<![CDATA[Writing in English as a Foreign Language (EFL) represents one of the most demanding competencies for Indonesian junior high school students, particularly those enrolled in Islamic boarding schools (pesantren) where instructional time is constrained by dual academic and religious curricula. This study examined the effectiveness of the Microselection Learning Strategy in improving descriptive text writing skills among ninth-grade female students at Khoirotul Usroh Islamic Boarding School, Stabat, Langkat Regency, North Sumatra, Indonesia. A Classroom Action Research (CAR) design based on the Kemmis and McTaggart cyclical model was employed across two cycles over five weeks, involving all 36 female students as participants through a saturated sampling technique. Data were collected using pre-tests and post-tests evaluated with a five-dimension rubric (title, identification, description, vocabulary, and closing), supplemented by observation sheets, semi-structured interviews, questionnaires, and documentation. Quantitative data were analyzed using descriptive percentage statistics; qualitative data followed Miles and Huberman's reduction–display–conclusion framework. Results showed a dramatic improvement in classical mastery: from 13.89% in the pre-test to 50.00% after Cycle I, and reaching 100.00% at the conclusion of Cycle II. The class mean score rose from approximately 46.9 to 87.1 points. Qualitative findings corroborated these gains, documenting improvements in vocabulary selection, sentence cohesion, paragraph organisation, and student confidence. The Microselection strategy proved highly effective as a scaffolded, cognitive-theory-grounded approach to EFL writing instruction in a boarding school context.]]>
