Attrition among urban engineering students at Stevens Institute of Technology was studied. At the time of the student's college entry, various ability, achievement, personality, interest, and biographical factors were analyzed in an effort to predict student persistence, withdrawal, or academic failure. For the 381 male students who entered the college in September 1969, 259 graduated and received their B.S. degree, 82 failed, and 40 voluntarily left for nonacademic reasons and transferred. Among the findings are the following: the average Scholastic Aptitude Test scores on the verbal and mathematics tests were lowest for the fail group and highest for the group that left the college and presumably transferred elsewhere; there were no significant differences between the three groups on the Brown-Holtzman Survey of Study Habits and Attitudes; students who persisted to graduate differed statistically from transfers on 22 of the occupational scales of the Strong Vocational Interest Blank for Men; significant differences between those who graduated and those who transferred were found on the F, Mf, and Sc scales of the Minnesota Multiphasic Personality Inventory (MMPI) (transfers obtained higher mean scores); the fail group obtained higher mean scores than the persisters on the Ma, Pd, Sc and F scales of the MMPI; and greater number of students who failed came from homes in which the parents were widowed, divorced, separated, or both deceased. An extensive literature review, a bibliography, and a sample student personal data record questionnaire are included. (SW)