Continuing the systematic determination of the electron temperature of HII regions using the Balmer and/or Paschen discontinuities by Guseva et al. (2006ApJ...644..890G) we focus here on 3.6m ESO telescope observations of a large new sample of 69 HII regions in 45 blue compact dwarf (BCD) galaxies. This data set spans a wide range in metallicity (Z_{sun}_/60<Z<Z_{sun}_/3) and, combined with the sample of 47 HII regions from Guseva et al. (2006ApJ...644..890G), yields the largest spectroscopic data set ever used to derive the electron temperature in the H^+^ zone. In the same way as in Guseva et al. (2006ApJ...644..890G) we have used a Monte Carlo technique to vary free parameters and to calculate a series of model spectral energy distributions (SEDs) for each HII region. The electron temperature in the H^+^ zones was derived from the best fitting synthetic and observed SEDs in the wavelength range ~3200-5100{AA}, which includes the Balmer jump. On the base of the present large spectroscopic sample we find that in hot (Te(H^+^)>11000K) HII regions the temperature of the O^2+^ zone, determined from doubly ionised oxygen forbidden lines, does not differ statistically from the temperature of the H^+^ zone. Thus, we confirm and strengthen the finding by Guseva et al. (2006ApJ...644..890G). We emphasize that due to a number of modelling assumptions and the observational uncertainties for individual objects, only a large, homogeneous sample, as the one used here, can enable a conclusive study of the relation between Te(H^+^) and Te(OIII).