- Mathematics
- Class 11
- Project 2
2Ten Questions to Any Number: The Power of Halving
One student thinks of a number from 1 to 1,000. The other may only ask "Is it higher or lower than ...?" Most beginners need many guesses, but a clever strategy always finds the number in at most 10 questions. Students discover the halving strategy (binary search), connect it with powers of 2 and logarithms, and then use it to find a word in a dictionary and a name in a sorted list.
- Class and age
- Class 11 · 16–17
- Activity type
- Individual or group
- Time needed
- 45 minutes
- Difficulty
- Easy
- Estimated budget
- Almost free
Mathematical concepts
- Halving the search space
- Powers of 2 and base-2 logarithms
- Worst-case number of steps
- Why sorting makes searching fast
Learning objectives
- Discover and explain the binary search strategy.
- Connect the number of questions to log₂ n.
- Compare linear and binary search.
- Apply the idea in everyday searching.
Real-life connection
Computers find a record among millions in a fraction of a second because data is kept sorted and searched by halving. Programmers also use halving to find the line where a bug starts.
Materials required
- Paper and pencil
- A dictionary and a sorted list of names (for example the class register)
- Calculator with a log function
Step-by-step instructions
- Play five rounds in pairs, counting questions. Record the largest number of questions needed.
- Share strategies. Which strategy guarantees the fewest questions?
- Use halving: always ask about the middle of the remaining range. Play again.
- Make a table: after k questions, how many numbers could remain? (1,000, 500, 250, ...)
- Find the smallest k with 2ᵏ ≥ 1,000. Check with log₂ 1,000 ≈ 9.97.
- How many questions for 1 to 1,000,000? For every person in Bangladesh?
- Time a search for a word in a dictionary by flipping pages one at a time versus by halving.
- Explain why the list must be sorted for halving to work.
Mathematical explanation
Each well-chosen question cuts the possibilities in half, so after k questions at most 1,000 ÷ 2ᵏ numbers remain. Because 2¹⁰ = 1,024 is more than 1,000, ten questions are always enough. In general the number of questions is the smallest whole number at least log₂ n. For a million numbers that is 20, and even for about 170 million people it is only 28. Checking items one by one (linear search) could take n steps. Halving only works when the list is in order, which is why sorting (project 1) and searching go together.
Expected project output
Game records, a halving table, a comparison of linear and binary search times and a one-paragraph explanation.
Skills developed
- Strategic thinking
- Logarithms in context
- Algorithm comparison
- Clear explanation
Assessment criteria
- The halving strategy is used correctly.
- The table and log₂ calculation are correct.
- Linear and binary search are compared with data.
- The need for sorting is explained.
Extension challenge
Play a version where the answerer may lie once. How many questions are now needed? Design a strategy.
Safety and inclusion
No special safety needs.