Jurnal Pendidikan

e-ISSN: 2443-3586 | p-ISSN: 1411-1942

Website http://jurnal.ut.ac.id/index.php/jp

Open access under CC BY NC SA Vol. 25, No. 1, 2024, 41 – 48

Copyright © 2022, the author(s) DOI: 10.33830/jp.v25i1.6406.2024

Meta-Analysis: The Effect of Self-Regulated Learning (SRL) on

Student Ability Levels

Dewi Sri Susanti1*, Rania Salsabilla1, Rahmi Hidayati2

1 Universitas Lambung Mangkurat, Banjabaru, Indonesia
2 Universitas Borneo Lestari, Banjarbaru, Indonesia

*Corresponding Author: [email protected]

Abstract: This study aims to identify the Effect Size and Publication Bias of the research results of
the relationship of Self Regulated Learning (SRL) to the Student’s Mastery Level with various
conclusions. This research is also used to identify the level of positive correlation by Meta-Analysis.
The methodology used in this research is a meta-analysis study of 14 research results. In this study,
the Effect Size values were obtained in the range of -0.13 to 1.22. Based on the estimated correlation
values, which currently stand at 0.498 with a confidence interval of 0.284 to 0.613, it can be said
that there is a positive relationship between Student’s Mastery Level and Self Regulated Learning
(SRL). In addition, by using the Egger's Test, a p-value of 0.166 > (α = 0.05) was obtained, so that
Publication Bias was not detected by the Meta-Analysis study. In general, it can be stated that Self
Regulated Learning (SRL) has a statistically significant positive influence on the level of student
mastery. The greater the motivation to learn independently, the better the students' academic
mastery.

Keywords: self-regulated learning, student’s mastery level, publication bias.

INTRODUCTION

The role of Human Resources (HR) in supporting the development of a country is

the main key in the era of cross-country competition. The current era of the Industrial
Revolution 4.0 demands the fulfillment of the needs of qualified, independent, and
technologically literate Human Resources (HR) (Rohida, 2018). Of the several factors that
influence the improvement of the quality of Human Resources (HR), one of them is the
role of education undertaken by everyone. Policy improvements in the implementation
of education are needed to ensure the success of education programs from early
childhood to higher education. A good education program is a form of future investment
that determines the advancement of civilization and the readiness of human resources to
compete internationally (Desmawan, et al., 2023). 5 (five) components determine
education policy, namely the existence of goals, planning, implementation programs,
decisions, and positive impacts caused (Elwijaya, Mairina, & Gistituati, 2021).
One indicator of the advancement of education can be measured by the Student
Mastery Level, which is the level of student success after experiencing the learning
process (Idrus, 2019). The level of student mastery can be identified through the
Academic Achievement Index, which comes from the results of learning efforts that a

41 Jurnal Pendidikan, Vol. 25, No. 1, 2024

person makes optimally. Many studies have discussed and analyzed the Student Mastery
Level, from the point of view of cognitive aspects and psychomotor aspects. But not a few
reviews from the affective aspect, one of which is by paying attention to internal factors,
namely the existence of Self Regulated Learning patterns.
Self-Regulated Learning (SRL) which can be called independence in learning is the
ability of someone who has knowledge of effective learning strategies and knows the
application of this knowledge so that they can regulate themselves in learning (Ruswana,
2018). The ability to learn independently will form students or students have good
motivation and ability to learn throughout life (Oishi, 2020). The self-learning process is
a learning model that provides opportunities for students to take initiative in setting
goals and creativity according to their learning styles (Pritandhari & Wibawa, 2021).
There have been many studies that discuss the relationship between Self Regulated
Learning (SRL) and students' ability to understand the topic or material being studied. In
some of these studies, it is stated that there is a close relationship between the two
variables with various correlation measures. This fosters the desire to analyze the results
of these studies to measure whether there is Publication Bias in the research data and
explore the existence of more general conclusions from studies that examine the
relationship between Student Mastery Level and learning techniques using Self Regulated
Learning (SRL) (Alhadi & Supriyanto, 2017). The statistical method that is widely used to
carry out the analysis process is Meta-Analysis, one of which is used by (Sari & Hardini,
2020).
Meta-analysis is a quantitative statistical method that analyzes quantitative data
from the results of studies that have been conducted to identify the alignment of results
in studies (Retnawati, 2018). Meta-analysis uses statistical measures obtained from
several studies on the same topic to determine whether the results studied are consistent.
In the study of the relationship between SRL and the level of student mastery, in addition
to identifying the pattern of data distribution, this research is also used to confirm the
existence of a strong relationship between the influence of Self Regulated Learning (SRL)
on student ability or student mastery level through Meta-Analysis. By applying meta-
analysis, it can be shown whether there is publication bias from the results of these
studies.

RESEARCH METHODS

The research began with a literature study of research publications on the effect of
Self Regulated Learning (SRL) on Student Mastery Levels from several scientific articles
that can be accessed online for the last 10 (ten) years. There are several
keywords/phrases used for the literature study search, namely the effect of self-regulated
learning on student mastery level, the effect of self-regulated learning on academic
achievement, self-regulated learning, the effect of self-regulation, self-regulated learning
strategies, meta-analysis of self-regulated learning and correlation between self-
regulated learning and academic achievement. Access to this journal article is carried out
through a website that provides information on research results, namely Google Scholar.
The search process resulted in 35 (thirty-five) suitable research papers published
in scientific journals, proceedings, and theses. Furthermore, the research was selected
based on the following criteria:
a. Primary research/studies consisting of experimental studies or surveys that
examine the Effect of Self Regulated Learning (SRL) on Student Mastery Levels.
b. A research report in a primary study that analyzes the statistical measure of
Correlation () as a measure of the direct relationship between the Effect of Self
Regulated Learning (SRL) on Student Mastery Level.

Jurnal Pendidikan, Vol. 25, No. 1, 2024 42

 The next step is to compile a resume of research results and identify research
results which include categorizing the suitability of criteria and summarizing the
statistical measures obtained. After the identification process of the research sample is
fulfilled, the correlation value and test results of each study are examined. Then the Effect
Size and Standard Error Effect Size were determined based on the correlation coefficient
value obtained from the study. The determination of the Effect Size value is based on the
correlation coefficient value between the two variables obtained based on the reporting
of the correlation between two continuous variables in the research results. Here are
some of the equations used. (Retnawati, 2018).
(1)

(2)
where:
F = test statistics
N = sample size
r = effect size

Then the Effect Size transformation is determined which is obtained from equation 2 :

(3)
And the variance of the Effect Size transformation is as follows:

(4)
The Meta Correlation Analysis method consists of two choices of analysis models,
namely the Fixed Effect Model and the Random Effect Model. The model chosen adjusts
to the characteristics of the research data. In data that tends to vary, a random effect
model can be used, while in data that tends to be uniform, a fixed effect model can be
used.
After determining the right model according to the identification of variance
heterogeneity, the calculation of Summary Effect is carried out which consists of several
stages. The following stages are: calculation of the weight of each study, tabulation of
data, calculation of the weighted average effect, determination of the confidence interval,
calculation of the p-value and transformation of the weighted average effect value and its
confidence interval into a correlation. Data distribution patterns will be presented
graphically in the form of Forest Plot and Funnel Plot. Then the last stage of the Meta-
analysis Procedure is the interpretation of the analysis results.
All stages of data analysis in this study used one of the Open Sources software,
namely JASP 0.16.3 software, which can be obtained at the following link: https://jasp-
stats.org. The operation of this software is relatively easy and is widely used in several
social research fields.

RESULTS AND DISCUSSION

Identification of Research Samples

From the search results through the Google Scholar search engine, 35 (thirty-five)
research titles were obtained that explained the relationship between SRL and student
mastery. However, among these 35 titles, there are only 14 (fourteen) research results
that have conformity with the criteria, while 21 other research results do not have
conformity with the criteria. Eleven of the 21 studies that did not fit the criteria consisted
of six studies that measured the relationship using regression analysis without including

43 Jurnal Pendidikan, Vol. 25, No. 1, 2024

the correlation coefficient and five other studies focused on discussing the final results of
the assessment rather than the relationship between variables. The remaining four
studies were identified as more suitable for using Meta Contracts Analysis and Meta Pre-
Post Contracts Analysis. The last six consisted of five research reports that used non-
student research objects and one research result that was not presented by the official
scientific media.

Transformation of Effect Size Value and Effect Size Standard Error

Table 1 presents 14 (fourteen) research results that examine the relationship
between Self Regulated Learning (SRL) and Student Mastery Level with the correlation
coefficient obtained and the transformation of the effect size value and standard error
effect size.

Table 1. Correlation coefficients and effect size of research results

Correlation
Sample Standard Error
No Authors coefficients Effect Size
Size (n) of Effect Size
(r)
1 (Supriyanto, 2015) 0.290 20 0.29857 0.24254
(Hasanah, Maria, &
2 0.445 187 0.47845 0.07372
Lutfianawati, 2019)
(Ruswana &
3 0.649 33 0.77357 0.18257
Zamnah, 2018)
(Sukmawansyah,
4 Hakim, & Hartono, 0.42 73 0.44769 0.11952
2019)
5 (Darmiany, 2016) 0.277 74 0.28443 0.11868
6 (Andiani, 2018) 0.690 33 0.84796 0.18257
(Ritonga, Vianty, Alwi,
7 Suhendi, & Sofendi, 0.104 175 0.10438 0.07625
2021)
(Meilisa, Sari, & Kasih,
8 0.218 197 0.22156 0.07180
2021)
(Ramadhani & Laili,
9 0.376 84 0.39539 0.11111
2021)
10 (Syahniar, 2018) 0.517 87 0.57224 0.10911
(Ningrum, Kumara, &
11 0.277 219 0.28443 0.06804
Prabandari, 2018)
12 (Edi, 2016) -0.13 76 -0.13074 0.11704
13 (Karuniawati, 2017) 0.84 67 1.22117 0.12500
14 (Isnayasari, 2018) 0.032 80 0.03201 0.11396

Based on Table 1, it can be shown that the value of the correlation measure r in 14
(fourteen) research papers varies in the weak, medium, and strong categories. In addition
to the relationship measured by positive correlation, there is also a measure of
relationship with negative correlation. A positive correlation is a unidirectional
relationship that means if Self Regulated Learning (SRL) increases, the level of student
mastery also increases. Conversely, a negative correlation is a relationship that is in the
opposite direction, namely if Self Regulated Learning (SRL) increases, the level of student
mastery decreases.
Table 1 also presents the effect size value, which is in the range of -0.13074 to
1.22117, while the standard error effect size value is obtained in the range of 0.06804 to
0.24254. The effect size value and standard error effect size of each research result are
used as the basis for data analysis using the Meta Correlation Analysis procedure.

Jurnal Pendidikan, Vol. 25, No. 1, 2024 44

Results of Meta Anlaysis
The meta-analysis begins by selecting the effect size measurement model through
heterogeneity of variance testing. It can be seen that the research data as many as 14
(fourteen) article titles have varying relationship measurement coefficients. This is
shown in the results of the variance of residual heterogeneity test, where the Glejser test
statistical value is greater than the distribution value at the degree of freedom 13 and the
p-value < 0.0001 (it is less than the significance level of test 0.05). This indicates that 14
(fourteen) research results are heterogeneous in variance.
By proving that the variance of research data is heterogeneous, the Random Effect
Model is the right model to use in the analysis, one of the models is the DerSimonian-
Laird method. This method was used to estimate the average Effect Size of 14 research
results and obtained an estimated value of 0.404. The calculation of the 95% confidence
interval produces confidence limits in the range of 0.243 to 0.564, where this range
contains positive values. Visualization of the Effect Size and Summary Effect values from
each study is presented in the form of a Forest Plot Figure 1.

Figure 1. Forest Plot of research data using the Random-Effect Model

Based on the Forest Plot in Figure 1, the results of the Effect Size value of 14
(fourteen) research results show a value close to the Summary Effect result of 0.40 with
a range of 0.24 to 0.56. The Effect Size value of the analyzed research shows a varied
value, which is between the ranges of -0.13 to 1.22. Most of the data boxplots are to the
right of the effect size 0 upright line, which indicates that the dominant research results
produce a positive correlation coefficient range. Only one study written by (Edi, 2016)
provides a range that tends to be negative. Visually, it can be concluded that the research
results tend to show a positive effect. To be sure, the meta-analysis stage will be carried
out in the following section.
The Meta-Analysis stage begins by transforming the weighted average effect value
( ) into the form of the Correlation coefficient value ( ). Transformation of the
weighted average effect value ( ) into the form of the Correlation Coefficient value ( )
obtained a value of 0.498 which indicates a "Medium" level of correlation. Transformation

45 Jurnal Pendidikan, Vol. 25, No. 1, 2024

of the confidence interval ( ) results in a range of values from 0.284 to 0.613. The range
does not contain zero or the correlation is at positive values, which indicates a positive
relationship between the two variables, SRL and student mastery level. This means that
an increase in self-regulated learning will be followed by an increase in the level of
student mastery in learning. This is in line with research (Sari & Hardini, 2020) which
states an identical conclusion. In contrast to the results of research (Wardani, Munir,
Lestari, & Anam, 2023) which produced a negative correlation between self-regulated
learning and students' grammar achievement in English language learning.
The Funnel Plot is drawn on a cartesian plane, where the flat axis shows the Effect
Size value and the upright axis shows the Standard Error value obtained from the results
of the analyzed studies. The center line shows the symmetry axis of the Summary Effect
distribution of the observed studies. Meanwhile, the scatter points in the Funnel Plot
show the distribution of the Effect Size value against the Standard Error. Funnel Plot is
used to visually detect the presence of Publication Bias. If there is no Publication Bias then
the Funnel Plot forms a symmetrical point distribution, otherwise there is Publication
Bias. The Funnel Plot in Figure 2, shows that there are 8 (eight) points to the left of the
symmetry axis and 6 (six) points to the right. At first glance, it can be said that the data is
relatively symmetrically distributed. However, drawing this conclusion is not easy
because it is only a subjective visual conclusion. So it is necessary to conduct a formal test,
namely Egger's Test to complete the conclusion about data distribution.

Figure 2. Funnel Plot of research data using the Random-Effect Model

Egger's Test results in a z value of 1.385 and a p-value of 0.166. The p-value is
greater than α = 0.05, leading to the conclusion that there is no Publication Bias in the
research data. Based on Table 2 which presents the heterogeneity estimate, it is known
that the value has an estimate value of 88.057%. The value which is greater than
25% and close to 100%, indicates that the "true" Effect Size value in all studies tends to
be homogeneous (Retnawati, 2018). By proving that the results are homogeneous both
graphically and through formal tests, it can be concluded that there is a positive
relationship between SRL and students' mastery level. The correlation coefficient of

Jurnal Pendidikan, Vol. 25, No. 1, 2024 46

0.498 shows a relationship at a moderate level, where an increase in Self Regulated
Learning motivation will be followed by an increase in ability of student of the material
studied.

Table 2. Heterogeneity estimation of research results

Confidence Interval of 95%
Estimation
Lower Bound Upper Bound
τ² 0.078 0.049 0.306
τ 0.280 0.221 0.553
I² (%) 88.057 82.097 96.650
H² 8.373 5.586 29.852

CONCLUSIONS

The meta-analysis conducted on the fourteen observed research results showed the
same or similar conclusion, which was that there was a statistically significant effect of
Self Regulated Learning (SRL) on Student Mastery Level. Visual and formal testing shows
that there is no publication bias in these studies. The level of relationship between the
two variables is 0.498 which falls into the medium category. In general, it can be stated
that Self Regulated Learning (SRL) has a real positive influence on the level of student
mastery. The greater the motivation to learn independently, the better the students'
academic mastery.

REFERENCES

Alhadi, S., & Supriyanto, A. (2017). Self Regulated Learning Concept: Student Lerning
Progress. Ahmad Dahlan University National Seminar on Guidance and Counseling, 2,
51-62.
Andiani, S. (2018). The Relationship between Academic Achievement and Self-Regulation
Strategies in Learning in Deaf Students. Calyptra: Jurnal Ilmiah Mahasiswa
Universitas Surabaya, 6(2), 26 - 39.
Darmiany. (2016). Self-regulated Learning of First-Year Elementary School Teacher
Education (PGSD) Students. Jurnal Psikologi Pendidikan dan Konseling, 2(1), 72-83.
Desmawan, D., Cahyaningdyah, F. A., Darwin, R., Putri, S. S., Rizqina, A., & Ikhsanudin.
(2023). Analysis of the Role of Education on the Quality of Human Resources to
Improve Community Productivity in DKI Jakar. Wawasan: Jurnal Ilmu Manajemen,
Ekonomi dan Kewirausahaan, 1(2), 214-224.
Edi, R. (2016). Correlation Between Self-Regulated Learning and Academic Achievement
ofChemistry Education Students of FKIP Sriwijaya University. Proceedings of the 2nd
SULE-IC 2016 (pp. 965-976). Palembang: Unsri.
Elwijaya, F., Mairina, V., & Gistituati, N. (2021). Basic Concepts of Education Policy. JRTI
(Jurnal Riset Tindakan Indonesia), 6(1), 67-71.
Hasanah, U., Maria, S., & Lutfianawati, D. (2019). The Relationship between Self-
Regulation in Learning and Learning Achievement in 2016 Batch Students of the
Faculty of Medicine, Malahayati University. Psyche: Jurnal Psikologi, 1(1), 78-91.
Idrus, L. (2019). Evaluation in the Learning Process. ADAARA: Jurnal Manajemen
Pendidikan Islam, 9(2), 920-935.
Isnayasari, J. (2018). The Correlation Between Self Regulating and Listening Achievement
of Undergraduate EFL Students of UIN Raden Fatah Palembang. Palembang: UIN

47 Jurnal Pendidikan, Vol. 25, No. 1, 2024

 Raden Fatah.
Karuniawati, M. (2017). Correlation between Self-Regulation Learning Strategy and
Receptive Vocabulary Size of English Department Students at Universitas Brawijaya.
Malang: Universitas Brawijaya.
Meilisa, O. D., Sari, H., & Kasih, L. C. (2021). The Relationship between Self Regulated
Learning and Academic Achievement of Nursing Students during the Pandemic
Period. Jurnal Ilmiah Mahasiswa Keperawatan, 5(2), 104-111.
Ningrum, R., Kumara, A., & Prabandari, Y. (2018). The relationship between self-regulated
learning and academic achievement of undergraduate medical students. 3rd Annual
Applied Science and Engineering, (pp. 89-103). Bandung.
Oishi, I. R. (2020). The Importance of Self-Study for Learners in Higher Education. Jurnal
IKRA-ITH Humaniora, 4(2), 50-55.
Pritandhari, M., & Wibawa, F. A. (2021). Analysis of Online-based Independent Learning
during the Covid-19 Pandemic. Jurnal Lentera Pendidikan, 6(1), 31-44.
Ramadhani, I. W., & Laili, N. (2021). The Relationship Between Self Regulated Learning
and Student Academic Achievement di Universitas Muhammadiyah Sidoarjo,
Indonesia. Indonesian Journal of Islamic Studies, 5, 6-11.
Retnawati, H. A. (2018). ntroduction to Meta-analysis. Jakarta: Parama Publishing.
Ritonga, O., Vianty, M., Alwi, Z., Suhendi, D., & Sofendi. (2021). The Relationship of Self-
Regulation Learning in Online Learning with Student Learning Achievement
Program Studi Pendidikan Bahasa Inggris Universitas Sriwijaya. Prosiding Seminar
Nasional UNIMUS, 4, pp. 601-615. Semarang.
Rohida, L. (2018). The Influence of the Industrial Revolution Era on Human Resource
Competencies. Jurnal Manajemen Bisnis Indonesia, 6(1), 114-136.
Ruswana, A. M. (2018). Correlation between Self-Regulated Learning and Students'
Mathematical Understanding Ability. Jurnal Pendidikan Matematika, 7(3), 381–388.
doi:https://doi.org/10.31980/mosharafa.v7i3.143
Ruswana, A. M., & Zamnah, L. N. (2018). The Relationship between Student Self-
Confidence and Mathematical Comprehension Ability in Kapita Selecta Mathematics
IV Lecture. Jurnal Teori dan Aplikasi Matematika, 2(1), 57-61.
Sari, A. R., & Hardini, A. T. (2020). Meta Analysis of the Effect of Problem Based Learning
Model on Mathematics Learning Outcomes. Jurnal Ilmiah Pendidikan Profesi Guru,
3(1), 1-8.
Sukmawansyah, A. T., Hakim, L., & Hartono, R. (2019). Relationship between Self-
Regulated Learning and Student Academic Achievement Fakultas Psikologi
Universitas Teknologi Sumbawa. Jurnal Psimawa, 1(1), 22-25.
Supriyanto. (2015). The Relationship Between Self-Regulated Learning and Academic
Achievement in College Students Semester I Prodi Psikologi Universitas
Pembangunan Jaya. Jurnal Universitas Pembangunan Jaya, 2(2), 49-61.
Syahniar. (2018). The correlation between self regulated and learning achievement of
underachiever students. International Conferences on Education, Social Sciences and
Technology (pp. 95-99). Padang: UNP Padang.
Wardani, A. D., Munir, A., Lestari, L. A., & Anam, S. (2023). Self-Regulated Learning
Strategies And Their Relationship To Grammar Achievement Of Undergraduate
English Department Students. LLT Journal: A Journal on Language and Language
Teaching, 26(2), 634-649.

Jurnal Pendidikan, Vol. 25, No. 1, 2024 48