The inference mediation hypothesis (IMH) assumes that individual difference factors that affect reading proficiency have direct and indirect effects on comprehension outcomes, with the indirect effects involving inference processes. The present study tested the IMH in a diverse sample of two and four-year college students in a task that emphasizes comprehension of the passage (traditional assessment) and a task that emphasizes complex problem solving (SBA). Participants were administered assessments of foundational skills that support reading, inference generation, a traditional assessment of comprehension proficiency, and a scenario-based reading assessment. The results support the IMH. However, the strength of the indirect relationships depended on the type of reading performance assessment. Coherence building inferences partially mediated the relationship for both assessments. However, elaborative inferences only partially mediated the relationship for the scenario-based assessment. The results are discussed in terms of theories of purposeful reading and implications for understanding college readiness. [This paper will be published in "Contemporary Educational Psychology."]