Recently, narrow particle size distributions, as measured by sieve analysis, of crushed waste glass were used as a replacement for Portland cement in concrete. Their chemical reactivity was successfully studied as a function of this measure of particle size. Differences between sieve analysis and laser diffraction measures of particle size prompted this current re-analysis. Extremely careful sieving was used to divide the crushed waste glass particles into 0-25 mu m, 25-38 mu m, and 63-75 mu m sieve size ranges, but laser diffraction did not agree with these particle size cutoffs. We use these same materials to try and understand the discrepancies between particle size as measured by laser diffraction and sieve analysis by using X-ray computed tomography followed by spherical harmonic analysis to measure the three-dimensional particle shape and size, as well as the length (L), width (W), and thickness (T) of each particle. We show how laser diffraction and X-ray CT results, along with sieve analyses, can be quantitatively related for these crushed waste glass particles in the approximate size ranges considered. In contrast to previous speculation, the particle width W does not have to correspond closely to the sieve opening - the correspondence depends on overall particle shape. In addition, we demonstrate how many particles are needed to analyze in order to achieve stable averages and distributions of the L/W, W/T, and L/T aspect ratios, which approximately define particle shape. These results have implications for how particle size is measured and interpreted in the cement and concrete and other industries.