Overview
This unit moves from describing data you already have (statistics) to predicting outcomes you don't have yet (probability); starting with the basic rules, then building up to the two named distributions the IB tests most: binomial and normal.
Core Content
Basic Probability Rules
$$P(A \cup B) = P(A) + P(B) - P(A \cap B)$$ $$P(A \cap B) = P(A)\cdot P(B) \text{ only if } A \text{ and } B \text{ are independent}$$ $$P(A|B) = \frac{P(A \cap B)}{P(B)}$$
Venn diagrams are best for "and/or" questions involving overlapping events; tree diagrams are best for sequential events (e.g. drawing twice without replacement).
"Independent" and "mutually exclusive" are opposites in spirit, not synonyms: independent events can both happen without affecting each other's probability, while mutually exclusive events can never both happen at once (so $P(A \cap B) = 0$).
The Binomial Distribution
Models the number of successes in $n$ independent trials, each with success probability $p$:
$$P(X=k) = \binom{n}{k}p^k(1-p)^{n-k}, \qquad E(X) = np$$
Your GDC's binompdf/binomcdf functions compute this directly.
The Normal Distribution
Models continuous data that clusters symmetrically around a mean $\mu$ with standard deviation $\sigma$. The 68-95-99.7 rule gives a quick sanity check: about 68% of data falls within 1 standard deviation of $\mu$, 95% within 2, and 99.7% within 3.
Why This Matters
The normal distribution is the natural extension of the "spread around the mean" idea from Descriptive Statistics and Correlation, where $\sigma$ was just a summary number, here it becomes the shape parameter of an entire distribution you can use to calculate probabilities.
Exam Application
Paper 2 expects you to use your GDC's distribution functions (normalcdf, invNorm, binompdf) rather than tables, know which function answers "probability up to a value" versus "value given a probability" (the inverse). A common trap: binomial questions asking for "at least" or "at most" require summing a range, not a single binompdf value, use binomcdf and adjust for whether the boundary is included.
Misconception Breakdown
Students often calculate $P(X \le k)$ when the question asks for $P(X < k)$ (or vice versa) with a discrete distribution like binomial. Why? With continuous distributions like normal, $P(X<k)$ and $P(X \le k)$ are identical, so students assume the same is true for binomial. But binomial is discrete; $P(X < k) = P(X \le k-1)$, and mixing these up changes the answer.
Quick Revision
- $P(A \cup B) = P(A)+P(B)-P(A\cap B)$.
- Independent: $P(A\cap B)=P(A)P(B)$. Mutually exclusive: $P(A\cap B)=0$.
- Binomial: $E(X)=np$; use for counting successes in fixed, independent trials.
- Normal: symmetric around $\mu$; 68-95-99.7 rule for quick estimates.