Investigating the psychometric features of a locally designed computational thinking assessment for elementary students

ABSTRACT

Background and Context

Educators and researchers are interested in building the computational thinking (CT) skills of K-12 students. However, the availability of language-agnostic assessments for lower elementary graders remains limited.

Objective

We present preliminary insights into the reliability and validity of the Computational Thinking Assessment for Elementary Students (CTAES), a new assessment to measure 3^rd-5^th grade students’ CT abilities.

Method

The CTAES was administered to 222 3^rd-5^th grade students. We conducted Rasch analyses, focusing on dimensionality, separation characteristics, and differential item functioning.

Findings

The CTAES appears to be unidimensional, primarily assessing students’ CT skills. Students with lower CT proficiency demonstrate lower likelihood of correctly responding to assessment items compared to peers with higher CT proficiency levels. Preliminary evidence suggests that the assessment does not exhibit bias based on gender or racial/ethnic background.

Implications

Initial findings suggest that the CTAES holds promise as a reliable and valid assessment tool, although there remain opportunities for further refinement.

KEYWORDS:

• Computational thinking
• assessment
• elementary education
• Rasch

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Lai (2021) provides a more extensive history of CT assessments and explains the tension between domain-general and domain-specific assessments; relatedly, Tang et al. (2020) provide a systematic review of CT assessments as of 2017.

2. The assessment has since been validated and published (Rachmatullah et al., 2020).

3. The CTAES can be accessed at: https://nyu.qualtrics.com/jfe/form/SV_4NPL0BfuUQmQcZg

Additional information

Funding

The work was supported by the National Science Foundation .