Statistics Chapter 1: Week 1- September 9th, 2026
Course Structure, Grading, and Learning Objectives
Course Accommodations and Equity Policies:
Learning Services assists students requiring accommodations by drafting official documentation detailing reasonable adjustments for specific learning challenges.
For non-urgent equity or diversity concerns, communication should be directed to the Diversity office or SPRA via email.
For urgent safety situations, emergency services must be contacted immediately.
Grading Breakdown and Assessment Structure:
Test 1: Weight of .
Test 2: Weight of .
Quizzes and Assignments: Designed to assess functional understanding using software tools (primarily Excel, with potential integration of additional computational software or programming languages).
Group Project: Requires students to collaboratively solve a structured business problem based on their domain interests.
Final Exam: Comprehensive evaluation covering all course content.
Academic Standards and Prior Learning:
Passing Grade: For degree-level students, the passing threshold is . (In contrast, diploma programs typically carry a passing requirement).
PL&AR (Prior Learning Assessment and Recognition): Students who have taken equivalent coursework may apply for transfer credit by consulting the department coordinator or program chair.
Core Learning Outcomes:
Collect and interpret data using formal statistical procedures to convert raw data into actionable information.
Analyze data using numerical summaries and computational tools.
Incorporate concepts of random variables and probability into structured decision-making processes.
Apply various probability distributions to model real-world business scenarios.
Analyze concepts of sampling error and sampling distributions to infer characteristics of target populations.
Formulate decision rules and execute hypothesis testing to interpret outcomes in business contexts.
Required Tools and Resources:
Textbook: Primary reading materials are linked directly on D2L / Brightspace.
Software: Microsoft Excel (required); statistical software such as Minitab or suitable alternatives under evaluation.
Hardware: A dedicated business financial calculator (standardized from prerequisite financial mathematics coursework).
Weekly Workload and Submission Policies:
Regular Cadence: Every week involves a new assignment or assessment to ensure continuous engagement.
Assignment Deadlines: Weekly assignments/quizzes are released following class and are due the following Wednesday at midnight ().
Late Policy: A grace window extends to Friday of the same week to accommodate scheduling constraints without falling behind.
Course Schedule Outline:
Week 1: Introduction to Sampling and Data Collection.
Week 2: Descriptive Statistics (Data display and numerical summaries).
Week 3: Advanced Descriptive Statistics (including Arithmetic Mean and Geometric Mean).
Week 4: Probability Topics (Terminology, Independent Events, and Mutually Exclusive Events).
Week 5: Discrete Random Variable Distributions.
Week 6: Test 1 (In-class, written paper format, covering Chapters 1–4; scheduled for approximately within the class block).
Week 7: Continuous Random Variable Distributions.
Week 8: Normal Distribution and the Central Limit Theorem.
Week 9: Test 2 (In-class, written paper format, covering Chapters 5–7).
Week 10: Confidence Intervals.
Week 11: Hypothesis Testing.
Week 12: Comprehensive Course Review, Final Group Project Submissions, and Final Examination.
Fundamentals of Statistics and Descriptive vs. Inferential Analysis
Definition of Statistics:
Statistics is the science of collecting, analyzing, interpreting, and presenting data.
It is a multi-stage methodology applicable across diverse fields including finance, computer science, medicine, and general management.
Two Major Branches of Statistics:
Descriptive Statistics:
Focuses on organizing, summarizing, and displaying data in a clear, meaningful format.
Examples: Daily weather forecasts displaying rain probabilities, athlete performance stats, national employment figures, and local crime summaries.
Inferential Statistics:
Focuses on interpreting data summaries, drawing conclusions, making predictions, and supporting formal decision-making regarding a larger population based on sample data.
Examples: Using historical pricing trends and tariff impact data to justify purchasing decisions before price escalations occur; evaluating academic performance metrics to determine eligibility for rewards.
Probability and Theoretical Limits:
Definition: Probability is the mathematical tool used to quantify and study randomness.
Sample Size Impact on Empirical Probability:
Small sample sizes yield high variability relative to theoretical expectations (e.g., flipping a fair coin times may not yield exactly heads and tails).
Large numbers of trials stabilize results toward theoretical probabilities (e.g., flipping a fair coin times yields empirical results extremely close to the theoretical value of or ).
Sufficient data points are required to make reliable statistical inferences and prevent over-generalization from outlier events.
Key Statistical Terminology: Populations, Samples, Parameters, and Statistics
Population:
The complete collection of all individuals, objects, measurements, or items under study.
Populations are not restricted to human groups; they can include physical items such as manufactured water bottles, commercial assets, or financial transactions.
Sample:
A representative subset drawn directly from the target population.
Sampling is essential because evaluating an entire population directly is frequently prohibitive due to excessive time constraints and financial costs.
Parameter vs. Statistic:
Parameter:
A fixed numerical value that describes a specific characteristic of an entire Population.
Memory Rule: Parameter matches Population.
Example: The true average GPA of all students who graduated from a university last year.
Statistic:
A numerical value that describes a specific characteristic of a Sample.
Memory Rule: Statistic matches Sample.
Example: The average GPA calculated from a group of randomly selected graduating students.
A sample statistic serves as a point estimate for the corresponding unknown population parameter.
Data Classifications and Variables: Quantitative vs. Qualitative
Variables:
A variable (typically denoted by capital letters such as or ) is a characteristic or attribute of interest that can be determined for each member of a population or sample.
Datum vs. Data:
Datum: A single, individual value or response recorded for a variable.
Data: The plural set of values collected for a variable across subjects.
Categorical (Qualitative) Data:
Consists of non-numerical attributes, labels, or categories that place individuals into distinct groups.
Mathematical calculations (e.g., calculating an arithmetic mean) cannot be meaningfully performed on categorical data.
Examples: Shirt sizes (Small, Medium, Large), hair color, blood type, star ratings (1-star to 5-star hotel classifications).
Quantitative Data:
Consists of numerical measurements or counts where standard mathematical operations are meaningful.
Discrete Quantitative Data:
Data resulting from counting, where values are restricted to distinct, separate integers (no fractional units exist).
Examples: The number of books carried in a backpack (), the number of cans of soup on a shelf, the number of annual visits to a physician.
Continuous Quantitative Data:
Data resulting from precise measurements along a continuous scale, which can take on any real decimal value within a given interval.
Examples: Body weight in kilograms (), elapsed time in hours (), soup package weights.
Basic Mathematical Summaries:
Arithmetic Mean: The sum of all numerical values divided by the total number of observations ():
Proportion: The ratio of the count of a specific subset category () to the total sample size ():
Methods of Data Visualization
Pie Charts:
Circular charts divided into wedge sectors representing relative proportions of a whole ().
Best suited for displaying discrete categorical data with mutually exclusive categories that sum strictly to .
Bar Graphs:
Visual displays using rectangular bars whose lengths/heights are proportional to category counts or percentages.
Required when categories overlap, when individual subjects belong to multiple groups, or when total percentages exceed .
Pareto Charts:
A specialized form of a bar graph where categorical bars are arranged in strict descending order of frequency from tallest (left) to shortest (right).
Handling Incomplete or Missing Data:
When sample datasets do not cover of a population (e.g., survey responses total ), an explicit "Other" or "Unaccounted" category must be included to ensure full representation.
Sampling Methodologies and Potential Sampling Errors
Requirement for Representativeness:
A sample must possess the exact relevant characteristics of the population from which it is drawn to avoid invalid statistical inferences and subsequent financial or strategic losses.
Standard Probability Sampling Methods:
Simple Random Sampling: Every member of the population has an equal and independent probability of selection (e.g., using a computer program to generate random student ID numbers).
Stratified Sampling: The target population is divided into distinct subgroups (strata) based on specific attributes, and a random sample is drawn from every individual stratum (e.g., selecting players from an 8–10 age group, players from an 11–12 age group, and players from a 13–14 age group).
Cluster Sampling: The population is divided into separate geographic or organizational units (clusters); specific clusters are randomly chosen, and all or a sample of elements within those selected clusters are evaluated (e.g., surveying all HR personnel across specific high-tech companies out of a industry sector).
Systematic Sampling: Selecting elements at regular, fixed numerical intervals from an ordered sampling frame, following a random starting point (e.g., interviewing every patient on a hospital admission list).
Non-Probability Sampling:
Convenience Sampling: Selecting easily accessible subjects. Highly prone to severe bias (e.g., surveying immediate classmates in a classroom regarding general retail consumption habits).
Sampling Conditions:
With Replacement: Selected units are returned to the population pool prior to drawing the next unit, allowing a single element to be selected multiple times.
Without Replacement: Selected units are removed from the population pool upon selection. Sampling without replacement on very small populations drastically alters selection probabilities and distorts results.
Errors and Biases in Data Collection:
Sampling Error: The natural variability and inherent statistical discrepancy between a sample statistic and the true population parameter. Increased sample sizes reduce sampling error.
Non-Sampling Error: Data distortions caused by human or operational factors unrelated to the sampling process (e.g., defective measuring instruments, flawed survey questions, data entry errors, non-response).
Sampling Bias: Systematically over-representing or under-representing specific population segments (e.g., evaluating general public interest in baseball by surveying attendees at a professional baseball stadium; store managers only giving feedback surveys to satisfied customers).
Confounding / Lurking Variables: External variables that influence the measured variables, masking true underlying cause-and-effect relationships.
Operational Ambiguity: Failure to clearly define variables (e.g., studying the impact of "exercise" without defining exact physical intensity, duration, or workout type).
Levels of Data Measurement and Frequency Distributions
The Four Levels of Measurement:
Nominal Scale:
Qualitative categories, names, labels, or attributes without any natural quantitative order.
Statistical Capability: Count frequencies only.
Examples: Eye color, food types, company department names.
Ordinal Scale:
Qualitative categories that follow a clear, logical ranking or order, but differences between data points cannot be measured quantitatively.
Examples: Customer service satisfaction ratings (Terrible, Poor, Neutral, Good, Excellent), course letter grades ().
Interval Scale:
Ordered numerical data with precise, measurable distances between values, but no absolute zero point (zero does not indicate the total absence of the quantity).
Examples: Temperature measured in Fahrenheit or Celsius ( does not mean "no heat").
Ratio Scale:
Ordered numerical data with equal intervals and a true absolute zero point (where zero represents the complete absence of the measured attribute).
Statistical Capability: Full arithmetic operations including meaningful ratios ( is twice as heavy as ).
Examples: Exam percentage scores, income in dollars, weight, age.
Frequency Tables and Metrics:
Frequency (): The raw count of observations belonging to a specific data interval or category.
Relative Frequency (): The proportion of total observations () falling into a given category:
Cumulative Relative Frequency (): The running total sum of consecutive relative frequencies up to the current class row. The final row entry must always equal ().
Principles of Experimental Design and Research Ethics
Experimental Variables:
Independent (Explanatory) Variable: The condition or factor intentionally manipulated by the researcher.
Dependent (Response Variable): The metric measured by the researcher to determine the impact of changing the independent variable.
Control Mechanisms:
Control Group: A baseline comparison group that receives no active experimental manipulation.
Placebo: An inactive, neutral treatment administered to control subjects to eliminate psychological bias (e.g., serving a healthy burger wrapped in fast-food packaging to evaluate brand perception).
Blinding Techniques:
Single-Blind Study: Participants are unaware of whether they are receiving the active treatment or a placebo, preventing participant expectation bias.
Double-Blind Study: Neither the study participants nor the evaluating researchers know who receives active treatments versus placebos, eliminating both participant expectation bias and researcher measurement bias.
Limitations: Blinding is impossible if treatments involve obvious sensory inputs (e.g., distinguishing between a strong floral scent mask and an unscented mask).
Class Exercises and Case Studies
Case Study 1: First-Year Student School Supplies Study:
Scenario: A college surveys randomly selected first-year students to determine average annual expenditure on non-book school supplies. Individual recorded responses included values such as \150$, \200$, and \225$.\n * *Population*: All first-year students attending the college.\n * *Sample*: The 100 surveyed first-year students.\n * *Parameter*: The overall average dollar amount spent on non-book school supplies by *all* first-year students at the college.\n * *Statistic*: The average dollar amount spent on non-book school supplies calculated strictly from the 100 surveyed students.\n * *Variable*: The dollar amount spent on non-book school supplies by an individual first-year student.\n * *Data*: The actual set of collected numerical responses (e.g., \150$, \200$, \225$).
Case Study 2: Graduate Student GPA Study:
Scenario: Evaluating academic performance metrics of college graduates.
Population: All students who graduated from the college last year.
Sample: A randomly selected subset group of last year's graduating students.
Parameter: The true average cumulative GPA of all students who graduated from the college last year.
Statistic: The average GPA calculated from the sampled subset of graduating students.
Variable: The cumulative GPA of a single graduating student.
Data: Specific recorded GPA figures (e.g., ).
Case Study 3: Automotive Crash Safety Study:
Scenario: Impact testing on randomly selected test vehicles crashed into a concrete wall at to evaluate driver-side dummy head injuries.
Population: All commercial vehicles equipped with front driver-side test dummies.
Sample: The selected vehicles evaluated in the crash