The study examined the psychometric structure of Basic Needs Satisfaction in General Scale (BNSG-S) with a sample of 495 college students. It compared the original 21-item version (Gagné, 2003) with the shortened 16-item version (Johnston & Finney, 2010), both with and without method effects. A series of confirmatory factor analyses (CFA) suggested a three-factor CFA model with method effect using the 16-item version showed acceptable model-data fit. However, the high correlations among factors posed threats to the discriminant validity. To address the multicollinearity among the factors, our study proposed and tested a second-order CFA model with method effect based on the 16-item shortened version of BNSG-S, which provided good model-data fit. Additionally, Cronbach's alpha and omega coefficients demonstrated satisfactory reliability. These results supported a higher-order factor structure of the 16-item version, suggesting the needs satisfaction may be interconnected and influenced by a broader overarching factor.