Vertical profiles of droplet effective radius (<i>r</i><sub>e</sub>) in shallow convective clouds are investigated using results from large-eddy simulations (LES) for clean (aerosol mixing ratio <i>N</i><sub>a</sub>=25 mg<sup>−1</sup>), intermediate (<i>N</i><sub>a</sub>=100 mg<sup>−1</sup>), and polluted (<i>N</i><sub>a</sub>=2000 mg<sup>−1</sup>) conditions. Cloud-top <i>r</i><sub>e</sub> for cloud populations comprising clouds with different heights and at different stages of their development are used to construct a vertical profile of <i>r</i><sub>e</sub>. For the polluted and intermediate cases where precipitation is negligible, the constructed <i>r</i><sub>e</sub> profiles represent the in-cloud <i>r</i><sub>e</sub> profiles fairly well with a low bias (about 10%). For the clean, drizzling case the in-cloud <i>r</i><sub>e</sub> can be very large and highly variable and profiling based on cloud-top <i>r</i><sub>e</sub> is less useful. The differences in <i>r</i><sub>e</sub> profiles between clean and polluted conditions derived in this manner are however, distinct. The subadiabatic characteristics of the simulated cumulus clouds are investigated to reveal the effect of mixing on <i>r</i><sub>e</sub> and its evolution. Results indicate that as polluted and moderately polluted clouds develop into their decaying stage, the subadiabatic fraction <i>f</i><sub>ad</sub> becomes smaller, representing a higher degree of mixing, and <i>r</i><sub>e</sub> becomes smaller (~10%) and more variable. However, for the clean case, smaller <i>f</i><sub>ad</sub> corresponds to larger <i>r</i><sub>e</sub> (and larger <i>r</i><sub>e</sub> variability), reflecting the additional influence of droplet collision-coalescence and sedimentation on <i>r</i><sub>e</sub>. Profiles of the vertically inhomogeneous clouds as simulated from the LES and those of the vertically homogeneous clouds are used as input to a radiative transfer model to study the effect of cloud vertical inhomogeneity on shortwave radiative forcing. For clouds that have the same liquid water path (LWP), <i>r</i><sub>e</sub> of a vertically homogeneous cloud must be about 76–90% of the cloud-top <i>r</i><sub>e</sub> of the vertically inhomogeneous cloud in order for the two clouds to have the same shortwave radiative forcing.