Conditional Probability Using Venn Diagrams & Contingency Tables

This section introduces independent events and how to test for independence. Two events A and B are independent if the occurrence of one does not affect the probability of the other. Mathematically, events A and B are independent if P(A ∩ B) = P(A) × P(B). The instructor demonstrates how to test for independence by calculating both sides of the equation and checking if they are equal. If they are equal, the events are independent; otherwise, they are dependent.

Conditional probability restricts the sample space to outcomes where a given event has occurred. The formula P(A|B) = P(A ∩ B) / P(B) calculates the probability of A given B. Two events are independent if the occurrence of one does not affect the probability of the other. For independent events, P(A ∩ B) = P(A) × P(B). To test independence, compare the product of individual probabilities with the intersection probability. If they are equal, events are independent; otherwise, they are dependent. This mathematical test provides a definitive way to determine whether events influence each other.

This section covers the fundamental concepts of independent and dependent events. Two events are independent if the occurrence of one event does not affect the probability of the other event. Two events are dependent if the occurrence of one event affects the probability of the other event. Two events A and B are independent if and only if P(A ∩ B) = P(A) × P(B). This means the probability of both events occurring together equals the product of their individual probabilities. If this equality holds, the events are independent; otherwise, they are dependent. To verify whether two events are independent, you calculate P(A ∩ B) and compare it with P(A) × P(B). When events are independent, the conditional probability P(A|B) equals P(A), meaning knowing event B has occurred does not change the probability of event A occurring.

Two events A and B are independent if knowing whether or not one event has occurred does not change the probability that the other event will happen. Mathematically, events are independent if P(A|B) = P(A) and P(B|A) = P(B). To test independence, calculate the conditional probability of one event given the other and compare it to the unconditional probability of that event. If they are equal, the events are independent; if different, they are dependent.

To test if events are independent, multiply P(A) by P(B) and compare to P(A and B). If equal, events are independent. For conditional probability P(A|B), if it equals P(A), events are independent. For complementary events, first find the complement probability by subtracting from 1, then apply the same multiplication test. These mathematical tests confirm whether one event affects the probability of another event occurring.

The general multiplication rule states P(A and B) = P(A) × P(B|A), applicable to both independent and dependent events. For independent events, this simplifies to P(A) × P(B). Tree diagrams visualize sequential decisions with branching probabilities. Example: If P(snooze) = 0.60 and P(late|snooze) = 0.30, then P(snooze and late) = 0.18. Total probability of late combines all pathways: P(late) = 0.18 + (0.40 × 0.10) = 0.22.

The general multiplication rule states that the probability of two events occurring in sequence is found by multiplying the probability of the first event by the conditional probability of the second event given the first occurred. Tree diagrams provide a visual method to break down complex probability problems into sequential stages, where each branch represents a possible outcome with its corresponding probability. To find the probability of a specific outcome, multiply the probabilities along the branches leading to that outcome. To find the total probability of an event that can occur through multiple paths, add the probabilities of all those paths. Conditional probability is calculated as the probability of the intersection of two events divided by the probability of the condition event.

The general multiplication rule states P(A and B) = P(A) × P(B|A), where P(B|A) is the conditional probability of B given A has occurred. Tree diagrams visually represent multi-stage chance processes by creating branches for all possible outcomes at each stage, with probabilities on each branch. To find the probability of a specific sequence, multiply along the path. To find the total probability of an event occurring through multiple pathways, add the probabilities of all such pathways. The conditional probability formula P(A|B) = P(A and B) / P(B) calculates the probability of A given B has occurred. These tools are essential for analyzing dependencies between events in real-world scenarios like medical testing, survey data, and everyday chance processes.

The general multiplication rule states that P(A and B) = P(A) × P(B|A) or equivalently P(A and B) = P(B) × P(A|B). This rule allows you to calculate the probability of two events occurring together by multiplying the probability of one event by the conditional probability of the other event given the first has occurred. Tree diagrams are visual tools that help map out all possible outcomes by representing each choice as a branch. The probability of reaching the end of any branch is found by multiplying the probabilities along that path. This method is particularly useful for problems involving multiple sequential choices or dependent events.

A tree diagram visually represents the multiplication rule. The first level branches represent the first event with their probabilities. The second level branches represent the second event given the first event occurred, with their conditional probabilities. The probability of a specific path (sequence of events) is the product of probabilities along that path. For example, P(AB) = P(A) × P(B|A), and P(A^cB) = P(A^c) × P(B|A^c). Tree diagrams can handle complementary events. For event A, the branches are A and A^c with probabilities P(A) and P(A^c) = 1 - P(A). For each branch, the second level shows conditional probabilities for event B and B^c. Calculate all path probabilities: P(AB) = P(A) × P(B|A), P(A^cB) = P(A^c) × P(B|A^c), P(AB^c) = P(A) × P(B^c|A), and P(A^cB^c) = P(A^c) × P(B^c|A^c). When events are mutually exclusive, the probability of their union is the sum of their individual probabilities.

Conditional probability is essential for interpreting medical test accuracy and business analytics. Medical tests with 90% sensitivity and 99% specificity do not mean a positive result means 90% probability of having the disease—the actual probability depends on disease prevalence. A 2010 New York Times study found doctors incorrectly estimated breast cancer probability after positive tests despite 90% accuracy, because they ignored the low base rate. In business, conditional probability enables targeted marketing by calculating conversion rates for specific customer segments. The lift metric measures campaign effectiveness by comparing targeted conversion rates to baseline rates. Medical tests require minimum prevalence thresholds to be reliable, explaining why PCR tests were initially only used in hospitals and later approved for home use as prevalence increased.

Sensitivity is the proportion of true positives among all actual diseased individuals, calculated as true positives divided by (true positives + false negatives). For example, with 85 true positives and 15 false negatives, sensitivity is 85/100 = 0.85 or 85%. Specificity is the proportion of true negatives among all actual healthy individuals, calculated as true negatives divided by (true negatives + false positives). For example, with 95 true negatives and 5 false positives, specificity is 95/100 = 0.95 or 95%. Positive predictive value (PPV) is the probability that a positive test result indicates actual disease, calculated as true positives divided by (true positives + false positives). Negative predictive value (NPV) is the probability that a negative test result indicates no disease, calculated as true negatives divided by (true negatives + false negatives).

In diagnostic testing, sensitivity (true positive rate) is useful for ruling out a disease because high sensitivity means few false negatives, while specificity (true negative rate) is useful for confirming a disease because high specificity means few false positives; the predictive values (positive and negative) depend on disease prevalence and represent the post-test probability of having or not having the disease, with likelihood ratios serving as synthetic tools that combine sensitivity and specificity to quantify how much a test changes the probability of disease.

Sensitivity measures how well a test identifies diseased individuals: TP/(TP+FN). Specificity measures how well a test identifies non-diseased individuals: TN/(TN+FP). Positive Predictive Value measures the probability that a positive test result indicates true disease: TP/(TP+FP). Negative Predictive Value measures the probability that a negative test result indicates true absence of disease: TN/(TN+FN). These four metrics provide complementary information about test performance and should be interpreted together for comprehensive evaluation.

This section covers diagnostic test evaluation: (1) Four categories - True Positive, False Positive, True Negative, False Negative; (2) Sensitivity = P(Test+|Disease+) = TP/(TP+FN); (3) Positive Predictive Value (PPV) = P(Disease+|Test+) = TP/(TP+FP); (4) Specificity = P(Test-|Disease-) = TN/(TN+FP); (5) Key distinction - sensitivity/specificity condition on disease status (vertical calculation), predictive values condition on test result (horizontal calculation); (6) Sensitivity/specificity depend on test quality; predictive values depend on disease prevalence.