Paired wooden child figures with different colored observation cards. Original AI-generated artwork produced for GreenCalc; not a measured diagram or empirical data display.
The information source belongs in the problem statement
Under an idealized model with two independent children and probability one-half for each of two categories, the ordered outcomes BB, BG, GB and GG are equally likely. Here B and G are labels in a probability exercise. This model is not a real-family prediction or a description of the full complexity of sex and gender.
Compare three observation rules
If a family is selected from those with at least one boy, the allowed outcomes are BB, BG and GB. Only one of those three has two boys, so the probability is one-third. If the older child is specified to be a boy, only BB and BG remain, giving one-half. If a child is selected uniformly from the two children and observed to be a boy, BB produces the observation twice as often as either mixed ordered family; the conditional two-boy probability is again one-half.
| Information protocol | P(two boys | information) |
|---|---|
| At least one boy in selected family | 1/3 |
| Older child is a boy | 1/2 |
| Uniformly selected child observed as boy | 1/2 |
Do not silently rewrite a conversation as a random experiment
A parent volunteering a statement may follow no specified random disclosure rule. The family-selection and disclosure process must be modeled before a numerical answer is justified. “I met a boy from a two-child family” could describe several different sampling schemes. Treating every sentence containing “a boy” as the at-least-one condition discards the observation mechanism.
Vary assumptions explicitly
The Boy or Girl Paradox calculator offers named protocols and an adjustable modeled boy probability. Its answer is conditional on those settings. The independent-tag guide explores the Tuesday variant. For another observation-pathway puzzle, compare Bertrand's Box Paradox; there too, counting surviving labels without weighting how they generated the observation gives the wrong model.
Check the model behind your result
Use the Boy or Girl Paradox for the supported calculation and review its stated assumptions before interpreting the output. Find related tools in the Statistics category, or browse the GreenCalc homepage for other calculation topics.
By GreenCalc Editorial Team · Updated 2026-10-26