MGU Mathematics Series — High School Statistics

Part I: Data, Questions, and Statistical Thinking

Data · Variability · Populations · Samples · Bias · Ethics

Chapter 1 of 14

MGU Mathematics Series · High School Statistics · Library Textbook Edition

Chapter 1:
What Statistics
Studies

Data, variability, uncertainty, individuals, variables, populations, samples, parameters, statistics, observational studies, experiments, descriptive and inferential reasoning, bias, and data ethics — the conceptual foundation every subsequent chapter requires

■ Math Level: High School Statistics ■

■ 14 Sections • Part I of 5 ■

Follows Algebra I and Geometry · Prepares students for Chapter 2 through Chapter 14

Chapter 1 — Front Matter

Purpose and Overview

This chapter is the opening gateway into statistics. Before introducing a single formula or graph, it builds the foundation that the rest of the book rests on: what statistics is, why it matters, and how it thinks. Statistics is not a collection of calculation tricks. It is a discipline for reasoning carefully about data, variability, uncertainty, and evidence. Students who finish this chapter should understand that the central questions of statistics are not only "What is the answer?" but also "How reliable is the answer?" and "What does the data actually suggest?"

By the end of this chapter, you should be able to explain what statistics studies, distinguish the key types of data and variables, identify populations and samples, recognize the difference between observational studies and experiments, and describe the role of bias and ethics in statistical work. These ideas will carry through every topic in this book.

Opening Question

A school district surveys 200 students and reports that 68% feel safe on campus. A second survey of 150 different students at the same school reports 52% feel safe. Both surveys were conducted in the same month.

Which result should you trust? Should you trust either one? What would you need to know to decide? This is a genuine statistical question — the kind statistics is built to help you think through. By the end of this chapter, you will have the vocabulary and framework to begin answering it. By the end of this book, you will have the tools to answer it rigorously.

14Sections

14Objectives

20Key Terms

8+Worked Examples

Learning Objectives

1Explain what statistics studies and how it differs from pure arithmetic or algebra.
2Define data, variability, and uncertainty in statistical terms.
3Identify individuals and variables in a data set.
4Distinguish between categorical and quantitative variables.
5Distinguish between a population and a sample.
6Distinguish between a parameter and a statistic.
7Distinguish between observational studies and experiments.
8Distinguish between descriptive and inferential statistics.
9Identify whether a question is a statistical question.
10Explain what bias is and how it differs from variability.
11Apply basic principles of data ethics to interpret results responsibly.
12Recognize where statistics appears in science, finance, medicine, and public life.
13Identify common early mistakes in statistical thinking.
14Prepare to organize and display data using tables and graphs.

Chapter 1 — Contents

Sections at a Glance

1.1What Statistics StudiesDescribing data; drawing conclusions; how statistics differs from pure computation; algebra and arithmetic connection
1.2Data, Variability, and UncertaintyData as recorded observations; variability as the central fact; uncertainty as quantified, not eliminated
1.3Individuals and VariablesIndividuals as the units studied; variables as characteristics that differ; rows and columns in a data set
1.4Categorical and Quantitative VariablesCategorical labels vs. numerical measurements; discrete vs. continuous; warning about numbers that are actually categories
1.5Populations and SamplesPopulation as the group of interest; sample as the subset studied; census; why the distinction matters
1.6Parameters and StatisticsParameters describe populations; statistics describe samples; standard notation: μ, σ, p vs. x̄, s, p̂
1.7Observational Studies and ExperimentsObservational studies measure without intervening; experiments use random assignment; confounding variables; causation
1.8Descriptive Statistics and Inferential StatisticsDescriptive work summarizes data at hand; inferential work draws conclusions beyond the sample
1.9Statistical QuestionsStatistical questions expect variability; non-statistical questions have single fixed answers; recognizing the distinction
1.10Bias, Variability, and ContextBias as systematic error; types of bias; why larger samples cannot fix bias; context shapes meaning
1.11Data Ethics and Responsible InterpretationPrivacy; informed consent; honest representation; overgeneralization; limits of models; who is not represented
1.12Statistics in Science, Finance, Medicine, and Public LifeClinical trials; polling; economic modeling; forensics; sports analytics; quality control
1.13Common Early Statistics MistakesSeven recurring conceptual errors; explanations and corrections; prevention strategies
1.14Preparing to Organize and Display DataShape, center, spread, unusual features; preview of Chapters 3–5; the two problem-solving frameworks

Section 1.1

What Statistics Studies

Statistics is the science of learning from data. That description sounds simple, but it contains a great deal. To learn from data means more than reading numbers off a table. It means asking where the data came from, what they measure, how much they vary, whether they represent what we actually care about, and what conclusions they do — and do not — support.

Consider a basic example. Suppose you want to know the average height of students at your school. You could measure everyone and calculate the mean. That is arithmetic. But suppose you want to know the average height of all teenagers in your state. You cannot measure everyone. You will have to work with a smaller group — a sample — and use that sample to make an educated estimate. That estimation process, including all the uncertainty it carries, is the heart of statistics.

What Statistics Studies	Description	Example
Describing data	Organizing, displaying, and summarizing data honestly — shape, center, spread, unusual features	Average test score of 30 students in a class
Drawing conclusions	Using a sample to estimate or test claims about a larger population, while quantifying uncertainty	Estimating the average test score of all students in a district from a sample

What Statistics Is Not

Statistics is not the same as finding the average of a list of numbers. It is not data entry, and it is not software operation. Statistics is the reasoning that wraps around the numbers — the questions asked before the data are collected, the decisions made about how to collect them, the care taken in displaying and interpreting them, and the intellectual honesty applied when drawing conclusions.

Practice 1.1Concept Check

In your own words, explain why finding an average is not the same as doing statistics.
A student says, "Statistics is just math with data." Write two or three sentences explaining what this description is missing.
Give one example of a situation from daily life where statistical reasoning would help you make a better decision. What would the data be? What question would statistics help answer?

Section 1.2

Data, Variability, and Uncertainty

Data are measurements or observations recorded about individuals or things. Data can take many forms: survey responses, measurements, counts, categories, rankings, timestamps, or any other recorded observation. A single data value is almost never very informative. It is the pattern across many values — how they are distributed, how they vary, and how they compare — that reveals something meaningful.

Variability is the central fact that makes statistics necessary. If every person in a population were identical, you could measure one person and know the whole population. But in the real world, people's heights differ. Their test scores differ. Their opinions differ. Variability is everywhere, and statistics exists precisely to measure, characterize, and reason about it.

Uncertainty is the consequence of variability. Because individuals differ and because samples are not perfect mirrors of their populations, statistical conclusions always carry some degree of uncertainty. Statistics does not eliminate uncertainty — it quantifies it. When a pollster reports that 53% of survey respondents prefer a candidate "with a margin of error of plus or minus 3 percentage points," they are expressing a measured, quantified form of uncertainty. That is not a weakness of the method; it is the method working correctly.

Student Note — Variability Is Not a Problem to Eliminate

Beginning students sometimes believe that variability means something went wrong — that if you had better data or a larger sample, the variability would disappear. This is a misconception. Variability is a real property of the world. A larger sample gives a more precise estimate of variability, but it does not eliminate it. Statistics teaches you to work honestly with variability rather than pretending it does not exist.

Practice 1.2Concept Check

Why is a single data value usually not very informative by itself?
Give an example of variability in each of the following contexts: (a) heights of plants in a garden, (b) arrival times of buses at a stop, (c) scores on a quiz.
A news report says a survey found that 61% of people support a new law, "with a margin of error of ±4 percentage points." What does the margin of error tell you about the uncertainty of the result?

Section 1.3

Individuals and Variables

When statisticians collect data, they record information about a collection of individuals. An individual is not necessarily a person. In statistics, an individual is any unit being studied. A set of individuals might be students in a high school, cars produced in a factory, hospitals in a region, countries in a database, trees in a forest, or days in a year.

The thing you record about each individual is called a variable. A variable is any characteristic that can take different values from one individual to another. In a typical data set, the individuals are the rows, and the variables are the columns. Each cell in the table records the value of one variable for one individual.

Example 1.3.1

Identifying individuals and variables in a data set

SituationA researcher collects data from five students. For each student, she records: name, grade level, hours studied per week, final exam score, and whether the student participated in a tutoring program.
IdentifyThe individuals are the five students — the units being studied. Each student occupies one row in the data set.
IdentifyThe variables are the five characteristics recorded: name, grade level, hours studied per week, final exam score, and tutoring participation. Each variable occupies one column.
Display Data table layout: Rows = individuals (one student per row) Columns = variables (name | grade | hours | score | tutoring) Each cell = one student's value for one variable

Individuals: five students. Variables: name, grade level, hours studied per week, exam score, tutoring participation.

Practice 1.3Identifying Individuals and Variables

For each scenario below, identify the individuals and list at least three variables that might be recorded.

A health clinic records information about each patient visit.
A car manufacturer tracks information about each vehicle sold.
A social scientist studies smartphone use among teenagers.
An agricultural researcher compares the growth of plants under different conditions.

Section 1.4

Categorical and Quantitative Variables

Not all variables are alike. Some variables describe which category or group an individual belongs to. Others record a numerical measurement. This distinction matters because different types of variables are displayed and analyzed differently.

Categorical variables (also called qualitative variables) record group membership, category, type, or label. The values of a categorical variable are not numbers used for arithmetic — they are names or labels. Even when categories are represented by numbers — such as using 1 for freshman, 2 for sophomore, 3 for junior, and 4 for senior — those numbers are labels, not measurements.

Quantitative variables (also called numerical variables) record measurements or counts that can be meaningfully used in arithmetic. Quantitative variables are further divided: discrete variables take whole-number values obtained by counting, while continuous variables can take any value within a range and are obtained by measuring.

Variable Type	Values Are	Examples
Categorical	Names, labels, or group memberships — arithmetic does not apply	Eye color, favorite season, political party, yes/no responses
Quantitative — Discrete	Whole numbers from counting	Number of siblings, number of cars in a lot, text messages sent
Quantitative — Continuous	Any value in a range, from measuring	Height, weight, temperature, time, income

Student Warning — Numbers That Are Actually Categories

ZIP codes, jersey numbers, ID numbers, and phone numbers are all made of digits, but they are categorical variables — not quantitative ones. You would not calculate the average ZIP code of a neighborhood. Before treating any numerical-looking value as a quantitative variable, ask yourself: would arithmetic on these numbers produce a meaningful result? If not, the variable is categorical.

Example 1.4.1

Identifying variable types

ProblemA survey collects: (a) age in years, (b) preferred social media platform, (c) hours per day on social media, (d) home state, (e) subscriber status. Classify each variable.
ThinkFor each variable, ask: are the values labels or numbers used in arithmetic? If arithmetic, do the values come from counting (discrete) or measuring (continuous)?
Analyze (a) Age in years → quantitative, discrete (whole years) (b) Platform preferred → categorical (names of platforms) (c) Hours per day → quantitative, continuous (measured) (d) Home state → categorical (state names are labels) (e) Subscriber status → categorical (subscribed / not subscribed)

Quantitative: (a) and (c). Categorical: (b), (d), and (e).

Practice 1.4Variable Classification

Classify each variable as categorical or quantitative. If quantitative, state whether it is discrete or continuous.

Number of movies watched per month
Genre of favorite movie
Race finish time in minutes
Political party registration
Weight of a newborn in kg
Letter grade in a course
Number of customer complaints per week
Country of birth
Blood pressure in mmHg
Student ID number

Section 1.5

Populations and Samples

One of the most important distinctions in statistics is between the group you are interested in and the group you actually study.

The population is the entire group of individuals about which you want to draw a conclusion. The population is defined by the question being asked. A sample is the subset of the population that you actually observe, measure, or survey. A census is the special case in which every individual in the population is measured — valuable when possible, but often impractical.

Key Idea

The population depends on the question, not on the study. If a researcher wants to know about all adults in the United States, the population is all U.S. adults. If a different researcher wants to know about adults in one city, the population is all adults in that city. Always ask: for this question, who counts as part of the population of interest?

Example 1.5.1

Identifying population and sample

ProblemA school principal wants to know whether students at Jefferson High School support a new cell phone policy. She surveys every student in three randomly selected homeroom classes.
ThinkThe population is the full group the principal wants to draw a conclusion about. The sample is the group actually surveyed.
Analyze Population: all students at Jefferson High School Sample: students in the three randomly selected homeroom classes
CheckThe sample is a subset of the population — it includes only some students at Jefferson High School, not all of them.

Population: all students at Jefferson High School. Sample: the three surveyed homeroom classes.

Practice 1.5Population and Sample

For each scenario, identify (a) the population, (b) the sample, and (c) what is being measured or recorded.

A pollster surveys 800 registered voters in a congressional district and finds that 54% support Candidate A.
A manufacturer tests 50 light bulbs from a production run of 10,000. Average lifetime: 1,150 hours.
A researcher surveys 300 adults selected from a city's tax records about commuting habits. She finds 38% commute by public transit.
A marine biologist collects 200 fish from a lake and measures their lengths. Average length: 24 centimeters.
A teacher calculates the average score on a quiz for the 28 students who took it today.

Section 1.6

Parameters and Statistics

Because samples are used to draw conclusions about populations, statisticians must carefully distinguish between numbers that describe populations and numbers that describe samples.

A parameter is a number that describes a characteristic of a population. Parameters are typically fixed — for a given population at a given time, there is one true value of a parameter, even if we do not know what it is. A statistic (used as a noun referring to a number, not the field of study) is a number that describes a characteristic of a sample. Because different samples from the same population will typically produce different statistics, statistics vary from sample to sample. The goal of inferential statistics is to use a sample statistic to estimate or test claims about an unknown population parameter.

Quantity	Population (Parameter)	Sample (Statistic)
Mean	μ (mu)	x̄ (x-bar)
Standard deviation	σ (sigma)	s
Proportion	p	p̂ (p-hat)
Size	N	n

Memory Aid — Statistics vs. Parameters

Both "sample" and "statistic" begin with the letter S. Both "population" and "parameter" begin with the letter P. When a value is calculated from a sample, it is a statistic. When it describes a population, it is a parameter.

Example 1.6.1

Parameter or statistic?

ProblemIn a survey of 400 U.S. adults, 62% reported owning a smartphone. Is 62% a parameter or a statistic?
ThinkWas the 62% calculated from the entire population (all U.S. adults) or from a sample (400 selected adults)?
Analyze 62% was calculated from a sample of 400 adults. → It is a statistic: p̂ = 0.62 The population parameter p is the true proportion of all U.S. adults who own a smartphone. That value is unknown. The survey was conducted precisely to estimate it.

62% is a statistic (p̂ = 0.62). The corresponding population parameter p is unknown and is what the survey estimates.

Practice 1.6Parameter or Statistic?

For each value, state whether it is a parameter or a statistic and explain your reasoning.

A survey of 1,200 likely voters finds that 47% support a ballot measure.
The average income of all 85 employees at a small company is $61,000.
A sample of 60 jars of peanut butter from a factory shows an average weight of 16.3 ounces.
The proportion of all U.S. residents who have a college degree, as recorded in a complete census.

Section 1.7

Observational Studies and Experiments

Another fundamental distinction in statistics is how data are produced. The method of data collection determines what kinds of conclusions can be drawn.

An observational study measures or records variables without intervening in what happens. Researchers observe individuals as they are, without assigning them to any condition, treatment, or group. Observational studies can reveal associations between variables — but they cannot establish that one variable causes another.

An experiment deliberately imposes a treatment or condition on individuals and measures the response. Researchers assign individuals to groups, with some receiving the treatment of interest and others serving as a control. The key word is random assignment. When individuals are randomly assigned to treatment groups, the groups are likely to be similar to each other at the start, and any observed difference in outcomes after the treatment can more credibly be attributed to the treatment itself.

Association Does Not Mean Causation

People who sleep with their shoes on are more likely to wake up with headaches. Does sleeping with shoes cause headaches? No — both behaviors are caused by a hidden third factor. This kind of hidden third factor is called a confounding variable. Because observational studies do not randomly assign individuals to conditions, they cannot rule out confounding variables. This is why even very strong associations in observational studies cannot prove that one thing causes another.

Example 1.7.1

Observational study or experiment?

Situation AResearchers track 500 runners over three years. They record each runner's weekly mileage and observe how often each experiences knee injuries. No runner is told how far to run.
Analyze AResearchers observe runners as they are, without assigning them to any mileage group. This is an observational study. It can reveal an association between mileage and injuries, but it cannot prove that high mileage causes injuries — perhaps runners with stronger joints naturally run more and also have fewer injuries.
Situation BResearchers randomly assign 100 patients with a skin condition to two groups. One group receives a new topical cream; the other receives a placebo cream. After four weeks, improvement rates are compared.
Analyze BResearchers impose a treatment (cream vs. placebo) using random assignment. This is an experiment. Because assignment was random, the two groups should be similar at the start; any difference in outcome can more credibly be attributed to the cream.

Situation A: observational study. Situation B: experiment with random assignment.

Practice 1.7Observational Study or Experiment?

A researcher surveys college students about how many hours they sleep per night and their GPA. No conditions are assigned to students.
An agricultural company randomly assigns half of 60 farm plots to a new fertilizer and the other half to the existing formula. Crop yields are compared after one growing season.
A researcher reviews five years of medical records to compare heart disease rates among smokers and non-smokers.
A school randomly selects 4 of 8 classrooms to receive a new reading program. The other 4 classrooms use the existing program. Reading scores are compared at year's end.
For each of the observational studies above: identify one confounding variable that could explain any observed association without causation.

Section 1.8

Descriptive Statistics and Inferential Statistics

The work of statistics can be divided into two broad areas.

Descriptive statistics involves organizing, summarizing, and presenting data in a way that is informative and honest. It does not go beyond the data at hand — it describes what was observed. Descriptive tools include frequency tables, bar graphs, histograms, dot plots, box plots, scatter plots, measures of center (mean, median, mode), and measures of spread (range, interquartile range, standard deviation).

Inferential statistics involves using data from a sample to draw conclusions about a larger population. Inference acknowledges that the sample is not perfect and that conclusions carry uncertainty. Inferential tools include confidence intervals, hypothesis tests, predictions from models, and probability-based reasoning about uncertainty.

Both areas are essential. You cannot do inferential statistics responsibly without first doing good descriptive work. And descriptive statistics alone is not enough when you want to draw conclusions that go beyond the sample.

Example 1.8.1

Descriptive or inferential?

ProblemClassify each task as primarily descriptive or inferential: (a) A teacher calculates the average quiz score of the 30 students who took it. (b) Based on 1,200 surveyed voters, a news organization estimates that 57% of all voters in the state support a measure. (c) A researcher tests whether a new teaching method led to higher exam scores than the traditional method.
ThinkAsk whether each task summarizes data at hand (descriptive) or draws a conclusion about a larger group (inferential).
Analyze (a) Teacher calculates average for 30 students who actually took quiz → Descriptive (summarizes those 30 students only) (b) Sample proportion (57%) used to estimate all voters in state → Inferential (conclusion extends beyond the sample) (c) Researcher tests a claim about whether the method caused improvement → Inferential (conclusion is meant to apply beyond this study)

(a) Descriptive. (b) Inferential. (c) Inferential.

Section 1.9

Statistical Questions

Not every question is a statistical question. A statistical question is one that expects variability in the answers and uses data to address that variability. Non-statistical questions have a single correct answer that does not vary from individual to individual.

Non-Statistical Question	Statistical Question
What is the sum of angles in a triangle?	How many hours per night do high school students sleep?
How many students are in this classroom right now?	What proportion of adults in this city exercise three times per week?
What is the boiling point of water at sea level?	Does the new medication reduce blood pressure more than the existing one?

Example 1.9.1

Statistical or non-statistical?

ProblemClassify: (a) How tall is the Eiffel Tower? (b) How tall are the students in this high school? (c) What score did Mia receive on the last test? (d) What scores did students in this class receive on the last test? (e) Does drinking more water improve athletic performance?
Analyze (a) The Eiffel Tower has one height → Non-statistical (b) Different students have different heights → Statistical (c) Mia received one score → Non-statistical (d) Different students received different scores → Statistical (e) Different people respond differently → Statistical

Non-statistical: (a), (c). Statistical: (b), (d), (e).

Practice 1.9Statistical Questions

Classify each as a statistical question or non-statistical question. For statistical questions, describe the variability that must be addressed.

How many days are in a leap year?
How many hours do students study per week?
What is the capital of Canada?
Do students who eat breakfast score higher on tests?
What is the speed of light?
How long does it take customers to place an order?
What is 15 × 8?
What proportion of U.S. adults currently smoke?
Did Anita score higher than Marcus on the last quiz?
How do recovery times differ by surgical method?

Section 1.10

Bias, Variability, and Context

Two words appear constantly in statistics and are often confused: bias and variability. Understanding both is essential for evaluating how trustworthy a statistical result is.

Variability means that values differ from one individual to another, or that a sample statistic differs from the population parameter it is meant to estimate. Some variability is natural and unavoidable. When you take a sample, you do not expect it to perfectly match the population — but you do expect it to be close enough to be useful.

Bias is a systematic tendency to produce estimates that are consistently too high or consistently too low. A biased study does not just vary around the truth — it misses the truth in one direction. Bias is particularly dangerous because it cannot be fixed by collecting more data. A larger sample that uses a biased method is still biased.

Type of Bias	Description	Example
Sampling bias	Some groups are more likely to be included than others	Surveying only students in AP classes to estimate overall student opinion
Response bias	Individuals give inaccurate answers due to question wording or social pressure	Asking "Don't you agree that the new policy is fair?" instead of a neutral question
Non-response bias	Non-respondents differ systematically from respondents	Only motivated, engaged students return survey forms, making results unrepresentative
Undercoverage	Some members of the population cannot be reached by the sampling method	A phone survey misses people without phones

Bias vs. Variability

Increasing sample size reduces variability — estimates become more consistent from sample to sample. But increasing sample size does NOT fix bias. A biased method applied to a very large sample still produces biased results. Getting more data from the wrong place does not help.

Practice 1.10Identifying Bias

A school surveys students about campus safety by placing survey boxes near the cafeteria. Only students who pass by that area and choose to respond fill out the survey. What type of bias might affect the results?
A researcher asks: "Should the government waste money on programs that don't work?" What type of bias does this question introduce?
A magazine asks readers to mail in responses to a question about preferred vacation destinations. What type of bias is introduced, and in what direction might the results be skewed?
Explain why collecting 5,000 responses using a biased method does not produce a more trustworthy result than 500 responses collected using the same biased method.

Section 1.11

Data Ethics and Responsible Interpretation

Statistics affects people's lives. Survey results influence elections. Clinical trial data shape treatment decisions. Economic data guide policy. Algorithmic predictions determine who gets loans, who receives parole, and who gets flagged for additional scrutiny. Because the stakes are real, the ethics of data collection and interpretation matter.

Privacy. Data about individuals should be collected, stored, and used in ways that protect personal information. Participants in studies have a right to know what is being collected and how it will be used.

Informed consent. In research involving human subjects, participants should be informed about the nature of the study and give their consent before data are collected. They should be able to withdraw without penalty.

Honest representation. Data displays, summary statistics, and reported results should represent the data accurately. Selecting only the favorable results, using misleading scales, or describing a sample statistic as if it were a population parameter are all forms of dishonest representation.

Overgeneralization. A result based on one sample should not be applied to groups that were not represented. A study conducted on college students may not generalize to the general population.

MGU Library Connection

Data ethics is a recurring theme throughout this textbook and the MGU curriculum
Related material in MGU Library: Research Methods — study design and IRB review
Related material: Public Policy — data in evidence-based decision-making
Related material: Data Science Foundations — algorithmic fairness and data governance

Practice 1.11Data Ethics

A company collects detailed purchasing data from app users and sells it to third-party advertisers without explicitly notifying users, noting it is covered by the terms of service. What ethical principles are at stake? What would responsible data practice look like?
A clinical researcher reports only the studies that showed positive results from a new treatment and ignores the studies that showed no effect. What is this practice called, and why is it ethically problematic?
A study is conducted only on male participants but its conclusions are presented as applying to all adults. Identify the ethical and statistical problem.

Section 1.12

Statistics in Science, Finance, Medicine, and Public Life

Statistics is embedded in the work of nearly every field that uses evidence to make decisions.

In science: Scientists use statistics to design experiments, analyze results, and decide whether observed effects are real or could be explained by chance. Climate scientists analyze temperature records over decades to identify trends. Ecologists use sampling methods to estimate animal populations.

In medicine: Clinical trials use random assignment and statistical analysis to test whether treatments are effective. Epidemiologists use observational data to identify risk factors for disease. Public health agencies use surveillance systems to detect outbreaks early.

In finance: Banks use statistical models to assess loan risk. Investors analyze historical price data to identify trends. Insurance companies use actuarial statistics to set premiums based on estimated risk.

In public life: Polls estimate public opinion before elections. Census data determine congressional representation and federal funding distribution. Sports analysts use statistics to evaluate player performance. Courts use forensic statistics to evaluate the strength of evidence.

Section 1.13

Common Early Statistics Mistakes

Students beginning their study of statistics tend to make a small set of conceptual errors repeatedly. Recognizing them early will save time and prevent confusion throughout the course.

Mistake 1 — Confusing a statistic with a parameter

"68% of teenagers prefer online video." (Based on a sample of 400 teens)

A survey result is not a population truth. It is an estimate, subject to uncertainty. The accurate statement is: "In our sample, 68% preferred online video. We estimate the true proportion is near 68%."

Mistake 2 — Believing larger samples always eliminate bias

"We got 2 million responses, so our result must be accurate."

In 1936, a famous magazine predicted the wrong presidential winner using over two million responses — because the sample overrepresented wealthy, phone-owning Americans. More data did not fix the flawed sampling method.

Mistake 3 — Treating association as causation

"Ice cream sales and drowning rates both rise in summer. Ice cream causes drowning."

Both are caused by a common factor: hot weather and outdoor swimming. Association does not prove causation. A well-designed experiment with random assignment is needed to establish causal conclusions.

Mistake 4 — Ignoring variability

"Both groups had the same average, so the methods produced the same results."

Two groups can have the same mean but very different distributions. One group might have scores tightly clustered near the average while the other has widely varying scores. Always examine spread, not just center.

Mistake 5 — Overgeneralizing from small or unrepresentative samples

"Our study of 12 college volunteers proves that all humans respond this way."

A study's conclusions are only valid for the population from which the sample was drawn. A sample of 12 volunteers is not a representative sample of all humans.

Mistake 6 — Confusing statistical significance with practical importance

"The result was statistically significant, so the new drug has a large medical benefit."

A very large sample can detect tiny differences that are statistically significant but trivially small in real life. A drug that lowers cholesterol by an average of 1 mg/dL may be statistically significant but clinically meaningless.

Mistake 7 — Failing to define the population clearly

"Most people prefer..." (without specifying which people)

Always ask: which population? All people in the world? Adults in the U.S.? Customers of a specific company? The conclusions of a study are only valid for the population from which the sample was drawn.

Section 1.14

Preparing to Organize and Display Data

Chapter 1 has introduced the big picture of statistical reasoning. The next chapters will begin developing specific tools and methods. In Chapters 3 and 4, you will learn how to organize data into tables and display it with graphs. Different variables require different displays: categorical variables call for frequency tables, bar graphs, segmented bar graphs, and two-way tables; quantitative variables call for dot plots, stem-and-leaf plots, histograms, box plots, and cumulative frequency displays.

When examining any data display, you will practice describing four features consistently:

Four Features of Every Distribution

Shape — Is the distribution symmetric, skewed, or roughly uniform? Are there multiple peaks?

Center — Where do values tend to fall? What is a typical value?

Spread — How much do the values vary? Tightly clustered or widely dispersed?

Unusual features — Are there outliers far from the main cluster? Gaps or clusters that suggest something interesting?

Two Problem-Solving Frameworks for This Course

Statistical Investigation: Ask → Plan → Collect → Organize → Analyze → Interpret → Communicate. This is the full cycle used throughout the book for larger investigations.

Individual Calculations: Read → Represent → Calculate → Check → Interpret. Use this for specific computations and analyses: understand the context first, choose a representation, show your work, verify reasonableness, translate the answer back into plain language.

Practice 1.14Readiness Check

A student says: "I already know how to make bar graphs and calculate averages, so I know descriptive statistics." What important ideas does this student still need to learn?
Rank the following four features of a distribution in the order you would typically describe them when writing about a graph. Explain your ranking: spread, shape, unusual features, center.
A researcher collects data on 50 patients: age, weight, blood pressure, diagnosis (hypertensive or not), and number of medications. Before graphing anything, what questions should the researcher ask about each variable?

Chapter 1 — Back Matter

Chapter Summary

Sec 1.1Statistics is the science of learning from data. It asks not only "What is the answer?" but also "How reliable is the answer?" and "What does the data suggest?" It differs from arithmetic because it requires judgment about collection, display, interpretation, and honest conclusions.
Sec 1.2Data are recorded observations about individuals. Variability — the natural tendency of values to differ — is the central fact that makes statistics necessary. Uncertainty is the consequence of variability; statistics quantifies uncertainty rather than eliminating it.
Sec 1.3Individuals are the units being studied. Variables are characteristics that take different values across individuals. In a data table, individuals are rows and variables are columns.
Sec 1.4Categorical variables record group membership or labels; arithmetic does not apply. Quantitative variables record numerical measurements or counts. Quantitative variables are discrete (counting) or continuous (measuring). Numbers that label rather than measure are categorical, even if they look numerical.
Sec 1.5The population is the full group of interest; the sample is the subset actually studied. A census studies every individual in the population. The population depends on the research question, not on the study design.
Sec 1.6Parameters describe populations (μ, σ, p); statistics describe samples (x̄, s, p̂). The goal of inferential statistics is to use a sample statistic to estimate or test claims about an unknown population parameter.
Sec 1.7Observational studies measure without intervening and can reveal association but cannot establish causation. Experiments impose treatments with random assignment and can support causal conclusions. Confounding variables are hidden third factors that explain observed associations without causation.
Sec 1.8Descriptive statistics summarizes data at hand using graphs, tables, and numerical measures. Inferential statistics draws conclusions beyond the sample using confidence intervals, hypothesis tests, and probability. Both are essential; descriptive work must precede responsible inference.
Sec 1.9Statistical questions expect variability in the answers and require data to address them. Non-statistical questions have fixed, single correct answers. Recognizing statistical questions shapes how studies are designed and how claims are evaluated.
Sec 1.10Variability is natural and unavoidable. Bias is systematic error that pushes estimates consistently away from the truth. Larger samples reduce variability but do not fix bias. Context shapes the meaning of every statistical result.
Sec 1.11Data ethics requires privacy, informed consent, honest representation, and awareness of who is and who is not represented. Statistical results affect real people and real decisions. Responsible interpretation acknowledges limits, uncertainty, and context.
Sec 1.12Statistics appears in science, medicine, finance, public policy, law, sports, and virtually every field that uses evidence to make decisions. The same core skills — recognizing data types, designing studies, interpreting results honestly — apply across all of these domains.
Sec 1.13Seven common early mistakes: confusing statistics with parameters; believing larger samples fix bias; treating association as causation; ignoring variability; overgeneralizing from small samples; confusing statistical significance with practical importance; failing to define the population clearly.
Sec 1.14Data displays are examined for four features: shape, center, spread, and unusual features. Two problem-solving frameworks used throughout the book: Ask–Plan–Collect–Organize–Analyze–Interpret–Communicate for full investigations; Read–Represent–Calculate–Check–Interpret for individual problems.

Key Terms Review

statistics

The science of learning from data by reasoning carefully about variability, uncertainty, and evidence.

data

Measurements or observations recorded about individuals; the raw material of statistical analysis.

variability

The natural tendency of values to differ from one individual to another, or from one sample to another.

uncertainty

The consequence of variability; statistical conclusions always carry some degree of uncertainty, which can be quantified.

individual

The unit being studied in a data set; not necessarily a person — could be a plant, a car, a country, or a day.

variable

A characteristic that takes different values from one individual to another.

categorical variable

A variable whose values are names, labels, or group memberships. Arithmetic does not apply to categorical values.

quantitative variable

A variable that records a numerical measurement or count that supports meaningful arithmetic.

discrete variable

A quantitative variable that takes whole-number values obtained by counting.

continuous variable

A quantitative variable that can take any value in a range, obtained by measuring.

population

The full group of individuals about which you want to draw a conclusion. Defined by the research question.

sample

The subset of the population actually observed, measured, or surveyed.

census

A study that includes every individual in the population.

parameter

A number that describes a characteristic of a population. Written with Greek letters: μ, σ, p.

statistic

A number that describes a characteristic of a sample. Written as: x̄, s, p̂.

observational study

A study that measures variables without intervening or imposing any treatment. Can reveal association but not causation.

experiment

A study that imposes a treatment using random assignment to establish causal conclusions.

confounding variable

A hidden third variable that causes both the explanatory and response variables to change, creating a false impression of causation.

descriptive statistics

Statistical work that organizes and summarizes the data at hand without drawing conclusions beyond it.

inferential statistics

Statistical work that uses sample data to draw conclusions about a larger population, always with acknowledged uncertainty.

statistical question

A question that expects variability in the answers and requires data to address that variability.

bias

A systematic tendency to produce estimates that are consistently too high or too low. Cannot be fixed by increasing sample size.

data ethics

The set of principles guiding responsible collection, use, and interpretation of data, including privacy, consent, and honest representation.

Concept Review Questions

In your own words, explain why statistics is more than arithmetic.
What is the difference between variability and bias? Can a study have high variability but no bias? Can it have low variability but still be biased? Explain both cases.
Why is it important to distinguish between a sample and a population?
What is a parameter? What is a statistic? Why do statisticians use different symbols for the two?
Explain in your own words why observational studies cannot establish causation.
What makes a question a statistical question? Give one example of a statistical question and one example of a non-statistical question on a topic of your choice.
What is the difference between descriptive statistics and inferential statistics? Give one concrete example of each.
Why does increasing the size of a biased sample not fix the bias?
Give two examples of ethical issues that can arise in statistical work.
What does it mean for a conclusion to overgeneralize? Give an example.

Variable Identification Practice

For each scenario, list at least two variables that would be recorded. For each variable, state whether it is categorical or quantitative, and if quantitative, state whether it is discrete or continuous.

A health clinic records information about each patient's visit.
A car manufacturer tracks information about each vehicle sold.
A social media researcher studies how teenagers use different platforms.
An agricultural researcher compares the growth of plants under different conditions.
A financial analyst tracks monthly stock prices for a portfolio of ten companies.
A city government tracks information about service requests received in a week.

Population and Sample Practice

For each scenario, identify (a) the population, (b) the sample, (c) the statistic being reported, and (d) the corresponding population parameter.

A pollster surveys 800 registered voters in a congressional district and finds that 54% support Candidate A.
A school nurse measures the blood pressure of every student who visits the health office in October. The mean blood pressure is 112/72 mmHg.
A manufacturer tests 50 light bulbs from a production run of 10,000. The average lifetime of the tested bulbs is 1,150 hours.
A researcher surveys 300 adults from a city's tax records about commuting habits. She finds 38% commute by public transit.
A marine biologist collects 200 fish from a lake and measures their lengths. The average length is 24 centimeters.

Statistical Question Practice

Classify each as a statistical question or a non-statistical question. For each statistical question, explain what source of variability must be addressed.

How many days are in a leap year?
How many hours per week do students in this school spend on homework?
What is the capital of Canada?
Do students who eat breakfast score higher on tests?
What is the speed of light in a vacuum?
How long does it take customers to place orders?
Did Anita score higher than Marcus on the quiz?
What proportion of U.S. adults currently smoke?
What temperature does water freeze at, at standard conditions?
How do recovery times differ by surgical method?

Real-World Applications

Application 1

Interpreting a survey result

ScenarioA national news organization reports: "A recent survey shows that 71% of Americans support stricter environmental regulations."
Questions(a) Identify the population and the sample as best you can from the report. (b) Is 71% a parameter or a statistic? How do you know? (c) What additional information would you want before fully trusting this result? (d) Could the result be biased? Describe at least one way bias could have been introduced.

Application 2

Identifying study type

Scenario AResearchers track 500 runners over three years, recording weekly mileage and knee injury frequency. No runner is assigned a mileage level.
Scenario BResearchers randomly assign 100 patients to receive either a new skin cream or a placebo cream. Improvement rates are compared after four weeks.
Scenario CAn agricultural company randomly assigns half of 60 farm plots to a new fertilizer and the other half to the existing one. Yields are compared after the growing season.
QuestionsFor each scenario: (a) Is it an observational study or an experiment? (b) If observational, identify one confounding variable. (c) If experimental, identify the treatment and the control condition.

Error Analysis

Each of the following statements contains a statistical error or misconception. Identify the error and explain how the statement should be corrected.

Problem 1

"We surveyed 10,000 shoppers at our stores. With that many responses, we can be confident our results apply to all consumers everywhere."

What is wrong with this reasoning? Identify the specific statistical concept being misapplied.

Problem 2

"Students who drink more water score higher on tests. Therefore, drinking water causes higher test scores."

Identify the logical error. Name one possible confounding variable. What kind of study would be needed to support a causal conclusion?

Problem 3

"The average income of the 15 employees surveyed is $58,000. This means the average income at the company is $58,000."

What distinction is the speaker missing? What notation correctly describes each value?

Problem 4

"This study had a very small p-value, so the new drug has a large and meaningful medical benefit."

Identify which of the seven common mistakes this represents. What should the speaker have said instead?

Problem 5

"The population of our study is the 250 adults who responded to our online survey."

What is the population actually described here? What is the sample? Rewrite the statement correctly.

Chapter 1 Checkpoint

Complete without assistance — assesses all key skills and concepts from Chapter 1

Section 1 — Populations, Samples, Parameters, and Statistics

A state health department wants to know the average number of hours per week adults in the state exercise. They randomly select 500 adults from driver's license records and survey them by phone. The average reported exercise time among respondents is 3.4 hours per week.

(a) What is the population? (b) What is the sample? (c) Is 3.4 hours a parameter or a statistic? (d) What symbol represents the sample mean? (e) What symbol represents the unknown population mean? (f) Identify one potential source of bias.

Section 2 — Variable Classification

Classify each variable as categorical or quantitative. If quantitative, state whether it is discrete or continuous.

(a) Number of movies watched last month (b) Genre of favorite movie (c) Race finish time in minutes (d) Political party registration (e) Weight of a newborn in kilograms (f) Letter grade in a course (A, B, C, D, F)

Section 3 — Observational Study or Experiment?

A student wants to study whether listening to music while studying affects quiz scores. She decides to have students who normally listen to music keep doing so, and students who normally study in silence continue that way. She then compares quiz scores.

(a) Is this an observational study or an experiment?
(b) Why can the student not conclude that listening to music causes higher or lower quiz scores?
(c) How would the study need to be redesigned to support a causal conclusion?

Section 4 — Statistical Questions

Classify each as a statistical question or non-statistical question.

(a) How far is it from New York City to Los Angeles?
(b) How much time do high school students spend on social media each day?
(c) What is 7 × 8?
(d) How does healthcare spending vary across countries?
(e) Did the football team win last night's game?

Section 5 — Error Analysis

A researcher reports: "Children in our study who watched more than two hours of television per day had lower reading scores than children who watched less. This proves that watching television reduces reading ability."

(a) What type of study would produce this kind of result?
(b) Identify the error in the researcher's conclusion.
(c) Name one confounding variable that could explain the association without television causing lower reading scores.
(d) What kind of study would be needed to support a causal conclusion?
A national survey reports that 63% of adults in the country support increased public transportation funding. A reporter writes: "The majority of the population supports more transportation funding." Evaluate this claim carefully. What is accurate? What is problematic?

Looking Ahead

Chapter 2: Collecting Data

Chapter 1 gave you the conceptual vocabulary for everything that follows. The ideas introduced here — populations and samples, parameters and statistics, observational studies and experiments, descriptive and inferential statistics, bias and variability — will appear again and again throughout this book and in every statistics course you take after this one.

Chapter 2 builds directly on Section 1.7 by developing the major sampling methods, study designs, and sources of bias in full detail. You will learn why simple random samples are the gold standard, how stratified, cluster, and systematic samples work, and what makes a study design capable of supporting a causal conclusion. The key insight of Chapter 2 is that statistics begins before calculation — how data are gathered determines what conclusions can responsibly be drawn.

Chapters 3 through 5 develop the descriptive statistics tools introduced in Section 1.8. Chapters 12 and 13 develop inferential statistics — confidence intervals and hypothesis tests — which formalize the ideas of uncertainty, estimation, and decision-making introduced here.