<![CDATA[The background of this research departs from the ethnic diversity of students in Ternate which has the potential to affect student retention. This research aims to assess the impact of various learning models on the retention of multiethnic students in Ternate City Junior High School. This type of research is a quasi experiment. The design used in this study is the Pretest-Posttest Nonequivalent Control Group Design. The total sample of 76 students consisted of 20 ethnic Ternate, 27 Makian, and 29 Tidore ethnicities. The Research Instrument consists of observation sheets and test questions. The data collection technique consists of (pretest) and (posttest). Data analysis used descriptive statistics to show the description of student retention. The analysis was continued using covariance analysis (ANACOVA). The results of the follow-up test showed that the retention of students in the STAD learning model group combined with NHT was not significantly different from the STAD model, however, it was significantly different from conventional learning. Ethnicity has a significant effect on the retention of multiethnic students. The retention of Ternate ethnic students is not significantly different from the Tidore ethnicity but is significantly different from the Makian ethnicity. The interaction of learning models and ethnicity has a significant effect on student retention. The STAD model combined with NHT has not been able to make a significant contribution to the retention of Ternate, Makian and Tidore ethnic students, but when viewed from the corrected average score of STAD combined with NHT ethnic Ternate, Makian and Tidore have the highest score when compared to other learning models.]]>
Copyrights © 2025
