社会认知复杂性的量化指标及其关系

在Kelly个人构念理论的基础上,Bieri提出用构念的分化程度衡量个体社会认知复杂性的思路,并具体用从不同构念上评价每个角色时评分的匹配个数作为复杂性的最化指标。此后,研究者从不同分析思路提出了一系列新的复杂性指标,如根据对构念的相关分析所得组内相关、RMS、SMC等,基于因素分析所得第一因素解释率、主要因素个数及其解释率等,冲突分析所得整体冲突百分比,此外还有MPP、LPP、ASO等各种Fiedler分数。本研究以402名初中一二年级青少年为被试。采用库格测验收集数据,以Gridstat4．0、Excel、SPSSl0．0等软件统计数据,分析了上述各种复杂性指标的内在关联,为同类研究选择使用这些指标提供了依据。

Indexes of Social Cognitive Complexity and Their Relationship

Based on the personal construct theory of George Kelly,Bieri(1955) introduced the concept of "social cognitive complexity" and defined it in terms of the differentiation of(number of constructs in) an individual's construct.After this,a lot of new indexes were developed,which included(1) intra-class correlation,RMS,SMC based on correlation analysis,and(2) factor number,the explanatory power or eigenvalue of the first Repgrid factor and the percentage of variance accounted for by main factors cumulatively or averagely based on factor analysis,and(3) overall percentage of conflict in Repgrid based on conflict anlysis,and(4) Fiedler scores such as MPP,LPP and ASO.By means of the repertory grid instrument(Repgrid),a sample of 402 seventh and eighth graders was investigated,and the correlation of the above social cognitive complexity indexes were examined by statistics programs such as Gridstat4.0,Excel,SPSS10.0.It was found that RMS,intra-class correlation,the eigenvalue of the first Repgrid factor,overall percentage of conflict in Repgrid,and the percentage of variance accounted for by main factors averagely were better indexes of social cognitive complexity,and the alternative ones included factor number, MPP,ASO,etc.However,LPP and SMC were not that good.The results of the present study offered new evidences for the selection of complexity indexes.