Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.



Rossman review
Posted:
Apr 26, 1996 1:29 PM


> > Book Review... > > > Workshop Statistics: Discovery with Data > Allan Rossman, Dickinson College, Pennsylvania > $25, 1995, 452pp., 0387944974,Jones and Bartlett, 8008320034, > custserve@jbpub.com > > Over the past few years I have written articles and given conference > presentations describing ways statistics courses should be redesigned > to better facilitate student learning. I have encouraged the use of > active learning strategies, problem solving using real data sets of > interest to students, small group discussions, writing assignments, > and appropriate technology. Allan Rossman's new introductory text, > Statistics Workshop, appears to have incorporated all of these important > components in a lowcost softcover textbook. > > This book is distinguished from other introductory textbooks in several > important ways. There is a focus on "big ideas" rather than a large > collection skills, definitions, explanations, and techniques. Instead > of presenting students with methods of analyzing data and descriptions > of concepts, Workshop Statistics is designed to provoke students to > discover concepts themselves. When skills are presented it is because > there is a reason to use them: to carry out the next step in solving > a problem by analyzing data. Rossman explains that pages of expository > material or solved examples are omitted to emphasize "the idea that > students construct their own knowledge of statistical ideas as they > work through the activities." While most textbooks contain problem > sets at the end of the chapter, each focusing on a different (and > often contrived) set of data, the homework activities in this text > extend the types of activities conducted in class, examining a few > real data sets in more detail. An emphasis on "writing to learn" is > also embedded in the activities by continually asking students to > write about different aspects of data analysis and to summarize what > they have learned about a particular topic. > > Workshop Statistics is divided into six units, each consisting of three > to five topics. Over 100 data sets are either generated by students or > included in tables (soon to be available on a data disk). Students > combine data explorations by hand with use of a computer or calculator > to generate graphs and statistics. > > The first unit, Topic 1: "Exploring Data: Distributions," illustrates the > handson data exploration approach. This unit is divided into five topics, > each appropriate for one class session. On the first day of class students > make predictions and gather data about themselves (gender, political views, > opinions of statistics). They are guided through activities where they learn > how to generate and analyze data to describe the number of states and > countries they have visited. By the end of this first activity students > have learned to distinguish different types of variables and data and how > to construct and describe dotplots and histograms. In the next section > students are asked to compare and examine dotplots, allowing ideas to > emerge of center, spread, shape, and outliers as distinguishing features > of these plots. Topics 3 and 4 introduce more details on measures of center > and spread, and the fifth topic challenges students to integrate and use > these techniques in comparing distributions. > > The next unit of the text explores relationship between two variables with > topics on graphing bivariate data, correlation and regression, and tables of > categorical data. Students are first shown scatterplots of data where they > develop their own ideas of association between variables. Later they analyze > data on space shuttle oring failures, peanut butter cost and quality, and > cars' fuel efficiency. By examining and manipulating different scatterplots > while viewing correlation coefficients, students construct ideas of how > different factors influence the correlation between variables. In an > activity comparing the average number of television sets per person and > life expectancy for 22 countries, students discover that correlation does > not imply causation. > > The third unit, "Randomness", is quite a departure from the typical chapter > on probability in most introductory textbooks. Instead, this unit introduces > sampling distributions and then focuses on the normal distribution and the > central limit theorem. As Rossman explains in the introduction, there is no > formal treatment of probability. Instead, ideas of probability are introduced > in the context of simulation and random variation and as they apply to methods > of statistical inference. > > Topic 4 introduces statistical inference first through confidence intervals > for proportions and then through significance tests. Topic 5 includes > important aspects of designing experiments and extends inference to > comparisons of two proportions. In the last unit, Inference From Data: > Measurements, student learn to make inferences involving one or two > population means. > > This textbook not only looks exciting, it also works well with students. > I know because I used an earlier version of Workshop Statistics in a > class and have also observed Allan Rossman teaching a class at Dickinson > College. Students typically work in pairs or groups on the data exploration > activities after a wholeclass orientation to the day's topic. While Allan > teaches his class in a room equipped with Macintosh computers, allowing > students to easily analyze data using Minitab, the format of the book > allows students to use any type of computer or graphing calculator either > during or outside of class. I have found students to enjoy the activities > and appreciate the varied and interesting data sets. Although students > initially resist writing long verbal descriptions, they eventually learn > how to do this and appear to value the process. > > I encourage all statistics educators to seriously consider using this > innovative text. I believe that it provides an exemplary instructional > approach that should enable more students to overcome their fears about > learning statistics and to become statistically literate. > > Reviewed by Joan Garfield > University of Minnesota > Minneapolis, Minnesota

_   Robert W. Hayden   Department of Mathematics /  Plymouth State College   Plymouth, New Hampshire 03264 USA  *  Rural Route 1, Box 10 /  Ashland, NH 032179702  ) (603) 9689914 (home) L_____/ hayden@oz.plymouth.edu fax (603) 5352943 (work)



