In this study, the R. W. Lent, S, D. Brown, and G. Hackett (1994) social cognitive model of interest development was applied to the prediction of interest in research. Participants included 184 counseling psychology doctoral students. Results of a path analysis indicated that 5 variables were significant predictors of interest in research: research outcome expectations, research self-efficacy beliefs, investigative interests, artistic interests, and age. In addition, several factors indirectly affected interest in research through effects on research self-efficacy-the research training environment, Holland Investigative interests, and year in program. Research training environment, Holland Investigative interests, and research self-efficacy beliefs also had an impact on interest in research indirectly through their effects on research outcome expectations.The scientist-practitioner model for psychology training programs has been endorsed since the first major training conference in counseling psychology (American Psychological Association, 1952). However, a growing body of literature indicates that counseling psychology graduate students and professionals lack interest in research and maintain low levels of research productivity (e.g., Royalty & Magoon, 1985;Royalty & Reising, 1986). Further, such behavior is not specific to counseling psychologists; Gelso (1993) maintained that a low research productivity rate exists among doctoral graduates in all professional psychology programs.Numerous investigations have attempted to identify the factors that lead to increased interest in research and research productivity. Support has been found for various factors, including personality characteristics (e.g., Kahn &

The Research Self-Efficacy Scale (RSES; Greeley, et al., 1989) was completed by 177 doctoral students from a wide variety of disciplines. Factor analysis of the RSES indicated four primary factors: Conceptualization, Early Tasks, Presenting the Results, and Implementation. Hierarchical regression analyses focused on 136 subjects from the original sample and indicated that three subscales of the RSES (Early Tasks, Conceptualization, and Implementation) accounted for unique variance in the prediction of interest in research involvement. The number of years in graduate school and involvement in research activities contributed significantly to the prediction of research self-efficacy.

A program titled "The Images Within" was implemented and evaluated in three sites on the east coast of the United States. This school curriculum uses art work developed by children of alcoholics to stimulate classroom discussions of the problems of parental alcohol abuse. The evaluation with 278 experimental and 310 control students indicated increased knowledge about the effects of alcohol and improved skills in coping with alcohol problems and help seeking behavior. Process data indicated that related programs were initiated, students were stimulated by the program and teachers were positive in their perceptions of the program. Schools implementing this program need to have developed referral networks to handle the individual concerns that are expressed as a result of participation in "The Images Within."

No abstract