Twin-screw extrusion processes are commonly refined on laboratory-scale extruders then scaled-up to manufacturing systems. When using twin-screw extrusion to compound filler into a polymer, the dispersion of the filler must be considered during scale-up. In this work, two scale-up methods are evaluated for how accurately they scale dispersion as measured by the Residence Stress Distribution, an experimental method that quantifies stress developed in a twin-screw extruder. The first scaleup method evaluated is the industry-standard scaling based on maintaining equivalent volumetric flow rate across extruder sizes. Volumetric scaling is compared to a second, novel scale-up method, the percent drag flow rule, which maintains the same degree of fill in the strongest dispersive screw elements on all extruder sizes. Both scale-up rules have been used to scale between three extruder sizes and have been evaluated for how accurately the larger extruders recreate the dispersive mixing of the smallest machine. Results indicate that the percent drag flow scale-up more accurately maintains dispersive mixing behavior than the volumetric scaling. Furthermore, percent drag flow scale-up resulted in all three extruder sizes behaving similarly to changes in operating conditions. These results indicate that percent drag flow scale-up is a valid technique to scale real industrial processes. (C) 2016 Society of Plastics Engineers