STEM education plays a special role in the development of soft skills, as it not only integrates scientific knowledge but also fosters students’ critical thinking, creativity, collaboration, and communication skills through project-based and research activities. However, traditional statistical methods are insufficient to clarify the mechanism of this influence. In this context, the use of modern approaches, such as the Structural Equation Model (SEM), allows for an accurate determination of the causal relationship between STEM education and soft skills. The aim of this study was to identify the impact of STEM education level on soft skills indicators and to assess the causal relationship between them based on the Structural Equation Model (SEM).

The research was conducted in a quasi-experimental format. The level of STEM engagement was measured using STEM_1–STEM_5 indicators, while soft skills indicators were assessed across four factors (communication, collaboration, creativity, and problem-solving) through 12 indicators. The application of the SEM revealed that the relationship between STEM education and soft skills is statistically significant. The model fit indices met the recommended criteria (χ²/df = 1.01, RMSEA = 0.009, CFI = 0.995, GFI = 0.963), indicating a high level of fit to the data. The direct effect of the STEM factor on the development of soft skills was found to be strong (β ≈ 0.70, p < 0.001), confirming that STEM approaches make a substantial contribution to enhancing students’ communication, collaboration, creativity, and problem-solving skills. This study was funded by the Science Committee of the Ministry of Science and Higher Education of the Republic of Kazakhstan (Grant No. BR28713097).