Electronic Theses and Dissertation

Kemampuan menyelesaikan masalah merupakan salah satu komponen krusial dalam proses pembelajaran matematika. Namun, sebagian besar siswa masih merasa kesulitan dalam menyelesaikan soal yang memiliki indikasi pemecahan masalah matematika. Penelitian ini bertujuan untuk mengetahui kemampuan siswa dalam menyelesaikan masalah matematis melalui penerapan model Problem Based Learning (PBL) berbantuan Artificial Intelligence (AI), khususnya dengan menggunakan platform ChatGPT. Penelitian ini menggunakan pendekatan kuantitatif dengan desain pre-eksperimen tipe One-Shot Case Study. Subjek penelitian adalah siswa kelas VII-1 SMP Muhammadiyah Kualasimpang. Instrumen yang digunakan berupa soal tes. Hasil analisis data menunjukkan bahwa rata-rata nilai Soal tes siswa adalah 70,536, yang melebihi Kriteria Ketuntasan Minimum (KKM) sebesar 65. Hasil uji normalitas menggunakan uji Shapiro-Wilk menunjukkan bahwa data berdistribusi normal. Uji hipotesis menggunakan One Sample t-test menghasilkan nilai thitung > ttabel (3,011512>1,703) pada taraf signifikansi 5%,. Dengan demikian, dapat disimpulkan bahwa kemampuan siswa menyelesaikan masalah matematis melalui penerapan model problem based learning (PBL) berbantuan AI mencapai ketuntasan.

Kata kunci: Problem Based Learning (PBL), Artificial Intelligence (AI), ChatGPT, kemampuan menyelesaikan masalah matematis.

Problem solving ability is one of the crucial components in the mathematics learning process. However, most students still find it difficult to solve problems that have indications of mathematical problem solving. This study aims to determine students' ability to solve mathematical problems through the application of the Problem Based Learning (PBL) model assisted by Artificial Intelligence (AI), especially by using the ChatGPT platform. This study used a quantitative approach with a One-Shot Case Study type pre-experiment design. The research subjects were students of class VII-1 SMP Muhammadiyah Kualasimpang. The instrument used was a test question. The results of data analysis showed that the average score of students' test questions was 70.536, which exceeded the Minimum Completeness Criteria (KKM) of 65. The normality test results using the Shapiro-Wilk test show that the data is normally distributed. Hypothesis testing using One Sample t-test resulted in a tcount> ttable value (3.011512>1.703) at the 5% significance level. Thus, it can be concluded that students' ability to solve mathematical problems through the application of problem-based learning (PBL) model assisted by AI reached mastery. Keywords: Problem Based Learning, Artificial Intelligence, ChatGPT, mathematical problem-solving ability.