Business Statistics: A First Course

Citation

Levine, D. M., Szabat, K. A., & Stephan, D. F. (2019). Business Statistics: A First Course (8th ed.). Pearson.

Chapter Summary

The eighth edition (copyright 2020) has an introductory "First Things First" chapter, fourteen numbered chapters in the printed book, a fifteenth chapter on quality management available online, appendices, and answers to self-test and selected even-numbered problems; the printed text runs to about 650 pages. Every chapter opens with a "Using Statistics" business scenario that is revisited at the end, and closes with a summary, key equations and terms, "Checking Your Understanding" questions, chapter review problems and cases, including a continuing case (Managing Ashland MultiComm Services) and "Digital Cases" that ask students to judge claims in business documents. Each chapter also has separate software guides for Excel, JMP and Minitab, and, for selected chapters, Tableau. The DCOVA framework (define, collect, organize, visualize, analyze) is used to structure the analyses.

First Things First

An opening chapter on why statistics matters, how business analytics and "big data" are changing the subject, and how to get started with software. The preface notes that it is available as a free online download so that students can start before the course.

Chapter 1: Defining and Collecting Data

Types of variables and measurement scales, populations and samples, data sources, the main sampling methods (simple random, systematic, stratified, cluster), data cleaning and preprocessing, and the types of survey error, with a section on ethical issues in surveys.

Chapters 2–3: Organizing and Visualizing Variables; Numerical Descriptive Measures

Chapter 2 covers summary and contingency tables, frequency distributions, bar, pie, Pareto and doughnut charts, histograms and polygons, scatter and time-series plots, visualising a mix of variables (including treemaps and sparklines), filtering and querying data, and the pitfalls of misleading charts. Chapter 3 covers measures of central tendency, variation and shape, quartiles and boxplots, population measures with the empirical rule and Chebyshev's theorem, and covariance and correlation.

Chapters 4–6: Basic Probability; Discrete Probability Distributions; The Normal Distribution

Probability concepts, conditional probability, decision trees, independence, Bayes' theorem and counting rules (Chapter 4); expected value and the binomial and Poisson distributions (Chapter 5); and the normal distribution, including how to evaluate whether data are approximately normal with a normal probability plot (Chapter 6).

Chapters 7–9: Sampling Distributions; Confidence Interval Estimation; Fundamentals of Hypothesis Testing: One-Sample Tests

The sampling distributions of the mean and the proportion, with the central limit theorem (Chapter 7); confidence intervals for a mean and a proportion and the determination of sample size (Chapter 8); and the logic of hypothesis testing, the Z and t tests for a mean, one-tail tests, the p-value approach and the Z test for a proportion (Chapter 9).

Chapters 10–11: Two-Sample Tests and One-Way ANOVA; Chi-Square Tests

Comparisons of two independent and two related means, two proportions and two variances, followed by one-way analysis of variance (Chapter 10); and chi-square tests for differences between two or more proportions and for independence (Chapter 11). From these chapters on, results are presented in tabular summaries that pair each statistical result with the business conclusion it supports.

Chapters 12–13: Simple Linear Regression; Multiple Regression

Chapter 12 develops the regression equation, measures of variation, the assumptions of regression, residual analysis, the Durbin-Watson statistic, inference about the slope and correlation, confidence and prediction intervals, and pitfalls. Chapter 13 extends this to several predictors, adjusted r², the overall F test, residual analysis, inference about coefficients, and dummy variables and interaction terms.

Chapter 14: Business Analytics

An all-new chapter in this edition that introduces descriptive analytics (dashboards), predictive analytics for prediction, classification, clustering and association, text analytics and prescriptive analytics, using JMP, Minitab, Tableau and Excel with Power BI.

Chapter 15 (online): Statistical Applications in Quality Management

Control charts for proportions, counts and the mean and range, process capability, total quality management and Six Sigma.

Key Concepts

First Things First and Chapters 1–3: Data and Description

• DCOVA framework: define the variables, collect the data, organize, visualize and analyze it, used as the structure for each analysis.
• Sampling methods and survey errors: how samples are drawn and how coverage, nonresponse, sampling and measurement errors arise.
• Descriptive measures: centre, variation and shape of a distribution, summarised by statistics and the boxplot.

Chapters 4–7: Probability and Sampling

• Conditional probability and Bayes' theorem: updating probabilities when information is known.
• Binomial, Poisson and normal distributions: the standard models for counts and for continuous measurements.
• Central limit theorem: the sampling distribution of the mean approaches normality as sample size grows, whatever the population's shape.

Chapters 8–11: Inference

• Confidence interval: a range of plausible values for a population parameter at a stated level of confidence.
• Hypothesis test and p-value: a procedure for deciding whether sample evidence is strong enough to reject a null hypothesis.
• ANOVA and chi-square tests: methods for comparing several means, or proportions and categorical variables.

Chapters 12–14: Regression and Analytics

• Least-squares regression and residual analysis: fitting a line and checking the assumptions behind it.
• Dummy variables and interactions: including categorical predictors and combined effects in multiple regression.
• Predictive analytics: methods such as classification and clustering used to find patterns in large data sets.

Critical Analysis

Strengths

• Interpretation over calculation. The authors state that they emphasise interpreting results, evaluating assumptions and deciding what to do when assumptions fail, rather than manual computation. The tabular summaries introduced in this edition show this in practice: each test result is set beside the decision it supports.
• Business context throughout. Each topic is presented through examples from accounting, finance, management and marketing, every chapter is built around a Using Statistics scenario, and the continuing and Digital Cases ask students to work with realistic data and to spot misuse of statistics.
• Flexible software support. Modular guides for Excel (with the authors' PHStat add-in), JMP, Minitab and Tableau let a course use whichever program it has without the software crowding out the statistics.
• Ample practice. "Learning the Basics" and "Applying the Concepts" problems, review problems and answers to self-test and many even-numbered problems support independent study.

Limitations

• A one-semester scope. Time-series forecasting and nonparametric tests are not among the printed chapters, and quality management is online only; the preface points to a library of separate online topics and chapters, but courses needing these subjects should check what is available or supplement the book.
• Software guides add bulk. Several separate software guides follow each chapter, so the printed book is long for its content, and students must learn to ignore the guides for programs they do not use.
• Business analytics is introductory. Chapter 14 is a survey of methods carried out in software rather than a treatment of how they work.

Where It Fits in the Course

This is a natural first textbook for the business statistics course: a one-semester treatment from data collection to multiple regression with a strong software emphasis. Charles Wheelan's Naked Statistics can be read before or alongside it as a non-technical account of what the tools are for and how they mislead; it has no exercises or software, so it complements rather than replaces Levine. Sharpe, De Veaux and Velleman's Business Statistics is a full-length alternative that also stresses thinking before computing and communicating results, and goes further into regression modelling and data mining; students who finish Levine and want more depth can move to it.

Ian Jacques's Mathematics for Economics and Business supplies the algebra and calculus background that more formal statistics needs; Levine itself requires only high-school algebra. Newbold and colleagues' Statistics for Business and Economics is a comprehensive text covering probability, inference, regression and time series in more mathematical detail, a sensible next step for students heading towards econometrics or quantitative finance.

How to Use This Book

High-school algebra is enough; no prior statistics is assumed. A typical one-semester course covers First Things First and Chapters 1–13, adding Chapter 14 where analytics is part of the syllabus.

• Core: Chapters 1–3 (data and description), 6–9 (normal distribution, sampling distributions, confidence intervals, hypothesis testing) and 12–13 (regression). These carry most course assessments.
• Supporting: Chapters 4–5 on probability can be covered selectively in courses focused on inference; Chapters 10–11 complete the standard set of tests.
• Optional: Chapter 14 on business analytics, and the online Chapter 15 for students in operations or quality management.

Choose one software program and follow only its guide for each chapter. Read each Using Statistics scenario first, then return to it at the end of the chapter to check that the analysis answers the business question. Work the "Learning the Basics" problems before "Applying the Concepts", use the answers at the back for self-test problems, and attempt at least one case per chapter: the Digital Cases are particularly useful practice in judging statistical claims.