Tipping the Scales: How Design Characteristics, Rounding, and Reversal of NASA-TLX Rating Scales Impact Workload Measurement 101 views

The safety and efficiency of human-machine system interactions are highly dependent on Mental Workload (MWL): a critical human factors construct frequently evaluated using the National Aeronautical and Space Administration's Task Load Index (NASA-TLX). Despite its widespread use, NASA-TLX faces several criticisms. This includes a lack of standardized scale design, inconsistent data transformations (e.g., rounding), and confusion arising from the tool's reversed "Performance" scale anchors. This dissertation addresses the listed shortcomings by (1) evaluating the effects of response format and rounding on NASA-TLX ratings, (2) identifying how scale design characteristics impact participants’ ratings, and (3) analyzing how anchor label variations in the Performance dimension contribute to confusion and misresponse. Analyses revealed that NASA-TLX rating distributions were significantly affected by response format and rounding practices, with the former potentially impacting how significant results manifest in experimental analyses. Several scale design attributes were evaluated based on correlation with objective performance metrics, and a preferred visual design format was selected. Finally, this work determined that the reversed Performance scale contributes to confusion and misresponse. Ultimately, this work provides actionable design guidelines for NASA-TLX rating scales, informs its proper use in human research, and enhances both accuracy and reliability of MWL measurement.

PHD (Doctor of Philosophy)

Workload; Pyschometrics; Scale Design; Subjective Workload Assessment; Human Factors Engineering; NASA-TLX

English

2026-03-27