A retrospective methodology was used to assess if there are any statistically significant relationships between university students who are in a teacher preparation program or who indicate an interest in becoming a teacher (preservice teachers) compared with those who indicate they are not interested in becoming a teacher (nonteaching students) regarding significant adults recalled while the subjects were in grades K-12. The students are compared across groups and within groups regarding demographic variables such as age, ethnicity, and gender as well as the characteristics of the significant adults who are identified. The preferred characteristics of significant teachers are described in some detail. In general, the findings do not support the primary hypothesis that there are significant differences in the frequency and description of significant adults when preservice teachers are compared to nonteaching students. Preservice teachers do select teachers as significant adults, but these findings may have more practical than statistical significance. The results, particularly with regard to gender and ethnicity, do provide the basis for drawing useful conclusions which have implications for teacher recruitment and for future research. (Author)