This study is the first to analyze how trust in science is affected by motivated reasoning through a large-scale information experiment. We study a representative sample from Germany and focus on evidence regarding the impact of migration on native labor-market success. We find that trust in science increases when randomly provided evidence aligns with prior beliefs but decreases when it contradicts them. These findings highlight the role of belief-confirmation bias in shaping attitudes toward scientific evidence The study underscores the importance of understanding how individuals interpret and respond to scientific information, particularly in the context of policy-making and public discourse.