Background: The Scored Patient-Generated Subjective Global Assessment (PG-SGA) is a multidimensional tool to assess malnutrition and risk factors. The objectives of this study are to determine the validity of the Thai version of the Scored PG-SGA (Thai PG-SGA) and examine the correlations with selected nutritional parameters. Methods: This observational analytic study included 195 cancer patients aged greater than 18 years at a university-affiliated hospital in Bangkok, Thailand. All patients were assessed for nutritional status by Thai PG-SGA in comparison to subjective global assessment (SGA). Anthropometry, body composition, and hand grip strength were evaluated. Results: According to PG-SGA global assessment categories, 39% (75) of 195 cancer patients were well nourished, 27% (53) were moderately malnourished and 34% (67) of patients were severely malnourished. Thai PG-SGA had a sensitivity of 99.1% and a specificity of 86.0% at predicting SGA classification. PG-SGA numerical scores were significantly different between well-nourished and malnourished groups (4.2 ± 2.4 Vs 16.3 ± 4.9; p < 0.001). The PG-SGA scores, nutritional status assessed by PG-SGA, and nutritional status assessed by SGA were correlated with weight, % weight loss in one month, body mass index, body fat, and hand grip strength (p < 0.001) respectively. Conclusions: Thai PG-SGA showed high sensitivity and good specificity in predicting malnutrition in Thai cancer patients. This tool demonstrated the correlations with anthropometric parameters, body composition, and muscle strength.