Situational Awareness Rating Technique (SART)
Also known as: SART
The Situational Awareness Rating Technique (SART), developed by Robert Taylor in 1990 for the NATO Advisory Group for Aerospace Research and Development (AGARD), is a subjective post-task measurement instrument for assessing an operator's degree of situational awareness (SA)—the perception of elements in the environment, understanding of their meaning, and projection of their future state. SART is widely used in aviation, military operations, emergency response, and human-factors research to evaluate system designs, training effectiveness, and task demands that enable or impair operator situational awareness.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use SART to evaluate system designs (cockpit layouts, control-room interfaces, command-center dashboards) that affect operator information availability and cognitive load. Administer to compare alternative interface designs (e.g., does integrating system information into a single display improve SA relative to separated displays?). Use in training and expertise research to show that expert operators achieve higher SA despite task complexity. Also use in incident/accident investigation: low SART scores reported by operators may identify where information design, training, or workload management failed. Particularly valuable in high-stakes domains (aviation, nuclear power, military, emergency services) where SA directly links to safety. Less suitable if SA has not been empirically linked to actual task performance in your domain (validate construct relevance first).
Strengths & limitations
- Domain-specific theoretical grounding: SART operationalizes the three-level SA model (Endsley) that is foundational to human factors; directly ties measurement to cognitive mechanisms underlying safe task performance.
- Validated in aviation and complex operations: Decades of use in cockpit design, military command-and-control, and emergency response; published norms and sensitivity to interface changes.
- Captures multidimensional SA: Separates perception (information availability) from comprehension (mental model formation) and projection (anticipatory ability), enabling diagnostic feedback on where SA breaks down.
- Brief administration: 10 items, 5–10 minutes post-task, making it feasible for repeated measurement across design iterations.
- Actionable for interface redesign: Low Perception scores drive information architecture changes; low Comprehension scores motivate training or display labeling; low Projection scores indicate need for decision-support or alerts.
- Subjective and retrospective: Relies on self-report of awareness post-task; operators may overestimate their comprehension or projection ('I would have noticed if...'), introducing optimism bias.
- Not predictive of actual performance in isolation: High SART does not guarantee correct decisions; must be paired with task outcome data (accuracy, response time, safety metrics).
- Vulnerable to task difficulty confounds: High cognitive load may depress SART regardless of interface design quality; without controlling workload, it's unclear whether low SART reflects poor design or inherent task difficulty.
- Scoring ambiguity: Unweighted vs. weighted scoring conventions vary across studies, hindering cross-study comparison; no universal agreement on which is 'correct.'
- Requires sufficient task exposure: SART validity depends on operators experiencing a task scenario complex enough to exercise SA components; trivial tasks may yield ceiling SART scores.
- Limited cross-domain validation: Most validation is in aviation; generalizability to other high-stakes fields (medical, military, emergency dispatch) requires domain-specific testing.
Frequently asked
Should I use weighted or unweighted SART scoring?
Unweighted (sum all 10 items directly, range 10–70) is simpler and more comparable to published norms. Weighted (SA = Perception+Comprehension+Projection minus Workload minus Stress) theoretically accounts for the idea that high workload undermines SA. Both are used; report whichever you choose, and ideally both for transparency. Unweighted is more standard in publications; start there.
Can SART predict actual situational awareness (e.g., whether operator will catch an anomaly)?
SART is a snapshot of perceived awareness post-task, not a perfect predictor of real-time behavior. A pilot may report high SART but still miss a warning light if distracted at the critical moment. Use SART to compare designs and flag areas for improvement, but validate against objective SA measures (probe questions during task, knowledge tests, actual error detection rates) to confirm it predicts real awareness.
What sample size do I need for SART?
For design comparison (Design A vs. B), aim for n≥15–20 per design to detect meaningful differences. For research publication or expert vs. novice comparison, n≥20–30. Small samples (n<10) can identify gross differences (e.g., new interface improves SART by 15 points) but lack power for subtle effects. Paired designs (same operators, both designs) require smaller n due to reduced between-subject variance.
Should SART be administered individually or in groups?
Individual administration (one operator, one assessor) is preferred to allow interview-style follow-up and clarification. Group administration (e.g., classroom of trainees rating a scenario) is feasible if instructions are clear and standardized; however, peer influence may occur (operators anchoring on each other's responses). For critical evaluations (design decisions), use individual administration.
How do I handle the fact that operators overestimate their own awareness?
SART is subjective and subject to optimism bias (operators believe they were more aware than they actually were). Mitigate by: (1) collecting objective performance data (accuracy, response time, error detection) and comparing to SART ratings; (2) using probe questions or knowledge tests during task to verify comprehension; (3) qualitatively interviewing operators ('You rated comprehension 6/7; walk me through how you understood the situation at the critical moment'). Use SART for relative design comparison (Design A SART vs. Design B SART) rather than interpreting absolute scores.
Sources
- Taylor, R. M. (1990). Situational awareness rating technique (SART): The development of a tool for aircrew systems design. In AGARD-CP-478 (pp. 3/1–3/17). NATO Advisory Group for Aerospace Research and Development. link ↗
How to cite this page
ScholarGate. (2026, June 3). Situational Awareness Rating Technique (SART). ScholarGate. https://scholargate.app/en/human-factors/situational-awareness-rating
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- NASA Task Load IndexHuman Factors↔ compare
- Operator Performance Assessment ScaleHuman Factors↔ compare
- Team Situation Awareness ScaleHuman Factors↔ compare
- Workload ProfileHuman Factors↔ compare