父母是和学前儿童联系极为紧密的社交对象，亲子冲突是儿童和父母消极互动模式的体现，这种不良互动往往受到消极家庭氛围的影响。在家庭中，父母抚养压力作为父母的消极心理感受往往表现在家庭生活中，这种压力可能营造了不良的氛围，会促使儿童与父母产生更多冲突。随着儿童在幼儿园的成长，父母抚养压力和亲子冲突都可能会有所发展变化。同时这种压力带来的不良氛围还可能影响儿童社会性发展，儿童社会技能就是会受到父母抚养压力影响的社会性发展变量之一(Anthonya et al., 2005; Neece & Baker, 2008)。社会技能体现着儿童处理人际关系的能力(Bornstein, Hahn, & Haynes, 2010; Raver & Zigler, 1997)，因此抚养压力还可能通过儿童社会技能对亲子冲突进一步造成影响。在这一框架下，本研究选取412名学前儿童为被试，对其在小班、中班、大班时的三年追踪测查数据进行分析。抚养压力、亲子冲突、儿童社会技能均由儿童的主要抚养人报告。本研究探究学前儿童父母抚养压力和亲子冲突的发展，具体考察了在三年期间父母抚养压力和亲子冲突的发展趋势以及不同群体的发展轨迹差异。并且本研究对抚养压力和亲子冲突二者间的因果关系进行检验，在纵向因果关系的基础上进行每一年的横向预测作用分析。最后，本研究检验了儿童社会技能在父母抚养压力影响亲子冲突过程中的中介作用。本研究发现：

1.学前儿童从小班到大班的三年期间，父母的抚养压力和亲子冲突都表现为线性下降的发展趋势。并且抚养压力和亲子冲突的初始水平、下降速度均存在显著的个体差异。

2.抚养压力和亲子冲突发展的初始水平存在群体之间的显著差异。高学历父母其抚养压力和亲子冲突初始水平都显著低于低学历父母，高家庭收入父母其抚养压力和亲子冲突初始水平都显著低于低家庭收入父母。父亲的抚养压力初始水平显著低于母亲。但对于抚养压力和亲子冲突的发展速度未发现以上群体之间的差异。

3.父母抚养压力对亲子冲突在纵向和横向层面具有预测作用。纵向层面，小班时的父母抚养压力对中班时的亲子冲突单向预测作用显著，中班时的父母抚养压力对大班时的亲子冲突单向预测作用显著；横向层面，小班、中班、大班的每个时间点上，父母抚养压力均能够显著预测同一时间点的亲子冲突；此外，抚养压力的发展初始水平显著预测亲子冲突发展初始水平，抚养压力的发展速度显著预测亲子冲突发展速度。

4.儿童社会技能的中介作用显著。具体表现为儿童小班时的父母抚养压力通过中班时的儿童社会技能进一步影响大班时的亲子冲突的纵向中介作用；以及小班时的儿童社会技能初始水平在父母抚养压力初始水平对亲子冲突初始水平的影响过程中的横向中介作用。

Parents are social partners who are closely connected to preschool children. Parent-child conflict is the negative interaction style between children and parents, which is often affected by the negative family atmosphere. In the family, as negative psychological feelings of parents, parenting stress is often expressed in the daily life. This stress may create a bad atmosphere, which may create much more conflicts between children and parents. As children grow up in preschool, parenting stress and parent-child conflicts may both develop and change. Meanwhile, the bad atmosphere caused by parenting stress may also affect children's social development. Social skills of children are one of the social development variables which are affected by parenting stress(Anthonya et al., 2005; Neece & Baker, 2008).Social skills reflect children's ability of handling interpersonal relationship(Bornstein, Hahn, & Haynes, 2010; Raver & Zigler, 1997).Therefore, parenting stress may further affect parent-child conflict through social skills of children. Under this framework, the present research selected 412 preschool children as participants and analyzed the three-year longitudinal data of each grade in preschool. Parenting stress, parent-child conflict, and children's social skills were all reported by children’s primary caregivers. The present research explored the development of parenting stress and parent-child conflict, and specifically investigated the development trends of parent-child stress and parent-child conflict during the three-year period as well as group differences of development trajectories. Besides, the causal relationship between parenting stress and parent-child conflict was tested. The widthwise forecasting effect in each year was tested on the basis of longitudinal causal relationship. Finally, the present research examined the mediating effect of children’s social skills in the process of parenting stress affecting parent-child conflict. Findings are as follows:

1. During three years in preschool from grade 1 to grade 3, parenting stress and parent-child conflict both showed linear downtrends. In addition, there were significant individual differences in the initial levels and decline rates of parenting stress and parent-child conflict.

2. The initial levels of parenting stress and parent-child conflict significantly differed between different groups. The initial levels of parenting stress and parent-child conflict of parents with high education were significantly lower than that of parents with low education. The initial levels of parenting stress and parent-child conflict of parents with high household income were significantly lower than that of parents with low household income. The initial level of fathers’ parenting stress was significantly lower than that of mothers. However, no such group difference was found in the development rate of parenting stress and parent-child conflict.

3. Parenting stress can predict parent-child conflict at the longitudinal level and the cross-sectional level. At the longitudinal level, parenting stress at grade 1 unidirectionally predicted parent-child conflict at grade 2 in preschool; Parenting stress at grade 2 unidirectionally predicted parent-child conflict at grade 3 in preschool; At the cross-sectional level, parenting stress always significantly predicted parent-child conflict at the same time of each grade in preschool, Moreover, the initial level of parenting stress significantly predicted the initial level of parent-child conflict. The development rate of parenting stress significantly predicted the development rate of parent-child conflict.

4. The mediating effect of children's social skills is significant. Specifically, the parenting stress at grade 1 in preschool longitudinally affected the parent-child conflict at grade 3 through children’s social skills at grade 2. And the initial level of children’s social skills widthwise mediated the effect of the initial level of parenting stress on the initial level of parent-child conflict at grade 1 in preschool.