Sukajaya, I Nyoman (2016) Serious Game Berbasis Taksonomi Bloom: Sebuah Pendekatan Alternatif Penilaian Pembelajaran Matematika Berbantuan Teknologi Informasi. Doctoral thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
2211301003-Disertation.pdf Download (4MB) | Preview |
Abstract
Pada penelitian ini dikembangkan sebuah pendekatan baru dalam penilaian
pembelajaran matematika berbantuan serious game dengan melibatkan komponen
pengetahuan geometri bangun datar jajar genjang dan komponen pedagogi
yakni taksonomi belajar menurut Bloom. Pendekatan penilaian ini diusulkan sebagai
alternatif baru merekam data pembelajar yang representatif mewakili karakteristik
individu mereka. Serious game yang diimplementasikan dikembangkan
mengikuti kerangka teknis serious game yang konstruktivis. Tantangan di serious
game didistribusikan ke dalam tiga level domain kognitif Bloom yang dimplementasikan
di SD: kemampuan mengingat (C1), memahami (C2), dan menerapkan
(C3). Serious game juga mengatur level kesukaran tantangan secara dinamis
berdasarkan pengalaman pemain pada tantangan sebelumnya. Pengaturan level
kesukaran secara dinamis ditujukan agar pemain tidak cepat frustrasi atau bosan
dalam permainan.
Serious game yang diimplementasikan dalam penilaian sudah melalui uji
penerimaan dan uji tanggapan dari pengguna. Klasifikasi data permainan dilakukan
melalui metode Bayes Net (BN), Naïve Bayes (NB), dan J48. Dalam melakukan
klasifikasi, penerapan tiga metode digabungkan dengan dua opsi tes: crossvalidation
dan percentage split. Klasifikasi di masing-masing perlakukan dikerjakan
dalam sepuluh ulangan melibatkan sub data yang dipilih acak sebagai data
pengujian. Hasil pengujian menunjukkan: (1) dari delapan skenario pengujian
penerimaan pengguna diperoleh bahwa keseluruhan masukan skenario pengujian
memberikan luaran yang sesuai harapan; (2) rata-rata skor respon pengguna yang
dikumpulkan menggunakan kuesioner skala Likert dengan lima opsi terletak
dalam interval kategori respon positif (59,93); (3) persentase kebenaran klasifikasi
tertinggi yang dihasilkan pada klasifikasi data permainan adalah 88,5% yang
dihasilkan melalui penerapan metode J48 yang digabungkan dengan opsi tes
percentage split = 85%. Kategori agreement pada penerapan metode J48 dengan
opsi tes percentage split adalah Baik (78%).
Dari tiga bentuk pengujian disimpulkan bahwa selayaknya metode J48
dipilih dalam melakukan klasifikasi data permainan serta penilaian melibatkan
serious game berbasis taksonomi Bloom dijadikan alternatif dalam penilaian pembelajaran
materi geometri bangun datar untuk siswa SD kelas 5.
====================================================================================
We developed a new approach for mathematics' learning assessment
applying a serious game which is called BoTySeGa. This approach was proposed
as an alternative way for recording learners' data which are representative to understand
the characteristic of learners. The game implemented in assessment
involves the three aspects of a serious game: game, knowledge, and pedagogy. We
involve the plane geometry of parallelogram for the 5th elementary students and
Bloom's taxonomy successively as knowledge and pedagogy aspects of the game.
The serious game was developed following the serious game constructivist
framework for children’s learning. Inside the game; the challenges are distributed
into the first three of Bloom's cognitive domain which are implemented in
elementary school: remember, understand, and apply. The game system adjusts
dynamically challenge's level of difficulty in consideration with players'
experience on the previous challenge. This approach is designed to bring players
far away of boredom and frustration.
The serious game applied in the proposed assessment has been tested
through user acceptance testing and players' respond to the usage of the game in
assessment. Gameplay data are classified through three methods i.e.: Bayes Net,
Naïve Bayes, and J48. Each method is conducted with two testing options: crossvalidation
and percentage split. The classification in each treatment is done in ten
times of repetition. Test results show that: (1) user acceptance testing involving 85
learners shows that BoTySeGa has fulfilled the learning assessment requirement,
(2) the average score of players' response recorded utilizing five-points Likerttype
of questionnaire is 59,93, it falls within "Positive" category, (3) the highest
percentage of correctly classification of gameplay data is 88,5% which is calculated
through classification applying method J48 with percentage split testing
option 85%. Level agreement of classification is 0.78 which is Good.
Based on testing results, we suggest the use of J48 method for the
classification of gameplay data and support the implementation of mathematics’
learning assessment utilizing Bloom's taxonomy-based serious game as an
alternative assessment in learning.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Additional Information: | RDE 003.8 Suk s |
| Uncontrolled Keywords: | serious game; taksonomi Bloom; Penilaian pembelajaran; Matematika; Bloom taxonomy; learning assessment; Mathematics |
| Subjects: | Q Science > QA Mathematics > QA269 Game theory |
| Divisions: | Faculty of Industrial Technology > Electrical Engineering > 20001-(S3) PhD Thesis |
| Depositing User: | Anis Wulandari |
| Date Deposited: | 29 May 2017 04:20 |
| Last Modified: | 27 Dec 2018 07:48 |
| URI: | http://repository.its.ac.id/id/eprint/41399 |
Actions (login required)
 |
View Item |