Selective subject attrition in a longitudinal study of head-injured veterans

Recent longitudinal studies of aging have suggested that intellectually inferior subjects selectively drop out during the course of the investigation. This issue of subject attrition was addressed in a neuropsychological study of 314 veterans who had sustained head or peripheral nerve injuries in World War II. These men were examined originally by Teuber and his colleagues at New York University in the 1940s and 1950s, and the veterans' participation was sought for a 40-year follow-up study at the Massachusetts Institute of Technology. In the current follow-up study, the veterans were subdivided according to their response to recent recruitment letters: those who agreed to be retested, those with whom recent contact was established but who declined to participate, those who did not reply to our correspondence, and those who were deceased. Those who agreed to be retested in the 1980s had significantly more education and achieved significantly higher Army General Classification Test scores both before and after injury than did the untested cohort. Even with the selective subject attrition, however, the remaining sample represented a broad range of intellectual competence.