File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Development and validation of a computational thinking test for lower primary school students

TitleDevelopment and validation of a computational thinking test for lower primary school students
Authors
KeywordsAssessment
Computational thinking
Early childhood education
Evidence-centred design
Primary school
Issue Date18-May-2023
PublisherSpringer
Citation
Educational Technology Research and Development, 2023, v. 71, n. 4, p. 1595-1630 How to Cite?
Abstract

Computational thinking (CT) has permeated primary and early childhood education in recent years. Despite the extensive effort in CT learning initiatives, few age-appropriate assessment tools targeting young children have been developed. In this study, we proposed Computational Thinking Test for Lower Primary (CTtLP), which was designed for lower primary school students (aged 6-10). Based on the evidence-centred design approach, a set of constructed-response items that are independent of programming platforms was developed. To validate the test, content validation was first performed via expert review and cognitive interviews, and refinements were made based on the comments. Then, a large-scale field test was administered with a sample of 1st-3rd graders (N = 1225), and the data was used for psychometric analysis based on both classical test theory (CTT) and item response theory (IRT). The CTT results provided robust criterion validity, internal consistency, and test-retest reliability values. Regarding IRT results, a three-parameter logistic model was selected according to the item fit indices, based on which fair item parameters and test information reliability were generated. Overall, the test items and the whole scale showed proper fit, suggesting that CTtLP was a suitable test for the target group. Analyses of the test performance were then put forward. Results reported that students' performance improved with grade level, and no gender difference was detected. Based on the test responses, we also identified children's challenges in understanding CT constructs, indicating that students tended to have difficulty in understanding loop control and executing multiple directions. The study provides a rigorously validated diagnostic test for measuring CT acquisition in lower primary school students and demonstrates a replicable design and validation process for future assessment practices, and findings on the difficulties children faced in CT conceptual understanding could shed light on CT primary and early childhood education.


Persistent Identifierhttp://hdl.handle.net/10722/339714
ISSN
2023 Impact Factor: 3.3
2023 SCImago Journal Rankings: 1.706
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZhang, SH-
dc.contributor.authorWong, GKW-
dc.date.accessioned2024-03-11T10:38:47Z-
dc.date.available2024-03-11T10:38:47Z-
dc.date.issued2023-05-18-
dc.identifier.citationEducational Technology Research and Development, 2023, v. 71, n. 4, p. 1595-1630-
dc.identifier.issn1042-1629-
dc.identifier.urihttp://hdl.handle.net/10722/339714-
dc.description.abstract<p>Computational thinking (CT) has permeated primary and early childhood education in recent years. Despite the extensive effort in CT learning initiatives, few age-appropriate assessment tools targeting young children have been developed. In this study, we proposed Computational Thinking Test for Lower Primary (CTtLP), which was designed for lower primary school students (aged 6-10). Based on the evidence-centred design approach, a set of constructed-response items that are independent of programming platforms was developed. To validate the test, content validation was first performed via expert review and cognitive interviews, and refinements were made based on the comments. Then, a large-scale field test was administered with a sample of 1st-3rd graders (N = 1225), and the data was used for psychometric analysis based on both classical test theory (CTT) and item response theory (IRT). The CTT results provided robust criterion validity, internal consistency, and test-retest reliability values. Regarding IRT results, a three-parameter logistic model was selected according to the item fit indices, based on which fair item parameters and test information reliability were generated. Overall, the test items and the whole scale showed proper fit, suggesting that CTtLP was a suitable test for the target group. Analyses of the test performance were then put forward. Results reported that students' performance improved with grade level, and no gender difference was detected. Based on the test responses, we also identified children's challenges in understanding CT constructs, indicating that students tended to have difficulty in understanding loop control and executing multiple directions. The study provides a rigorously validated diagnostic test for measuring CT acquisition in lower primary school students and demonstrates a replicable design and validation process for future assessment practices, and findings on the difficulties children faced in CT conceptual understanding could shed light on CT primary and early childhood education.</p>-
dc.languageeng-
dc.publisherSpringer-
dc.relation.ispartofEducational Technology Research and Development-
dc.subjectAssessment-
dc.subjectComputational thinking-
dc.subjectEarly childhood education-
dc.subjectEvidence-centred design-
dc.subjectPrimary school-
dc.titleDevelopment and validation of a computational thinking test for lower primary school students-
dc.typeArticle-
dc.identifier.doi10.1007/s11423-023-10231-2-
dc.identifier.scopuseid_2-s2.0-85159674048-
dc.identifier.volume71-
dc.identifier.issue4-
dc.identifier.spage1595-
dc.identifier.epage1630-
dc.identifier.eissn1556-6501-
dc.identifier.isiWOS:000989906400001-
dc.publisher.placeNEW YORK-
dc.identifier.issnl1042-1629-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats