Context: Graduate medical education (GME) programs must develop curriculum to ensure scholarly activity among trainees and faculty to meet accreditation requirements and to support evidence-based medicine.Objective: Test whether research-related needs and interests varied across four groups: primary care trainees, specialty trainees, primary care faculty, and specialty faculty.Design: We surveyed a random sample of trainees and faculty in Kaiser Permanente Southern California's GME programs. We investigated group differences in outcomes using Fisher exact and Kruskal-Wallis tests.Main Outcome Measures: Research experiences, skills, barriers, motivators, and interests in specific research skills development.Results: Participants included 47 trainees and 26 faculty (response rate = 30%). Among primary care faculty, 12 (71%) reported little or no research experience vs 1 (11%) for specialty faculty, 14 (41%) for primary care trainees, and 1 (8%) for specialty trainees (p < 0.001). Submission of research to the institutional review board, an abstract to a conference, or a manuscript for publication in the previous year varied across groups (p = 0.001, p = 0.003, and p < 0.001, respectively). Overall self-reported research skills also differed across groups (p < 0.001). Primary care faculty reported the lowest skill level. Research barriers that differed across groups included other work roles taking priority; desire for work-life balance; and lack of managerial support, research equipment, administrative support, and funding.Conclusion: Faculty and trainees in primary care and specialties have differing research-related needs that GME programs should consider when designing curricula to support scholarly activity. Developing research skills of primary care faculty is a priority to support trainees' scholarly activity.