Advances in Social Science, Education and Humanities Research, volume 303

International Conference of Primary Education Research Pivotal Literature and Research UNNES 2018 (IC PEOPLE UNNES 2018)

Effect of Science-Technology-Society
Approaches to The Science Process Skills
Elementary School Students
Nurul Iman Nurahman Siti Mualimah Regina Lichteria Dadan Djuanda
Basic Education, Basic Education, Panjaitan Primary School Teacher
Postgraduate University Postgraduate University Primary School Teacher Education, University of
State of Semarang, State of Semarang, Education, University of Education Sumedang,
Semarang, Central Java, Semarang, Central Java, Education Sumedang, Bandung, West Java,
Indonesia Indonesia Bandung, West Java, Indonesia
Indonesia

Abstract—The first observation at some Natural Sciences learning not only making
schools indicated learning process of Natural Sciences students understand the concept, but how the concept
just concept-oriented without observation or applied can be applied in the daily life of students. That is why
on the real nature. That was cause students Science natural sciences learning always doing observation or
process skills did not increased. This study aims to experiments activities. Observation and experiments
find out effect of Science-Technology-Society (STS) which students do not only build the concept but also
Approaches to the science process skills. This study could improve science process skills and scientific
used pretest-posttest control group design was attitudes. It is certainly required learning process in
conducted at Putat Elementary School involved 74 accordance with the principles of Natural Sciences study.
students 5th grade which was divided into two But in fact, Natural Sciences learning process in
groups. The first group was experiment class (VB) some schools not in ideal conditions for achieving goals.
and the second was control class (VA). The results The condition is evidenced by the lack of students
with calculation the difference of two mean used science process skills which becomes one of the Natural
confidence level 95% show average score of Science Sciences learning goals. Things that caused
process skills between experiment class and control improvements of sciences process skills and scientific
class has different. Experiment class with STS attitudes of students not optimal which demonstrated by
approaches learning indicated better enhancement to studies and assessments undertaken by PISA (Program
Science process skills compared control class with for International Student Assessment) in 2012 that
conventional learning. The STS approaches was measures students sciences literacy skills. Indonesia was
implemented in one week at water cycle learning. one of country who participated in that study and
Study instrument used science process skills test, assessments got an average score of science literacy
learning outcomes test, and observation classes. The ability of 382. With that score, Indonesia was only
results show that STS approaches could improving ranked 64 out of 65 countries that participated in the test.
students Science process skills and making students The position was only one rank from the last position.
more exciting on Natural Sciences learning. These results indicated that the ability of Students
Keywords—STS approaches, science process Natural Sciences still low in the scope of science
skills, learning approaches, learning outcomes. literacy. The lack of development science process skills
aspects due to less innovative learning. Natural Sciences
1. INTRODUCTION learning at the schools just give priority to concept-
Education is attempt to developing better personal oriented and the teacher only use conventional learning
skills and attitudes. Education Another describing method like discourse and question and answer methods
education as an stage that has an institutional with the [6]. That was one of some aspects which caused students
aims to develop skills in knowledge, personal skills, and science process skills does not have better
attitudes [17]. With that goal, learning activities for improvements.
students should using meaningful learning. That Natural Sciences concepts obtained through the
meaning not only making students memorize the theory sciences process will last long and could be used to
but also applied that theory in the real life. Learning overcome problems in the real life [12]. It should be
activities will be more meaningful if knowledge is found more meaningful if the theory and practice implemented
by students themselves, not just from teacher each other in a balanced with the orientation on
informations [12]. Through direct experience, students improvement science process skills and scientific
will be better understanding the concept, especially in attitudes. Natural Sciences learning objectives at schools
Natural Sciences. on Kurikulum Tingkat Satuan Pendidikan (KTSP)
confirms that learnings is expected to instill and develop

Copyright © 2019, the Authors. Published by Atlantis Press.

This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). 18
 Advances in Social Science, Education and Humanities Research, volume 303

students knowledge, skills, and attitudes and raise implemented use the STS approaches learning while the
awareness to appreciate the natural surrounding (BNSP, control class was not given treatment or implemented
2006). To realize that, teachers need innovative learning use the conventional learning as usual [9].
which connect theory with environmental problems. It is Research populations was superior groups
reinforced with Piaget cognitive theory, elementary elementary schools in district Sedong. From six schools
students is in the range of 6-12 years old in the pre- belonging to superior groups, conducted a draw to
operative and formal operational phase. At that phase, determined the sample and obtained by 5th grade
students have not been able to think abstractly and still students of Putat Elementary School. After that, 74
refer to concrete experiences (Piaget in Asy`ari, 2006). students from Putat Elementary School divided into two
Natural Sciences learning for elementary schools should groups namely the experiment class (VB) and Control
be able to provide hands-on experiences to develop their class (VA).
capabilities, so they can explore and understand nature Research instruments on this study used
as a whole [12]. quantitative analysis consisted of science process skills
Based on the problem studies above, Science- and learning outcomes test. Validity, reliability, level of
Technology-Society (STS) approaches become one of difficulty, and appropriateness of questions tested first to
solutions in improving science process skills and 70 students. Validity and reliability were used to test the
learning outcomes. Natural Sciences learning connected feasibility of the questions to be tested measure students
to daily life through related technology products, will abilities.
make students feel that Natural Sciences learning is Calculation of validity used correlation
useful to learn [11]. Application of STS approaches coefficient formula product moment raw score by
could make students applied concept of Natural Sciences Pearson. After calculation, correlation coefficient
at daily life with regard the technology that is growing at interpreted using validity coefficient by Guilford [13].
their society. The application of STS approaches in Nine questions of science process skills and seven
Natural Sciences learning is done in several stages. It questions of learning outcomes were tested. Test results
starting with invitation, exploration, solutions, and show 6 of 9 science process skills questions had high
applications [1]. Invitation stage begins with giving validity while 3 other questions had medium validity. 5
students the ongoing problems or issues in the of 7 learning outcomes questions had high validity, one
environment that could observe by students. Exploration question had very high validity and one more question
stage giving students the opportunity to understand had medium validity.
concepts with teacher guidance through observation and Calculation of realibility use reliability coefficient
problem analysis. Problem identification from by Cronbach Alpha. After calculation, correlation
exploration can be taken by students with reading coefficient interpreted using reliability coefficient by
articles from newspapers, listening radio or watching Guilford [13]. Test results show reliability of science
news from television. At the solution stage, students process skills questions had correlation coefficient
begin to develop the concept they have gained with the reached 0,84 and based on coefficient correlation was in
new concept that fit with environmental conditions. It very high category. Reliability of learning outcomes
becomes the student references to analyzing the cause of questions had correlation coefficient reached 0,79 and
the problem and looking for alternative solutions that can based on coefficient correlation was in high category.
be used to solve the problem. After they find alternative Difficulty level of science process skills and
solutions, at the application stage, they try to applying learning outcomes questions calculated by Microsoft
the solution as a provision to be applied in the students Excel for Windows. Test results show 8 of 9 science
environment. On this stage also do the stabilization of process skills questions had medium level and one
the concept that students are not mistaken in question had difficult level. All of learning outcomes
understanding the new concept. questions had medium level. Appropriateness of science
Based on the above exposure, this study aims to process skills and learning outcomes questions
find out the improvement students science process skills calculated by Microsoft Excel for Windows. Test results
and learning outcomes in the water cycle concept with show 6 of 9 science process skills questions had good
STS approaches learning and conventional learning. This appropriateness, 2 questions had very good
study also observed the correlation between science appropriateness, and one more question had sufficient
process skills and students learning outcomes. appropriateness. All of learning outcomes questions had
good appropriateness.
II. METHOD Other research instruments used on this study was
This study used experiment method which have teacher performances and students activity observations.
two compared groups selected at random. Research Teacher performances which observation was
design used pretest-posttest control group design. compatibility of planning and implementing STS
Following description of the research design. First steps approaches learning and conventional learning. Students
research was divide students into two class and activity observation was done to know response of
determined control class and experiment class (A), students during following STS approaches learning and
continued with given pretest to both classes (0). After conventional learning.
that, experiment class was given treatment that

19
 Advances in Social Science, Education and Humanities Research, volume 303

Research procedure on this study had four stages. and students activity observation. After all instruments
The stage starts with preparation, implementation, data were analyzed then drawing conclusions from the initial
processing, and data analysis. Preparation stage start hypothesis whether it was accepted or rejected.
with set problems and topics for learnings, prepared
instruments, consulted topics and instruments to experts, III. RESULTS AND DISCUSSION
validity instruments test, improvement instruments, Pretest data shows the difference average between
permit to the school where the research was conducted, the experiment class and the control class. The average
and consultation with teacher about technical research pretest of science process skills the experiment class was
implementation. Implementation stage start with pretest 20,05 while the control class was 17.83. There was the
to experiment class and control class aims to measure difference amount 2,22. After the pretest result of both
students initial ability from both classes. The next classes, the first step was normality test to that results.
activity was the learning according to the schedule that Analysis was done by Lilliefors test (Kolmogorov-
has been set. Experiment class did the learning with STS Smirnov).
approaches and control class did the conventional As shown in table 1, the results show normality
learning for one week. During the learning process, test of the experiment class pretest has P-value (Sig.)
teacher performance and students activities are observed. 0,046 with Lilliefors test. That point less than α = 0,05.
In the end of learning, students were given posttest to Therefore, pretest data for the experiment class is
determined the effect of STS approaches and abnormal distribution. Furthermore, control class pretest
conventional learnings to the students science process has P-value (Sig.) 0,200 with Lilliefors test. That point
skills and learning outcomes. more than α = 0,05. Therefore, pretest data for the
After all learning activites were completed and all control class is normal distribution. Because one of the
instruments are filled, data was processed and analyzed. data is abnormal distribution, then the next step is two
Pretest and posttest data from STS approaches and mean difference test with Mann-Whitney or U-test. It is
conventional learning was processed and analyzed by because the data being tested is unrelated data or
quantitative analysis consisted of normality test, independent. Table 2 display the results two mean
homogeneity test, two mean difference test, and difference test of both classes pretest has P-value (Sig.2-
correlation coefficient test. Things that must be tailed) 0,553 with Mann-Whitney or U-test. That point
considered if one or both groups data were abnormal P-value (Sig.2-tailed) more than α = 0,05. Based on
distribution, it was not followed by homogeneity test but these results, there is no difference in initial ability
used non-parametic statistical test with Mann-Whitney between the experiment class and the control class.
or U-test. Analysis was also did on teacher performance
Table 1. Normality test of pretest and posstest data science process skills
Tests of Normality Tests of Normality
Class Kolmogorov-Smirnova Class Kolmogorov-
Smirnova
Statist df Sig. Statist df Si
ic ic g.
Pretes Experiment .144 38 .046 Posttest Experiment .190 38 .0
t 01
Control .121 36 .200* Control .155 36 .0
29
a. Lilliefors Significance Correction a. Lilliefors Significance Correction
*. This is a lower bound of the true significance.

Table 2. Two mean difference test of pretest and posttest data science process skills
Test Statisticsb Test Statisticsb
Pretest Posttest
Mann-Whitney U 629.500 Mann-Whitney U 200.500
Wilcoxon W 1.296E3 Wilcoxon W 866.500
Z -.594 Z -5.238
Asymp. Sig. (2-tailed) .553 Asymp. Sig. (2-tailed) .000
a. Based on 10000 sampled tables with starting a. Based on 10000 sampled tables with starting seed 743671174.
seed 1314643744.
b. Grouping Variable: Class b. Grouping Variable: Class
test of the experiment class posttest has P-value (Sig.)
Posttest data shows the difference average between
both classes. The average pretest of science process skills the 0,001 with Lilliefors test. That point less than α = 0,05.
experiment class was 61,61 while the control class was 35.50. Therefore, posttest data for the experiment class is
There was the difference amount 26,11. After the posttest abnormal distribution. Furthermore, control class
result of both classes, the next step was normality test by posttest has P-value (Sig.) 0,029 with Lilliefors test. That
Lilliefors test (Kolmogorov-Smirnov). On table 1, normality point less than α = 0,05. Therefore, posttest data for the

20
 Advances in Social Science, Education and Humanities Research, volume 303

control class is abnormal distribution. Because both of Table 4. Normality test of pretest and posttest data learning
the data is abnormal distribution, then the next step is outcomes
two mean difference test with Mann-Whitney or U-test. Tests of Normality Tests of Normality
Table 2 display the results two mean difference test of Clas Kolmogorov- Clas Kolmogorov-
both classes pretest has P-value (Sig.2-tailed) 0,000 with s Smirnova s Smirnova
Mann-Whitney or U-test. That point P-value (Sig.2- Statistic df Si Stati df Si
tailed) less than α = 0,05. Based on these results, there is g. stic g.
difference in final ability between the experiment class Prete Expe .136 3 .0 Po Expe .202 3 .0
and the control class. st rime 8 7 stt rime 8 0
To see the effect of STS approaches and nt 4 est nt 0
conventional learning to the science process skills, Cont .179 3 .0 Cont .083 3 .2
rol 6 0 rol 6 0
Wilcoxon test was conducted on both classes. Data
5 0*
processing and analyzing to answer the hypothesis is a. Lilliefors Significance Correction a. Lilliefors Significance
done with the SPSS for Windows. Correction
Table 3 show two mean difference pretest and *. This is a lower bound of
posttest experiment class with Wilcoxon test. The results the true significance.
display points of P-value (Sig.1-tailed) = 0,000 or less
than α = 0,05. Based on these results, there is difference Table 5. Two mean difference test of pretest and posttest data
between initial ability and final ability of science process learning outcomes
skills at experiment class. Table III also show two mean
Test Statisticsb Test Statisticsb
difference pretest and posttest control class with
Pret Post
Wilcoxon test. The results display points of P-value est test
(Sig.1-tailed) = 0,000 or less than α = 0,05. Based on Mann-Whitney U 671. Mann-Whitney U 213.
these results, there is difference between initial ability 500 500
and final ability of science process skills at control class. Wilcoxon W 1.41 Wilcoxon W 879.
2E3 500
Table 3. Wilcoxon test results of data pretest and posttest Z - Z -
science process skills experiment and control class .136 5.10
3
Test Statistics Test Statistics Control Asymp. Sig. (2-tailed) .892 Asymp. Sig. (2-tailed) .000
Experiment Classb,c Classb,c
a. Based on 10000 sampled a. Based on 10000 sampled
Posttest Posttest tables with starting seed tables with starting seed
– – 334431365. 299883525.
Pretest Pretest
b. Grouping Variable: Class b. Grouping Variable: Class
Z -5.375a Z -4.940a
Asymp. .000 Asymp. .000
Sig. (2- Sig. (2- As shown in table 4, the results show normality
tailed) tailed) test of the experiment class pretest has P-value (Sig.)
a. Based on negative a. Based on negative 0,074 with Lilliefors test. That point more than α = 0,05.
ranks. ranks. Therefore, pretest data for the experiment class is normal
b. Wilcoxon Signed b. Wilcoxon Signed distribution. Furthermore, control class pretest has P-
Ranks Test Ranks Test value (Sig.) 0,005 with Lilliefors test. That point less
c. Based on 10000 c. Based on 10000 than α = 0,05. Therefore, pretest data for the control
sampled tables with sampled tables with class is abnormal distribution. Because one of the data is
starting seed 2000000. starting seed abnormal distribution, then the next step is two mean
299883525.
difference test with Mann-Whitney or U-test. Table 5
display the results two mean difference test of both
Analysis data learning outcomes classes pretest has P-value (Sig.2-tailed) 0,892 with
The average pretest of learning outcomes the Mann-Whitney or U-test. That point P-value (Sig.2-
experiment class was 30,66 while control class was tailed) more than α = 0,05. Based on these results, there
30,89. There was the difference amount 0,23. After the is no difference in initial ability between the experiment
pretest result of both classes, the first step was normality class and the control class.
test to that results. Analysis was done by Lilliefors test Posttest data shows the difference average
(Kolmogorov-Smirnov). between both classes. The average pretest of learning
outcomes the experiment class was 73,13 while the
control class was 50.56. There was the difference
amount 22,57. After the posttest result of both classes,
the next step was normality test by Lilliefors test
(Kolmogorov-Smirnov). On table 4, normality test of the
experiment class posttest has P-value (Sig.) 0,000 with
Lilliefors test. That point less than α = 0,05. Therefore,
posttest data for the experiment class is abnormal

21
 Advances in Social Science, Education and Humanities Research, volume 303

distribution. Furthermore, control class posttest has P- Table 6 show two mean difference pretest and
value (Sig.) 0,200 with Lilliefors test. That point more posttest experiment class with Wilcoxon test. The results
than α = 0,05. Therefore, posttest data for the control display points of P-value (Sig.1-tailed) = 0,000 or less
class is normal distribution. Because one of the data is than α = 0,05. Based on these results, there is difference
abnormal distribution, then the next step is two mean between initial ability and final ability of learning
difference test with Mann-Whitney or U-test. Table 5 outcomes at experiment class. Table 6 also show two
also display the results two mean difference test of both mean difference pretest and posttest control class with
classes posttest has P-value (Sig.2-tailed) 0,000 with Wilcoxon test. The results display points of P-value
Mann-Whitney or U-test. That point P-value (Sig.2- (Sig.1-tailed) = 0,000 or less than α = 0,05. Based on
tailed) less than α = 0,05. Based on these results, there is these results, there is difference between initial ability
difference in final ability between the experiment class and final ability of learning outcomes at control class.
and the control class.
To see the effect of STS approaches and Analysis data observation
conventional learning to the learning outcomes,
Wilcoxon test was conducted on both classes just like Table 7. Teacher performance observation results
science process skills data analysis.
Class Learning Learning Learning Averages Inter
1 2 3 pretation
Table 6. Wilcoxon test results of data pretest and posttest Experiment Class
learning outcomes experiment and control class
Planning 90% 95% 95% 93% Very
Test Statistics Experiment Test Statistics Control Classb,c good
Classb,c Implemen 90% 93% 95% 93% Very
Posttest – Posttest – ting good
Pretest Pretest
Z -5.382a Z -4.942a
Control Class
Planning 95% 95% 95% 95% Very
Asymp. Sig. (2- .000 Asymp. Sig. (2- .000
tailed) tailed) good
a. Based on negative ranks. a. Based on negative ranks. Implemen 90% 93% 95% 93% Very
ting good
b. Wilcoxon Signed Ranks Test b. Wilcoxon Signed Ranks Test

c. Based on 10000 sampled c. Based on 10000 sampled

tables with starting seed tables with starting seed
2000000. 299883525.

Teacher performance observation was measured observation format in planned and implemented
used observation format. But, there was difference conventional learning. Based on table 7, teacher
between the observation format experiment class and the performance averages in planning on control class is 95%
observation format control class. The experiment class better than experiment class with 93%, nevertheless
uses the observation format in planned and implemented teacher performance averages in implementation on both
STS approaches learning while the control class uses the classes is equal with 93%.
Table 8. Students activity observations results

Experiment Class
Activities Participation Motivation Cooperation Friction
Learning 1
Total 59 54 60 74
% 52% 47% 53% 65%
Averages 54%
Interpretation C
Learning 2
Total 67 64 70 89
% 59% 56% 61% 78%
Averages 64%
Interpretation B
Learning 3
Total 78 70 71 96
% 68% 61% 62% 84%
Averages 69%
Interpretation B

22
 Advances in Social Science, Education and Humanities Research, volume 303

Control Class
Activities Participation Motivation Cooperation Friction
Learning 1
Total 48 47 52 65
% 44% 44% 48% 60%
Averages 49%
Interpretation C
Learning 2
Total 50 55 58 74
% 46% 51% 54% 69%
Averages 55%
Interpretation C
Learning 3
Total 55 68 59 86
% 51% 63% 55% 80%
Averages 62%
Interpretation B

Students activites observation was measured used by an increase in averages scores of the science process
observation format. The goal is to measure students skills and learning outcomes. Science process skills
responses to the learning process. Aspects measured in averages scores of experiment class increase from 20,05
the observation are students participation, motivation, to 61,61 while learning outcomes averages scores
cooperation, and friction. Based on table VIII, students increase from 30,66 to 73,13. In the control class,
activites averages experiment class on the meeting 1 is science process skills averages scores increase from
54% better than control class with 49%. It is also seen at 17,83 to 35,50 while learning outcomes averages scores
the final meeting, students activites averages experiment increase from 30,89 to 50,56. These results also show
class on the meeting 3 is 69% better than control class that water cycle concept learning with STS approaches
with 62%. However, students activities averages from better than conventional learning in improving students
both classes have an increase at each meeting. science process skills and learning outcomes though
This study to find out a better effect for intial ability score of learning outcomes control class
improvements of students science process ability and was better than experiment class by the difference point
learning outcomes by using STS approaches learning 0,23.
and conventional learning. Learning process was done The next analysis examines the related between
on the concept of water cycle in 5th grade conducted in science process skills and learning outcomes. Correlation
one week. The study invoved 74 students who were test used coefficient correlation test (Pearson
divided into two groups. First group was experiment Correlation) on science process skills posttest data and
class (VB) which did the STS approaches learning and learning outcomes posttest data. If Pearson correlation
second group was control class (VA) which did the was positive then there was positive correlation between
conventional learning. Analysis data start with initial science process skills and learning outcomes. Moreover,
ability of students science process skills and learning the higher of Pearson values also the higher correlation.
outomes in both classes. That was done by analyzing Based on Pearson correlation results, obtained positive
pretest data from both classes with Mann-Whitney test results with 0,676 points in the experiment class, and
which level of trust reach 95%. Based on the test, there positive results with 0,575 points in the control class.
was no difference in science process skills between These results show that improved in science process
experiment class and control class. Similar results were skills will be followed also by improved learning
also obtained on students learning outcomes after being outcomes.
tested using Mann-Whitney test. It shows both classes
have the same ability. IV. CONCLUSION
After doing each learning for one week, all Based on the exposure of the research results, it
students from both classes were tested again used can be concluded that STS approaches learning and
Wilcoxon test on pretest and posttest. The results conventional learning has an effect to students science
obtained there was a difference between initial ability process skills and learning outcomes at water cycle
and final ability from both classes in science process concept in 5th grade elementary schools. Both learning
skills and learning outcomes. These results indicate that can improve science process skills. But STS approaches
there was an improvement in science process skills and learning improve students science process skills better
learning outcomes from both classes. That was supported than conventional learning. Similar results were also

23
 Advances in Social Science, Education and Humanities Research, volume 303

found on learing outcomes analysis data. Both learning [6] Lestari, Ni. Md. E., dkk. 2013. Pengaruh
can improve learning outcomes. But STS approaches Pendekatan Sains Teknologi Masyarakat (STM)
learning improve students learning outcomes better than terhadap Hasil Belajar IPA Siswa Class IV SD di
conventional learning. Gugus III Kecamatan Tegallalang. Mimbar
The results also indicated a correlation between PGSD: E-Journal Universitas Pendidikan
science process skills and learning outcomes. It was Ganesha. 1(1).
reinforced by Pearson correlation between science [7] Maryanto dan Purwanto. 2009. Ilmu Pengetahuan
process skills and learning outcomes with positive Alam 5 untuk SD/MI Class 5. Depdiknas: Jakarta.
results. These results show that improved in science [8] Maulana. 2008b. Konsep Dasar dan Aplikasi
process skills will be followed also by improved learning Statistika serta Teori Distribusi Peluang. Royyan
outcomes. Press: Subang.
Research results show that Natural Sciences [9] Maulana. 2009. Memahami Hakikat, Variabel,
learning at water cycle concept with STS approaches dan Instrumen Penelitian Pendidikan dengan
better than conventional learning in improving students Benar. Learn2live ‘n Live2learn: Bandung.
science process skills and learning outcomes. It was [10] OECD-PISA. 2013. PISA 2012 Results in Focus:
because STS approaches learning not only concept- What 15-years-old Know and What They Can Do
oriented, but also giving students direct experience with What They Know Volume VII. Not
though the atual problems that occur in their Published.
environment. With problem identification can improve [11] Poedjiadi, A. 2005). Sains Teknologi Masyarakat
students ability to observe the surrounding environment. Model Pembelajaran Kontekstual Bermuatan
Moreover, STS approaches could make Natural Science Nilai. PT. Remaja Rosdakarya: Bandung.
learning more meaningful and improve students ability [12] Sujana, A. 2012. Pendidikan IPA Teori dan
to applied concept, process skills, creativity, and Praktik. Rizqi Press: Bandung.
scientific attitudes on appreciate technology product and [13] Suherman, E. dan Sukjaya, Y. 1990. Petunjuk
responsible for the problems that arise in the Praktis untuk Melaksanakan Evaluasi Pendidikan
environments [1]. The use of continuous and diverse Matematika. Wijayakusumah 157: Bandung.
evaluation method could encourage students to be [14] Sukardi. 2005. Metodologi Penelitian Pendidikan
serious in following the learning because the value is not Kompetensi dan Praktiknya. PT Bumi Aksara:
only about cognitive, but also participation and Jakarta.
creativity. [15] Surapranata, S. 2009. Analisis, Validitas,
Reliabilitas dan Interpretasi Hasil Tes
REFERENCES Implementasi Kurikulum 2004. PT. Remaja
Rosdakarya: Bandung.
[1] Asy’ari, M. 2006. Penerapan Pendekatan Sains-
[16] Suwintara, I Pt., dkk. 2013. Pengaruh Pendekatan
Teknologi-Masyarakat dalam Pembelajaran Sains
STM terhadap Sikap Ilmiah dan Hasil Belajar
di Sekolah Dasar. Universitas Sanata Dharma:
pada Pembelajaran IPA Siswa Class IV SD
Yogyakarta. Gugus V Kecamatan Sawan. Mimbar PGSD: E-
[2] Badan Standar Nasional Pendidikan. 2006. Journal Universitas Pendidikan Ganesha. 1(2).
Panduan Penyusunan Kurikulum Tingkat Satuan
[17] Syah, M. 2010. Psikologi Pendidikan. PT.
Pendidikan Jenjang Pendidikan Dasar dan
Remaja Rosdakarya: Bandung.
Menengah. BP Dharma Bakti: Jakarta.
[18] Undang-undang Republik Indonesia Nomor 20
[3] Bundu, P. 2006. Penilaian Keterampilan Proses Tahun 2003. Sistem Pendidikan Nasional
dan Sikap Ilmiah dalam Pembelajaran Sains-SD,
(SISDIKNAS). Fokus Media: Bandung.
Depdiknas: Jakarta.
[19] Wardana, I. K., dkk. 2013. Pengaruh Model
[4] Departemen Pendidikan Nasional. 2007. Kajian
Kontekstual terhadap Keterampilan Proses Sains
Kebijakan Kurikulum Mata Pelajaran IPA.
dan Hasil Belajar Sains pada Siswa Class IV SD
Depdiknas: Jakarta.
Gugus V Dr. Soetomo. Mimbar PGSD: E-Journal
[5] Kariadinata, R. dan Abdurahman, M. 2012. Universitas Pendidikan Ganesha. 1(3).
Dasar-dasar Statistik Pendidikan. CV Pustaka
Setia: Bandung.

24