Write your message
Volume 20, Issue 97 (3-2021)                   علوم روانشناختی 2021, 20(97): 23-38 | Back to browse issues page

XML Persian Abstract Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

soltani N, vahedi S, bayrami M, zarghami M H. Use of network data analysis to improve algorithm of computerized adaptive tests. علوم روانشناختی. 2021; 20 (97) :23-38
URL: http://psychologicalscience.ir/article-1-942-en.html
Professor,PhD in educational psychology, Department of educational Sciences, University of Tabriz, Tabriz, Iran
Abstract:   (342 Views)
Background: Network methodology makes it possible to study several goals at the same time and has capabilities leads to providing appropriate analytical tools. There are studies on the use of network data analysis in the field of educational assessment but there is less research that has practically constructed an adaptive test based on network theory or presented its algorithm. Is it possible to use network science to build adaptive tests? Aims: This study was performed with the aim of improving the adaptive test algorithm based on network data analysis on language learners of language teaching centers and institutions in Tehran. Method: This research was descriptive and its main focus was on the analysis of test questions using network data analysis technique. The statistical population included all students of the top ten language teaching institutes in Tehran in 2019 who had the ability to read English at a minimum level. Among them, 1556 people were selected by convenience sampling method. The tool used was the 140-item English Vocabulary Size Test (VST) (Nation & Waring, 1997). The analysis of test questions was performed through classical test theory, item response theory and network analysis based on Fruchterman-Ringold algorithm. Results: The parameters of the test questions were calculated based on classical test theory and item respons theory. The relationship map between the questions was drawn based on Fruchterman-Ringold algorithm and statistical test of partial correlation with alpha modulation (p< 0/01), network centrality parameters and adaptive test algorithm were extracted. Conclusions: Utilizing the benefits of network data analysis techniques such as results visibility, providing simple and comprehensive reports and considering the importance of the questions in the structure of the communication network between the questions have led to the improvement of adaptive testing methodes.
Full-Text [PDF 842 kb]   (154 Downloads)    
Type of Study: Research | Subject: Special
Received: 2020/09/3 | Accepted: 2021/03/30 | Published: 2021/02/20

Add your comments about this article : Your username or Email:
CAPTCHA

Send email to the article author


Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

© 2021 CC BY-NC 4.0 | Journal of psychologicalscience

Designed & Developed by : Yektaweb