Comprehensive Study Notes: Elementary Statistics Eighth Edition - Probability and Discrete Probability Distributions

Conditional Probability and the Multiplication Rule

Conditional Probability Concepts and Definitions

  • Conditional Probability: The probability of an event occurring, given that another event has already occurred.

  • Symbolized as P(B∣A)P(B \mid A), which is read as "the probability of event BB, given event AA".

  • In sequential events where the order of occurrence matters, event AA is the given event (known first), and event BB is the subsequent event whose outcome is evaluated.

  • Formula for Conditional Probability:   P(B∣A)=P(A and B)P(A)P(B \mid A) = \frac{P(A \text{ and } B)}{P(A)}   where P(A)≠0P(A) \ne 0

Conditional Probability Calculations

Speeding Ticket Example
  • Scenario: Police records indicate that 47%47\% of drivers stopped for speeding were male and were issued a ticket. Overall, 52%52\% of all drivers stopped were male.

  • Define Events:

    • Event AA: Driver is male

    • Event BB: Driver receives a ticket

  • Given Quantities:

    • P(A)=0.52P(A) = 0.52

    • P(A and B)=0.47P(A \text{ and } B) = 0.47

  • Probability Calculation:   P(B∣A)=P(A and B)P(A)=0.470.52≈0.904P(B \mid A) = \frac{P(A \text{ and } B)}{P(A)} = \frac{0.47}{0.52} \approx 0.904

Self-Driving Vehicle Survey Example
  • Scenario: A survey conducted by The Harris Poll asked 970970 U.S. adults whether they have ever ridden as a passenger in a self-driving vehicle.

Conditional Probability Table for Self-Driving Vehicle Survey
  • Survey Data Contingency Table:

    • Ages 18−6418-64: Yes =202= 202, No =549= 549, Total =751= 751

    • Ages 65+65+: Yes =23= 23, No =196= 196, Total =219= 219

    • Totals: Yes =225= 225, No =745= 745, Grand Total =970= 970

  • Task: Find the probability that an adult is 1818 to 6464 years old, given that the adult has ridden as a passenger (said yes) in a self-driving vehicle.

  • Solution Process:

    • The given condition narrows the sample space strictly to the adults who answered "Yes". Thus, the total sample space size becomes 225225

    • Within this restricted group of 225225 adults, the number of individuals aged 1818 to 6464 is 202202

    • Calculation:     P(18–64∣Yes)=202225≈0.898P(\text{18--64} \mid \text{Yes}) = \frac{202}{225} \approx 0.898

Independent and Dependent Events

  • Independent Events: Two events are independent when the occurrence of one event does not affect the probability of the occurrence of the other event.

    • Mathematical Condition for Independence:     P(B∣A)=P(B)P(B \mid A) = P(B)     or     P(A∣B)=P(A)P(A \mid B) = P(A)

    • When events are independent, the given condition has no impact and can be ignored.

  • Dependent Events: Events that are not independent. The occurrence of one event changes the probability of the second event.

Examples of Dependent Events
  • Card Drawing Example: Selecting a King (AA) from a standard deck of 5252 playing cards, not replacing it, and then selecting a Queen (BB).

    • There are 44 Kings and 44 Queens in a deck.

    • Probability of selecting a King first: P(A)=452P(A) = \frac{4}{52}

    • Not replacing the King leaves 5151 cards remaining (33 Kings, 44 Queens, and 4444 other cards).

    • Probability of selecting a Queen second: P(B∣A)=451P(B \mid A) = \frac{4}{51}

    • Because P(B∣A)≠P(B)P(B \mid A) \ne P(B), the events are dependent.

  • Driving and Accident Example:

    • Event AA: Driving over 85 mph85\,\text{mph}

    • Event BB: Getting in a car accident

    • Driving over 85 mph85\,\text{mph} significantly increases the likelihood of getting in an accident. Thus, Event AA changes the probability of Event BB, making them dependent.

The Multiplication Rule

  • General Multiplication Rule: The probability that two events AA and BB will occur in sequence is:   P(A and B)=P(A)⋅P(B∣A)P(A \text{ and } B) = P(A) \cdot P(B \mid A)

  • Multiplication Rule for Independent Events: When events AA and BB are independent, the rule simplifies to:   P(A and B)=P(A)⋅P(B)P(A \text{ and } B) = P(A) \cdot P(B)

  • The independent rule can be extended to any number of sequential independent events:   P(A and B and C… )=P(A)⋅P(B)⋅P(C)…P(A \text{ and } B \text{ and } C \dots) = P(A) \cdot P(B) \cdot P(C) \dots

Sequential Applications of the Multiplication Rule
  • Two Cards Drawn Without Replacement:

    • Task: Find the probability of selecting a King and then a Queen from a standard deck of 5252 cards without replacement.

    • Events are dependent.

    • Calculation:     P(K and Q)=P(K)⋅P(Q∣K)=452⋅451=162652≈0.006P(K \text{ and } Q) = P(K) \cdot P(Q \mid K) = \frac{4}{52} \cdot \frac{4}{51} = \frac{16}{2652} \approx 0.006

  • Coin Toss and Die Roll:

    • Task: Find the probability of tossing a Head on a coin and then rolling a 66 on a six-sided die.

    • Events are independent.

    • Calculation:     P(H and 6)=P(H)⋅P(6)=12⋅16=112≈0.083P(H \text{ and } 6) = P(H) \cdot P(6) = \frac{1}{2} \cdot \frac{1}{6} = \frac{1}{12} \approx 0.083

Principles of "At Least" vs. "At Most" and Complement Rules

  • At Least One: Represents 11 or more outcomes (1,2,3,4,5,…1, 2, 3, 4, 5, \dots).

  • At Most One: Represents 11 or fewer outcomes (00 or 11).

  • Complementary Probability Principle for "At Least One":   P(at least one)=1−P(none)P(\text{at least one}) = 1 - P(\text{none})   where P(all possible outcomes)=1P(\text{all possible outcomes}) = 1

Car Tire Defect Example
  • Scenario: A set of 44 tires is purchased. According to the manufacturer, 3%3\% (0.030.03) of all tires produced are defective, and 97%97\% (0.970.97) are not defective. Individual tire defects represent independent events.

  • Probability that none is defective (all good):   P(none defective)=(0.97)⋅(0.97)⋅(0.97)⋅(0.97)=(0.97)4≈0.8853P(\text{none defective}) = (0.97) \cdot (0.97) \cdot (0.97) \cdot (0.97) = (0.97)^4 \approx 0.8853

  • Probability that all four tires are defective (all bad):   P(all defective)=(0.03)⋅(0.03)⋅(0.03)⋅(0.03)=(0.03)4=0.00000081P(\text{all defective}) = (0.03) \cdot (0.03) \cdot (0.03) \cdot (0.03) = (0.03)^4 = 0.00000081

  • Probability that at least one tire is defective:   P(at least 1 defective)=1−P(none defective)=1−0.8853=0.1147P(\text{at least 1 defective}) = 1 - P(\text{none defective}) = 1 - 0.8853 = 0.1147

  • Probability that at least one tire is not defective:   P(at least 1 good)=1−P(all defective)=1−0.00000081=0.99999919P(\text{at least 1 good}) = 1 - P(\text{all defective}) = 1 - 0.00000081 = 0.99999919

Medical Residency Match Example

  • Background Data: According to the National Resident Matching Program, 1932619326 U.S. MD medical school seniors applied to residency programs and submitted preference lists.

  • Of these seniors, 1810818108 were matched with residency positions.

  • Unmatched seniors: 19326−18108=121819326 - 18108 = 1218

  • Probability of matching:   P(A)=1810819326≈0.937P(A) = \frac{18108}{19326} \approx 0.937

  • Among matched seniors, 75.6%75.6\% (0.7560.756) received one of their top three residency choices (P(B∣A)=0.756P(B \mid A) = 0.756).

  • Matched seniors who did not receive one of their top three choices: 100%−75.6%=24.4%100\% - 75.6\% = 24.4\%

  • Probability Calculation: Find the probability that a randomly selected senior was matched with a residency position AND it was one of their top three choices.   P(A and B)=P(A)⋅P(B∣A)=0.937⋅0.756≈0.708P(A \text{ and } B) = P(A) \cdot P(B \mid A) = 0.937 \cdot 0.756 \approx 0.708

  • Assessment of Unusualness: In statistics, an event is considered unusual if its probability is less than or equal to 0.050.05. Because 0.708>0.050.708 > 0.05, it would not be unusual for a randomly selected senior to match with one of their top three choices; it is highly likely.

The Addition Rule and Mutually Exclusive Events

Mutually Exclusive (Disjoint) Events

  • Mutually Exclusive: Two events AA and BB are mutually exclusive if they cannot occur at the same time. They have no outcomes in common.

    • Disjoint sets: A∩B=∅A \cap B = \emptyset

  • Not Mutually Exclusive: Two events AA and BB can occur at the same time, sharing one or more common outcomes.

Classification Examples
  • Example 1: Event AA = Randomly select a male student; Event BB = Randomly select a nursing major.

    • Classification: Not mutually exclusive (Joint). Some male students are nursing majors.

  • Example 2: Event AA = A person has blood type A; Event BB = A person has blood type O.

    • Classification: Mutually exclusive (Disjoint). A person cannot have both blood type A and blood type O simultaneously.

The Addition Rule

  • General Addition Rule: For any two events AA and BB, the probability that AA or BB will occur is:   P(A or B)=P(A)+P(B)−P(A and B)P(A \text{ or } B) = P(A) + P(B) - P(A \text{ and } B)

  • Addition Rule for Mutually Exclusive Events: If events AA and BB are mutually exclusive, P(A and B)=0P(A \text{ and } B) = 0, so the rule simplifies to:   P(A or B)=P(A)+P(B)P(A \text{ or } B) = P(A) + P(B)

  • This rule extends to any number of mutually exclusive events:   P(A or B or C… )=P(A)+P(B)+P(C)…P(A \text{ or } B \text{ or } C \dots) = P(A) + P(B) + P(C) \dots

Practical Addition Rule Examples

Election Survey Example
  • Scenario: A random sample of 400400 voters was surveyed exiting the polls regarding two ballot issues: Issue 1 and Issue 2.

  • Contingency Table:

    • Yes on Issue 1: Yes on Issue 2 =120= 120, No on Issue 2 =80= 80, Row Total =200= 200

    • No on Issue 1: Yes on Issue 2 =60= 60, No on Issue 2 =140= 140, Row Total =200= 200

    • Column Totals: Yes on Issue 2 =180= 180, No on Issue 2 =220= 220, Grand Total =400= 400

  • Question 1: Probability a voter voted Yes on Issue 1.   P(Yes Issue 1)=200400=0.50P(\text{Yes Issue 1}) = \frac{200}{400} = 0.50

  • Question 2: Probability a voter voted Yes on BOTH Issue 1 and Issue 2.   P(Yes Issue 1 and Yes Issue 2)=120400=0.30P(\text{Yes Issue 1 and Yes Issue 2}) = \frac{120}{400} = 0.30

  • Question 3: Probability a voter voted Yes on Issue 1 OR Yes on Issue 2.   P(Yes 1 or Yes 2)=P(Yes 1)+P(Yes 2)−P(Yes 1 and Yes 2)P(\text{Yes 1 or Yes 2}) = P(\text{Yes 1}) + P(\text{Yes 2}) - P(\text{Yes 1 and Yes 2})   P(Yes 1 or Yes 2)=200400+180400−120400=260400=0.65P(\text{Yes 1 or Yes 2}) = \frac{200}{400} + \frac{180}{400} - \frac{120}{400} = \frac{260}{400} = 0.65

Standard Deck Card Draw Example
  • Task: Select a card from a standard deck of 5252 cards. Find the probability that the card is a 44 (Event AA) or an Ace (Event BB).

  • Evaluation: There are 44 fours and 44 Aces in a deck. A card cannot be both a 44 and an Ace simultaneously, so the events are mutually exclusive.

  • Calculation:   P(4 or Ace)=P(4)+P(Ace)=452+452=852≈0.154P(4 \text{ or Ace}) = P(4) + P(\text{Ace}) = \frac{4}{52} + \frac{4}{52} = \frac{8}{52} \approx 0.154

Die Rolling Example
  • Task: Roll a six-sided die. Find the probability of rolling a number less than 33 or rolling an odd number.

  • Define Events:

    • Event AA (less than 33): {1,2}  ⟹  P(A)=26\{1, 2\} \implies P(A) = \frac{2}{6}

    • Event BB (odd number): {1,3,5}  ⟹  P(B)=36\{1, 3, 5\} \implies P(B) = \frac{3}{6}

    • Event A and BA \text{ and } B (less than 33 AND odd): {1}  ⟹  P(A and B)=16\{1\} \implies P(A \text{ and } B) = \frac{1}{6}

  • Evaluation: The outcome 11 satisfies both events, so they are not mutually exclusive.

  • Calculation:   P(less than 3 or odd)=P(less than 3)+P(odd)−P(less than 3 and odd)P(\text{less than 3 or odd}) = P(\text{less than 3}) + P(\text{odd}) - P(\text{less than 3 and odd})   P(less than 3 or odd)=26+36−16=46≈0.667P(\text{less than 3 or odd}) = \frac{2}{6} + \frac{3}{6} - \frac{1}{6} = \frac{4}{6} \approx 0.667

Monthly Sales Volume Example
  • Scenario: Frequency distribution of monthly sales volumes over the past 33 years (3636 months total):

    • $0−$24999\$0 - \$24999: 3 months3\,\text{months}

    • $25000−$49999\$25000 - \$49999: 5 months5\,\text{months}

    • $50000−$74999\$50000 - \$74999: 6 months6\,\text{months}

    • $75000−$99999\$75000 - \$99999 (Event AA): 7 months7\,\text{months}

    • $100000−$124999\$100000 - \$124999 (Event BB): 9 months9\,\text{months}

    • $125000−$149999\$125000 - \$149999: 2 months2\,\text{months}

    • $150000−$174999\$150000 - \$174999: 3 months3\,\text{months}

    • $175000−$199999\$175000 - \$199999: 1 month1\,\text{month}

  • Task: Find the probability that the representative sells between $75000\$75000 and $124999\$124999 next month.

  • Evaluation: Events AA and BB are mutually exclusive because monthly sales cannot fall into two distinct bracket ranges at the same time.

  • Calculation:   P(A or B)=P(A)+P(B)=736+936=1636≈0.444P(A \text{ or } B) = P(A) + P(B) = \frac{7}{36} + \frac{9}{36} = \frac{16}{36} \approx 0.444

Blood Type and Rh Factor Example
  • Scenario: A blood bank cataloged donor blood types and Rh factor status over 55 days, totaling 409409 donors:

    • Rh-Positive: Type O =156= 156, Type A =139= 139, Type B =37= 37, Type AB =12= 12, Total =344= 344

    • Rh-Negative: Type O =28= 28, Type A =25= 25, Type B =8= 8, Type AB =4= 4, Total =65= 65

    • Totals: Type O =184= 184, Type A =164= 164, Type B =45= 45, Type AB =16= 16, Grand Total =409= 409

  • Problem 1: Find the probability a donor has Type O or Type A blood.

    • Events are mutually exclusive (a donor cannot have both Type O and Type A).   P(Type O or Type A)=184409+164409=348409≈0.851P(\text{Type O or Type A}) = \frac{184}{409} + \frac{164}{409} = \frac{348}{409} \approx 0.851

  • Problem 2: Find the probability a donor has Type B blood or is Rh-negative.

    • Events are not mutually exclusive (a donor can have Type B blood and be Rh-negative simultaneously; there are 88 such donors).   P(Type B or Rh-neg)=P(Type B)+P(Rh-neg)−P(Type B and Rh-neg)P(\text{Type B or Rh-neg}) = P(\text{Type B}) + P(\text{Rh-neg}) - P(\text{Type B and Rh-neg})   P(Type B or Rh-neg)=45409+65409−8409=102409≈0.249P(\text{Type B or Rh-neg}) = \frac{45}{409} + \frac{65}{409} - \frac{8}{409} = \frac{102}{409} \approx 0.249

Additional Card Selection Exercises
  • From a standard deck of 5252 cards:

    • (a) Select a Red suit or a 55:

    • Red cards =26= 26, Fives =4= 4, Red Fives =2= 2     P=2652+452−252=2852≈0.538P = \frac{26}{52} + \frac{4}{52} - \frac{2}{52} = \frac{28}{52} \approx 0.538

    • (b) Select a Club or a Queen:

    • Clubs =13= 13, Queens =4= 4, Club Queens =1= 1     P=1352+452−152=1652≈0.308P = \frac{13}{52} + \frac{4}{52} - \frac{1}{52} = \frac{16}{52} \approx 0.308

    • (c) Select a 44 or a Face Card:

    • Fours =4= 4, Face Cards (J, Q, K) =12= 12, Fours that are Face Cards =0= 0     P=452+1252−052=1652≈0.308P = \frac{4}{52} + \frac{12}{52} - \frac{0}{52} = \frac{16}{52} \approx 0.308

Summary Table of Fundamental Probability Rules

Type / Rule

Verbal Description

Mathematical Notation / Formula

Classical Probability

Outcomes in sample space are known and equally likely.

P(E)=Number of outcomes in event ENumber of outcomes in sample spaceP(E) = \frac{\text{Number of outcomes in event } E}{\text{Number of outcomes in sample space}}

Empirical Probability

Outcome frequencies estimated from observed trials.

P(E)=Frequency of event ETotal frequency=fnP(E) = \frac{\text{Frequency of event } E}{\text{Total frequency}} = \frac{f}{n}

Range of Probabilities

Probability of any event lies between 00 and 11, inclusive.

0≤P(E)≤10 \le P(E) \le 1

Complementary Events

Set of all sample space outcomes not in event EE.

P(E′)=1−P(E)P(E') = 1 - P(E)

Multiplication Rule

Probability of two events occurring in sequence.

Dependent: P(A and B)=P(A)⋅P(B∣A)P(A \text{ and } B) = P(A) \cdot P(B \mid A)
Independent: P(A and B)=P(A)⋅P(B)P(A \text{ and } B) = P(A) \cdot P(B)

Addition Rule

Probability of at least one of two events occurring.

General: P(A or B)=P(A)+P(B)−P(A and B)P(A \text{ or } B) = P(A) + P(B) - P(A \text{ and } B)
Mutually Exclusive: P(A or B)=P(A)+P(B)P(A \text{ or } B) = P(A) + P(B)

Advanced Counting Principles, Permutations, and Combinations

Factorials

  • Factorial (n!n!): The product of all consecutive positive integers from nn down to 11.   n!=n⋅(n−1)⋅(n−2)⋅(n−3)…3⋅2⋅1n! = n \cdot (n-1) \cdot (n-2) \cdot (n-3) \dots 3 \cdot 2 \cdot 1

  • Special Factorial Definitions:

    • 1!=11! = 1

    • 0!=10! = 1

  • Examples:

    • 5!=5⋅4⋅3⋅2⋅1=1205! = 5 \cdot 4 \cdot 3 \cdot 2 \cdot 1 = 120

    • 10!=10⋅9⋅8⋅7⋅6⋅5⋅4⋅3⋅2⋅1=362880010! = 10 \cdot 9 \cdot 8 \cdot 7 \cdot 6 \cdot 5 \cdot 4 \cdot 3 \cdot 2 \cdot 1 = 3628800

  • Sudoku Puzzle Example: A blank 9×99 \times 9 Sudoku grid requires filling digits 11 through 99 in each row, column, and 3×33 \times 3 grid. The number of unique ways to fill the first row of a blank grid is:   9!=9⋅8⋅7⋅6⋅5⋅4⋅3⋅2⋅1=362880 ways9! = 9 \cdot 8 \cdot 7 \cdot 6 \cdot 5 \cdot 4 \cdot 3 \cdot 2 \cdot 1 = 362880\,\text{ways}

  • StatCrunch Command: To calculate factorials in StatCrunch, use Fact(n) under Data -> Compute -> Expression (e.g., Fact(9) yields 362880362880).

Permutations

  • Permutation: An ordered arrangement of distinct items where order matters and no item can be repeated.

  • Key Requirements:

    1. All objects are distinct.

    2. No object can be repeated.

  • Permutation Formula: The number of permutations of nn distinct objects taken rr at a time (r≤nr \le n) is:   nPr=n!(n−r)!{}_n P_r = \frac{n!}{(n-r)!}

Permutation Examples
  • 4-Digit Security Code Example:

    • Problem: Find the number of ways to form a four-digit code using digits 00 through 99 with no repeated digits.

    • Parameters: n=10n = 10 digits, r=4r = 4 selected digits.

    • Hand Calculation:     10P4=10!(10−4)!=10!6!=10⋅9⋅8⋅7⋅6!6!=10⋅9⋅8⋅7=5040{}_{10} P_4 = \frac{10!}{(10-4)!} = \frac{10!}{6!} = \frac{10 \cdot 9 \cdot 8 \cdot 7 \cdot 6!}{6!} = 10 \cdot 9 \cdot 8 \cdot 7 = 5040

    • StatCrunch Method: Type Perm(10, 4) in the Expression box to obtain 50405040

  • Car Race Finishing Positions Example:

    • Problem: 3333 race cars start the Indianapolis 500. How many ways can the cars finish first, second, and third?

    • Parameters: n=33n = 33, r=3r = 3 (positions represent specific ordered outcomes).

    • StatCrunch Calculation: Perm(33, 3) yields 32736 ways32736\,\text{ways}

Permutations with Repeated Items (Distinguishable Permutations)

  • Used when some items in a set are identical, avoiding double-counting indistinguishable arrangements.

  • Distinguishable Permutations Formula: The number of distinguishable permutations of nn objects where n1n_1 are of type 1, n2n_2 are of type 2, …, and nkn_k are of type kk is:   n!n1!⋅n2!⋅n3!…nk!\frac{n!}{n_1! \cdot n_2! \cdot n_3! \dots n_k!}   where n1+n2+n3+⋯+nk=nn_1 + n_2 + n_3 + \dots + n_k = n

Repeated Permutation Examples
  • Subdivision Planning Example:

    • Scenario: A contractor plans a subdivision consisting of 66 one-story houses, 44 two-story houses, and 22 split-level houses (1212 houses total).

    • Parameter Breakdown: n=12n = 12, n1=6n_1 = 6, n2=4n_2 = 4, n3=2n_3 = 2

    • Hand Calculation:     12!6!⋅4!⋅2!=479001600720⋅24⋅2=13860 distinguishable ways\frac{12!}{6! \cdot 4! \cdot 2!} = \frac{479001600}{720 \cdot 24 \cdot 2} = 13860\,\text{distinguishable ways}

    • StatCrunch Method: fact(12)/(fact(6)*fact(4)*fact(2)) yields 1386013860

  • Password Construction Example:

    • Scenario: Miss Sennerrasson builds a password using all 1212 letters of her last name (Sennerrasson).

    • Letter Frequency Breakdown:

    • S =3= 3

    • N =3= 3

    • E =2= 2

    • R =2= 2

    • A =1= 1

    • O =1= 1

    • Total n=12n = 12

    • Calculation:     12!3!⋅3!⋅2!⋅2!⋅1!⋅1!=3326400 distinguishable passwords\frac{12!}{3! \cdot 3! \cdot 2! \cdot 2! \cdot 1! \cdot 1!} = 3326400\,\text{distinguishable passwords}

Combinations

  • Combination: A selection of rr objects from a group of nn distinct objects without regard to order and without repetition.

  • Combination Formula:   nCr=n!(n−r)!⋅r!{}_n C_r = \frac{n!}{(n-r)! \cdot r!}   where r≤nr \le n

Combination Examples
  • Highway Construction Bids Example:

    • Scenario: A state department of transportation receives 1616 bids for an interstate highway project and hires 44 companies.

    • Evaluation: The order in which the 44 companies are hired does not matter.

    • Hand Calculation:     16C4=16!(16−4)!⋅4!=16!12!⋅4!=16⋅15⋅14⋅13⋅12!12!⋅(4⋅3⋅2⋅1)=4368024=1820 combinations{}_{16} C_4 = \frac{16!}{(16-4)! \cdot 4!} = \frac{16!}{12! \cdot 4!} = \frac{16 \cdot 15 \cdot 14 \cdot 13 \cdot 12!}{12! \cdot (4 \cdot 3 \cdot 2 \cdot 1)} = \frac{43680}{24} = 1820\,\text{combinations}

    • StatCrunch Method: Type Comb(16, 4) to obtain 18201820

Summary of Counting Principles

Summary Table of Counting Principles

Principle

Description

Formula

Fundamental Counting Principle

If event 1 occurs in mm ways and event 2 occurs in nn ways, sequential occurrences equal m⋅nm \cdot n.

m⋅nm \cdot n

Permutations of nn Objects

Total ordered arrangements of nn distinct objects.

n!n!

Permutations of nn Objects taken rr at a time

Ordered selection of rr objects from nn distinct objects (r≤nr \le n).

nPr=n!(n−r)!{}_n P_r = \frac{n!}{(n-r)!}

Distinguishable Permutations

Ordered arrangements of nn objects with sub-groups of identical items.

n!n1!⋅n2!…nk!\frac{n!}{n_1! \cdot n_2! \dots n_k!}

Combinations of nn Objects taken rr at a time

Unordered selection of rr objects from nn objects (r≤nr \le n).

nCr=n!(n−r)!⋅r!{}_n C_r = \frac{n!}{(n-r)! \cdot r!}

Probability Applications Using Counting Principles

Student Advisory Board Example
  • Scenario: A board has 1717 members. Three members are selected at random for the distinct roles of Chair, Secretary, and Webmaster.

  • Evaluation: Because roles are distinct, order is important (Permutations).

  • Calculation:   Total outcomes=17P3=4080\text{Total outcomes} = {}_{17} P_3 = 4080   P(selecting specific 3 members)=14080≈0.0002P(\text{selecting specific 3 members}) = \frac{1}{4080} \approx 0.0002

  • StatCrunch Syntax: 1/Perm(17, 3)

Class Competition Selection Example
  • Scenario: A class has 3030 students. The teacher selects 33 students at random to represent the class at a competition.

  • Evaluation: Order does not matter (Combinations).

  • Calculation:   Total groups=30C3=4060\text{Total groups} = {}_{30} C_3 = 4060   P(you and your 2 best friends are selected)=14060P(\text{you and your 2 best friends are selected}) = \frac{1}{4060}

Playing Cards Hand Example
  • Scenario: Find the probability of being dealt 55 diamonds from a standard deck of 5252 cards.

  • Step 1 (Unrestricted total hands): Choosing any 55 cards from 5252:   52C5=2598960{}_{52} C_5 = 2598960

  • Step 2 (Restricted diamond hands): Choosing 55 diamonds from the 1313 diamonds in the deck:   13C5=1287{}_{13} C_5 = 1287

  • Step 3 (Probability Calculation):   P(5 diamonds)=13C552C5=12872598960≈0.0005P(5 \text{ diamonds}) = \frac{{}_{13} C_5}{{}_{52} C_5} = \frac{1287}{2598960} \approx 0.0005

Jury Selection Example
  • Scenario: A jury of 1212 members is selected from a jury pool of 7575 people (4040 men and 3535 women).

  • Task: Find the probability that the selected jury consists of exactly 77 men and 55 women.

  • Step 1 (Total unrestricted sample space): Selecting 1212 people from 7575:   75C12{}_{75} C_{12}

  • Step 2 (Favorable outcomes with gender restrictions):

    • Selecting 77 men from 4040: 40C7{}_{40} C_7

    • Selecting 55 women from 3535: 35C5{}_{35} C_5

    • Combined selections (Fundamental Counting Principle): 40C7⋅35C5{}_{40} C_7 \cdot {}_{35} C_5

  • Step 3 (Probability Formula):   P(7 men and 5 women)=40C7⋅35C575C12≈0.2317P(7 \text{ men and } 5 \text{ women}) = \frac{{}_{40} C_7 \cdot {}_{35} C_5}{{}_{75} C_{12}} \approx 0.2317

  • StatCrunch Syntax: (Comb(40,7)*Comb(35,5))/Comb(75,12)

Discrete Probability Distributions

Random Variables: Discrete vs. Continuous

  • Random Variable: A numerical value resulting from a random process, assigning numbers to outcomes of an uncertain event.

Classification of Probability Distributions
  • Discrete Random Variable:

    • Has a finite or countable number of possible outcomes.

    • Outcomes can be listed as discrete integers (0,1,2,3,…0, 1, 2, 3, \dots).

    • Example 1: x=number of asthma attacks a patient experiences in a monthx = \text{number of asthma attacks a patient experiences in a month} (x∈{0,1,2,3,… }x \in \{0, 1, 2, 3, \dots\}).

    • Example 2: x=number of correct answers on a 20-question quizx = \text{number of correct answers on a 20-question quiz} (x∈{0,1,2,…,20}x \in \{0, 1, 2, \dots, 20\}).

  • Continuous Random Variable:

    • Has an uncountable number of possible outcomes along a continuous measurement scale or interval.

    • Example 1: x=a person’s systolic blood pressurex = \text{a person's systolic blood pressure} (can take values such as 90.290.2, 120.7120.7, 134.9 mmHg134.9\,\text{mmHg}).

    • Example 2: x=amount of time spent studying for an exam in hoursx = \text{amount of time spent studying for an exam in hours} (can take values such as 1.01.0, 1.51.5, 2.67 hours2.67\,\text{hours}).

Discrete Probability Distributions

  • Lists every possible value the discrete random variable xx can assume, paired with its corresponding probability P(x)P(x).

  • Mandatory Conditions:

    1. The probability of each value lies between 00 and 11, inclusive:      0≤P(x)≤10 \le P(x) \le 1

    2. The sum of all individual probabilities equals 11:      ∑P(x)=1\sum P(x) = 1

Construction Steps for a Discrete Probability Distribution
  1. Construct a frequency distribution for the possible outcomes of random variable xx

  2. Sum all frequencies to find n=∑fn = \sum f

  3. Calculate the relative frequency/probability of each outcome:    P(x)=fnP(x) = \frac{f}{n}

  4. Verify that 0≤P(x)≤10 \le P(x) \le 1 for all xx and that ∑P(x)=1\sum P(x) = 1

Construction Examples
  • Coin Tossing Example (22 coins tossed, x=# headsx = \text{\# heads}):

    • Sample Space: HH (x=2x=2), HT (x=1x=1), TH (x=1x=1), TT (x=0x=0)

    • Probability Distribution Table:

    • x=0x = 0: Frequency f=1f = 1, Relative Frequency P(0)=14=0.25P(0) = \frac{1}{4} = 0.25

    • x=1x = 1: Frequency f=2f = 2, Relative Frequency P(1)=24=0.50P(1) = \frac{2}{4} = 0.50

    • x=2x = 2: Frequency f=1f = 1, Relative Frequency P(2)=14=0.25P(2) = \frac{1}{4} = 0.25

  • Passive-Aggressive Personality Inventory Example:

    • An industrial psychologist administers a test to 150150 employees scoring passive-aggressive traits on a scale from 11 (extremely passive) to 55 (extremely aggressive), with 33 being neutral.

    • Frequency & Probability Distribution Table:

    • x=1x = 1: f=24  ⟹  P(1)=24150=0.16f = 24 \implies P(1) = \frac{24}{150} = 0.16

    • x=2x = 2: f=33  ⟹  P(2)=33150=0.22f = 33 \implies P(2) = \frac{33}{150} = 0.22

    • x=3x = 3: f=42  ⟹  P(3)=42150=0.28f = 42 \implies P(3) = \frac{42}{150} = 0.28

    • x=4x = 4: f=30  ⟹  P(4)=30150=0.20f = 30 \implies P(4) = \frac{30}{150} = 0.20

    • x=5x = 5: f=21  ⟹  P(5)=21150=0.14f = 21 \implies P(5) = \frac{21}{150} = 0.14

    • Verification: ∑P(x)=0.16+0.22+0.28+0.20+0.14=1.00\sum P(x) = 0.16 + 0.22 + 0.28 + 0.20 + 0.14 = 1.00

Histogram of Passive-Aggressive Traits Distribution
  • Graphical Properties: The histogram display shows bar widths of 11, meaning bar area corresponds directly to outcome probability. The distribution shape is approximately symmetric.

Verification Exercises of Probability Distributions

  • Three-Day Rain Forecast Verification:

    • Table Data: x=0  ⟹  P(0)=0.216x = 0 \implies P(0)=0.216; x=1  ⟹  P(1)=0.432x = 1 \implies P(1)=0.432; x=2  ⟹  P(2)=0.288x = 2 \implies P(2)=0.288; x=3  ⟹  P(3)=0.064x = 3 \implies P(3)=0.064

    • Verification:

    • All probabilities lie in [0,1][0, 1]

    • ∑P(x)=0.216+0.432+0.288+0.064=1.000\sum P(x) = 0.216 + 0.432 + 0.288 + 0.064 = 1.000

    • Conclusion: Is a valid probability distribution.

  • Invalid Example 1 (Sum Condition Failure):

    • Table Data: x∈{5,6,7,8}x \in \{5, 6, 7, 8\} with P(5)=0.28P(5)=0.28, P(6)=0.21P(6)=0.21, P(7)=0.43P(7)=0.43, P(8)=0.15P(8)=0.15

    • Evaluation: ∑P(x)=0.28+0.21+0.43+0.15=1.07≠1\sum P(x) = 0.28 + 0.21 + 0.43 + 0.15 = 1.07 \ne 1

    • Conclusion: Not a probability distribution.

  • Invalid Example 2 (Range Condition Failure):

    • Table Data: x∈{1,2,3,4}x \in \{1, 2, 3, 4\} with P(1)=12P(1)=\frac{1}{2}, P(2)=14P(2)=\frac{1}{4}, P(3)=54P(3)=\frac{5}{4}, P(4)=−1P(4)=-1

    • Evaluation: P(3)=1.25>1P(3) = 1.25 > 1 and P(4)=−1<0P(4) = -1 < 0. Probabilities can never be negative or exceed 11

    • Conclusion: Not a probability distribution.

Mean, Variance, Standard Deviation, and Expected Value

Mean (Expected Value) Formula
  • The mean μ\mu of a discrete probability distribution is the weighted average of the outcomes:   μ=E(x)=∑[x⋅P(x)]\mu = E(x) = \sum [x \cdot P(x)]

  • Calculation for Passive-Aggressive Personality Scores:   μ=1(0.16)+2(0.22)+3(0.28)+4(0.20)+5(0.14)\mu = 1(0.16) + 2(0.22) + 3(0.28) + 4(0.20) + 5(0.14)   μ=0.16+0.44+0.84+0.80+0.70=2.94\mu = 0.16 + 0.44 + 0.84 + 0.80 + 0.70 = 2.94

  • Interpretation: The mean score 2.942.94 indicates that the average personality trait score is neither extremely passive nor aggressive, but sits slightly below neutral (33), closer to passive.

Variance and Standard Deviation Formulas
  • Variance (σ2\sigma^2): Measures the dispersion of outcomes around the mean:   σ2=∑[(x−μ)2⋅P(x)]\sigma^2 = \sum [(x - \mu)^2 \cdot P(x)]

  • Standard Deviation (σ\sigma):   σ=σ2=∑[(x−μ)2⋅P(x)]\sigma = \sqrt{\sigma^2} = \sqrt{\sum [(x - \mu)^2 \cdot P(x)]}

  • StatCrunch Results for Passive-Aggressive Personality Data:

    • Mean μ=2.94\mu = 2.94

    • Standard Deviation σ≈1.271\sigma \approx 1.271

    • Variance σ2≈(1.271)2≈1.615\sigma^2 \approx (1.271)^2 \approx 1.615

Expected Value in Decision Analysis: Raffle Tickets Example
  • Scenario: A raffle sells 15001500 tickets at $2\$2 each for four prizes: $500\$500, $250\$250, $150\$150, and $75\$75. You purchase one ticket.

  • Profit/Gain Outcomes (x=Prize−Ticket Costx = \text{Prize} - \text{Ticket Cost}):

    • Win $500\$500 Prize: Net Gain x=$500−$2=$498x = \$500 - \$2 = \$498, P(498)=11500P(498) = \frac{1}{1500}

    • Win $250\$250 Prize: Net Gain x=$250−$2=$248x = \$250 - \$2 = \$248, P(248)=11500P(248) = \frac{1}{1500}

    • Win $150\$150 Prize: Net Gain x=$150−$2=$148x = \$150 - \$2 = \$148, P(148)=11500P(148) = \frac{1}{1500}

    • Win $75\$75 Prize: Net Gain x=$75−$2=$73x = \$75 - \$2 = \$73, P(73)=11500P(73) = \frac{1}{1500}

    • Win No Prize (14961496 tickets): Net Gain x=$0−$2=−$2x = \$0 - \$2 = -\$2, P(−2)=14961500P(-2) = \frac{1496}{1500}

  • Expected Value Calculation:   E(x)=μ=498(11500)+248(11500)+148(11500)+73(11500)+(−2)(14961500)E(x) = \mu = 498\left(\frac{1}{1500}\right) + 248\left(\frac{1}{1500}\right) + 148\left(\frac{1}{1500}\right) + 73\left(\frac{1}{1500}\right) + (-2)\left(\frac{1496}{1500}\right)   E(x)=498+248+148+73−29921500=−20251500=−$1.35E(x) = \frac{498 + 248 + 148 + 73 - 2992}{1500} = \frac{-2025}{1500} = -\$1.35

  • Interpretation: On average, you can expect to lose $1.35\$1.35 for every raffle ticket purchased.

  • StatCrunch Procedure: Enter gain outcomes in column 1 and probabilities (in decimal format) in column 2. Go to Stat -> Calculators -> Custom. Select outcomes column as Values in and probability column as Weights in, then click Compute.

Binomial Probability Distributions

Characteristics of Binomial Experiments

  1. The experiment is repeated for a fixed number of independent trials (nn).

  2. Each trial has only two possible outcomes, classified as a Success (SS) or a Failure (FF).

  3. The probability of success (pp) is constant across all trials.

  4. The random variable xx represents the count of successes in nn trials.

Notation for Binomial Experiments
  • nn: Total number of repeated independent trials

  • pp: Probability of success in a single trial

  • qq: Probability of failure in a single trial (q=1−pq = 1 - p)

  • xx: Count of successes in nn trials (x∈{0,1,2,3,…,n}x \in \{0, 1, 2, 3, \dots, n\})

Common Real-World Binomial Examples
  • COVID-19 rapid test result (Positive / Negative)

  • Multiple-choice exam question (Correct / Incorrect)

  • Survey response (Yes / No)

  • Job applicant response (Accept / Reject)

Binomial Experiment Identification Examples

  • Surgical Procedure Example:

    • Scenario: A surgical procedure has an 85%85\% (0.850.85) chance of success. A doctor performs the procedure on 88 patients. The random variable xx represents the number of successful surgeries.

    • Analysis:

    • 88 independent trials (n=8n = 8)

    • Two outcomes per patient (Success vs. Failure)

    • Constant success probability (p=0.85p = 0.85, q=1−0.85=0.15q = 1 - 0.85 = 0.15)

    • Count of successes x∈{0,1,2,3,4,5,6,7,8}x \in \{0, 1, 2, 3, 4, 5, 6, 7, 8\}

    • Conclusion: Is a binomial experiment.

  • Marble Selection Example:

    • Scenario: A jar contains 55 red, 99 blue, and 66 green marbles. You randomly select 33 marbles from the jar without replacement. The random variable xx counts the number of red marbles.

    • Analysis:

    • Without replacement, the probability of drawing a red marble changes on each draw (first draw p=520p = \frac{5}{20}; second draw probability depends on first draw outcome).

    • Trials are dependent.

    • Conclusion: Not a binomial experiment.

Binomial Probability Formula and Methods

  • Binomial Probability Formula: The probability of getting exactly xx successes in nn trials is:   P(x)=nCx⋅px⋅qn−x=n!(n−x)!⋅x!⋅px⋅qn−xP(x) = {}_n C_x \cdot p^x \cdot q^{n-x} = \frac{n!}{(n-x)! \cdot x!} \cdot p^x \cdot q^{n-x}

Rotator Cuff Surgery Example
  • Scenario: Rotator cuff surgery has a 90%90\% (0.900.90) success rate. The surgery is performed on 33 patients. Find the probability of the surgery being successful on exactly two patients.

  • Parameters: n=3n = 3, p=0.90p = 0.90, q=0.10q = 0.10, x=2x = 2

  • Method 1: Tree Diagram & Multiplication Rule:

    • Possible outcomes with x=2x = 2 successes: SSF, SFS, FSS

    • P(SSF)=(0.90)(0.90)(0.10)=0.081P(\text{SSF}) = (0.90)(0.90)(0.10) = 0.081

    • P(SFS)=(0.90)(0.10)(0.90)=0.081P(\text{SFS}) = (0.90)(0.10)(0.90) = 0.081

    • P(FSS)=(0.10)(0.90)(0.90)=0.081P(\text{FSS}) = (0.10)(0.90)(0.90) = 0.081

    • Sum: 0.081+0.081+0.081=3(0.081)=0.2430.081 + 0.081 + 0.081 = 3(0.081) = 0.243

  • Method 2: Binomial Formula:   P(2)=3C2⋅(0.90)2⋅(0.10)3−2P(2) = {}_3 C_2 \cdot (0.90)^2 \cdot (0.10)^{3-2}   P(2)=3!(3−2)!⋅2!⋅(0.81)⋅(0.10)=3⋅(0.81)⋅(0.10)=0.243P(2) = \frac{3!}{(3-2)! \cdot 2!} \cdot (0.81) \cdot (0.10) = 3 \cdot (0.81) \cdot (0.10) = 0.243

  • Method 3: StatCrunch Procedure:

    • Select Stat -> Calculators -> Binomial

    • Enter n=3n = 3, p=0.90p = 0.90, and set P(X=2)P(X = 2). Click Compute to obtain 0.2430.243

Binomial Distribution Construction and Application Examples

Pew Research Online Usage Example
Pew Research Online Usage Survey Bar Chart
  • Data: Pew Research survey indicates 48%48\% (0.480.48) of U.S. adults go online "several times a day".

  • Scenario: 66 adults are selected at random (n=6n = 6, p=0.48p = 0.48, q=0.52q = 0.52).

  • Probability Distribution Construction (x∈{0,1,2,3,4,5,6}x \in \{0, 1, 2, 3, 4, 5, 6\}):

    • P(0)=6C0(0.48)0(0.52)6=1⋅1⋅0.01977≈0.020P(0) = {}_6 C_0 (0.48)^0 (0.52)^6 = 1 \cdot 1 \cdot 0.01977 \approx 0.020

    • P(1)=6C1(0.48)1(0.52)5=6⋅(0.48)⋅(0.03802)≈0.109P(1) = {}_6 C_1 (0.48)^1 (0.52)^5 = 6 \cdot (0.48) \cdot (0.03802) \approx 0.109

    • P(2)=6C2(0.48)2(0.52)4=15⋅(0.2304)⋅(0.07312)≈0.253P(2) = {}_6 C_2 (0.48)^2 (0.52)^4 = 15 \cdot (0.2304) \cdot (0.07312) \approx 0.253

    • P(3)=6C3(0.48)3(0.52)3=20⋅(0.11059)⋅(0.14061)≈0.311P(3) = {}_6 C_3 (0.48)^3 (0.52)^3 = 20 \cdot (0.11059) \cdot (0.14061) \approx 0.311

    • P(4)=6C4(0.48)4(0.52)2=15⋅(0.05308)⋅(0.2704)≈0.215P(4) = {}_6 C_4 (0.48)^4 (0.52)^2 = 15 \cdot (0.05308) \cdot (0.2704) \approx 0.215

    • P(5)=6C5(0.48)5(0.52)1=6⋅(0.02548)⋅(0.52)≈0.079P(5) = {}_6 C_5 (0.48)^5 (0.52)^1 = 6 \cdot (0.02548) \cdot (0.52) \approx 0.079

    • P(6)=6C6(0.48)6(0.52)0=1⋅(0.01223)⋅1≈0.012P(6) = {}_6 C_6 (0.48)^6 (0.52)^0 = 1 \cdot (0.01223) \cdot 1 \approx 0.012

  • Verification: ∑P(x)=0.020+0.109+0.253+0.311+0.215+0.079+0.012=0.999\sum P(x) = 0.020 + 0.109 + 0.253 + 0.311 + 0.215 + 0.079 + 0.012 = 0.999 (differs from 11 due strictly to rounding).

Digital Device Fatigue Example
  • Scenario: Survey by The Harris Poll shows 75%75\% (0.750.75) of U.S. adults experience digital device fatigue. You randomly select 100100 adults.

  • Problem: Find the probability that exactly 66 experience digital device fatigue.

  • Parameters: n=100n = 100, p=0.75p = 0.75, x=66x = 66

  • StatCrunch Calculation: In the Binomial Calculator, set n=100n = 100, p=0.75p = 0.75, P(X=66)P(X = 66). Result =0.01116782= 0.01116782

U.S. Economy Problem Survey Example
  • Scenario: Survey by Ipsos finds 22%22\% (0.220.22) of U.S. adults state the economy, unemployment, and jobs are the most important national problem. Randomly select 44 adults (n=4n = 4, p=0.22p = 0.22, q=0.78q = 0.78).

  • Case 1: Exactly two respond yes (x=2x = 2):   P(2)=0.17667936P(2) = 0.17667936

  • Case 2: At least two respond yes (x≥2x \ge 2):   P(x≥2)=P(2)+P(3)+P(4)=0.21224368P(x \ge 2) = P(2) + P(3) + P(4) = 0.21224368

  • Case 3: Fewer than two respond yes (x<2x < 2):   P(x<2)=P(0)+P(1)=0.78775632≈0.788P(x < 2) = P(0) + P(1) = 0.78775632 \approx 0.788

Cancer Survivors Graphing Example
  • Scenario: American Cancer Society reports 64%64\% (0.640.64) of cancer survivors are 6565 years or older. Select 66 survivors at random (n=6n = 6, p=0.64p = 0.64, q=0.36q = 0.36).

  • Probability Values:

    • P(0)=0.002P(0) = 0.002

    • P(1)=0.023P(1) = 0.023

    • P(2)=0.103P(2) = 0.103

    • P(3)=0.245P(3) = 0.245

    • P(4)=0.326P(4) = 0.326

    • P(5)=0.232P(5) = 0.232

    • P(6)=0.069P(6) = 0.069

Cancer Survivors Probability Histogram
  • Identification of Unusual Events: An event is statistically unusual if its probability is less than 0.050.05. Here, x=0x = 0 (P=0.002P = 0.002) and x=1x = 1 (P=0.023P = 0.023) are unusual outcomes.

Mean, Variance, and Standard Deviation of Binomial Distributions

Formulas
  • Mean:   μ=n⋅p\mu = n \cdot p

  • Variance:   σ2=n⋅p⋅q\sigma^2 = n \cdot p \cdot q

  • Standard Deviation:   σ=n⋅p⋅q\sigma = \sqrt{n \cdot p \cdot q}

June Cloudiness Example in Pittsburgh, PA
  • Scenario: NOAA data indicates that in Pittsburgh, Pennsylvania, about 56%56\% (0.560.56) of days in June are cloudy. June has 3030 days (n=30n = 30, p=0.56p = 0.56, q=0.44q = 0.44).

  • Mean Calculation:   μ=n⋅p=30⋅(0.56)=16.8 days\mu = n \cdot p = 30 \cdot (0.56) = 16.8\,\text{days}

  • Variance Calculation:   σ2=n⋅p⋅q=30⋅(0.56)⋅(0.44)=7.392≈7.4\sigma^2 = n \cdot p \cdot q = 30 \cdot (0.56) \cdot (0.44) = 7.392 \approx 7.4

  • Standard Deviation Calculation:   σ=7.392≈2.7 days\sigma = \sqrt{7.392} \approx 2.7\,\text{days}

  • Interpretation: On average, Pittsburgh experiences 16.816.8 cloudy days during June, with a standard deviation of about 2.72.7 days.

  • Usual Range Analysis (μ±2σ\mu \pm 2\sigma):

Usual Range Interval Diagram
  • Lower Bound:     16.8−2(2.7)=16.8−5.4=11.4 days16.8 - 2(2.7) = 16.8 - 5.4 = 11.4\,\text{days}

  • Upper Bound:     16.8+2(2.7)=16.8+5.4=22.2 days16.8 + 2(2.7) = 16.8 + 5.4 = 22.2\,\text{days}

  • Evaluation of Unusual Outcomes:

    • A June with 1111 or fewer cloudy days is unusual.

    • A June with 2323 or more cloudy days is also unusual.