Abstract :
Randomized Control Group Pretest-Posttest design. The population in this study
were all eighth grade students of SMP Negeri 3 Rambah Samo with samples of
class VIII A and class VIII B. The sample classes were obtained using saturated
sample techniques. Whereas to determine the experimental class and control class
is randomly by drawing. In the experimental class a problem based learning
(PBL) model and a control class were applied in conventional learning. The
instrument of this study was a test instrument consisting of pretest and posttest. To
test the hypothesis, a statistical analysis was conducted to test the similarity of the
average N-Gain score between the experimental group and the control group.
Before the hypothesis testing is carried out the prerequisite test is done, namely
the normality test with the lilliefors test, the results of which are two sample
classes with normal distribution and the homogeneity test using the F test which
results in the two sample classes having homogeneous variances. Furthermore,
the hypothesis test is performed using the t test, the calculation results obtained by
tcount> t table is 4.16> 2.011 for ? = 0.05 then H0 is rejected. So that it can be
concluded that there is an increase in mathematical communication skills of class
VIII students of SMP Negeri 3 Rambah Samo through the Problem Based
Learning (PBL) model.