The 143 men were part of an original random sample of 450 men who were released from two London prisons between 1966 and 1969 and who were among the one-third of prisoners who refused the opportunity of using the job-finding service. The offenders had all stated their plans for employment as training when they refused the help. The frequency and the seriousness of their reconvictions were both considered. Of the total sample, 29 percent were not reconvicted after 12 years. A consistent pattern emerged when previous convictions were considered. About one-fourth of those with 1 or 2 previous convictions were reconvicted within 12 years, compared to 78 percent of those with 3 to 9 previous convictions and 94 percent of those with 10 or more previous convictions. Differences between those with different types of stated employment plans following release were strongly affected by the previous conviction rates even at the 1-year followup. The apparent importance of particular types of stated plans for employment was really an artifact of failing to control for previous convictions. The study group was also diverse in terms of seriousness of convictions. Just under one-quarter (22 percent) had subsequent court appearances that were neither frequent nor serious, whereas 48 percent had serious and/or frequent reconvictions. The men's responses to the original offer of help in finding employment were rational in most cases. Half the offenders had no reconvictions or only minor ones. The other half probably made an equally realistic and rational choice in refusing employment assistance. Essentially they were already committed to a criminal career. The 12-year followup period provided a much more complete picture than the 1-year period. Data tables and eight references are listed.