Abstract

The purpose of this study is to test the effectiveness of PjBL model by using Science KIT to improve students' confidence and problem-solving skills. This research is done by using a quasi-experimental design with a type of non-equivalent control group design. The research subjects were 50 fifth grade students from Elementary School 02 Krompeng and Elementary School 01 Kalirejo. Data on students’ self-confidence and problem-solving skills assessment are collected through observation and tests. Using N-Gain analysis and t-test. Research results show that there are different results in terms of self-confidence attitudes between groups of students who follow PjBL model with Science KIT and the other groups of students who follow PjBL learning model, obtained a value (tvalue = 37.12 > ttable  = 1.68), there are differences in learning outcomes as well, in terms of problem-solving skills between groups of students who follow PjBL learning model with Science KIT with groups of students who follow PjBL learning model, it is proven post-test result obtained in experimental class is 74.96 while in the control class is 60.72 (tvalue = 19.55 > ttable = 1.71). The results of this research are beneficial for teachers as an overview of condition self-confidence and problem-solving skills that can be taken into consideration in planning the lesson.