The aims of this research is to: 1) To find out how is the implementation of teaching speaking by using Problem-Based Learning model at elevent class of SMAN 1 Grogol?, 2) To find out the effect of teaching speaking using Problem-Based Learning at elevent class of SMAN 1 Grogol, 3) To find out the effect of the teaching speaking without using Problem-Based Learning at elevent class of SMAN 1 Grogol, 4) To analyze the differences between students who are taught using Problem-Based Learning model from without using Problem-Based Learning in the teaching speaking.
This research uses a quantitative research methods. The sample of this research used two classes, class is XI- 4 as the experimental class and XI-5 as a controll class. The sample was chosesn by purposive sample, where the school chose the class through the teacher with berbagai macam pertimbangan. The data collection technique of this research is pre-test and post-test. The pre-test is used to determine the initial value of students before the using Problem-Based Learning model , and the post-test is used to see the final value of students after the using Problem-Based Learning model. For instruments, validation and reliability tests are carried out first before being given to the sample. And to analyze the data, normality, homogeneity, and t-test will be conducted.The results show that : 1) What is the effect of teaching speaking using Problem-Based Learning model at elevent class of SMAN 1 Grogol?, and 2) What is the effect of the teaching speaking without using Problem-Based Learning model at elevent class of SMAN 1 Grogol? 3) Is there any significant differences between students who are taught using Problem-Based Learning model from without using Problem-Based Learning in the teaching speaking?
The results show that: 1) There is an effect of teaching spaking by using Problem-Based Learning model on eleventh class students of SMAN 1 Grogol. The results are seen from the descriptive calculations where the average between the pre-test and post-test of the experimental class is 18.53. 2) There is no significant effect of teaching speaking without using Problem-Based Learning model on eleventh class students of SMAN 1 Grogol. The results hows that the student’s average score only increased 1.33 after using the Problem-Based Learning model. 3) There is a significant difference between using Problem-Based Learning model and Discovery Learning. The sig. (2-tailed) shows 0.005 <0.05, which means there is a significant difference in the average student learning outcomes between the experimental class post-test and the control class post-test after using the Problem-Based Learning model with other model (Discovery Learning). The researcher concluded that there was a significant difference between using Problem-Based Learning model and Discovery Learning.

Key words: Effectiveness, teaching speaking, teaching model, problem base-learning.