展开模型对EPQ反应数据的拟合调查

心理与行为研究 ›› 2009, Vol. 7 ›› Issue (4): 300-303.

展开模型对EPQ反应数据的拟合调查

邓稳根¹,戴海琦²,黄海³,戴慧群¹

1 赣南师范学院教育科学学院,赣州 341000;
2 江西师范大学心理学院,南昌 330022;
3 中国地质大学马克思主义学院,武汉 430074

收稿日期:2008-09-22 出版日期:2009-12-20 发布日期:2015-01-21
通讯作者: 戴海琦,男,江西师范大学心理学院教授,博士生导师。Email: dhaiqi@eyou.com。
作者简介:邓稳根,男,赣南师范学院教育科学学院讲师,博士。
基金资助:
本研究得到江西省高校人文社会科学规划基金XL1003项目的资助

INVESTIGATION ON THE FIT OF UNFOLDING IRT MODELS TO RESPONSE DATA OF EPQ

Deng Wengen¹,Dai Haiqi²,Huang Hai³,Dai Huiqun¹

1 School of Education Science, Gannan Teachers′ College, Ganzhou, 341000;
2 School of Psychology, Jiangxi Normal University, Nanchang, 330022;
3 Marxism Institute, China University of Geoscience, 430074

Received:2008-09-22 Online:2009-12-20 Published:2015-01-21

摘要/Abstract

摘要： 调查了拓广等级展开模型（GGUM）对EPQ反应数据的拟合情况。结果发现：（1）E量表最符合GGUM单维性的假设,即得到2个线性主成分,项目在这2个主成分上的负荷略微呈现一个半圆形的模式;（2）约束GGUM的阈限参数在项目间相等而得到的模型A在对EPQ的数据进行分析时是最简洁而有效的;（3）模型A和2PLM相比,前者在E和P上的拟合显著优于后者,在N上,两个模型提供了相似的拟合度。结果表明展开模型比优势模型更适合用于分析EPQ的数据。

关键词: 拓广等级展开模型, 人格, EPQ, 模型-数据拟合

Abstract: The fit of generalized graded unfolding model(GGUM)to the EPQ was investigated. Results suggested that (1)the assumption of unidimensionality was completely fulfilled in the E scale, and two principal components were remained by parallel analysis, and the loading of items to these components formed a fan-like pattern resembling a semicircle,(2)the constrained model(which was recorded as model A)of GGUM was best among four nested models, and(3)both model A and 2PLM provided similarly good fit to the N scale, but when applied to E or P scale, Model A showed better fit than 2PLM. Therefore, unfolding models rather than dominant counterparts are more useful in personality fields.

Key words: generalized graded unfolding model, personality, EPQ, model-data fit

中图分类号:

G449.1

邓稳根,戴海琦,黄海,戴慧群. 展开模型对EPQ反应数据的拟合调查[J]. 心理与行为研究, 2009, 7(4): 300-303.

Deng Wengen,Dai Haiqi,Huang Hai,Dai Huiqun. INVESTIGATION ON THE FIT OF UNFOLDING IRT MODELS TO RESPONSE DATA OF EPQ[J]. Studies of Psychology and Behavior, 2009, 7(4): 300-303.

参考文献

[1] Andrich D, Luo G. A hyperbolic cosine latent trait model for unfolding dichotomous single-stimulus responses. Applied Psychological Measurement, 1993, 17(3): 253-276
[2] Andrich D. A hyperbolic cosine latent trait model for unfolding polytomous responses: Reconciling Thurstone and Likert methodologies. British Journal of Mathematical and Statistical Psychology, 1996, 49(2): 347-365
[3] Roberts J S, Laughlin J E. A unidimensional item response theory model for unfolding responses from a graded disagree-agree response scale. Applied Psychological Measurement, 1996, 20(3): 231-255
[4] Roberts J S, Donoghue J R, Laughlin J E. A general item response theory model for unfolding unidimensional polytomous responses. Applied Psychological Measurement, 2000, 24(1): 3-32
[5] Chernyshenko O S, Stark S, Chan K Y, et al. Fitting item response theory models to two personality inventories: Issues and insights. Multivariate Behavioral Research, 2001, 36(4): 523-562
[6] Stark S, Chernyshenko O S, Drasgow F, et al. Examining assumptions about item responding in personality assessment: Should ideal point methods be considered for scale development and scoring? Journal of Applied Psychology, 2006, 91(1): 25-39
[7] Hoijtink H. A latent trait model for dichotomous choice data. Psychometrika, 1990, 55(4): 641-656
[8] Cliff N, Collins L M, Zatkin J, et al. An ordinal scaling method for questionnaire and other ordinal data. Applied Psychological Measurement, 1988, 12(1): 83-97
[9] van Schuur W H. Nonparametric unidimensional unfolding for multicategory data//Freeman J R. Political analysis: Vol 4. Ann Arbor MI: University of Michigan Press. 1993:41-74
[10] 郭庆科, 周晶. Likert 量表分析中不同IRT 模型的有效性. 心理学探新, 2004, 24(3): 67-70
[11] O′Connor B P. SPSS and SAS programs for determining the number of components using parallel analysis and Velicer′s MAP test. Behavior Research Methods, Instrumentation, and Computers, 2000, 32(3): 396-402
[12] Davision M L. On a metric, unidimensional unfolding attitudinal and developmental data. Psychometrika, 1977, 42(4): 523-548
[13] Nandakumar R, Hotchkiss L, Roberts J S. Attitudinal data: dimensionality and start value for estimating item parameters. The annual meeting of the American Educational Research Association, New Orleans, LA, 2002
[14] DeMars C E. Type I error rates for the generalized graded unfolding model fit indices. Applied Psychological Measurement, 2004, 28(1): 48-71
[15] Drasgow F, Levine M V, Tsien S, et al. Fitting polytomous item response theory models to multiplechoice tests. Applied Psychological Measurement, 1995, 19(2): 143-165