國立虎尾科技大學 |

Using the Rasch Model in a Computer Adaptive Testing Application to Enhance the Measurement Quality of Emotional Intelligence.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Using the Rasch Model in a Computer Adaptive Testing Application to Enhance the Measurement Quality of Emotional Intelligence./
作者:	Hobson, Ernest Guy.
面頁冊數:	1 online resource (224 pages)
附註:	Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
Contained By:	Dissertations Abstracts International83-02B.
標題:	Research. -
電子資源:	click for full text (PQDT)
ISBN:	9798744425616

Using the Rasch Model in a Computer Adaptive Testing Application to Enhance the Measurement Quality of Emotional Intelligence.
Hobson, Ernest Guy.

Using the Rasch Model in a Computer Adaptive Testing Application to Enhance the Measurement Quality of Emotional Intelligence. - 1 online resource (224 pages)

Source: Dissertations Abstracts International, Volume: 83-02, Section: B.

Thesis (Ph.D.)--University of Johannesburg (South Africa), 2015.

Includes bibliographical references

Background: The use of objective measurement via Rasch models in self-report scales has increased in the recent past, which in turn has enabled the development of computer adaptive tests (CATs). Despite significant advances in both objective measurement and computer adaptive applications few studies have ventured into the personality domain and the emergent field of emotional intelligence. The development of CATs of personality attributes holds advantages such as the reduction in items used for the assessment, reduction of respondent fatigue, and greater cooperation from respondents in the assessment. Research purpose: The aim of this study was to develop a computer adaptive test of the trait Self-control sub-scale of a trait-based emotional intelligence inventory (Trait Emotional Intelligence Questionnaire: TEIQue). Secondary objectives were to examine the functioning of the CAT by (a) comparing the CAT with a static version, and (b) to establish a practical approach to developing a computer adaptive solution to existing static fixed format self-report inventories. Research design: Participants were 681 working South African adults who participated in a local validation research project of the TEIQue. All respondents completed an informed consent form indicating willingness to participate in the research process. The self-control scale consists of 31 items. Each item employs a 7-point Likert-type response format. The data analysis entailed three steps. Step 1 focussed on establishing a benchmark based on the static measurement of trait based self-control. The analysis entailed calibrating and composing a core self-control item bank by means of a Rasch rating scale analysis. The Rasch analysis focussed on how well the observed data fitted the measurement model and identifying items that meet the criteria for inclusion into the item bank. The data were analysed for individual item fit, unidimensionality, local independence, and differential item functioning across gender. Items that did not meet the criteria for good fit were excluded from the item bank. The rating scale analysis yielded 16 items that met the requirements of the Rasch model. These items constituted the static item bank. Step 2 involved obtaining person measures and standard errors for the static item bank. Step 3 involved a post hoc CAT simulation of the responses of the 681 persons to the 16 items, where different item selection criteria and stopping rules were applied to obtain CAT based person measures and standard errors. All simulations were carried out with the Firestar software. The aim of this step was to evaluate the correspondence between static person measures (i.e. measures obtained with the full item bank) and adaptive person measures (i.e. measures obtained with the simulated CAT). Main findings: Overall, high correlations were found between person measures of the static inventory and the CAT version. With a 13 out of 16 item strategy a very high correlation (r = .97) was obtained. The 7 out of 16 item strategy yielded a satisfactory, but somewhat weaker result (r = .91). Results showed that on average about ten items were required to obtain a standard error of person measures < .40. Different initial item starting and subsequent item selection strategies yielded largely similar results with no one single strategy clearly appearing to be the best strategy.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798744425616Subjects--Topical Terms:

558735
Research.
Index Terms--Genre/Form:

554714
Electronic books.

Using the Rasch Model in a Computer Adaptive Testing Application to Enhance the Measurement Quality of Emotional Intelligence.
LDR:04721ntm a22003497 4500 001 1150766
005 20241029122310.5
006 m o d
007 cr bn ---uuuuu
008 250605s2015 xx obm 000 0 eng d
020 $a 9798744425616
035 $a (MiAaPQ)AAI28332601
035 $a (MiAaPQ)Johannesburg13971
035 $a AAI28332601
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Hobson, Ernest Guy. $3 1477339
245 1 0 $a Using the Rasch Model in a Computer Adaptive Testing Application to Enhance the Measurement Quality of Emotional Intelligence.
264 0 $c 2015
300 $a 1 online resource (224 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
500 $a Advisor: De Bruin, Gideon P.
502 $a Thesis (Ph.D.)--University of Johannesburg (South Africa), 2015.
504 $a Includes bibliographical references
520 $a Background: The use of objective measurement via Rasch models in self-report scales has increased in the recent past, which in turn has enabled the development of computer adaptive tests (CATs). Despite significant advances in both objective measurement and computer adaptive applications few studies have ventured into the personality domain and the emergent field of emotional intelligence. The development of CATs of personality attributes holds advantages such as the reduction in items used for the assessment, reduction of respondent fatigue, and greater cooperation from respondents in the assessment. Research purpose: The aim of this study was to develop a computer adaptive test of the trait Self-control sub-scale of a trait-based emotional intelligence inventory (Trait Emotional Intelligence Questionnaire: TEIQue). Secondary objectives were to examine the functioning of the CAT by (a) comparing the CAT with a static version, and (b) to establish a practical approach to developing a computer adaptive solution to existing static fixed format self-report inventories. Research design: Participants were 681 working South African adults who participated in a local validation research project of the TEIQue. All respondents completed an informed consent form indicating willingness to participate in the research process. The self-control scale consists of 31 items. Each item employs a 7-point Likert-type response format. The data analysis entailed three steps. Step 1 focussed on establishing a benchmark based on the static measurement of trait based self-control. The analysis entailed calibrating and composing a core self-control item bank by means of a Rasch rating scale analysis. The Rasch analysis focussed on how well the observed data fitted the measurement model and identifying items that meet the criteria for inclusion into the item bank. The data were analysed for individual item fit, unidimensionality, local independence, and differential item functioning across gender. Items that did not meet the criteria for good fit were excluded from the item bank. The rating scale analysis yielded 16 items that met the requirements of the Rasch model. These items constituted the static item bank. Step 2 involved obtaining person measures and standard errors for the static item bank. Step 3 involved a post hoc CAT simulation of the responses of the 681 persons to the 16 items, where different item selection criteria and stopping rules were applied to obtain CAT based person measures and standard errors. All simulations were carried out with the Firestar software. The aim of this step was to evaluate the correspondence between static person measures (i.e. measures obtained with the full item bank) and adaptive person measures (i.e. measures obtained with the simulated CAT). Main findings: Overall, high correlations were found between person measures of the static inventory and the CAT version. With a 13 out of 16 item strategy a very high correlation (r = .97) was obtained. The 7 out of 16 item strategy yielded a satisfactory, but somewhat weaker result (r = .91). Results showed that on average about ten items were required to obtain a standard error of person measures < .40. Different initial item starting and subsequent item selection strategies yielded largely similar results with no one single strategy clearly appearing to be the best strategy.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Research. $3 558735
650 4 $a Cognitive psychology. $3 556029
650 4 $a Psychology. $3 555998
650 4 $a Emotional intelligence. $3 558148
650 4 $a Personality. $3 556406
650 4 $a Quantitative psychology. $3 1182802
650 4 $a Cognitive ability. $3 1372497
650 4 $a Self control. $3 1477340
650 4 $a Principal components analysis. $3 639703
650 4 $a Information processing. $3 1473673
655 7 $a Electronic books. $2 local $3 554714
690 $a 0633
690 $a 0621
690 $a 0632
710 2 $a University of Johannesburg (South Africa). $3 1372785
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Dissertations Abstracts International $g 83-02B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28332601 $z click for full text (PQDT)