Dust attenuation A_FUV in magnitudes as a function of redshift. The thick solid red curve represent the fit (see text for details) of the average dust attenuation, as a function of redshift, determined in this work, using the Calzetti’s law. The A_FUV values derived from this fit in each redshift bin are listed in Table 1. The magenta dot-dashed curve is the average A_FUV determined in this work using the recipe in Meurer et al. (1999), based on the β slope. The green dotted curve is the same as the dot-dashed one, but here we use the recipe in Cortese et al. (2006), calibrated with normal star forming galaxies. The thin blue horizontal solid line is a constant A_FUV computed with the Calzetti’s law using one single typical value of E(B − V) (=0.13). The right y-axis shows the multiplicative factor to be applied to the observed luminosities, i.e., 10^0.4A_FUV. All the other symbols represent dust attenuations found in literature, as indicated in the labels: orange dashed line from Tresse et al. (2007) (obtained comparing not-corrected SFRDs with dust-corrected SFRDs from other works, see text for details); gray open diamonds and filled squares from the compilation in Hopkins (2004, H04) (only FUV and NUV determinations, diamonds and squares for a correction independent from or dependent on SFR, respectively); cyan filled upside-down triangles from Steidel et al. (1999, St99); thick gray crosses from Ouchi et al. (2004, O04); orange open and filled squares from Bouwens et al. (2009, B09) and Bouwens et al. (2011, B11) respectively (slightly shifted in redshift for clarity); blue filled circles from Takeuchi et al. (2005, T05), based on the 02h field GALEX-VVDS LFs (Arnouts et al. 2005) and the CDFS field Spitzer 15-μm LFs (Le Floc’h et al. 2005) at 0.2 < z < 1; arrows from Schiminovich et al. (2005, Sc05), also based on GALEX-VVDS survey; filled magenta triangles and magenta crosses from Reddy & Steidel (2009, R&S09) (triangles and crosses for a dust attenuation dependent on or independent from FUV luminosity, respectively).