Elementary Statistics Exam 1 Study Guide: Chapters 1–4 (Triola)

Chapter 1 – Introduction to Statistics

Big Ideas and Key Concepts

Statistics is the science of collecting, organizing, summarizing, and interpreting data. Understanding the distinction between populations and samples, as well as parameters and statistics, is fundamental. Study design, including awareness of bias and confounding, is crucial for valid conclusions.

• Population: The entire group being studied.

• Sample: A subset of the population used to draw conclusions.

• Parameter: A numerical measurement describing a population (often denoted by Greek letters such as μ, σ, p).

• Statistic: A numerical measurement describing a sample (often denoted by x̄, s, p̂).

• Variables: Can be quantitative (numerical) or qualitative (categorical).

• Discrete data: Countable values (e.g., number of students).

• Continuous data: Any value within an interval (e.g., time, weight).

• Binary data: Exactly two outcomes (e.g., yes/no).

Levels of Measurement

• Nominal: Names/labels; no order. Example: Eye color.

• Ordinal: Categories with order, but differences aren’t measurable. Example: Class rank.

• Interval: Numerical; equal spacing; no true zero. Example: Temperature in °C.

• Ratio: Numerical; equal spacing; true zero; ratios make sense. Example: Height, weight.

Ethics, Bias, and Stakeholder Influence

• Stakeholders: May influence study design or interpretation.

• Bias: Systematic error that makes results inaccurate (e.g., bad sampling, leading questions).

• Sampling bias: Sample is not representative due to undercoverage or nonresponse.

• Confounding variable: Related to both explanatory and response variables, distorting relationships.

Statistical vs Practical Significance

• Statistical significance: Indicates whether a result is unlikely due to chance (often using α = 0.05).

• Practical significance: Considers whether the effect size is meaningful in real life.

• Example: Large samples can yield statistically significant but practically unimportant results.

Chapter 2 – Exploring Data: Tables and Graphs

Frequency Distributions

Frequency distributions organize data into classes, showing how many values fall into each class. Key terms include class limits, class width, class midpoint, and class boundaries.

• Frequency: Number of values in a class.

• Class limits: Smallest and largest values allowed in a class.

• Class width: Difference between consecutive lower class limits.

• Class midpoint: $(\text{Lower class limit} + \text{Upper class limit}) / 2$

• Class boundaries: Numbers separating classes with no gaps (e.g., 0.5 below/above class limits).

• Relative frequency: $\text{RF} = \frac{\text{class frequency}}{n}$

• Cumulative frequency: Running total of frequencies up to a class.

Step-by-Step: Constructing a Grouped Frequency Distribution

1. Find minimum and maximum values.

2. Decide number of classes (usually 5–10).

3. Compute class width: $(\text{max} - \text{min}) / \text{number of classes}$, round up.

4. Choose first lower class limit.

5. Create classes by adding class width each time.

6. Count frequencies for each class.

7. Optional: Add relative and cumulative frequencies.

Example: Grouped Frequency Distribution

Data (minutes): 4, 10, 12, 15, 15, 20, 25, 30, 35, 45, 60, 80. Class width = 15, starting at 0.

Class	Frequency
0–14	3
15–29	4
30–44	2
45–59	1
60–74	1
75–89	1

Graphs and Their Uses

• Histogram: Quantitative data; bars touch; shows shape and outliers.

• Frequency polygon: Line graph connecting class midpoints.

• Ogive: Graph of cumulative frequency.

• Dotplot: Shows each value; good for small datasets.

• Stem-and-leaf plot: Displays distribution and exact values.

• Pareto chart: Categorical data; bars ordered high to low.

• Pie chart: Shows proportions; best with few categories.

• Time series graph: Data over time; shows trends.

• Scatterplot: Two quantitative variables; shows association.

Scatterplots and Correlation

• Positive association: As x increases, y increases.

• Negative association: As x increases, y decreases.

• Linear correlation coefficient (r): Measures strength and direction; $-1 \leq r \leq 1$.

• Statistical significance: Use critical value table or p-value; if $|r|$ is large enough, correlation is significant.

• Line of best fit: $y = mx + b$

Describing Distribution Shape

• Normal (bell-shaped): Symmetric, mound-shaped.

• Uniform: Frequencies equal across classes.

• Skewed right: Long tail to the right (mean > median).

• Skewed left: Long tail to the left (mean < median).

• Outliers: Values far from the rest; check with z-scores or boxplots.

Misleading Graphs

• Non-zero axis exaggerates differences.

• Changing scale or intervals can hide/exaggerate trends.

• 3-D effects distort perception.

• Cherry-picking time windows changes the story.

Chapter 3 – Descriptive Statistics

Measures of Center

Measures of center summarize the typical value in a dataset. The mean is sensitive to outliers, while the median is resistant.

• Mean (sample): $\bar{x} = \frac{\sum x}{n}$

• Mean (population): $\mu = \frac{\sum x}{N}$

• Median: Middle value when data are sorted.

• Mode: Most frequent value; can be unimodal, bimodal, or none.

• Midrange: $\text{MR} = \frac{\text{min} + \text{max}}{2}$

Example: Mean, Median, Mode, Midrange

Data: 5, 7, 7, 10, 12

• Mean: $(5+7+7+10+12)/5 = 41/5 = 8.2$

• Median: 7 (middle value)

• Mode: 7

• Midrange: $(5+12)/2 = 8.5$

Measures of Variation

• Range: $R = \text{max} - \text{min}$

• Sample variance: $s^2 = \frac{\sum (x - \bar{x})^2}{n - 1}$

• Sample standard deviation: $s = \sqrt{\frac{\sum (x - \bar{x})^2}{n - 1}}$

• Population variance: $\sigma^2 = \frac{\sum (x - \mu)^2}{N}$

• Population standard deviation: $\sigma = \sqrt{\frac{\sum (x - \mu)^2}{N}}$

Example: Sample Standard Deviation

Data: 2, 4, 4, 4, 5

• Mean: $\bar{x} = 3.8$

• Deviations: -1.8, 0.2, 0.2, 0.2, 1.2

• Squares: 3.24, 0.04, 0.04, 0.04, 1.44; sum = 4.80

• $s^2 = 4.80/4 = 1.20$

• $s = \sqrt{1.20} \approx 1.1$

Coefficient of Variation (CV)

• CV: $\text{CV} = \frac{s}{\bar{x}} \times 100\%$

• Used to compare variation between datasets with different units or means.

z-Scores and Unusual Values

• z-score: $z = \frac{x - \bar{x}}{s}$ or $z = \frac{x - \mu}{\sigma}$

• Indicates how many standard deviations x is from the mean.

• Rule of thumb: $|z| \geq 2$ is unusual; $|z| \geq 3$ is very unusual.

• Empirical Rule: For normal distributions: ~68% within 1σ, ~95% within 2σ, ~99.7% within 3σ.

Percentiles and Quartiles

• Percentile: Separates lowest k% of data from the rest.

• Quartiles: Q1 = 25th, Q2 = 50th (median), Q3 = 75th percentile.

Step-by-Step: Percentile Rank

1. Sort data.

2. Count values less than x (L).

3. Count values equal to x (E).

4. Percentile rank: $100 \times \frac{L + 0.5E}{n}$

Example: Percentile Rank

Data: 10, 12, 12, 15, 18, 20, 20, 22, 30, 35. x = 20.

• L = 5, E = 2, n = 10

• Percentile rank: $100 \times \frac{5 + 1}{10} = 60$

Step-by-Step: k-th Percentile Value

1. Sort data.

2. Compute locator: $L = \frac{k}{100} \times n$

3. If L is whole, average L-th and (L+1)-th values; else, round up to next integer and use that position.

Example: 75th Percentile

Data: 10, 12, 12, 15, 18, 20, 20, 22, 30, 35. 75th percentile.

• L = $0.75 \times 10 = 7.5$ → round up to 8th position.

• 8th value is 22; P75 = 22.

Five-Number Summary and Boxplots

• Five-number summary: min, Q1, median (Q2), Q3, max.

• IQR: $\text{IQR} = Q3 - Q1$

• Outlier fences: Lower fence = $Q1 - 1.5 \cdot \text{IQR}$; Upper fence = $Q3 + 1.5 \cdot \text{IQR}$

• Boxplot: Box from Q1 to Q3, line at median, whiskers to most extreme non-outlier values.

Chapter 4 – Probability

Probability Basics

Probability quantifies the likelihood of an event. The sample space is the set of all possible outcomes.

• Probability of event A: $P(A) = \frac{\text{# ways A occurs}}{\text{total # outcomes}}$ (for equally likely outcomes)

• $0 \leq P(A) \leq 1$

• Sample space: All possible outcomes.

Complement Rule

• Complement: $P(A^c) = 1 - P(A)$

• "At least one" often uses complement: $1 - P(\text{none})$

Addition Rule

• General addition rule: $P(A \cup B) = P(A) + P(B) - P(A \cap B)$

• Mutually exclusive: If A and B cannot happen together: $P(A \cup B) = P(A) + P(B)$

• $P(A \cap B) = 0$ for mutually exclusive events.

Conditional Probability and Independence

• Conditional probability: $P(A | B) = \frac{P(A \cap B)}{P(B)}$ (if $P(B) > 0$)

• Multiplication rule: $P(A \cap B) = P(A) \cdot P(B | A) = P(B) \cdot P(A | B)$

• Independent events: $P(A | B) = P(A)$

• Independent multiplication: $P(A \cap B) = P(A) \cdot P(B)$

Step-by-Step: "At Least One" with Independent Trials

1. Identify single-trial success probability p.

2. Compute failure probability q = 1 − p.

3. For n trials: $P(\text{no successes}) = q^n$

4. $P(\text{at least one success}) = 1 - q^n$

Example: At Least One Success

Free throw success probability p = 0.30, n = 5 shots.

• q = 0.70

• $P(\text{no successes}) = 0.70^5 = 0.16807$

• $P(\text{at least one}) = 1 - 0.16807 = 0.83193 \approx 0.832$

Counting Methods

• Multiplication rule: If one action can occur in m ways and another in n ways, total ways = m·n

• Factorial: $n! = n \cdot (n-1) \cdot (n-2) \cdots 2 \cdot 1$, $0! = 1$

• Permutations: $P(n, r) = \frac{n!}{(n-r)!}$ (order matters)

• Combinations: $C(n, r) = \frac{n!}{r! (n-r)!}$ (order does not matter)

• Check that $n \geq r$

Diagnostic/Contingency Tables

• Read row, column, and grand totals.

• Convert counts to probabilities by dividing by grand total.

• Conditional probabilities: restrict to the 'given' group.

• Example: $P(\text{Disease} | \text{Positive}) = \frac{\text{true positives}}{\text{all positives}}$

Quick Problem Routines

• Mean: Add values $(\sum x)$, divide by n.

• Median: Sort, pick middle (or average two middles).

• Std dev (sample): Mean → deviations → square → sum → divide by (n−1) → sqrt.

• CV: Compute s and x̄, then $(s/x̄) \times 100\%$.

• z-score: $(x−\text{mean})/s$.

• Grouped frequency table: Choose class width, build classes, tally frequencies, (optional) RF/CF.

• At least one (independent): $1 - (1-p)^n$.

• Union probability: Use addition rule: $P(A)+P(B)-P(A\cap B)$.

• Conditional probability: $P(A|B)=P(A\cap B)/P(B)$.

• Perm vs comb: Order matters? yes→$P(n,r)$; no→$C(n,r)$.

Symbol Cheat Sheet

Symbol	Meaning
n	Sample size
N	Population size
x	A single data value
Σx	Sum of all data values
x̄	Sample mean
μ	Population mean
s	Sample standard deviation
s²	Sample variance
σ	Population standard deviation
σ²	Population variance
min, max	Smallest and largest data values
R	Range = max − min
MR	Midrange = (min + max)/2
z	z-score = (x − mean)/std dev
Q1, Q2, Q3	Quartiles (25th, 50th, 75th percentiles)
IQR	Interquartile range = Q3 − Q1
p	Probability of success
p̂	Sample proportion
P(A)	Probability of event A
Aᶜ	Complement of A (not A)
A ∪ B	Union: A or B (or both)
A ∩ B	Intersection: A and B
P(A|B)	Conditional probability of A given B
!	Factorial
P(n,r)	Permutations
C(n,r)	Combinations

Common Wording and Their Statistical Meaning

• "at least one": $1 - P(\text{none})$

• "exactly one": Count specific cases (often $n \cdot p \cdot q^{n-1}$ in binomial)

• "A or B": Union; use addition rule

• "A and B": Intersection; use multiplication rule

• "given": Conditional probability; restrict to the given group

• "random": Every member has equal chance of selection

• "unbiased": No systematic error; method doesn’t favor outcomes