<![CDATA[This study examined the implementation and learning improvement associated with the discovery learning model assisted by simple molecular teaching aids in supporting in-depth learning of functional groups among grade XII students at SMA Negeri 2 Maginti. In-depth learning in this study refers to students' ability to visualize molecular structures, distinguish functional groups, and explain the relationship between functional group structure and compound characteristics rather than merely memorize names or formulas. The study employed a quasi-experimental one-group pretest-posttest design involving 29 students. The intervention used low-cost molecular models made from cork and satay skewers to represent atoms and chemical bonds during the discovery-learning stages of stimulation, problem identification, data collection, data processing, verification, and generalization. Data were collected through conceptual understanding tests, student activity observation sheets, student response questionnaires, and classroom documentation. Data analysis was conducted descriptively by calculating the mean score, mastery percentage, student activity percentage, student response percentage, and normalized gain (N-Gain). The findings showed that the average pretest score of 46.00 increased to 85.00 in the posttest, while learning completion improved from 10.34% to 89.66%. The N-Gain value reached 0.72, which falls into the high category. Student activity during learning was high to very high, particularly in constructing molecular structures (96.55%), discussing differences among functional groups (89.66%), and drawing conclusions from observations (86.21%). Student responses also showed high agreement that the teaching aids helped visualize molecular structures and clarify functional group concepts. These results indicate that discovery learning assisted by simple molecular teaching aids can support conceptual improvement and active participation in functional group learning, especially in schools with limited facilities; however, the findings should be interpreted within the limits of a one-group design without a control class.]]>
