This study employed a mixed-design approach and the Many-Facet Rasch Measurement (MFRM) framework to investigate whether rater bias occurred between the onscreen scoring (OSS) mode and the paper-based scoring (PBS) mode. Nine human raters analytically marked scanned scripts and paper scripts using a six-category (i.e., six-criterion) rating rubric. Interviews with these raters were then conducted to gather their reflections concerning the marking experiences in the two modes. Software program FACETS was employed to estimate raters' scores, with the results indicating that (a) four raters marked the scanned scripts more severely than they did the paper scripts, whereas the remaining four raters exhibited an opposite tendency; (b) the only rater whose composite scores were comparable between the two modes still awarded harsher scores to the category of Mechanics in PBS mode; and (c) the category of Mechanics was scored more harshly in OSS mode, whereas Topic Development/Support was scored more severely in PBS mode. Analyses of quantitative and qualitative data indicated that computers affected raters' marking behaviors and their assessment, assuming a role in causing scoring bias. Implications for rater training were provided.