This note suggests a solution to the problem of achieving randomization in experimental settings where units deemed eligible for treatment "trickle in," that is, appear at any time. The solution permits replication of the experiment in order to test for time-dependent effects. (Author/CTM)

Flexibility in evaluative research design does not necessitate the abandonment of randomly constructed comparison groups. Three designs are reviewed which provide at least the option of randomization while maintaining great flexibility. The strengths and weaknesses of the designs are discussed. (Author)

A major problem in social science research is that of successfully carrying out the random assignment of persons to experimental and control groups. In this study a computer-based random assignment procedure operated successfully on a weekly basis for 17 consecutive weeks in a program serving over 360 ex-offenders. (CTM)

Evaluations involving nonrandom assignment to treatment or control groups are vulnerable to an accidental or intentional confounding of a selection effect with the treatment effect. Two techniques, discriminant analysis and base expectancy analysis, permit separate estimation of the selection and treatment effects in the final results. (Author/CTM)

The possible biasing effects of attrition of subjects in longitudinal experiments is a problem to which social science researchers and evaluators rarely attend. A study of research on Project Follow Through, however, revealed few differences in family income or pretest scores, between students who dropped out of treatment and of comparison groups.…