Two-Way Tables and Conditional Probability

A two-way table has rows for one category and columns for another, with counts filling the cells. From one table you can read three different kinds of probability. A joint probability is one cell divided by the grand total, like the chance someone is a teenager who chose option Y. A marginal probability uses a row or column total divided by the grand total, like the chance someone is a teenager at all. A conditional probability goes further: it restricts you to one row or column entirely, and asks what fraction of that restricted group has some other property.

The word "given" in a problem tells you which group to lock onto first. "Given that the student passed" means you throw out everyone who didn't pass, before you ask anything else. The denominator becomes the total for passers only, found by adding the passing counts across every category row. The numerator is the count that satisfies both the given condition and the new condition. This is why order matters: the chance of clouds given rain restricts to rainy days first, while the chance of rain given clouds restricts to cloudy days first. Same two cells can be involved, but different totals sit on the bottom, so the two answers are usually different numbers.

This skill sits next to other data-analysis topics because the same table can also produce percent changes, basic probabilities, and relative frequencies depending on what's asked. The habit to build is always the same: find the phrase after "given" or "among," identify that group's total, and use only that total as the denominator.