Representativeness of two sampling procedures for an internet intervention targeting cancer-related distress: a comparison of convenience and registry samples.

Internet interventions often rely on convenience sampling, yet convenience samples may differ in important ways from systematic recruitment approaches. The purpose of this study was to evaluate potential demographic, medical, and psychosocial differences between Internet-recruited and registry-recruited cancer survivors in an Internet-based intervention. Participants were recruited from a cancer registry (n = 80) and via broad Internet outreach efforts (n = 160). Participants completed a set of self-report questionnaires, and both samples were compared to a population-based sample of cancer survivors (n = 5,150). The Internet sample was younger, better educated, more likely to be female, had longer time since diagnosis, and had more advanced stage of disease (p's < .001), and the registry-sample was over-represented by men and those with prostate or other cancer types (p's < .001). The Internet sample also exhibited lower quality of life and social support and greater mood disturbance (p's < .001). Understanding how convenience and systematic samples differ has important implications for external validity and potential for dissemination of Internet-based interventions.