Design Matters: Scale Design and Rounding Ratings Impacts NASA-TLX Results

The NASA Task Load Index (NASA-TLX) is a widely used subjective measure of mental workload (MWL), but its visual design and postcollection data manipulation (rounding) practices lack standardization across studies. These inconsistencies may bias results. This secondary analysis of a prior within-subjects experiment examined whether different response formats (Knob, Slider, Ask) affect TLX ratings. Friedman tests revealed significant differences between formats across all dimensions except Effort, regardless of rounding. Wilcoxon Signed Ranks tests further showed that rounding significantly alters the data distribution for each sub-scale. Parametric comparisons of the data between different task-difficulty conditions suggest that the experimental findings can be influenced by the response format used to collect ratings. These findings demonstrate how rounding and interface design can meaningfully influence TLX results. Therefore, greater consistency in documentation, implementation, and reporting is needed to improve reliability and reduce bias in future studies. We call for further research into these issues