<![CDATA[The purposes of the research are: (1) to know the influence of constructivist-metacognitive model towards students? cognitive learning outcome; and (2) to know the influence of constructivist-metacognitive model towards students? retention. This research was quasy experiment with posttest only nonequivalent control group design by using the experimental class (application of constructivist-metacognitive model) and control class (without application constructivist-metacognitive model). Research population were all students at the eleventh grade of science class at SMAN 3 Surakarta in academic year 2014/2015. The sampling technique was cluster random sampling, so the researcher chose XI MIPA 6 as the experimental class with 33 students and XI MIPA 5 as the control class with 32 students. Data were collected by testing and non-testing technique. Data were analyzed by t-test.Based on the t-test analysis, it is acquired that p-value of students? cognitive learning outcome is less than 0,050 (p (0,000) < 0,050), so H0 is denied. In summary, there is significant difference on cognitive learning outcome between the control class and the experimental class. T-test analysis about students? retention is acquired that p-value is less than 0,050 (p (0,016) < 0,050), so H0 is denied. It can be concluded that there is significant difference on students? retention between the control class and the experimental class.The conclusions of this research are: (1) constructivist-metacognitive model affects students? cognitive learning outcome; and (2) constructivist-metacognitive model affects students? retention. Keywords : constructivist-metacognitive model, cognitive learning outcome, students? retention]]>
