The results presented in this paper are part of a process to analyse systematically the sources of uncertainty in X-ray stress determination. They concern one part of the effects of temperature variations which could intervene either as random fluctuations or as a monotonic drift during the acquisition. The proposed formulation is in agreement with the recommendations of the ISO guide on the expression of uncertainty (GUM). It was found that the effect is usually negligible for laboratory experiments which are often temperature controlled and for most materials. However the uncertainty can reach 20 MPa for austenitic steels and a temperature drift of 2 K.