The proportion of essential tremor (ET) cases that can be attributed to genetic factors is unknown; estimates range from 17–100%. One possible reason for this variability is that clinic and community cases may differ with regard to family history of ET. This is because clinic patients are self-selected and represent as few as 0.5% of all ET cases. Our goal was to determine whether ET cases ascertained from a clinic differed from those ascertained from a community in terms of the family history information that they provided. Subjects (57 clinic, 64 community) underwent a family history interview. Clinic cases were 4.73 times more likely to report an affected relative than were community cases. We conclude that there was a substantial difference between our clinic and community ET cases in terms of the information they provided regarding their family history. Selection and reporting biases could have accounted for this difference. Because of these biases, the source of the cases must be taken into consideration when investigators are trying to synthesize the widely variable results of studies that have estimated the genetic contribution to ET. © 2001 Movement Disorder Society.