A two-year longitudinal investigation was conducted. The purposes of this study were to investigate the relative importance of affiliative tendency, personal trait desirability, physical attractiveness, attitude similarity, and intelligence in determination of sociometric status at two different times in one academic year, and to determine the stability of sociometric status when the students remain in the same class over the academic year and when they enter a relatively new class in the following year. The subjects were 160 male and 353 female 7th-graders from 12 junior high schools in Taipei, Taiwan. In the second year, 31 males and 64 females in 10 schools were followed-up. During the 3rd through 4th week of the first year, the subjects were given a series of instruments that included the Revised Interpersonal Judgment Scale, Personal Facts Scale, Personal Trait Rating Scale, the Attitude Questionnaire for Junior High School Students, and the Revised AGCT. The same procedure was repeated at the end of the academic year. In the second year, the Revised Interpersonal Judgment Scale was administered during the 3rd through 4th week, and again at the end of the academic year. Multiple regression methods were the principle forms of analysis. The results indicate that, at the early stage of acquaintance, personal trait desirability and subjective physical attractiveness are the most important determinants of sociometric status for both males and females. At the later stage, those two variables continue to be the most important determinants for males, while for females, personal trait desirability, subjective physical attractiveness, intelligence, and affiliative tendency are the most important ones. Scholastic achievement increases prediction at the later stage for both males and females. If the membership of a class remains constant, the correlations between the socio metric status measured at the early stage and the late stage range from .42 to .62. If the membership changes so that only a small proportion of the original sample remains in the new group, the correlations of sociometric status at the corresponding stages of the two academic years range from .55 to .65.