Heuristic Evaluation Notes

Heuristic Evaluation Overview

  • Definition: Heuristic evaluation is a usability inspection method developed by Jakob Nielsen to identify usability problems in a design.
  • Purpose: Provides rapid feedback and is considered a high-value approach to uncovering usability issues.

Evaluation Methods for Prototypes/Software

  1. Empirical Evaluation

    • Involves real users assessing the product.
    • Requires formality; users interact directly with the software.

  2. Formal Evaluation

    • Utilizes models and formulas to derive usability metrics.
    • Helps predict user interface performance based on behavioral models.

  3. Automated Evaluation

    • Employs software tools to measure usability.
    • Effective for low-level usability checks but challenging for higher-level evaluations.

  4. Critique from Experts

    • Feedback from peers or experts based on established heuristics.
    • Useful throughout different design stages, especially before user testing or redesigns.

Timing for Design Critiques

  • Before User Testing: Identify and resolve minor issues beforehand.
  • Before Redesigning: Use critique to understand effective aspects of the design.
  • When Problems are Known: Gather evidence to articulate and prioritize issues.
  • Before Product Release: Finalize improvements through structured peer review.

Key Steps in Heuristic Evaluation

  1. Establish Clear Goals: Begin with defined objectives for the evaluation even if outcomes are unexpected.
  2. Gather Evaluators: Utilize a small group (3-5) of evaluators to examine usability principles.
  3. Independent Evaluation
    • Evaluators review the user interface separately and identify usability issues based on heuristics.
  4. Aggregate Findings: Discuss findings after evaluations, leveraging the wisdom of crowds to identify common and unique issues.

Ten Design Heuristics (by Jakob Nielsen)

  1. Show system status: Keep users informed about what is happening.
  2. Familiar metaphors/language: Use familiar visual and textual elements to facilitate understanding.
  3. Control & freedom: Users should easily undo actions.
  4. Consistency: Ensure uniformity across the interface.
  5. Error prevention: Design systems to minimize the chances of user errors.
  6. Recognition over recall: Facilitate information retrieval rather than forcing memorization.
  7. Flexibility & efficiency: Allow users to tailor interactions to their needs.
  8. Aesthetic & minimalist design: Avoid unnecessary information which can overwhelm users.
  9. Error recovery: Help users recognize, diagnose, and recover from errors.
  10. Assistance: Provide help and documentation as needed.

Evaluator’s Process

  • Evaluators are typically given tasks to complete while engaging with the design:
    • Use design iteratively to uncover usability principles and issues.
    • Focusing individually maximizes diverse input on usability problems.

Importance of Multiple Evaluators

  • Using multiple evaluators increases the likelihood of identifying problems, but the return diminishes with more evaluators.
  • An optimum group size (3-5 evaluators) effectively balances cost and discovery of usability issues.

Cost-Effectiveness of Heuristic Evaluation

  • A small group can identify a substantial number of usability problems quickly and cost-effectively.
  • Example: A case study showed a benefit-cost ratio of 48, indicating high returns on instigating changes based on heuristic evaluations.

Heuristic Evaluation vs. User Testing

  • Speed: Heuristic evaluations can be completed more quickly than user testing.
  • Pre-Interpreted Results: Evaluators highlight specific issues and their proposed fixes directly.
  • Accuracy: User testing may identify more accurate problems because it involves actual user interaction.
  • Complementary Methods: It is valuable to employ both evaluations and user testing for comprehensive insights.

Phases of Heuristic Evaluation

  1. Pre-evaluation Training: Equip evaluators with necessary context and knowledge.
  2. Evaluation: Conduct individual interface assessments.
  3. Severity Rating: Rank problems based on frequency, impact, and persistence.
  4. Debriefing: Discuss findings and potential improvements with the design team.

Assigning Severity Ratings

  • Severity Scale:
    • 0: No usability issue
    • 1: Cosmetic problem
    • 2: Minor usability issue
    • 3: Major usability issue
    • 4: Usability catastrophe

Example Severity Rating

  • Issue: Unable to edit personal weight in health app.
  • Severity: 2
  • Heuristics Violated: Control & freedom
  • Description: Initial weight entry is non-editable, hindering usability after initial setup.

Debriefing Sessions

  • Conduct meetings with evaluators and the development team to:
    • Discuss general UI issues and gather qualitative feedback.
    • Explore major problem areas and brainstorm improvement strategies.