be seen in the following table: 

Table 1. Average comparison of 3 subjects 

SUBJECT B. INDONESIA IPA MATHEMATICS 

AMOUNT 2347.47 2341.89 2603.6 

MEAN 60.19 60.05 66.76 

VAR 194.1513 107.4918 48.40722 

 

 
Based on the data above, it can be seen that the lowest average student score in the 

subjects is in science subjects, namely 60.05, while the highest average subject is in math 

subjects, namely 66.76. This shows that students' mathematics skills are better than science and 

Indonesian subjects. This shows that students' mathematics skills are better than science and 

Indonesian subjects. 

To see the correlation between the subjects, it can be seen in table 2, as follows: 

 
Table 2. Covariance Correlation Matrix 

B. INDONESIA 1 59.8121 38.06963 

IPA 59.8121 1 33.13578 

MATHEMATICS 38.06963 33.13578 1 
 B. Indonesia IPA Math 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 3



Based on the data above, the highest correlation value is 59.8121, namely between 

science and Indonesian subjects. While the lowest correlation is 33.13578 which is between math 

and science subjects. The data shows that the highest correlation between subjects is between 

Indonesian and science, while the weakest correlation value is between science and math. When 

looking at the data as a whole, we can conclude that the data is positive, this shows that students 

try to do maximum learning in all subjects tested. 

The prediction of subjects against other subjects can be seen with the inverse correlation 

in table 3 below: 

Table 3 Inverse correlation matrix 

INDONESIAN 
LANGUAGE 

0.00598919 -0.002383584 -0.003078555 

IPA 0.000489005 0.0115965 -0.008322629 

MATHEMATICS -0.005044904 -0.006063495 0.028776209 
 Bahasa Indonesia IPA Math 

 

 

D. Data Analysis 2019 - 2020 

 
The data analysis used in this section is the UN scores of 42 students in 2019-2020. The 

data used are students' national exam scores in B Indonesia, science, and math subjects. The 

correlation between these subjects can be seen in table 4 below: 

Table 4. Basic Calculation 

SUBJECT B INDONESIA IPA MATHEMATICS 

AMOUNT 3480 3451 3487 

MEAN 82.85714 82.16667 83.02381 

VARIAN 9.783972 8.288618 8.218931 

Based on the data above, the highest average score is 83.02 in math subjects. The subject 

with the lowest average score is 82.16667, namely in science subjects. The data explains that the 

highest student ability is in math subjects, while the lowest student ability is in science subjects, 

namely 82.16. The correlation between subjects can be seen in table 5 below: 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 4



Table 5. Variance-Covariance Correlation Matrix 

INDONESIAN 
LANGUAGE 

1 7.833333 8.265306 

IPA 7.833333 1 7.281746 

MATHEMATICS 8.265306 7.281746 1 

 Bahasa Indonesia IPA Math 

The data above explains that the highest correlation value is 8.265306, namely in 

Indonesian and mathematics subjects. While the lowest correlation value is 7.281746, namely in 

math and science subjects. The overall correlation value has a positive value, this shows that 

students put maximum effort in all subjects. 

E. Learning Process Control Analysis 

 

The data used in the analysis is a range of 11 classes in 2012 - 2023. The data used is data 

on student scores on the national exam consisting of scores in Indonesian, science, and math 

subjects. 

First analyze the process control by calculating the determinant value of the covariance 

matrix, can be seen in the following table: 

Table 6. determinant of covariance matrix and UCL value 

 
YEAR MDKO UCL 2. UCL 3 

 

2012 - 2013 28.84768 23.990 33.638 

2013 - 2014 18.15177 23.990 33.638 

2014 - 2015 0.000186 23.990 33.638 

2015 - 2016 3.05895 23.990 33.638 

2016 - 2017 0.000117 23.990 33.638 

2017 - 2018 1.45886 23.990 33.638 

2018 - 2019 0.002368 23.990 33.638 

2019 - 2020 0.049788 23.990 33.638 

2020 - 2021 0.005479 23.990 33.638 

2021 - 2022 0.01623 23.990 33.638 

2022 - 2023 0.024024 23.990 33.638 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 5



 
 

Based on the results of the data obtained, we can explain the value in 2012 - 2013 the 

learning process is not going well, this can be seen from the value of the determinant covariance 

matrix (MDKov) passing the UCL value limit of 2. If the data obtained will be used to control 

the learning process in the following year, the prediction data will be refined so that the data 

obtained is not above the UCL. The iteration process is used to refine the control by not 

including the values in 2012 - 2013. 

The analysis of iteration 2, namely data analysis for 2013 - 2023 can be seen in the 

following table: 

YEAR MDKO UCL 2 UCL 3 

2013 - 2014 18.15177 13.613 19.281 

2014 - 2015 0.000186 13.613 19.281 

2015 - 2016 3.05895 13.613 19.281 

2016 - 2017 0.000117 13.613 19.281 

2017 - 2018 1.45886 13.613 19.281 

2018 - 2019 0.002368 13.613 19.281 

2019 - 2020 0.049788 13.613 19.281 

2020 - 2021 0.005479 13.613 19.281 

2021 - 2022 0.01623 13.613 19.281 

2022 - 2023 0.024024 13.613 19.281 

DETERMINANT MATRIX VARIABILITY CONTROL 

MDKo UCL 2. UCL 3 

40 

35 

30 

25 

20 

15 

10 

5 

0 
2012 -2013 - 2014 - 2015 - 2016 - 2017 - 2018 - 2019 - 2020 - 2021 - 2022 - 

2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 6



 
 

The data above explains the variableity control line (MDKoV) and the PBM achievement 

control line (T^2) are on the bottom line of UCL 2 and UCL 3. This shows that the learning 

process for 10 years has been going well and in accordance with the learning outcomes. it can be 

concluded that the UN scores from 2013 - 2023 can be used as a controller of the learning 

process. 

F. Summary 

a. Correlation between subjects 

1. UN data 2017 - 2018 

a) Based on the data obtained, we can conclude that in the UN subjects namely B 

Indonesia, Science, and Mathematics at MIS Teladan GUPPI Tebing Tinggi 

school in 2017 - 2018, the average value in B Indonesia subject is 60.19 with a 

standard deviation of 13.93, science subject is 60.05 with a standard deviation of 

10.37, and math subject is 66.76 with a standard deviation of 6.96. 

b) The highest correlation between subjects in 2017-2018 was 59.8121, between 

science and Bahasa Indonesia. While the lowest correlation is 33.13578 which is 

between math and science subjects 

Chart Title 

25 

 
20    

15 

10 

 
5 

 
0 

2013 - 2014 - 2015 - 2016 - 2017 - 2018 - 2019 - 2020 - 2021 - 2022 - 
2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 

MDKo UCL 2 Ucl 3 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 7



2. UN score data 2019 - 2020 

a) Based on the data obtained, we can conclude that in the UN subjects, namely B 

Indonesia, Science, and Mathematics at MIS Teladan GUPPI Tebing Tinggi 

school in 2019-2020, the average score in B Indonesia subject is 82.85 with a 

standard deviation of 3.13, science subject is 82.17 with a standard deviation of 

2.88, and math subject is 83.02 with a standard deviation of 2.87. 

b) The highest correlation value is 8.265306, namely in Indonesian and mathematics 

subjects. While the lowest correlation value is 7.281746, namely in math and 

science subjects. 

b. Learning Process Control 

Data for 10 years on UN exam scores, namely 2017 - 2023, shows a mismatch 

between the learning process and the results of learning outcomes. this statement 

can be seen from the variable graph with the determinant of the covariance matrix 

which appears to exceed the limit of the UCL 2 value (uper control limit). After the 

2017 - 2018 data is excluded, the control graph is already in a position below the 

UCL. This statement shows that the data is valid to be used to show the learning 

process. 

 
 

Reference 

 
Antoro, B. (2023). Statistical Process Control (SPC) Analysis as a Method of Evaluating 

Students' Mathematics Learning Process. Cendekia Journal: Journal of Mathematics 

Education, 7(3), 2941-2954. https://doi.org/10.31004/cendekia.v7i3.2852 

Darodjat, D., & Wahyudhiana, W. (2015). Education Program Evaluation Model. XIV, 1-23. 

https://doi.org/10.30595/islamadina.v0i0.1665. 

Hasratuddin. (n.d.). Learning Process Control at SMPN 6 Medan. 

 

Ratnawulan, E., & Rusdiana, H. A. (2014). Evaluation of Learning. Pustaka Setia Bandung. 

 

Law Number 20 Year 2003.pdf. (n.d.). 

IJO - INTERNATIONAL JOURNAL OF MATHEMATICS (ISSN: 2992-4421 )

Volume 06 | Issue 12 | December 2023 | http://ijojournals.com/index.php/m/index 8


