Recommendation 4.2.8:
Help the user understand complex topics that are critical to the task
Benefits Users
Visual | Cognitive
Relevant W3C Guidance
Phase 1: Gather & Organize
User Research
- List out the topics that the chatbot is supporting. Ask users whether they think the topic is cconfusing or unclear. Ask objective questions (i.e., questions with correct answers) about the topics to learn about average knowledge. Alternatively, talk to government employees who may have a finger on the pulse of public knowledge.
Phase 2: Design & Implement
Design Question
-
How will the user find more information on complex topics?
- Including explanations for every topic will inflate content. Tooltips can be a way to provide explanations only when needed (but make sure they are accessible). Another option is to include a link to more information. Consider allowing the user to adjust the level of detail in the chatbot’s messages.
Tip
-
Always explain why the chatbot is asking for a response or information from the user if it is not obvious.
-
Add context to questions so they make sense on their own
-
Explain complex concepts and difficult words if those concepts and words are critical to the task.
-
Support magnification up to 200%.
Examples
-
Instead of “What is your adjusted net income?”, ask “What is the number in box 10B on form 998, your adjusted net income?”
-
Explain cause-and-effect relationships: “Did you pay for a dependent’s education this year? That may earn you a tax refund.”
-
When the chatbot must discuss or ask about a complex topic, offer the user a choice to learn more about the topic, either on an accessible webpage, in a short explainer video with captions, or in a subsequent chatbot message. Be sure to bring the conversation back to the original topic after explaining.
Phase 3: Test & Evaluate
Ask the User
-
Did the chatbot complex topics so that you could understand them?
- This question is subjective; use a Likert scale. This question should be asked post-study.