关键词: 儿童, 人格, 群组序列追踪设计, 潜变量增长曲线模型, 多层线性模型

Abstract: At present, researchers in western countries explored the construction and development of children personality maturely, but studies’ foundation in our country is relatively weak. Many studies used the western ?paradigms as references, or used the revised personality questionnaires which were applied in the western culture to assess children in our country directly. Researchers consider that personality is the psychological properties with cultural traits. If the study is conducted to explore the construction and development of children’s personality systematically, we should carry out it in the context of the entire Chinese culture, which can ensure the study’s conclusion reflecting the characteristics of Chinese children’s personality exactly. Above all, this paper contributes to the local research of construction and development of children’s personality in our country, such as the development characteristics of age and gender. First, we carried out qualitative research combining the teachers’ free description and selected the proper personality traits adjectives which can describe the children from the Wang Dengfeng’ adults personality traits adjectives. And integrated the two ways and referred to the related literatures for the qualitative research. A precise and completed descriptive personality construction of children was obtained. Second, on the basis of qualitative research, we made the teachers’ rating questionnaires, which was used to investigate the children’s personality in the quantitative ?paradigm, and established the rating personality construction of Chinese children. Combining qualitative and quantitative research, we established personality construction of preschool children which was made up of five dimensions, including intellectual characteristics, conscientiousness, pro-sociality; emotional stability, extraversion. The personality questionnaire was named as “preschool children personality rated by teachers” with fine reliability and validity which is made independently as assessing tools. Third, we adopted the longitudinal study with the cohort sequential design to explore the age and gender characteristic of preschool children’s personality development. Recruited 608 preschool children and classified into three cohorts which were named 3-3.5, 3.5-4,4-4.5 years old as participants. Using new personality questionnaire, we measured once every six months, measured four times in all. Next, the BIC in multiple group analysis of latent growth curve model (LGM) was applied to decide whether the three cohorts have all the same developmental trend. In other words, a single coherent trajectory across the observed age range is appropriate. If a single coherent trajectory exists, we will apply hierarchical linear modeling(HLM) to master and deal with data to further explore the age and gender characteristics of children personality development. The result showed that: (1) As for the age difference, the five dimensions of preschool children’s personality developed rapidly between 3 and 6 years old. And the developmental trend was quadratic curve. The critical period is 3 to 4 years old, and personality continued to develop between 4 and 5 years old, but the speed become slow and stable from 5 to 6 years old. (2) Considering the gender differences of children personality, the outcomes suggested that scores of conscientiousness and pro-sociality dimensions of preschool girls were significantly higher than preschool boys at the age of 3 years, but the developing speed had no difference from 3 to 6 years. In other words, the development level of girls’ conscientiousness and pro-sociality was always higher than boys’ at the age of 3 years. In conclusion, the five dimensions of preschool children’s personality developed with the quadratic curve, and became stable when children were 5-6 years. The two dimensions which are named conscientiousness and pro-sociality have dramatically gender differences between boys and girls at the age of 3 years old.

Key words: children, personality, cohort sequential longitudinal design, latent growth curve model, hierarchical linear modeling