Vibration tests consisting of heel impacts by a person, shaker impacts, and random shaker excitations were carried out on an 8.5 × 8.3 m steel-joist-concrete floor. Structural variations to the floor sample included edge conditions restrained and free; two slab thicknesses; changes in joist stiffness; and inclusion of transverse bracing. The lower four to six natural frequencies and associated mode shapes and damping values are determined. Damping for the sample is significantly increased by the presence of one or more persons on the floor. Viscous damping coefficients for people are evaluated, and comparisons presented between measured and calculated damping ratios, modelling people by simple dashpots. Predicted and measured acceleration responses due to heel impact, shaker impact, and shaker white-noise excitation are also compared.