PENERAPAN PEMBELAJARAN KOOPERATIF MODEL GROUP INVESTIGATION (GI) UNTUK MENINGKATKAN MOTIVASI DAN HASIL BELAJAR BIOLOGI

Model pembelajaran mempengaruhi motivasi dan hasil belajar Biologi siswa sekolah menengah atas. Berdasarkan observasi awal, diketahui bahwa pembelajaran Biologi di kelas X-1 SMA Negeri 7 Malang memiliki beberapa ciri, yaitu (1) motivasi belajar siswa cenderung kurang, (2) dalam menyelesaikan suatu tugas, siswa dengan karakteristik yang sama cenderung mengelompok sehingga terbentuk kelompok homogen yang tidak seimbang di kelas, dan (3) metode ceramah yang diterapkan oleh guru tidak menarik minat siswa sehingga mempengaruhi hasil belajarnya. Berdasarkan kondisi tersebut maka dilakukan penelitian dengan menggunakan pembelajaran kooperatif model Group Investigation (GI) untuk meningkatkan motivasi dan hasil belajar Biologi siswa. Jenis penelitian ini adalah penelitian tindakan kelas dengan menggunakan pendekatan kualitatif dan dirancang dalam dua siklus. Hasil penelitian menunjukkan penerapan pembelajaran kooperatif model GI dapat meningkatkan motivasi dan hasil belajar Biologi siswa. Motivasi belajar siswa meningkat sebesar 31, 48% yaitu dari 46, 15% dengan kategori cukup pada siklus I menjadi 77, 63% dengan kategori baik pada siklus II. Hasil belajar siswa juga mengalami peningkatan, terlihat dari nilai rata-rata siswa sebesar 81, 94% pada siklus I menjadi 89, 18% pada siklus II dan siswa yang tuntas belajar juga mengalami peningkatan dari 79, 49% pada siklus I menjadi 97, 44 % pada siklus II. Hasil angket juga menunjukkan bahwa siswa menyukai dan lebih termotivasi untuk belajar Biologi setelah mengikuti pembelajaran dengan pembelajaran kooperatif model GI.

A Use Case Complexity Based Approach to Prioritize Test Cases for Software Product Line Testing

Software Product Line (SPL) Engineering is software engineering tools, techniques and methods for the development of products sharing common features along with some variant features. A feature model (FM) represents the relationship among various features. Feature model is the main artifact that generates different products of SPL. Product testing is the necessary part of quality assurance. The combinatorial rise of the products occurs due to the large number of features resulting, SPL testing very complex. To reduce the complexity of testing, there are some techniques used in the literature like products selection, minimization and prioritization. In product selection, the scope of testing is reduced. Product minimization chooses a product that represents all the characteristics of products and eliminates redundancy but not feasible for revealing all faults and early fault detection. Product minimization and selection discard some products but prioritization covers all the products. Products Prioritization technique reorders the products based on feature coverage criteria or feature coupling complexity criteria. Prioritization reveals faults earlier that help in early fault correction, feedback, saves cost and time. In existing techniques of prioritization criteria, no one addressed the feature?s individual complexity to prioritize the products. In this present research, the researcher calculated use case complexity of individual features calculated followed by finding product?s complexity in order to prioritize product (Test Case).There are abstract and concrete features in a feature model. Concrete features can be mapped on use case descriptions because they provide the functional requirements of system. An algorithm proposed for finding complexity of UC that based on UC template attributes. It eventually prioritized the products based on individual feature?s complexity. To validate the work, e Shop case study used that was evaluated by generating feature model and valid products configurations with the help of the SPL tool. The e Shop data set obtained from SPLOT that was a free repository. Lists of prioritized products for existing criteria and proposed criteria obtained by running their respective algorithms. Effectiveness of the proposed criteria in comparison with the existing criteria judged by calculating average percentage of faults detected for each prioritized list of products. Proposed approach significantly improved the average percentage of faults detected. The results of case study show that different ordering of the same test cases gives considerable difference in the average percentage of faults detected.