Reviews
Here's what people are saying about Visual Statistics:
General comments about VS
Comments about the worktext
Comments about the data sets
Comments about specific modulesGeneral Comments About VS
The whole set of interactive material is "cool." I very much like the interactive capabilities of the software. … I think the material in VS is an excellent tool to complement other text and lecture material. … I still use lectures heavily in my class of 160 students, but I could envision moving away from lectures and relying more on lab exercises. VS would be a great tool for that.
John C. Dutton—North Carolina State University
The software modules are all excellent tools for illustrating various statistical concepts visually. They have all been well thought out, programmed effectively, and allow for a number of different kinds of explorations. If viewed as tools for an instructor to use and incorporate into presentations or to use as the basis for lab assignments tailored to the specific course being taught, they are wonderful. … I think this package of software tools is BY FAR the best of those that I have seen.
Roxy Peck—California Polytechnic State University- San Luis Obispo
The software program is impressive. … The more time I have spent with Visual Statistics, the more excited I am … The program makes as complete an integration of analytic and graphic information as I have seen.
Priscilla Chaffe-Stengel—California State University- Fresno
The visual approach in this software cannot be had in a traditional textbook. Visual Statistics is a valuable teaching tool and an excellent complement to a primary textbook.
Robert Hannum—University of Denver
I am very impressed by this package. The graphics are truly excellent and appealing. The software displays so quickly that there is no noticeable delay, and it is very interactive. ... Whoever did the graphical design layout did the best job I have ever seen. The screens are really attractive and logical, and I found it inviting. … This is probably the best of its kind that I have seen. It is miles ahead of Paul Velleman's ... in terms of illustrating concepts to students.
David C. Howell—University of Vermont
The authors have done a good job and have created a product that could play an important role in shaping how statistics will be taught in the classrooms of the 21st century.
Subha Chakraborti—University of Alabama
I really like this software and have used it in a second statistics course.
Patti Collings—Brigham Young University
I believe Visual Statistics provides unique capabilities (options, plots, displays, dynamic linking of items) that would be very useful in teaching the concepts covered by the modules. Visual Statistics also has the advantage over standard stat packages of being in module form, so that each module can be relatively easy to operate and still get the intended point across to the user without providing too much clutter that more general stat packages must provide to be for general use.
William Duckworth II—Iowa State University
The software has been well designed for ease of use by a novice.
Sue B. Schou—Idaho State University
… this software system is rather unique and I am not sure that I have ever reviewed or used anything like it. It is distinctly different in function and purpose to a standard statistical software package such as Minitab, SAS, or SPSS. I would rate the simulations as very well done …
Glenn W. Milligan—Ohio State University
Overall, the package covers everything necessary for an introductory statistics course. … the package is a great teaching tool.
Mark E. Rush—Washington & Lee University
The Worktext
The manual does what it is supposed to do, and does it well. It briefly introduces the concepts covered in the module (without the more detailed explanation found in a primary textbook) and explains how to use the software for that module. I give high marks for the exercises – they serve as a structured way to reinforce the concepts – and view them as a most valuable part of the Visual Statistics package. Some students will play with the software, but not think critically about the statistical concepts. The exercises help to guide the student. Also, kudos for the self-evaluation quiz and (especially) the glossary.
Robert Hannum—University of Denver
What I think that this project has, and the web stuff I have seen doesn't, is the workbook approach that gives them specific things to do and look for. I could assign this as homework, and be pretty confident that they would do it and would learn something. Just saying "Go look at this, cuz it's good for you" won't get you anywhere. … I particularly liked the one page overview [in the worktext]. In my own books I get down in when do histograms leave off and frequency distribution take over, and find that boring. This did a great job of dispensing with that issue in about three sentences.
David C. Howell—University of Vermont
The Overview of Concepts was succinct and to the point. It is mostly all that a student would need to be successful. I especially liked the basic learning exercises – they could be described as guide discovery exercises and would definitely enhance a student's learning. Congratulations!!!
Sue B. Schou—Idaho State University
The exercises provide a way to guide the students through the rich possibilities inherent in the interactive software.
John C. Dutton—North Carolina State University
The Worktext for the module provides a very condensed and accurate description of the related concepts. If used along side with a full-length textbook, students should benefit tremendously through the interactive environment.
Pin Ng—University of Illinois
The Illustration of Concepts are excellent. They are clear, well written, and seem to make good points. … The Advanced Learning Exercises are good. … The Self-Evaluation Quizzes, combined with the solutions and suggestions, are excellent – keeps up with the theme of helping the student learn by themselves. … Overall, I think the manual [Worktext] is well-written, i.e., the quality is high. … I think this is an excellent teaching/learning tool.
Robert K. Smidt—California Polytechnic State University- San Luis Obispo
The Visual Statistics Data Sets
I am very impressed with the quantity and quality of the data used. The data sets contain interesting examples which the students are drawn to. I had students go into the data sets and “play” in order to get some interesting information. What a great way for students to learn to work with data. … This is an outstanding software program which really helps the user understand the principles behind a number of statistical concepts.
Nancy C. Aiello—Northern Virginia Community College
The data sets are well chosen to provide the instruction of the concepts being demonstrated. I was able to move easily from one data set to another using the instructions within the module. The description of each data set including variable definitions is excellent. I especially appreciated the discussion on how the sampling was done.
Sue B. Schou—Idaho State University
The data sets are an important feature of the software. I like being able to switch quickly from data set to data set to illustrate different points. For example, it is nice to be able to show histograms that are symmetric, skewed, have outliers, etc., all in about 10 minutes of class time. It is important to have a context, but it is nice to have one that doesn’t take much explaining and that the students can relate to. They like trying to guess which states are the outliers in some of the displays. … The ability to use the data editor so that I can use my own data when I want to is a nice feature, although I don’t use it too often because the data sets provided are usually adequate to illustrate the important concepts.
Roxy Peck—California Polytechnic State University- San Luis Obispo
Wonderful. Both the choice of types of fields and nature of problems or quirks in the data sets make them interesting and relevant.
Robert K. Smidt—California Polytechnic State University- San Luis Obispo
Specific Modules
Visualizing Univarite Data Analysis
I am very much impressed how well the Chapter 1 module illustrates univariate data analysis. The ability to see the effect of varying the number of classes and/or class width in a histogram (or similar display) is a valuable feature. The four panel quadrant display is wonderful. The raw data and table of descriptive statistics are available while options to switch graph or histogram type make it easy to compare different displays and “see” the data from a variety of perspectives.
Robert Hannum—University of Denver
Good material! Fully integrates data, graphic and analytic summary. Will be very helpful to students.
Priscilla Chaffe-Stengel—California State University- Fresno
This module is one of my favorites…
Roxy Peck—California Polytechnic State University- San Luis Obispo
Visualizing Shapes of Distributions
Differences between discrete and continuous data are well demonstrated. This is difficult material to teach ordinarily and the authors have succeeded in making it easier to understand and interesting.
Sue B. Schou—Idaho State University
This module does a really good job of covering the concepts and objectives. The module is easy to use and nicely relates the concepts … I used the module in an introductory class and a majority of the students were better able to understand populations.
Nancy C. Aiello—Northern Virginia Community College
Visualizing Discrete Distributions
Good graphics. Excellent scenarios. Wonderful feature to superimpose approximation distributions over the discrete distribution – having the approximation distributions map over the discrete distribution shows students dynamically how approximations work and why they are of interest.
Priscilla Chaffe-Stengel—California State University- Fresno
The graphical display provides a clear presentation of the selected distribution. I am especially pleased with the ability to compute and plot as graphical overlays the various approximating distributions. The software even reports the degree of accuracy of the approximation fit. Besides understanding basic distribution shapes, the approximation display provides the best opportunity for learning this concept. … Grade A, the graphical display is excellent. The overlay plots with the approximating distributions are very well done. The ease of flexibility in adjusting distribution parameters provides for an effective in-class demonstration.
Glenn W. Milligan—Ohio State University
Visualizing Continuous Distributions
This module is a good introduction to the shape and characteristics of continuous distributions. The conceptual development in the learning examples follows a building pattern which should help students be more successful.
Sue B. Schou—Idaho State University
Visualizing Random Samples
The demonstration will help the student understand the inherent variation in shape from sample to sample. This is a valuable skill in that not all normal samples actually look normal. I also found that setting the base axis range to the population and then comparing the box plots could help motivate the concept of sample to sample variability. … Grade B+, good visuals. The template providing for 12 distributions in addition to the data sets and standard PDF’s is a nice addition.
Glenn W. Milligan—Ohio State University
I view this module as the best for teaching purposes. The template makes it very easy to change various attributes of the distributions and the learning examples are an excellent set of guided exercises. This module would be very helpful in making the student understand how sampling makes a difference in the shape of a sample set of data. The overview of concepts helps the student understand the importance of sampling.
Sue B. Schou—Idaho State University
Visualizing The Central Limit Theorem
The Central Limit Theorem is well demonstrated with this module. I especially liked the use of a non-normal population to sample from since “seeing is believing” for the average beginning statistics student.
Sue B. Schou—Idaho State University
This is another wonderful module for students to play with to better understand the Central Limit Theorem. I used this module in my introductory statistics class and the students quickly grasped the concepts.
Nancy C. Aiello—Northern Virginia Community College
Understanding the central limit theorem is a critical concept to master in any business statistics course. The present module is effective at motivating the sampling process from the parent population, and in developing an understanding of the rate of convergence to normality. … Grade A-, very good simulation visuals for the sampling and convergence process.
Glenn W. Milligan—Ohio State University
Visualizing Properties of Estimators
This module will help students understand the fundamental construction process for sampling distributions for several different parameters. The inclusion of medians in addition to means allows for a comparison of such effects as efficiency. I was most surprised to see a bootstrap simulation. … Grade A, this is an excellent set of simulation experiments that runs in real time. The notes reported in the “Analysis of Experiment” are very good and insightful. Overall, this is a high quality module.
Glenn W. Milligan—Ohio State University
This is an excellent chapter to demonstrate the basic properties of estimators. Excellent on all characteristics.
Robert K. Smidt—California Polytechnic State University- San Luis Obispo
Visualizing One-Sample Hypothesis Tests
Grade A, the display presenting the results for repeated confidence interval is excellent. I was delighted to discover the use of the display when the null hypothesis was false.
Glenn W. Milligan—Ohio State University
The module does an excellent job of allowing students to play with samples and better understand hypothesis testing.
Nancy C. Aiello—Northern Virginia Community College
I especially enjoyed working with this module. The templates are very well designed to demonstrate the concepts and the learning examples are well designed as well. I liked the four way split screen as well as the experiment showing the confidence intervals and normal curve. The confidence intervals appearing one at a time is a great effect since the student would find it more believable. Most of the software that I have seen places all 100 of the confidence intervals on the screen at once! This module effectively helps the student understand why one is able to reject the null hypothesis and also aids the understanding Type I error.
Sue B. Schou—Idaho State University
Visualizing Two-Sample Hypothesis Tests
Grade A
Glenn W. Milligan—Ohio State University
It is well done, particularly the four screen display. … Overall, a very good module.
Robert K. Smidt—California Polytechnic State University- San Luis Obispo
Visualizing Power and Type I/II Error
Great software. I use this module a lot in my presentations.
Roxy Peck—California Polytechnic State University- San Luis Obispo
Grade A, excellent graphics. The comparison of power for various sample sizes, alpha levels, and the effect of separation in parameter values is a very good learning tool. The graphic for reading the power level for the specified mean or proportion is very good. The ability to shade the relevant sampling distributions is a handy in-class display.
Glenn W. Milligan—Ohio State University
I think the Visual Statistics displays for the Type I and Type II errors, the power of the test and the critical values are very useful for teaching the power of a test. I particularly appreciate the authors’ use of different colors to shade and label the various elements.
Priscilla Chaffe-Stengel—California State University- Fresno
Visualizing Analysis of Variance
…it does a great job of showing just what is going on in a one-way ANOVA.
William Duckworth II—Iowa State University
Grade A+, the replication experiment is very well done. I could just about formulate an entire lecture around this module! One can even demonstrate the lack of statistical robustness in the presence of unequal sample sizes when homogeneity of variance or normality is violated. Very well done.
Glenn W. Milligan—Ohio State University
I appreciate the authors including demonstration and discussion of the underlying assumptions of ANOVA. [Its] inclusion in this module makes it more effective in helping a student develop complete understanding of ANOVA. This module is well designed in both the software and the learning examples. … This module helps visualize an ANOVA very well. In fact, I have not seen any software which does it any better. Congratulations.
Sue B. Schou—Idaho State University
Visualizing Bivariate Data Analysis
I like this module very much for illustrating outliers and influential points in regressions, and the difference between a point with a large residual and an influential point.
Roxy Peck—California Polytechnic State University- San Luis Obispo
Nice, nice, nice! The Do-It-Yourself section is especially useful with the ability to move data points around at will and to generate new samples with the push of a button. A nice selection of scenarios and data complements this module too.
William Duckworth II—Iowa State University
What a wonderful template to play with! The template for this module is very well done and covers all the changes that one wish to show for the concepts of correlation and regression. The learning example was well designed with step-by-step changes to show the effects on both correlation and regression.
Sue B. Schou—Idaho State University
My introductory students used this module and obtained an excellent understanding of correlation and simple regression analysis. This is an excellent module to incorporate into an introductory course and there are plenty of data sets to work with. The scenarios are useful in giving students examples of various positive and negative correlations. The presence of both an example section and the databases gives students an almost infinite variety of bivariate analyses to perform. My students spent hours investigating the databases and analyzing data of interest to them. Again, this is a wonderful way for students to discover the value of statistics.
Nancy C. Aiello—Northern Virginia Community College
Visualizing Simple Regression
I am impressed with the feature of replication. ... The module is easy to demonstrate in class without taking too much time.
Sung K. Ahn—Washington State University
In regression one concept that is not easy to explain from an intuitive standpoint is why the confidence bands around the regression line are curved. ... Most would just accept the fact and then forget it as soon as the exam is over. With Visual Statistics the student could quickly replicate the regression lines and the flare of the lines at the extremes. This process quickly illustrates the dependence of the slope and intercept. …The module is extremely easy to use. In the Do-It-Yourself tab a student can even create their own model and observe how a sample from a population will differ in the coefficients by a predictable amount.
Don R. Robinson—Illinois State University
Grade A, good graphics and corresponding display for inference results. The ability to modify the scatter plot to teach various aspects about the model and graphic are a real plus. I was particularly impressed by the replication (simulation) experiment. This is a marvelous way to teach the nature of interval estimation for the conditional mean.
Glenn W. Milligan—Ohio State University
Visualizing Regression Assumptions
I liked all the options provided [simulating models with (1) autocorrelated disturbances, (2) heteroskedastic disturbances, (3) disturbances having a non-normal distribution, (4) disturbances related to the independent variable, and (5) a non-zero mean].
Patti Collings—Brigham Young University
Visualizing Multiple Regression Analysis
This module is very well done. It illustrates major concepts graphically that are typically difficult to explain. … The examples and data sets available … are excellent since they demonstrate the major problems that occur in doing multiple regression.
Sue B. Schou—Idaho State University
The examples and databases are excellent. I really like the ease of use. It will not be necessary to spend much time teaching software use. … The module is very well done and easy to use.
John C. Dutton—North Carolina State University
Visualizing Regression Models
I really liked seeing the option of transforming the Y variable and or X variable and examining the new model. This should really help students understand what happens with a transformation.
Patti Collings—Brigham Young University
Visualizing Trends and Seasonality
This is another good module. While the individual components of a time series might be easy to explain, what happens when they are combined is more difficult. The ability to change the components and their magnitude was a tremendous help.
Robert K. Smidt—California Polytechnic State University- San Luis Obispo
... again, good graphics. I was delighted to discover that I can manipulate the error level in the data. This can be helpful as one develops the concept of the background random variation in a time series process. Optional statistical output includes measures of goodness-of-fit. It would be exciting to run this simulation in class while varying the model parameters and examining the effect on such measures.
Glenn W. Milligan—Ohio State University
Visualizing Statistical Quality Control
This is an excellent module which helps students to visualize quality control. It can be incorporated after students learn about means, sampling and standard deviations.
Nancy C. Aiello—Northern Virginia Community College
This module was the most fun to experiment with. The button for producing out of control SPC charts was great. I always have difficulty teaching these concepts because there just is not much real data which has the problems that I wish to show. After doing this module, a student would surely have clear understanding of statistical process control charts and how to interpret them.
Sue B. Schou—Idaho State University
Back to
What is
Visual Statistics?
 |
 |
|