ABSTRACT
Analytical solutions to the problem of causal order become particularly important when delinquency may be a cause as well as an effect. The mere fact of arrest, or adjudication is often thought to produce differences between delinquents and nondelinquents that were not present prior to the official action. There are several ways of solving the problem of causal order in cross-sectional studies, at least in part. A solution to the problem of causal order, at least in principle, is the longitudinal or panel study. In an ideal version of this design, the investigator would select a sample of infants and continually collect data on them until they became adults. The practical difficulties of panel studies are undoubtedly the major reason for their infrequent use in delinquency research. Once the data are collected, awareness of the problem should alert the analyst to possible ambiguity and stimulate a search for evidence.
