Surveying marginalized populations

According to interviews conducted throughout the first year of prison release with 122 formerly incarcerated men and women around 36 years of age, marginalized people may be unresponsive to recruitment efforts for social science research studies because they lack permanent domiciles and telephone numbers and face substance abuse and mental health issues, and the study suggests that selection bias can occur in surveys of poor and disadvantaged populations if such individuals are not retained in studies. - Read at PNAS.org

Article #16-04138: “Study retention as bias reduction in a hard-to-reach population,” by Bruce Western, Anthony Braga, David Hureau, and Catherine Sirois.