The effective delayed neutron fraction βeff is of primary importance for reactivity control of fissile systems and therefore for reactor design and safety analyses. Validation of βeff calculations is complicated by the limited availability of benchmark-quality data. This paper focuses on evaluation and analysis of βeff measurements with 252Cf-source pseudo-worth and noise methods performed at SNEAK 7A and SNEAK 7B assemblies in Germany in the 1970s. The experiments are thoroughly documented in the International Handbook of Evaluated Reactor Physics Benchmark Experiments and briefly presented in this paper. The measurements performed with the two different methods on SNEAK 7A and SNEAK 7B and other facilities systematically produce different values. Given that the noise approach presumes evolution of neutron field fluctuations in a one-point kinetic model, it was assumed that the discrepancies originate from spatial effects. A two-point kinetic model was tested to check this assumption. The paper demonstrates that the βeff measured with the noise method on the SNEAK 7A and SNEAK 7B facilities should be corrected while the 252Cf-source pseudo-worth measurement produces an accurate value.