Назад към всички

Quiz

// Design engaging quizzes with effective questions, scoring logic, and results that drive learning or conversions.

$ git log --oneline --stat
stars:1,933
forks:367
updated:March 4, 2026
SKILL.mdreadonly
SKILL.md Frontmatter
nameQuiz
descriptionDesign engaging quizzes with effective questions, scoring logic, and results that drive learning or conversions.

Situation Detection

ContextLoad
Knowledge assessment, exams, certificationstypes.md → Knowledge section
Personality quizzes, "Which X are you?"types.md → Personality section
Lead generation, marketing quizzestypes.md → Lead-gen section
Writing effective questionsquestions.md
Building quiz UI/UX, gamificationimplementation.md

Universal Rules

One concept per question. Double-barreled questions confuse and measure nothing. "Do you like pizza and exercise?" → Bad.

Wrong answers must be plausible. If correct answer is obvious by elimination, you're testing pattern recognition, not knowledge.

Results must feel personal. Generic outcomes kill engagement. "You got 7/10" loses to "You're an 80s Movie Expert — you caught references most people miss."

Progress visibility motivates. Show question count, progress bar, time remaining. Uncertainty creates anxiety and abandonment.


Quiz Types Quick Reference

TypeGoalTypical LengthResults
KnowledgeAssess learning10-20 questionsScore + feedback per answer
PersonalityEngagement, sharing5-12 questionsPersonality type/category
AssessmentDiagnose level/fit10-30 questionsDetailed report
Lead-genCapture email5-8 questionsResults gated behind email
TriviaEntertainmentAnyLeaderboard, social share

Question Design Checklist

  • Clear, unambiguous wording
  • One correct answer (or explicit multi-select instruction)
  • Distractors are plausible, not obviously wrong
  • No "all of the above" or "none of the above" (lazy design)
  • Avoid negatives ("Which is NOT...")
  • Test the concept, not reading comprehension

Scoring Patterns

Simple percentage: Correct/total × 100. Best for knowledge tests.

Weighted scoring: Some questions worth more. Good for prioritized competencies.

Branching outcomes: Answer combinations map to results. Used in personality quizzes.

Diagnostic rubric: Score across multiple dimensions. Best for assessments and skill evaluations.


Engagement Boosters

  • Immediate feedback after each answer (right/wrong + explanation)
  • Visual progress indicator
  • Streak rewards ("3 in a row!")
  • Time pressure (optional, increases excitement but also anxiety)
  • Social sharing of results
  • Leaderboards for competitive contexts

Red Flags

  • All correct answers in position B/C → Detectable pattern
  • Questions testing obscure trivia vs actual learning objectives
  • Results that don't connect to actions ("Now what?")
  • Too long with no progress indication → Abandonment
  • Mobile-unfriendly UI (tiny buttons, horizontal scroll)

When to Load More

SituationReference
Designing for specific quiz typetypes.md
Writing and reviewing questionsquestions.md
Building quiz flow, UI, toolsimplementation.md