Professional socialization was viewed as the interactive process by which an individual integrates a professional role into the self-concept through the acquisition and internalization of the requisite knowledge, skills, values, attitudes, and norms of the profession (Jacox, 1978; Moore, 1970). According to the Cohen (1981) model of professional socialization, students progress through the four developmental stages of unilateral dependence, negative/independence, dependence/mutuality, and interdependence as they advance through an educational program in nursing. The theoretical framework incorporated selected concepts of role theory and Perry's (1968) theory of intellectual and ethical development into Cohen's model.

The purpose of the study was to ascertain whether baccalaureate nursing students evidenced the proposed stages of professional socialization. The Professional Socialization Staging Scale (PS('3)), consisting of four subscale scores representative of the developmental stages, was administered concurrently to 422 students enrolled in eight clinical nursing courses offered at the University of Alabama School of Nursing. Instrument revision was accomplished using a cross-validation design, and the study hypotheses were tested with 214 subjects in the cross-validation subsample. Coefficient alpha values for the STAGE I through STAGE IV subscales were found to be .56, .72, .46, and .73, respectively.

Because all groups of students evidenced the interdependence stage, there was no overall relationship between the developmental stages and level of enrollment in the educational program. Thus the findings did not support the Cohen model. Significant findings were that beginning students were more dependent than were graduating students, that older students were less dependent and more highly interdependent than were younger students, and that students with concurrent work experience in nursing-related fields were more highly interdependent than were students without concurrent work experience. There were no significant differences in the stages among groups of students with differences on the variables of race, gender, marital status, previous work experience, or presence of immediate family members who were nurses.

Although the validity of the model is brought into question by the findings, the validity of the research instrument also is questionable. It is recommended that Cohen's model be further tested, using other research measurements of a revised version of the PS('3).