THE EFFECTIVENESS OF BARRETT'S AND BLOOM'S TAXONOMY IN WRITING THE PISA MODEL READING COMPREHENSION TEST

International assessments such as PISA, TIMSS, and PIRLS indicate that Indonesian students perform poorly, ranking below Malaysia and far behind Singapore. This issue is compounded by internal challenges, including teacher quality. To address this, a study was conducted to evaluate the effectiveness of the Barrett and Bloom model evaluation question grid in PISA-style reading comprehension questions. The research utilized the ADDIE Research and Development model, analyzing two sets of teacher-created questions. A senior teacher developed Package A using a reading comprehension model, while a junior teacher designed Package B with a non-argumentative Bloom grid. Data analysis was conducted using a t-test in SPSS. The findings revealed that Package B outperformed Package A, with t = -11.133 and a significance level of ≤ 0.05. Additionally, most senior teacher-written questions were HOTS-based, making them more challenging and requiring critical thinking. Conversely, junior teacher-created questions were easier, with fewer HOTS elements, resulting in better test scores. Thus, the study concludes that the Barrett Grid is more effective in designing PISA-style reading comprehension questions.