Thermal stresses caused by heating and non-uniform cooling of a small volume of metal give rise to residual stresses in welding. Residual stresses are one of the main problems related to the manufacture of metal components where welding processes are involved. Despite constant growth of computational techniques, there is still a need for experimental work aimed at establishing general principles of cause and effect relationship in the generation of residual stresses. However, there is great disagreement in the literature among experimental results. One of the likely reasons would be the difficulty of identifying and discretizing the effect of each variable that governs the phenomenon.The objective of the present work was to propose a methodological approach to be able to evaluate in an isolated way, but under a holistic view, the influence of the factors involved in the generation of residual stresses and then to make the own experimental evaluation of the same through the Diffraction of X-ray (XRD). Initially, a diagram was proposed to classify the factors that influence the generation of thermal stress at different levels (primary and secondary effects). Then, an original methodology to evaluate the influence of primary factors (heated metal volume and metallurgical transformation) on the generation of residual stresses was applied and evaluated. The whole process of measuring residual stresses by XRD was also systematically evaluated and a series of recommendations were proposed.The proposed methodology showed to be able to demonstrate that the welding energy is not a primary parameter, that is, able to govern directly the generation of residual stresses in steel evaluated. The systematic evaluation of the process of measurement of residual stresses by XRD has shown that it is necessary to quantify the previous state of the RT contained in the sheet before welding to obtain more reliable results, but this approach should only be done in regions where the metal heating is below its metallurgical transformation. In addition, it has been shown that sheet restraint influences the temporal evolution of residual stresses and that it is good practice to use a rigid fixation to ensure the quality of the results.