Transmission electron microscopy is used to determine the dissociation width of extended dislocations in n‐, si‐, and p‐GaAs. Results obtained by high‐resolution transmission electron microscopy (HRTEM) and the weak‐beam technique (WB) are compared. Reasons for the differences between the results obtained from WB and HRTEM and their influence on the determination of the stacking‐fault energy are discussed. The results are compared with previous investigations, which show a wide scatter of results despite the fact, that the stacking‐fault energy is an intrinsic material property. A number of parameters, such as the defect used for the determination, localization of the partial‐dislocation cores and the electrical characteristics of the material are considered to explain the discrepancies.