In this study, a comprehensive assessment of the uncertainty in the measurement of gross beta activity in the groundwater of a nuclear power plant was performed. Through a detailed analysis of the main steps in the determination process, the sources of uncertainty affecting the measurement results were identified and quantified, and the combined and expanded uncertainties were calculated. The results showed that the main factors affecting the determination of gross beta activity concentration in groundwater by the thick source method include sampling, sample recovery rate, correction factor of α-ray cross-talk in the β channel, specific activity of the reference material, and sample counting rate, collectively accounting for approximately 80.6%. The contributions of the specific activity of the reference material, the sample counting rate, and the spreading uniformity of the sample were approximately 35.4%. The effects of the sample size, residue quality, and data rounding were relatively small. The uncertainty introduced by sampling, which accounted for 13.5%, should not be underestimated.