What are the system requirements?
What are reviewer's comments on the software?
What topics does the program cover?
Can you tell me more about the free statistical and experimental design support?
Can you tell me more about the teaching experience of the software author?
What are some of the research publications of the software author?
Windows, any version
CDrom drive
Note: please temporarily disable any "pop-up killers" that you may have installed when running the program.
"A clear understanding of experimental design is a significant first step to prepare the budding brains for their later year interests in research. However, teaching experimental design remains a challenging realm for the statistics teachers. And there there is a dearth of good illustrative software that can facilitate this task. The Art of Experiment is an appreciative endeavour in this direction. The interactive approach adopted in the software can facilitate a great deal of work for the statistics teachers at high-school and undergraduate levels. Besides, it should sure make the statistics class a lot more fun and glamour filled even for the last benchers." Manoj Kumar,
Senior Research Scholar, School of Management, Indian Institute of Technology
"It is well-done, and the information is straight-forward and concise for students who need a really good introduction to the statistical concepts and tests covered." Rusty Sinclair, Training Director, Marco Polo State Administrator, Texas Computer Education Association
"Beginning with a simple example of the scientific method, The Art of the Experiment presents sophisticated statistical concepts and methods of analyzing data. The interactive nature of the software makes it a useful inclusion in secondary and undergraduate science classes." Noreen Neary, Science Teacher, Saint Vincent Academy, Newark, New Jersey
This program introduces the scientific method with a story about Vincent VanGogh contemplating sunflowers and wondering why they have different size flowers. It describes how he carries out a simple test to see if the type of soil might be a factor. Although not couched in the jargon of science, in his test VanGogh carries out the typical steps of an experiment.
After the introduction, the program uses four main sections to cover experimental design and statistical analysis:
I. Variables - Stating what you are interested in
II. Hypothesis - Stating the relationship between your variables
III. Rules - Stating what will constitute support for your hypothesis
IV. Tools - Understanding how statistics assess relationshipsI. Variables - Stating what you are interested in
The first step in carrying out an experiment is stating what variables you are interested in. This section defines the terms: variable, independent variable, dependent variable, data, N, sample, population, descriptive statistic and inferential statistic. The VanGogh flowers experiment is used to give examples of these terms.
II. Hypothesis - Stating the relationship between your variables
After selection of variables, the next step is to state how they are related in terms of a hypothesis. This section defines the terms: hypothesis, null hypothesis, alternative hypothesis, one-tailed/directional null & two-tailed nondirectional hypothesis.An exercise is presented that requires students to apply the terminology from sections I and II to a new experiment concerning the spiraling patterns of a snail and the survivability of its offspring.
III. Rules - Stating what will constitute support for your hypothesis
After the null and alternative hypothesis have been set up, the next step is to create rules that allow us to decide in favor of one or the other. The decision table is presented, which gives the logical basis for supporting the null or alternative hypothesis. The idea of type I and type II error are presented along with their associated probabilities, alpha and beta, and the concept of power.
An exercise is presented which reinforces the decision table concepts by having students place type I error, etc in the correct cells of the decision table.
The rules are further elaborated with the idea of establishing a critical value to determine the outcome of an experiment. This is presented in terms of a sample butterfly that had been given special nectar compared with a population of butterflies.
IV. Tools - Understanding how statistics assess relationships
Section III concerned the general rules of scientific evidence. This section gives the specific details concerning ANOVA and its test statistic, F, which is a primary tool used to assess whether an IV and DV are related. The logic of the F ratio is presented in terms of the related ideas of signal to noise, discriminability, overlap, due to chance vs due to our IV, variability accounted for to variability unaccounted for, and variability between to variability within. This is tied back into VanGogh's sunflowers. Two exercises follow in which students assess outcomes in terms of relative noise and relative signal.
The two types of variables (independent groups and repeated measures) are presented and compared with each other. The resultant designs (independent groups design, repeated measures design and mixed design) are presented.
The parts of an ANOVA source table are discussed. Ten exercises cap off the program. Students drag and drop icons on three different screens to indicate 1) whether the associated variables are independent groups or repeated measures, 2) to set up the source column of the source table, and 3) to set up the degrees of freedom column.
Statistical and experimental design support
As a free bonus that comes with purchase of the software, if you have questions about quantitative research methods, statistical analysis, scientific methods or experimental design please e-mail them to me. I regularly participate in e-mail listservs in stats and methods, so it's something I typically do anyway. In particular my areas of specialization are monte carlo methods, data transformation, bootstrapping & subsequent analysis techniques; but I can answer basic level questions in all areas of stats and methods. Just send me an e-mail--see the contact page. Of course, there are reasonable limits to this service. It's not intended as a means to avoid homework, ask really simple questions that a basic textbook or a quick search via google will answer or receive extensive consultation.
The author of the program, Jeff Rasmussen, PhD has taught at Malcolm Shabazz High School, Madison, WI; Lajo y Rodriguez, Madrid, Spain; Tulane University, New Orleans, LA; University of Arizona, Tucson, AZ; Purdue University Indianapolis, Indianapolis, IN. He has taught Graduate and Undergraduate Statistics, Advanced Technology of Measurement, Computer Applications in Psychology, Child and Adolescent Psychology, Introductory Psychology, Developmental Lab, Discovering Psychology, Foundations of Digital Technology.
Here is a partial listing of the publications in statistics.
Rasmussen, J. L. (1999). ANOVA multimedia: A Program for teaching ANOVA designs. In M. E. Ware & C. L. Brewer, (Eds). Handbook for teaching statistics and research methods (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.
Rasmussen, J. L. (1993). An evaluation of Shaffer's multiple comparison test. Methodika, 7, 44-52.
Rasmussen, J. L. & Dunlap, W. P. (1991). Dealing with nonnormal data: Parametric analysis of transformed data vs nonparametric analysis. Educational and Psychological Measurement. 51, 32-43.
Rasmussen, J. L. (1991). Data transformation and absenteeism. Methodika, 5, 47-62.
Rasmussen, J. L. (1991). SHAFHC: A FORTRAN implementation of Shaffer's multiple comparison procedure with HC enhancement. Psychometrika, 56, 153.
Rasmussen, J. L. (1989). Data transformation, type I error rate and power. British Journal of Mathematical and Statistical Psychology, 42, 203- 213.
Rasmussen, J. L. (1989). Parametric and nonparametric analysis of groups by trials design under variance- covariance inhomogeneity. British Journal of Mathematical and Statistical Psychology, 42, 91- 102.
Rasmussen, J. L. (1989). Analysis of Likert-scale data: A reinterpretation of Gregoire and Driver. Psychological Bulletin, 105, 167-170.
Rasmussen, J. L. (1989). Computer-intensive correlational analysis: Bootstrap and approximate randomization techniques. British Journal of Mathematical and Statistical Psychology, 42, 103-111.
Rasmussen, J. L., Heumann, K. A., Heumann, M. T. & Botzum, M. (1989). Univariate and multivariate groups by trials analysis under violation of variance-covariance and normality assumptions. Multivariate Behavioral Research, 24, 93-105.
Rasmussen, J. L. (1989). Distribution of the Job Descriptive Index: An evaluation of the effects of transformation. Educational and Psychological Measurement, 49, 89-98.
Rasmussen, J. L. (1989). A monte carlo evaluation of Bobko's ordinal interaction analysis technique. Journal of Applied Psychology, 74, 242-246.
Rasmussen, J. L. & Loher, B. T. (1988). Appropriate critical percentages for the Schmidt and Hunter meta-analysis procedure: A comparative evaluation of Type I error rate and power. Journal of Applied Psychology, 73, 683-687.
Moehle, K. A., Rasmussen, J. L. & Fitzhugh-Bell, K. B. (1988). Psychometric confirmation of neuropsychological theory. In J. M. Williams & C. J. Long (Eds.), Cognitive neuropsychology. New York: Plenum.
Rasmussen, J. L. (1988). "Bootstrap confidence intervals: Good or bad": Comments on Efron (1988) and Strube (1988) and further evaluation. Psychological Bulletin, 104, 297-299.
Rasmussen, J. L. (1988). An evaluation of small- sample statistics that test whether variables measure the same trait. Applied Psychological Measurement, 12, 177-187.
Rasmussen, J. L. (1988). Evaluating outlier identification tests: Mahalanobis D squared and Comrey Dk. Multivariate Behavioral Research, 23, 189-202.
Rasmussen, J. L. (1987). Parametric and bootstrap approaches to repeated measures designs. Behavior Research Methods, Instruments, and Computers, 19, 357-360. Rasmussen, J. L. (1987). Estimating correlation coefficients: Bootstrap and parametric approaches. Psychological Bulletin, 101, 136-139.
Rasmussen, J. L. (1986). An evaluation of parametric and nonparametric tests on modified and nonmodified data. British Journal of Mathematical and Statistical Psychology, 39, 213-220.
Rasmussen, J. L. (1986). Strategies for debugging control language of statistical software packages. Teaching of Psychology, 13, 36-38.
Rasmussen, J. L. (1985). Data transformation maximizing homoscedasticity and within-group normality. Behavior Research Methods, Instruments, and Computers, 17, 411-412.
Rasmussen, J. L. (1984). A FORTRAN program for statistical evaluation of pseudorandom number generators. Behavior Research Methods, Instruments, and Computers, 16, 63-64.
Rasmussen, J. L. (1984). A program for approximating R squared probability values in best subset multiple regression. Behavior Research Methods, Instruments, and Computers, 16, 61-62.
Rasmussen, J. L. & Dunlap, W. P. (1983). A program for simultaneously transforming two variables to maximize linear regression, homoscedasticity and normality. Behavior Research Methods and Instrumentation, 15, 357-358.
Dunlap, W. P. & Rasmussen, J. L. (1982). A program to determine data transformations maximizing linear regression. Behavior Research Methods and Instrumentation, 14, 357-358.
Rasmussen, J. L. (1982). DATIN, an interactive data entry program. Behavior Research Methods and Instrumentation, 14, 488.