NASA Task Load Index (NASA-TLX)

Raw NASA-TLX is the simple average of the six subscales (range 0–100). Weighted TLX applies importance weights to each subscale based on pairwise comparisons, then sums and normalizes (also 0–100). Both are valid; raw is faster and nearly as discriminative in group studies; weighted is theoretically preferred but adds 5–10 minutes of administration time. Choose based on study design and resources.

There is no universal threshold. Interpretation depends on task context and operator expertise. In safety-critical domains (aviation, surgery), scores >75 may warrant design revision or additional training. In non-critical tasks, scores of 60–70 may be acceptable. Always reference published domain-specific norms (e.g., commercial aviation, emergency medicine) when available.

NASA-TLX is designed for post-task or post-block assessment, providing a retrospective global rating. Online workload measures (e.g., SWAT, AraujoTracker using gaze or heart rate) exist for continuous real-time monitoring, but they differ methodologically. If continuous data is required, consider complementing post-task TLX with physiological measures (heart rate variability, pupil dilation) or secondary task performance.

NASA-TLX has been validated in Western adult populations (pilots, medical staff, researchers). Validity in children, elderly, non-English speakers, and non-Western cultural contexts is less established. Translations exist (German, French, Japanese), but cultural differences in scale interpretation and anchoring should be acknowledged. Pilot testing is recommended in new populations.

If one or two subscales are missing, mean-imputation (using the respondent's average of completed items) or list-wise deletion are acceptable if missingness is <10%. If more than two subscales are incomplete, discard that assessment. Avoid imputing from sample means; it biases subscale variance. Report handling method explicitly.