You are here

Workshop Statistics: Discovery with Data

Allan J. Rossman and Beth L. Chance
Key College Publishing
Publication Date: 
Number of Pages: 
Workshop Statistics
[Reviewed by
Sarah Boslaugh
, on

Workshop Statistics: Discovery with Data is a textbook for teaching elementary statistics which is based on the 1992 recommendations of the American Statistical Association/Mathematical Association of America Joint Committee on Undergraduate Statistics:

  • teach statistical thinking
  • employ more data and concepts, less theory and fewer recipes
  • foster active thinking.

This is the third edition of Workshop Statistics: the first edition (1996) was developed as part of the Workshop Mathematics Project, funded by the National Science Foundation and the Fund for the Improvement of Post-Secondary Education.

The Workshop Statistics method of teaching requires students to participate actively in the classes rather than following a lecture model. The usual concepts and procedures of introductory statistics are covered (from types of data through correlation and regression) but with a much greater emphasis on thinking through the material, and working with actual data and simulations, than on applying formulas and definitions. The basic pattern of most chapters is to begin with questions which require students to think about the concept, followed by simulation exercises (using provided materials including a random number table and a number of Java applets, or common objects such as index cards) before formal calculations are introduced.

Many professors teach statistics using simulations already, and have written applets or other simulation programs to allow students to perform procedures such as repeated sampling from a population with known parameters. Many applets are available on the internet for teaching purposes also. The great benefit of Workshop Statistics is that the simulation approach is built into the text, and the applets (supplied on a CD) are both easy to use and written specifically to support the examples used in the text.

Although I much prefer the simulation-based method of instruction (as I suspect do many statisticians), and think it works particularly well for students who are not mathematically-minded, it won’t work in every situation. For one thing, introductory statistics is often taught as a service course for other departments (for instance, medicine or engineering), and the faculty members in those departments may think that what their students should be learning is a series of procedures (how to do a t-test, chi-square test, etc.) rather than statistics as a way of thinking. Another problem is that the discovery approach is slower than simply applying formulas, and requires students to go through all the steps: this is fine for younger students whose full-time job is going to school, but a real problem for older students and working professionals who have more demands on their time. A third problem is that this text is not that useful as a reference, for instance if the student simply wants to look up a formula (which is a legitimate use of a textbook: I still use my statistics book for that purpose all the time). But for classes where Workshop Statistics is a good fit, it should be a very useful textbook.

There are, or will be soon, specialized editions of Workshop Statistics for teaching with Fathom and with graphing calculators, and the second edition also came in a version using Minitab, and had companion manuals for JMP and SPSS also: see the Key College web site for more information about these. The web site also includes support materials for students and teachers, which can be accessed with a code provided with each copy of Workshop Statistics.

Allan J. Rossman received his PhD in Statistics from Carnegie and is a Professor in the Department of Statistics at Cal Poly in San Louis Obispo. He is a Fellow of the American Statistical Association, President of the International Association for Statistical Education, and has served as chair of the ASA’s Sectional on Statistical Education and the ASA/MAA Joint Committee on Undergraduate Statistics. Beth Chance received her PhD in Operations Research from Cornell University and is Professor of Statistics at Cal Poly in San Louis Obispo. She is a Fellow of the American Statistical Association and the inaugural winner of the Waller Education Award for Excellence and Innovation in Teaching Introductory Statistics. Rossman and Chance together have co-edited STATS magazine and the Proceedings of the Seventh International Conference on Teaching Statistics.

Sarah Boslaugh ( is a Performance Review Analyst for BJC HealthCare and an Adjunct Instructor in the Washington University School of Medicine, both in St. Louis, MO. Her books include An Intermediate Guide to SPSS Programming: Using Syntax for Data Management (Sage, 2004), Secondary Data Sources for Public Health: A Practical Guide (Cambridge, 2007), and Statistics in a Nutshell (O'Reilly, forthcoming), and she is Editor-in-Chief of The Encyclopedia of Epidemiology (Sage, 2008).

Unit 1: Collecting Data and Drawing Conclusions

  • Data and Variables
  • Data and Distributions
  • Drawing Conclusions from Studies
  • Random Sampling 
  • Designing Experiments

Unit 2: Summarizing Data

  • Two-Way Tables
  • Displaying and Describing Distributions
  • Measures of Center
  • Measures of Spread
  • More Summary Measures and Graphs

Unit 3: Randomness in Data

  • Probability
  • Normal Distributions
  • Sampling Distributions: Proportions
  • Sampling Distributions: Means
  • Central Limit Theorem

Unit 4: Inference from Data: Principles

  • Confidence Intervals: Proportions
  • Tests of Significance: Proportions
  • More Inference Considerations
  • Confidence Intervals: Means
  • Tests of Significance: Means

Unit 5: Inference from Data: Comparisons

  • Comparing Two Proportions
  • Comparing Two Means
  • Analyzing Paired Data

Unit 6: Inferences with Categorical Data

  • Goodness-of-Fit Tests
  • Inference for Two-Way Tables

Unit 7: Relationships in Data

  • Graphical Displays of Association
  • Correlation Coefficient
  • Least Squares Regression
  • Inference for Correlation and Regression