How the dust continuum was taken into account in the analysis. Top panel shows the distribution of T_dust and β_RL for the SPT sample. In the bottom panel, we used the CIGALE mock analysis to check whether T_dust and β_RL can be correctly estimated using the available set of data for this SPT sample. In this mock analysis, we use the best models and each of the exact parameters used to build the models for each of the objects. We add the observed noise to the models, and refit the simulated data to re-estimate the same _dust and β_RL parameters. A good correlation between the ‘estimated’ and the ‘exact’ T_dust and β_RL suggests that we are able to derive them correctly.