<![CDATA[This research is aim to compare the students critical thinking skills who engaged in a problem-based learning model with students who engaged in conventional learning. This is experimental research, with posttest only control group design. The population is 50 of Accounting XI students in Widya Bhakti Vocational School in Ruteng, Flores. Samples grouping on nonprobability sampling, that is saturated sampling where all of this population members are used as samples. The sample in the experimental class was all 24 students of class XI Accounting 1 who received treatment problem-based learning, while the control class sample was all 26 students of class XI Accounting 2 who received conventional learning. Data is collected through tests in the form of description questions.A prerequisite test was carried out before testing the research hypothesis: the normality test and the homogeneity test. The results of these prerequisite test are data in two groups indicated as normally distributed and both are homogeneous. Then, the research hypothesis was tested using t-test. Obtained and on a significance level of and degrees of freedom. Because then is rejected while is acceptable. In this case, the average value of critical thinking skills in the experimental class, the average value of critical thinking skills in the control class. The meaning of the results of this analysis are: (1) when compared to conventional learning models, the problem-based learning model has a more positive and significant influence on the formation of students' critical thinking skills; (2) The value of t-count = 4.005 indicates that the average value of critical thinking skills of students using problem-based learning models is so higher than the average value of conventional learning students' critical thinking abilities.]]>
