CAJ | 학술논문

对2288个留守儿童样本用儿童长处与困难问卷（SDQ）进行施测，并对他们的情绪和行为问题特征进行探索性的潜在类别分析，发现留守儿童的情绪和行为问题存在明显的分组特征。统计指标支持了3个潜类别的模型，根据3个潜在类别在问卷各条目上的条件概率特征分别定义为：“适应困难组”，“行为冲动组”和“良好适应组”，3个潜在类别所占全体样本的比例分别为32%、41%和27%。进一步的分析发现：相比较良好适应组而言，适应困难组和行为冲动组有着显著的性别和年级水平效应，适应困难组和行为冲动组的男生所占比例更大；同时，这两个组的小学生所占的比例也更大。
대2288개류수인동양본용인동장처여곤난문권（SDQ）진행시측，병대타문적정서화행위문제특정진행탐색성적잠재유별분석，발현류수인동적정서화행위문제존재명현적분조특정。통계지표지지료3개잠유별적모형，근거3개잠재유별재문권각조목상적조건개솔특정분별정의위：“괄응곤난조”，“행위충동조”화“량호괄응조”，3개잠재유별소점전체양본적비례분별위32%、41%화27%。진일보적분석발현：상비교량호괄응조이언，괄응곤난조화행위충동조유착현저적성별화년급수평효응，괄응곤난조화행위충동조적남생소점비례경대；동시，저량개조적소학생소점적비례야경대。
It is reported that there are over 60 million left-behind children in rural China. As a lot of reports showed, most research focused on the mental problems of left-behind children, and the main method adopted by the researchers was factor-analysis . The analysis object of factor analysis are indexes or variables, and it analyzes samples according relevant indexes, assuming that the research variables are all continuous. But not every variable is continuous, there may be qualitative differences as well as quantitative differences between individuals＇ psychological characteristics. From person-centered perspective, this study aims to explore the characteristics of left-behind children＇s emotional and behavioral problems, and to figure out if there are qualitative differences between left-behind children groups. On the other hand, due to the urgent demand of intervention and compensation for left-behind children, this study aims to analyze group differences heterogeneity of left-behind children’s emotional and behavioral symptoms, to find out the intervention group which need help most and to carry out more effective intervention. The Children＇s Strengths and Difficulties Questionnaire （SDQ） were administered to 2288 children and exploring laten class analysis was adopted. Obvious grouping features appeared and the model of three latent classes were supported. The three classes were defined as “maladjustment group” “behavior impulse group” and “basically adapted group” respectively , their proportion of all the samples were 32%, 41% and 27%. To explore the impact of gender and grade level on latent classes of left-behind children’s emotional and behavioral problems characteristics , we built a model of multivariate logistic regression, which took the three latent classes extracted in the first step as the dependent variable, took gender and different grade level as independent variables. Further analysis find that compared with basically adapted group, maladjustment group and behavior impulse group have significant gender and grade effects, male students have more big proportion in the two groups,and the primary school children’s proportion in the two groups are higher. This study is the practical implications research in the field of the Left-Behind Children＇s emotional and the behavioral problems. This study has important practical significance, teachers and psychologists can give more targeted education and counseling according to children＇s different problem characteristics.