A Real Data Simulation Study of Computerized Adaptive Testing of Chinese Soldier Personality Questionnaire

A real data simulation study of computerized adaptive testing (CAT) of Chinese soldier personality questionnaire (CSPQ) was conducted to find out the feasibility of using count-down method to realize the CAT of personality questionnaire. The data was from 100,125 normal Chinese young people. All participants were male, and the mean age of the individuals was 18.3 (range 16-21, SD = 2.5). The data included two parts, one part was collected in 2005 including 47,550 subjects, the other was collected in 2006 including 52,575 subjects. The first part was used to compute the frequencies of item endorsement and the scalability coefficient H. According to the results three item presentation orders were determined. And the second part was used to compare these presentation orders. Results showed that the original order had the worst efficiency, the other two orders had similar results. But all orders can bring great improvement to efficiencies. From this study we can see count-down method can be used in the CAT of CSPQ. At last the test was divided into two stages. Items, which belong to at least two dimensions, were presented first, then the rest items were presented according to the third order. And under the two-stage testing, new method to realize CAT of CSPQ was put forward. But this is just a simulation study, the item savings and validity of the CAT of CSPQ have to be investigated in the further real testing.

[1]  Fritz Drasgow,et al.  Examining assumptions about item responding in personality assessment: should ideal point methods be considered for scale development and scoring? , 2006, The Journal of applied psychology.

[2]  Y. Ben-Porath,et al.  Comparability and validity of computerized adaptive testing with the MMPI-2. , 1995, Journal of personality assessment.

[3]  Y. Ben-Porath,et al.  Comparability of computerized adaptive and conventional testing with the MMPI-2. , 1991, Journal of personality assessment.

[4]  Jeffrey A Douglas,et al.  Computerized adaptive testing under nonparametric IRT models , 2006 .

[5]  D. D. Bickerstaff,et al.  Computerized adaptive testing , 2015 .

[6]  S Stark,et al.  Fitting Item Response Theory Models to Two Personality Inventories: Issues and Insights , 2001, Multivariate behavioral research.

[7]  C. G. Watson,et al.  Comparability of computer- and booklet-administered Minnesota Multiphasic Personality Inventories among primarily chemically dependent patients. , 1990 .

[8]  A. Dowson,et al.  Applications of computerized adaptive testing (CAT) to the assessment of headache impact , 2003, Quality of Life Research.

[9]  R. Meijer,et al.  Analyzing psychopathology items: a case for nonparametric item response theory modeling. , 2004, Psychological methods.

[10]  Steven P Reise,et al.  A Discussion of Modern Versus Traditional Psychometrics As Applied to Personality Assessment Scales , 2003, Journal of personality assessment.

[11]  S. van Buuren,et al.  Unidimensionality and reliability under Mokken scaling of the Dutch language version of the SF-36 , 2004, Quality of Life Research.

[12]  P. J. Ferrando The accuracy of the E, N and P trait estimates: an empirical study using the EPQ-R , 2003 .