Investigating The Reliability And Validity Of The Cognitive Style Based On The Data Of 363 Chinese Soldiers

Cognitive style,which consists of the wholist-analytic(WA)dimension and the verbal-imagery(VI)dimension,has been defined as an individual's particular and preferred way of organizing and processing information.Despite of a few studies have considered the reliability and the validity of the measurements of the two-dimensional cognitive style model,little has been done to explore their stability and validity on Chinese population.Hereby,the present study intended to bridge the gap by providing the reliability and validity data of two performance-based cognitive style tests(i.e.Verbal Imagery Cognitive Styles Test&Extended Cognitive Style Analysis-Wholistic Analytic Test,V1CS&E-CSA-WA).The test-retest and split-half reliability of VICS and E-CSA-WA was collected accordingly.Majority of the 363 participants(all male soldiers,age range 18?36 years)completed the tests twice,with an interval of 7-99 days.Results revealed that the two tests generally have the acceptable test-retest reliability,r(WA)=0.368(N=305),r(VI)=0.491(N=302).Meanwhile,the length of the test-retest interval had little impact on the reliability outcomes of CSA-WA(interval=7-16 days,r(WA)=0.367,n=176;interval=27?99 days,r(WA)= 0.370,n=129),however prolonged interval would slightly lower the reliability outcomes of VICS(interval=7-16 days,r(VI)=0.524,n=176;interval=27-99 days,r(?)= 0.444,n=129).The split-half reliability was calculated by comparing the performances of the first half and second half of each task,split-half r(WA)=0.111-0.393,split-half r(?)=0.527-0.695.Furthermore,cognitive style ratios that were calculated based on the median reaction times(r(WA)=0.292?0.393,r(?)=0.621-0.695)were significantly more reliable than those calculated based on the mean reaction times(r(WA)=0.111-0.237,r(?)=0.527-0.551).This study also investigated the relationship between the two dimensions of cognitive styls,as well as their relationship with other psychological traits,inluding intelligence(measured by odd items chosen from the Raven Advanced Progressive Matrices Test),personality(measured by Neuroticism Extraversion Openness Personality Inventory,NEO-PI),and emotional states(measured by Self-Rating Depression Scale&Self-Rating Anxiety Scale,SDS and SAS)in order to examine the tests' construct validity.Results found that the average of correlations between the VI and WA ratios were low(r = 0.048,range r =-0.034 to 0.066)and the correlations between the cognitive style ratios and intelligence,personality,depressive and anxious emotional states were all low(range r =-0.127 to 0.089).In conclusion,this study suggested that the stability and internal consistency of VICS and E-CSA-WA test were consistent with the results from previous research in western countries.Both tests are relatively reliable tools for measuring cognitive styles.Using median reaction times instead of using mean reaction times can make the calculated cognitive style ratios more reliable.Moreover,the WA dimension and the VI dimension of cognitive style are psychometrically independent from each other and from other psychological constructs including intelligence,personality and emotional states.The tests are suitable tools for measuring Chinese people's WA and VI cognitive styles.