Computational Thinking (CT) has gained a great interest over the last years as a major set of skills, which promotes new ways of thinking to students of all grades across all subjects. However, there is still no consensus on a particular way of its measurement. This study examines the construct and convergent validity, as well as the internal consistency and test-retest reliability of an adapted version of Computational Thinking Test (CTt) in Greek culture, firstly developed in Spain by Marcos Román- Gonzalez and his research team. The psychometric testing of the CTt engaged 85 primary students attending the 6th grade. Reliability as internal consistency and structure validity was measured from the score performance obtained from all the participants. In contrary, test-retest reliability was assessed with the participation of the first 32 participated students who completed again the CTt after two weeks, while the convergent validity was evaluated with the remaining 53 students who completed concurrently both the CTt and five relative tasks adopted from the Bebras international contest conducted in 2018- 19 and 2019-2020. Regarding convergent validity, the results showed statistically significant, positive and excellent correlation between the CTt and the selected Bebras Tasks (r=0.861). As for the structure validity of the tool, CTt total score had a positive, moderate to strong and statistically significant correlation with the score of each of the seven computational concepts that the CTt measures, with the r value ranging from 0.413 to 0.712. Finally, reliability as internal consistency of the CTt was measured by Cronbach’s Alpha which was α=0.708 showing a good reliability for the entire sample, while the testretest reliability for the total score of the CTt is excellent with an ICC value 0.943. Thus, the adapted version of CTt seems to be a reliable and valid CT assessment-tool for the Greek primary school students.
Psychometric testing of the adapted Computational Thinking Test in Greek Culture: ensuring its reliability, structure and convergent validity
UGOLINI F.C.
2022-01-01
Abstract
Computational Thinking (CT) has gained a great interest over the last years as a major set of skills, which promotes new ways of thinking to students of all grades across all subjects. However, there is still no consensus on a particular way of its measurement. This study examines the construct and convergent validity, as well as the internal consistency and test-retest reliability of an adapted version of Computational Thinking Test (CTt) in Greek culture, firstly developed in Spain by Marcos Román- Gonzalez and his research team. The psychometric testing of the CTt engaged 85 primary students attending the 6th grade. Reliability as internal consistency and structure validity was measured from the score performance obtained from all the participants. In contrary, test-retest reliability was assessed with the participation of the first 32 participated students who completed again the CTt after two weeks, while the convergent validity was evaluated with the remaining 53 students who completed concurrently both the CTt and five relative tasks adopted from the Bebras international contest conducted in 2018- 19 and 2019-2020. Regarding convergent validity, the results showed statistically significant, positive and excellent correlation between the CTt and the selected Bebras Tasks (r=0.861). As for the structure validity of the tool, CTt total score had a positive, moderate to strong and statistically significant correlation with the score of each of the seven computational concepts that the CTt measures, with the r value ranging from 0.413 to 0.712. Finally, reliability as internal consistency of the CTt was measured by Cronbach’s Alpha which was α=0.708 showing a good reliability for the entire sample, while the testretest reliability for the total score of the CTt is excellent with an ICC value 0.943. Thus, the adapted version of CTt seems to be a reliable and valid CT assessment-tool for the Greek primary school students.File | Dimensione | Formato | |
---|---|---|---|
2022 - Kakavas, Ugolini - ICERI Psychometric.pdf
non disponibili
Dimensione
303.45 kB
Formato
Adobe PDF
|
303.45 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.