<![CDATA[This research was motivated by the low level of students' descriptive text writing skills in elementary schools. This research aims to determine the effect of the Think Talk Write learning model assisted by KOMED media on students' descriptive text writing skills. The research approach used is quantitative with a quasi-experimental design (Quasi-Experimental) in the form of a Nonequivalent Control Group Design. Sampling was carried out using a purposive sampling technique, which determined students in classes III-A and III-B at SD Negeri Harjosari 02 as research subjects. Data collection uses test techniques (pretest and posttest) and non-test (observation, questionnaires, interviews and documentation). Data analysis was carried out through prerequisite tests (normality and homogeneity) as well as hypothesis testing using the independent sample t-test and simple linear regression test. The research results show that: (1) There is a significant difference in descriptive text writing skills between the experimental class and the control class, as evidenced by the results of the independent sample t-test which obtained a significance value of 0.000 < 0.05 so that H0 was rejected and Ha was accepted; (2) There is a significant influence on the use of the TTW learning model assisted by KOMED media on students' descriptive text writing skills, as evidenced by the results of a simple linear regression test which obtained a significance value of 0.000 < 0.05 so that H0 was rejected and Ha was accepted. The Think Talk Write learning model assisted by KOMED media has been proven to have a significant positive effect on improving the descriptive text writing skills of class III students.]]>
