Interview Prep

Boy or Girl Paradox Answer: 1 in 3, Not 1 in 2 [Solved]

Boy or girl paradox answer: two children silhouettes, 1 in 3 not 1 in 2

The boy or girl paradox answer is 1 in 3. A family has two children, and at least one of them is a boy. The chance that both children are boys is 1 in 3, not 1 in 2. This walkthrough shows why, then takes on the famous Tuesday boy twist.

The boy or girl paradox is one of the most reported probability questions in quant and tech interviews, and it is famous for splitting rooms of smart people. Most candidates answer 1 in 2 without thinking. The puzzle is also known as the two-child paradox, and it first reached a wide audience through Martin Gardner in 1959. It still catches people today because the trap is not the maths. The trap is the wording.

The Puzzle: Two Children, One Boy

Here is the question exactly as an interviewer would ask it.

The question: A family has two children. At least one of them is a boy. What is the probability that both children are boys?

Before you read on, pick an answer. Most people say 1 in 2, and they say it fast. The reasoning sounds solid: one child is a boy, so the other child is a boy or a girl, and each is equally likely. That reasoning is wrong, and the reason it is wrong is the whole point of the puzzle.

The Boy or Girl Paradox, Step by Step

To solve the boy or girl paradox, write down every possible family before you use the clue. With two children, and with boys and girls equally likely, there are four equally likely families. List the older child first, then the younger child.

Older child Younger child Family Fits the clue?
Boy Boy BB Yes
Boy Girl BG Yes
Girl Boy GB Yes
Girl Girl GG No

Now apply the clue. At least one child is a boy, so the GG family drops out. Three families are left, and each is equally likely. Only one of them, BB, has two boys. So the answer is 1 in 3.

Notice what the table does. It stops you from treating the two children as one known boy plus one unknown child. The clue does not point at a specific child. It rules out one family out of four, and the three families that survive do not split evenly between two boys and one boy.

Why 1 in 2 Feels Right and Is Wrong

The 1 in 2 answer comes from a quiet switch in the question. It answers a different question from the one asked. If the interviewer had said the older child is a boy, the answer would be 1 in 2. In that version, BG drops out as well as GG, because the older child is fixed as a boy. Two families are left, BB and GB, and only one has two boys.

So the two versions look almost the same, yet they give different answers. The difference is which families the clue removes. “At least one is a boy” removes one family. “The older one is a boy” removes two. That is why the exact wording matters, and that is why interviewers ask this question. They want to see whether you condition on the event you were actually given, or on the event you assumed.

Here is a second way to see it. Imagine 400 families with two children. About 100 are BB, 100 are BG, 100 are GB, and 100 are GG. Keep only the 300 families with at least one boy. Of those 300, only the 100 BB families have two boys. That is 100 out of 300, which is 1 in 3. The numbers are the same as the table, just bigger.

The Tuesday Boy Twist in the Boy or Girl Paradox

Once you have the 1 in 3 answer, the interviewer may add one more detail. It sounds like it should change nothing, and that is what makes it a great follow-up.

The twist: A family has two children. At least one of them is a boy born on a Tuesday. What is the probability that both children are boys?

The answer is 13 in 27, which is about 48 in 100. It is close to 1 in 2, but it is not 1 in 2. Here is why.

First, count the families that fit the clue. Each child can be a boy or a girl, born on any of seven days. That gives 14 combinations per child, and 196 pairs in total. Now count the pairs with at least one boy born on a Tuesday. There are 14 pairs where the older child is a Tuesday boy, and 14 where the younger child is a Tuesday boy. One pair is counted twice, the pair where both children are Tuesday boys. So 27 pairs fit the clue.

Next, count how many of those 27 pairs have two boys. Take the 14 pairs with an older Tuesday boy. In 7 of them, the younger child is a boy born on any day. Take the 14 pairs with a younger Tuesday boy. In 7 of them, the older child is a boy born on any day. One pair is counted twice, the pair with two Tuesday boys. So 13 of the 27 pairs have two boys. The answer is 13 in 27.

The Tuesday detail works because it makes the BB family more likely to produce the clue than the mixed families. A BB family has two chances to produce a Tuesday boy, while a BG or GB family has only one. So the clue points at BB families more often than it points at mixed families, and the answer moves from 1 in 3 toward 1 in 2. The rarer the detail, the closer the answer gets to 1 in 2.

That last point is worth saying out loud in the interview. If the detail were “born on a Tuesday in a leap year”, the answer would sit even closer to 1 in 2. The detail itself does not matter. What matters is how many chances each family has to produce it.

What the Interviewer Is Testing in the Boy or Girl Paradox

The interviewer is not testing whether you have memorised 1 in 3. They are testing whether you build the sample space before you answer. The candidates who do well write down the four cases first, then cross out the ones the clue removes. The candidates who struggle jump straight to the “other child” argument and never check which families survive.

Three habits matter most. First, state your assumptions: boys and girls equally likely, births independent. Second, condition on the exact event given, not on a tidier version of it. Third, check the answer against a count. The 400-family version of this puzzle takes thirty seconds, and it catches the 1 in 2 error before you say it out loud.

One more thing impresses interviewers. If you can explain why the answer changes when the wording changes, you show that you understand conditional probability, not just this puzzle. That is the skill the question is really testing, and it is the same skill behind the birthday paradox and the Monty Hall problem.

For the Tuesday boy analysis, see the treatment by Peter Lynch in the Irish Mathematical Society Bulletin (2011), which traces the puzzle from Martin Gardner to Gary Foshee’s 2010 Gathering for Gardner presentation. Read the paper (UCD repository).

Try Another Puzzle

Enjoyed this one? The birthday paradox is the next classic to try: in a room of 23 people, two of them probably share a birthday, and the maths is just as surprising. For a harder test of the same conditioning skill, the Monty Hall problem asks whether you should switch doors, and most rooms split on it.

Boy or Girl Paradox: FAQs

Is the boy or girl paradox answer really 1 in 3?

Yes, under the standard reading of the question. At least one child is a boy, which leaves three equally likely families, BB, BG, and GB. Only one of them has two boys.

Does the Tuesday boy change the answer?

Yes, it moves the answer to 13 in 27, about 48 in 100. The Tuesday detail makes BB families more likely to fit the clue, because they have two chances to produce a Tuesday boy.

What if the older child is a boy?

Then the answer is 1 in 2. Naming a specific child removes two families, BG and GG, and leaves BB and GB. Only one of the two has two boys.

Enjoying this? Add StudyFromNotes as a preferred source on Google to see our latest notes and updates more often:

Leave a Reply

Your email address will not be published. Required fields are marked *