Development of a Forced-Choice Measure of Typical-Performance Emotional Intelligence

被引:36
作者
Anguiano-Carrasco, Cristina [1 ]
MacCann, Carolyn [2 ]
Geiger, Mattis [3 ]
Seybert, Jacob M. [1 ]
Roberts, Richard D. [4 ]
机构
[1] Educ Testing Serv, Princeton, NJ 08541 USA
[2] Univ Sydney, Sydney, NSW 2006, Australia
[3] Univ Ulm, D-89069 Ulm, Germany
[4] Profess Examinat Serv, New York, NY USA
关键词
item response theory (IRT); emotional intelligence; forced-choice assessment; Thurstone's law of comparative judgment; TRADITIONAL STANDARDS; INCREMENTAL VALIDITY; SHORT-FORM; TRAIT; DISCRIMINANT; ABILITIES; IRT;
D O I
10.1177/0734282914550387
中图分类号
G44 [教育心理学];
学科分类号
0402 ; 040202 ;
摘要
Self-report ratings of emotional intelligence (EI) can be faked in high-stakes situations. Although forced-choice administration can prevent response distortion, it produces ipsative scores when scored conventionally. This study (n = 486) develops an 18-item EI rating scale assessing emotion perception, understanding, and management. We compare validity evidence for: (a) a single-stimulus rating scale; and (b) a forced-choice assessment scored with conventional methods versus item response theory (IRT) methods. The single-stimulus items showed acceptable fit to a three-factor solution, and the forced-choice items showed acceptable fit to the IRT solution. Correlations with criterion variables (ability and self-reported EI, Big Five personality, loneliness, life satisfaction, and GPA) were obtained for 283 participants. Correlations were in the expected direction for the single-stimulus and the IRT-based forced-choice scores. In contrast, the conventionally scored forced-choice test showed the expected correlations for emotion management, but not for emotion perception nor understanding. Results suggest that IRT-based methods for scoring forced-choice assessments produce equivalent validity to single-stimulus rating scales. As such, IRT-based scores on forced-choice assessments may allow EI tests to be used for high-stakes applications, where faking is a concern.
引用
收藏
页码:83 / 97
页数:15
相关论文
共 42 条
[1]  
Allen V, 2014, EDUC PSYCHOL HANDB, P162
[2]   Development of the situational test of emotional understanding - brief (STEU-B) using item response theory [J].
Allen, Veleka D. ;
Weissman, Alexander ;
Hellwig, Susan ;
MacCann, Carolyn ;
Roberts, Richard D. .
PERSONALITY AND INDIVIDUAL DIFFERENCES, 2014, 65 :3-7
[3]   Is there a validity increment for tests of emotional intelligence in explaining the variance of performance criteria? [J].
Amelang, Manfred ;
Steinmayr, Ricarda .
INTELLIGENCE, 2006, 34 (05) :459-468
[4]  
Bar-On R., 2000, BAR ON PARKER, V2000, P363
[5]   Strengths and limitations of ipsative measurement [J].
Baron, H .
JOURNAL OF OCCUPATIONAL AND ORGANIZATIONAL PSYCHOLOGY, 1996, 69 :49-56
[6]   The relationship between ipsatized and normative measures of personality [J].
Bartram, D .
JOURNAL OF OCCUPATIONAL AND ORGANIZATIONAL PSYCHOLOGY, 1996, 69 :25-39
[7]   Convergent, discriminant, and incremental validity of competing measures of emotional intelligence [J].
Brackett, MA ;
Mayer, JD .
PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN, 2003, 29 (09) :1147-1158
[8]   Relating emotional abilities to social functioning: A comparison of self-report and performance measures of emotional intelligence [J].
Brackett, Marc A. ;
Rivers, Susan E. ;
Shiffman, Sara ;
Lerner, Nicole ;
Salovey, Peter .
JOURNAL OF PERSONALITY AND SOCIAL PSYCHOLOGY, 2006, 91 (04) :780-795
[9]  
Brown A., 2010, THESIS U BARCELONA S
[10]   How IRT Can Solve Problems of Ipsative Data in Forced-Choice Questionnaires [J].
Brown, Anna ;
Maydeu-Olivares, Alberto .
PSYCHOLOGICAL METHODS, 2013, 18 (01) :36-52