Error Consequence and Recovery Coding

Description

Code each error in a user test by consequence (positive, negative, neutral), type and the user's recovery strategy, then prioritise redesign by negative consequences.

Counting errors treats them all alike; this scheme codes what they cost and how users respond. Each error is classed by consequence: positive (fails the aim but yields useful knowledge), negative (dead end, lost progress, a return to the start, or irreversible effect) or neutral (no effect). It is also typed as a commission, omission or wrong-action error, or a motor-control error. The participant's recovery strategy is coded as systematic exploration, trial and error, or rigid repetition of a failed action. Categories are fixed before testing, applied during observation, and used afterwards to remove negative-consequence errors first and to anticipate which users will struggle. Research behind it found strategy differences track device experience more than age. In safety-critical settings a formal scheme such as HFACS should be used instead.

Attributes

Loss type
Latent
Stage
Evaluation
Data
Mixed
Effort
Days

Steps

  1. Before testing, list expected errors and fix the coding categories for consequence, type and strategy.
  2. Recruit actual target users rather than colleagues.
  3. During sessions, log each error with its consequence: positive, negative or neutral.
  4. Code its type: commission, omission, wrong action or motor control.
  5. Code the participant's recovery strategy: systematic exploration, trial and error, or rigid repetition.
  6. After testing, prioritise redesign by removing negative-consequence errors first and use strategy profiles to anticipate problems for similar users.

Sources

  • 100 Things Every Designer Should Know About People (Weinschenk, 2011)

Also known as

  • Error consequence and recovery-strategy coding during user testing
  • Error consequence coding
  • Error recovery strategy coding
  • Positive/negative/neutral error classification

Categories

  • User Research
  • Usability & Evaluation