ISSN 1671-3710
CN 11-4766/R
主办:中国科学院心理研究所
出版:科学出版社

心理科学进展 ›› 2010, Vol. 18 ›› Issue (8): 1314-1320.

• • 上一篇    下一篇

心理学研究数据类型与统计方法——谈函数型数据分析的引入

焦璨;熊敏平;张敏强

  

  1. (1华南师范大学心理应用研究中心, 广州510631) (2 深圳大学师范学院心理学系, 深圳 518000)
  • 收稿日期:2009-12-28 修回日期:1900-01-01 出版日期:2010-08-15 发布日期:2010-08-15
  • 通讯作者: 张敏强

Different Types of Data and Statistical Methods in Psychology:Introduction of Functional Data Analysis

JIAO Can;XIONG Min-Ping;ZHANG Min-Qiang   

  1. (1 Psychological Application Research Center, South China Normal University, Guangzhou 510631, China)
    (2 Department of Psychology, Shenzhen University, Shenzhen 518000, China)
  • Received:2009-12-28 Revised:1900-01-01 Online:2010-08-15 Published:2010-08-15
  • Contact: ZHANG Min-Qiang

摘要: 心理学研究数据大致分为截面数据、时间序列数据和面板数据, 三种数据类型的分析方法及使用前提因数据属性不同而有所不同。心理学截面数据的统计方法过于依赖模型的线性结构和假设条件等, 在处理心理学面板数据中难以充分发挥统计方法的功用。函数型数据分析方法主要适用于面板数据处理, 特别适宜ERP、fMRI、发展心理等心理实验中存在时间序列的面板数据的统计分析, 为心理学研究提供了有力的新工具。

关键词: 心理研究方法, 截面数据, 时间序列数据, 面板数据, 函数型数据分析

Abstract: Data from psychological research can be roughly divided into three different types, they are cross-sectional data, time-series data and panel data. Methods for analyzing such data should be based on this research and the differentiation among them are briefly discussed. Methods for cross-sectional data in psychology are over depended on the linear structure and certain assumptions. Consequently, it ’s difficult to obtain benefits from those statistical methods when working with panel data in psychology. The lately developed functional data analysis method , which is mainly applied to panel data, has been mainly discussed in this research. According to the literature review, this method has been judged as a powerful new tool for psychological research, especially for time-series panel data from ERP、fMRI and developmental psychology experiments.

Key words: psychology research methods, cross-sectional data, time-series data, functional data analysis