
use a two-way table as a sample space to compute joint, marginal and conditional probabilities, apply the addition and complement rules, and decide whether two events are independent.
Instruction
Sample spaces and events. A sample space S is the set of all possible outcomes. An event is a subset of it. Events combine in three ways (G.CP.A.1): A or B (the union: A, B or both), A and B (the intersection: both), and not A (the complement: everything in S outside A).
Real data as a sample space. NOAA’s daily records for St. Louis Lambert International Airport give every winter day (December, January, February) from 2000 through 2024, paired with the day before. Call a day rainy if at least 0.01 inch of precipitation fell (melted snow counts). The two-way table of these 2,256 day-pairs:
| Rain today | Dry today | Total | |
|---|---|---|---|
| Rain yesterday | 294 | 394 | 688 |
| Dry yesterday | 396 | 1,172 | 1,568 |
| Total | 690 | 1,566 | 2,256 |
Relative frequencies (A1.DS.A.4 / A1.DS.A.4a / A1.DS.A.4b). A joint relative frequency divides a cell by the grand total: P(rain yesterday and rain today) = 294/2,256 ≈ 0.130. A marginal relative frequency uses a row or column total: P(rain today) = 690/2,256 ≈ 0.306. A conditional relative frequency restricts attention to one row or column.
Conditional probability (G.CP.A.3). The probability of A given B is the fraction of B’s outcomes that are also in A:
P(A | B) = P(A and B) / P(B)
P(rain today | rain yesterday) = 294/688 ≈ 0.427. P(rain today | dry yesterday) = 396/1,568 ≈ 0.253. After a rainy winter day, rain the next day is about 1.7 times as likely as after a dry one.
The addition rule (G.CP.A.6). P(A or B) = P(A) + P(B) − P(A and B). You subtract the overlap because adding P(A) and P(B) counts it twice. P(rain yesterday or rain today) = 688/2,256 + 690/2,256 − 294/2,256 = 1,084/2,256 ≈ 0.480. The complement rule, P(not A) = 1 − P(A), gives P(dry both days) = 1 − 0.480 = 0.520, which matches the table: 1,172/2,256.
Independence (G.CP.A.2, G.CP.A.3). Events A and B are independent if knowing one occurred does not change the probability of the other: P(A | B) = P(A). Equivalently, P(A and B) = P(A)·P(B). Check both ways:
- P(rain today | rain yesterday) = 0.427, but P(rain today) = 0.306. They are not equal.
- P(rain yesterday)·P(rain today) = 0.305 × 0.306 ≈ 0.093, but P(both) = 0.130. They are not equal.
Rain on consecutive days is not independent. In everyday language (G.CP.A.5): a weather system can take more than a day to cross Missouri, so one rainy day makes the next more likely. Forecasters call this persistence. Knowing that yesterday was rainy is real information about today.
Why it matters later. Many probability models assume independence, and the binomial model in Lesson 4.4 does. Real processes like weather often violate it, and the table is how you check. (With data, “not exactly equal” is not enough: small differences can be chance. Here 0.427 against 0.306, on 688 rainy-yesterday pairs, is far too large to be chance. You will test such tables formally with a chi-square test in Statistics B.)
📖 Read for this lesson
The reading opens over the lesson; close it to come back to exactly where you were. The checks below draw on it.
Formative check
Work through these before moving on. They are not graded — they tell you whether the standard below has landed yet.
From the table, P(rain today | rain yesterday) = 294/ ≈ 0.427, and P(rain today | dry yesterday) = 396/ ≈ 0.253.
What is P(rain yesterday OR rain today) for the 2,256 winter day-pairs?
Because P(rain yesterday) × P(rain today) ≈ 0.093 while P(both) ≈ 0.130, rain on consecutive winter days is not independent.
Use the summer (June-August) two-way table for St. Louis in this lesson’s reading (2,300 day-pairs, NOAA GHCN-Daily, retrieved 2026-11-10); the reading also has winter, spring and fall if you prefer another season. Compute P(rain today), P(rain today | rain yesterday) and P(rain today | dry yesterday), decide whether the events look independent, and explain the result to someone who has never studied probability in three sentences. Check your sentences with the checklist in the reading, or ask a parent or guardian to read them.
How confident are you that you can use a two-way table to compute conditional probabilities and check independence?
Practice
Work these on paper or in your notebook, then open Check your answer. Aim for all of Fluency and Application; try at least one Challenge.
Printable version: this unit’s practice workbook (PDF)
Fluency
Build speed and accuracy with the core skill.
- A survey records two yes/no questions. Yes to both: 33; yes to A only: 37; yes to B only: 51; no to both: 29. Find P(A), P(A and B), P(A | B), and decide whether A and B look independent.
Check your answer
Answer: P(A) = 7/15 ≈ 0.467; P(A and B) = 11/50 ≈ 0.220; P(A | B) = 11/28 ≈ 0.393; not independent, since P(A | B) ≠ P(A).Build the two-way table first: rows A yes/no, columns B yes/no. - A survey records two yes/no questions. Yes to both: 25; yes to A only: 24; yes to B only: 9; no to both: 15. Find P(A), P(A and B), P(A | B), and decide whether A and B look independent.
Check your answer
Answer: P(A) = 49/73 ≈ 0.671; P(A and B) = 25/73 ≈ 0.342; P(A | B) = 25/34 ≈ 0.735; not independent, since P(A | B) ≠ P(A).Build the two-way table first: rows A yes/no, columns B yes/no. - A survey records two yes/no questions. Yes to both: 48; yes to A only: 42; yes to B only: 40; no to both: 41. Find P(A), P(A and B), P(A | B), and decide whether A and B look independent.
Check your answer
Answer: P(A) = 10/19 ≈ 0.526; P(A and B) = 16/57 ≈ 0.281; P(A | B) = 6/11 ≈ 0.545; not independent in this data, since P(A | B) ≠ P(A); the difference is small, so in a wider population independence might be plausible.Build the two-way table first: rows A yes/no, columns B yes/no. - Two fair dice are rolled. Find P(sum = 5).
Check your answer
Answer: 1/9Count outcomes in the 36-outcome sample space. - Two fair dice are rolled. Find P(at least one die shows 3).
Check your answer
Answer: 11/36Count outcomes in the 36-outcome sample space. - Two fair dice are rolled. Find P(sum = 3).
Check your answer
Answer: 1/18Count outcomes in the 36-outcome sample space. - Two fair dice are rolled. Find P(sum = 12).
Check your answer
Answer: 1/36Count outcomes in the 36-outcome sample space. - Events A and B have P(A) = 0.4, P(B) = 0.5 and P(A and B) = 0.2. Find P(A or B) and decide whether A and B are independent.
Check your answer
Answer: P(A or B) = 0.4 + 0.5 − 0.2 = 0.7. Independent, because P(A)·P(B) = 0.2 = P(A and B). - If P(A) = 0.35, what is P(not A)?
Check your answer
Answer: 0.65.Complement rule.
Application
Use the skill in context. Show your reasoning.
- St. Louis, January days 2000-2024 (NOAA), paired with the day before: rain yesterday & rain today 101; rain yesterday & dry today 122; dry yesterday & rain today 121; dry yesterday & dry today 430. Find P(rain today), P(rain today | rain yesterday) and P(rain today | dry yesterday). Are the events independent?
Check your answer
Answer: P(rain today) = 222/774 ≈ 0.287; P(rain | rain yesterday) = 101/223 ≈ 0.453; P(rain | dry yesterday) = 121/551 ≈ 0.220. Not independent: rain is far more likely after a rainy day.Use row totals for the conditional probabilities. - Kansas City International, winter days (Dec-Feb) 2000-2024: rain-rain 185, rain-dry 310, dry-rain 309, dry-dry 1,452 (yesterday listed first). Compute P(rain yesterday or rain today) with the addition rule, and compare rain persistence at Kansas City with St. Louis’s winter value of 0.427.
Check your answer
Answer: P(either) = (495 + 494 − 185)/2,256 = 804/2,256 ≈ 0.356. At Kansas City, P(rain | rain yesterday) = 185/495 ≈ 0.374, lower than St. Louis’s 0.427 but far above Kansas City’s P(rain | dry yesterday) = 309/1,761 ≈ 0.175: rain persists at both, and Kansas City has fewer wet days overall (P(rain today) ≈ 0.219 against 0.306).Subtract the overlap once. - St. Louis summer days (Jun-Aug) 2000-2024, counting only heavy-rain days (PRCP at least 1.00 in): rain-rain 10, rain-dry 83, dry-rain 85, dry-dry 2,122. Find P(heavy rain today). Why should you be cautious about P(heavy rain today | heavy rain yesterday) here?
Check your answer
Answer: P(heavy rain today) = 95/2,300 ≈ 0.041. P(heavy rain | heavy rain yesterday) = 10/93 ≈ 0.108 rests on only 93 heavy-rain days (and only 10 repeats), so it is a very uncertain estimate.Conditional probabilities computed from small counts vary a lot.
Challenge
Stretch problems. Expect to think before you write.
- Show that if P(A | B) = P(A), then P(B | A) = P(B) (assuming both probabilities are positive).
Check your answer
Answer: P(A | B) = P(A) means P(A and B) = P(A)P(B). Then P(B | A) = P(A and B)/P(A) = P(A)P(B)/P(A) = P(B).Independence is symmetric.
Review
Keep earlier skills sharp.
- (Unit 3) A study finds that people who own dogs walk more. Is this an experiment or an observational study, and can it show that owning a dog causes more walking?
Check your answer
Answer: Observational; it cannot show causation (people who like walking may choose dogs). - (Unit 2) Suppose a regression of y on x has r² = 0.82. What does that mean?
Check your answer
Answer: About 82% of the variation in y is accounted for by the linear relationship with x.
Four Seasons of Day-Pairs at St. Louis: The Two-Way Tables for Winter, Spring, Summer and Fall
NOAA NCEI GHCN-Daily, St. Louis Lambert International Airport (USW00013994), daily precipitation 2000-2024, tabulated as rainy/dry day-pairs by season; retrieved November 10, 2026 · Public domain (works of the United States government)
Why this reading. The lesson builds one two-way table for St. Louis winter days and asks you to build the same table for another season. Building it from the NOAA file is a good exercise when the download works; the analysis should not depend on that. All four seasons at St. Louis Lambert are tabulated here from the same 9,132 days, with the winter table reproduced so you can see they were built the same way. The summer table is worked; spring and fall are yours.
How each table was built
NOAA GHCN-Daily, St. Louis Lambert International Airport (USW00013994), January 1, 2000 to December 31, 2024: 9,132 days, none missing. A day is rainy if PRCP ≥ 0.01 in; that measurement is the liquid water in rain or in melted snow, so a snowy day counts as wet. Each row of a table is a day-pair: a day that falls in the season, and the day before it. Winter (December, January, February) has 2,256 pairs rather than 2,257 because January 1, 2000 has no “yesterday” in the file. Spring (March-May) and summer (June-August) each have 92 days × 25 years = 2,300 pairs; fall (September-November) has 91 × 25 = 2,275. Retrieved November 10, 2026.
Winter (Dec-Feb), the lesson’s table
| Rain today | Dry today | Total | |
|---|---|---|---|
| Rain yesterday | 294 | 394 | 688 |
| Dry yesterday | 396 | 1,172 | 1,568 |
| Total | 690 | 1,566 | 2,256 |
Spring (Mar-May)
| Rain today | Dry today | Total | |
|---|---|---|---|
| Rain yesterday | 426 | 441 | 867 |
| Dry yesterday | 440 | 993 | 1,433 |
| Total | 866 | 1,434 | 2,300 |
Summer (Jun-Aug)
| Rain today | Dry today | Total | |
|---|---|---|---|
| Rain yesterday | 289 | 405 | 694 |
| Dry yesterday | 402 | 1,204 | 1,606 |
| Total | 691 | 1,609 | 2,300 |
Fall (Sep-Nov)
| Rain today | Dry today | Total | |
|---|---|---|---|
| Rain yesterday | 252 | 344 | 596 |
| Dry yesterday | 347 | 1,332 | 1,679 |
| Total | 599 | 1,676 | 2,275 |
Summer, worked
P(rain today) = 691 / 2,300 = 0.300: about three summer days in ten. P(rain today | rain yesterday) = 289 / 694 = 0.416. P(rain today | dry yesterday) = 402 / 1,606 = 0.250. After a rainy summer day, rain is about 1.66 times as likely as after a dry one (winter: 0.427 against 0.253, a factor of 1.69). Independence check the other way: P(rain yesterday) × P(rain today) = (694/2,300)(691/2,300) = 0.0907, while P(both) = 289/2,300 = 0.1257, about 1.39 times larger. Not independent.
A caution that matters for rarer events: here the conditional probability 289/694 rests on 694 rainy-yesterday pairs, which is plenty. If you narrowed the event to days with an inch or more of rain, the same kind of table would have only about a hundred pairs in its first row, and a difference from the marginal probability could easily be chance. A formal test (Statistics B) is what settles that. The four tables above rest on hundreds of pairs in every row.
What to do
- For spring and for fall, compute P(rain today), P(rain today | rain yesterday) and P(rain today | dry yesterday), and P(rain yesterday or rain today) by the addition rule. Check the last against the table directly.
- Decide for each season whether consecutive days look independent, and rank the four seasons by the ratio of the two conditional probabilities.
- Explain the summer result to someone who has never studied probability, in three sentences, without using the word “conditional”. Check your sentences against this list: one number for how often it rains, one for rain after a rainy day, one for rain after a dry day. Or ask a parent or guardian to read them and say whether they make sense.
Sources: NOAA National Centers for Environmental Information, GHCN-Daily (Daily Summaries), station USW00013994, St. Louis Lambert International Airport (the query used here) (ncei.noaa.gov)