Computational thinking (CT) is one of the essential skills for students in the 21st century, which has been widespread around the world. Therefore, the assessment of CT has become a hot topic. Increasing efforts to integrate CT assessment across all disciplinary backgrounds have lately progressed. The scale assessment is the most representative assessment tool. However, it cannot adequately reflect students’ preferences and ideas. The present study was aimed at proposing an improved scale-based assessment method, which could collect students’ fuzzy preferences with the support of Pythagorean fuzzy set (PFS). This method treats CT assessment as a multi-criteria decision-making (MCDM) problem. In the PFS environment, the whale optimization algorithm (WOA) is used for criteria weight computation, and students’ CT along with five sub-abilities are classified into five levels using the adapted Vlse Kriterijumska Optimizacija I Kompromisno Resenje (VIKOR) method. A total of 24 eighth-grade students participated in the CT assessment. And the experimental results indicate the feasibility of the method. Two sensitivity analyses were conducted to verify the effectiveness of the model. Furthermore, a correlation analysis between the method, traditional scales method, and teacher evaluations also confirmed it.