<![CDATA[This study aims to determine the differences in mathematical connection ability between students taught using the Discovery Learning model and those taught using direct instruction. This research is a quantitative study employing an experimental method with a factorial experiment design. The study was conducted at MTs Al Huda Pekanbaru during the 2025/2026 academic year. The research population consisted of all seventh-grade students at MTs Al Huda Pekanbaru. The sample was selected using a cluster random sampling technique, with class VII.A designated as the experimental class and class VII.C as the control class. Data collection techniques included tests, questionnaires, observations, and documentation. The research instruments comprised mathematical connection ability test questions (Pretest and Posttest) and an Adversity Quotient questionnaire. The data analysis technique used was a two-way ANOVA. Based on the results of the data analysis, it can be concluded that there is a difference in mathematical connection ability between students who learned using the Discovery Learning model and those who learned using direct instruction (Fcalculated = 12.95 > Ftable = 4.05); there is a difference in mathematical connection ability among students with high, moderate, and low Adversity Quotient (Fcalculated = 18.31 > Ftable = 3.20); and there is no interaction between the Discovery Learning model and Adversity Quotient on students' mathematical connection ability (Fcalculated = -4.35 < Ftable = 3.20).]]>
