top of page
Wooden trays of navy and sage tokens with a small selected group in a brass dish

SAT Conditional Probability: Choose the Group Before the Denominator

Oct 15
2 min read

The condition changes who can be selected


College Board includes probability and conditional probability in SAT Math. The arithmetic is often straightforward. The harder decision is which people or objects belong in the denominator after the question supplies extra information.


Words such as “among,” “of those,” and “given that” restrict the group. Before dividing anything, finish this sentence: “We are choosing only from…” That group becomes your new reference set.


Work with a clearly defined sample


In this original example, a club surveys 100 members. Forty attend evening sessions, 30 travel by bicycle, and 12 do both. A member is selected uniformly at random from the 100 members. The probability that the member both attends evening sessions and cycles is 12/100, or 0.12.


Now the question changes: given that the selected member attends evening sessions, what is the probability that the member cycles? Only the 40 evening attendees remain eligible. Of those, 12 cycle. The probability is 12/40 = 3/10 = 0.30.


Reverse the condition and the answer can change


Given that the member cycles, the probability of evening attendance is 12/30 = 2/5 = 0.40. The overlap is still 12, but the denominator is now the 30 cyclists. A shared numerator does not make the two conditional probabilities equal.


This distinction also explains why “both” is not the same instruction as “given.” “Both” identifies an intersection. A condition specifies the group within which you measure that intersection. Read the entire sentence before choosing a total.


Use the table's labels, not its largest number


In a two-way table, first locate the row or column named by the condition. Find its total, then identify the cell that also satisfies the requested outcome. If a total is missing, calculate it from the relevant entries rather than defaulting to the grand total.


Some tables show percentages instead of counts. Check what those percentages are percentages of: the whole sample, each row or each column. A row percentage cannot automatically be treated as a whole-sample percentage. The labels and stated sampling process determine the calculation.


Keep independence out of the problem unless it is justified


Do not multiply two overall probabilities merely because the question mentions two characteristics. In the club example, 0.40 × 0.30 = 0.12 happens to match the overlap proportion, but a different overlap would not. Use the information actually provided; do not assume independence.


For practice, write the eligible group, overlap and fraction before simplifying. Check that the numerator cannot exceed the restricted-group total and that the result lies between 0 and 1. Those checks catch errors, but the most important check remains conceptual: did you answer the condition in the correct direction?


Sources




Related Newton reading




Comments


Newton Learning System

bottom of page