Applied Statistics Support Site 

A Free Service for those Interested in Applied Statistics.  This site is always under construction.   

Prepared by Craig A. Stevens, PMP, CC and his Students

 

Links to Resources on this Website

  1. Research Links

  2. Stats Defined

  3. Descriptive Statistics

  4. Sampling Methods

  5. Using Excel

  6. Discrete Stats

  7. Probability

  8. Continuous Stats

  9. Regression

  10. Hypothesize Test

  11. Confidence Intervals

  12. ANOVA Tables

  13. Least Squares Point Estimates

  14. Test of Variability

  15. Non-parametric

  16. Non-Parametric Based on Ranks

  17. Non-parametric Probability

  18. Test of Means

  19. Sign Test

  20. Business Intelligence and Data Mining

  21. Six Sigma

Other Links

  1. Link to Six Sigma

  2. Basic Statistics

  3. Six Sigma Online Training - Six Sigma Online Certification
    Six Sigma Training by Six Sigma Online, Black Belt Certification, Green Belt Training, DFSS Design for 6 Sigma Courses & Certification "Best Six Sigma Training on the Web" Payment Plans Available!!

  4. http://math.about.com/library/weekly/aa020502a.htm  

  5. http://mathworld.wolfram.com 

  6. http://www.fourmilab.ch/rpkp/experiments/analysis/chiCalc.html 

  7. http://www.robertniles.com/stats/tests.shtml

  8. http://members.aol.com/johnp71/javastat.html#WhichAnalysis

  9. http://www.plantbio.ohiou.edu/epb/instruct/quantmet/lectures/pdf/lec7.pdf 

  10. http://home.ubalt.edu/ntsbarsh/excel/excel.htm#rFtest

  11. http://members.aol.com/johnp71/javasta2.html#General

  12. Links From Dr. James L. Schmidhammer University of Tennessee Knoxville

    General Statistics

    StatSoft Electronic Textbook

    Linear Regression

    Topics on Linear Regression from UCLA Web Site

    StatSoft Electronic Textbook

    Logistic Regression

    Topics on Logistic Regression from UCLA Web Site

    Introduction to Logistic Regression (Patel)

    Logistic Regression Analysis (Dayton)

    StatSoft Electronic Textbook

    Decision Trees

    Decision Trees for Predictive Modeling

    Tree Structured Data Analysis

    StatSoft Electronic Textbook

    Neural Networks

    Neural Networks and Statistical Models

    StatSoft Electronic Textbook

    Model Assessment

    Tutorial on Receiver Operating Characteristic (ROC) curves

     

 

 

The Great American Humorist, Mark Twain, Once said, "I have come loaded with Statistics." 

 

www.thinkexist.com. Found by LaSaundra Patton-Matlock, (UoPhx, MBA Stats, 2005.)

 

 

"In this class, out of all the other classes that I have taken, I really learned the most.  I appreciate you and your teaching style. You have a great sense of humor and I wish you well ..."

 

Lavon Grissom (UoPhx MBA Stats, 2007)

 

This is a site to support the TNU MHR 3040 and  UoPhx QNT 530, 531, 554, and Mgt 510) workshops.

Links to research data  http://www.westbrookstevens.com/links_to_others.htm 

 

What is Statistics? – A Study of how to best (a) Collect Data, (b) Describe and Summarize Data, and (c) Draw Practical Conclusions Based on Data. A way to: 

  • test theories and practices in some cases related to doing work 
  • determine the characteristics of a population by using a sample 
  • Use data to make inferences under uncertainty 

Big Picture - Science of amassing data, taking a portion of it, and seeing what that portion tells us about the whole 

Little Picture - actual statistics themselves statistics with little “s” is any number that represents something else

 

What is Statistics?

 

By Verta Session-Webb, Kourtney Tharpe, Terri-Jane Hammerle, Augustine Ozobu

(UoPhx 2008)

 

The word "statistics" comes in various definitions, but not one could be regarded as complete and absolute. It has been frequently referred to either as the quantitative or numerical information itself or the methods of dealing with the information. To facilitate understanding, many statisticians prefer the term "statistical data" for quantitative information and the methods of dealing with the information the "statistical methods." Statistics is a collection of procedures and principle for gaining and processing information in order to make decisions when faced with uncertainty. (Davis, Simon & Utts, 2002). This paper will concentrate the on the practical application of the following statistical concepts such as statistical data, standard deviation, mean, mode, and median concepts, and sampling techniques.

 

Statistical Methods

 

Statistical methods are usually divided into five basic steps:

 

1. Collection. Quantitative information supplies facts for solving problems involving numbers. After the identification of the problem, certain relevant facts that can be expressed numerically should be gathered. Statistical data can be classified, according to source, as primary and secondary data. The latter is the most convenient and economic way of obtaining information from published materials. When published data are not available for a particular study, primary data may become necessary. Collecting primary data by survey is usually a costly, tedious, and time-consuming process.

 

2. Organization. Secondary statistical data are usually in organized form. However, those that are collected from a survey need organization. The first step in organizing data is editing. This is done to correct omissions, inconsistencies, irrelevant answers, and erroneous computations. The next step is classifying, which is to decide the proper classifications in which the edited items will be grouped. This is a very important step since the succeeding steps are affected by given classifications. The last step is tabulation. In this step, similar items are counted and recorded according to proper classifications.

 

3. Presentation. In order to facilitate statistical analysis, data are presented in textual form, table, or graph. Textual presentation is convenient only for presenting a few items. When a large mass of data is involved, this becomes inefficient and burdensome, since detailed explanations and properties of data may have to be repeated many times. Users usually prefer Tables if they can be effectively constructed. A graph is a pictorial representation of data. It usually gives the user of data only an approximate value of the facts.

 

4. Analysis. This is about the analysis of a population or universe based on a sample study. There are numerous methods of analyzing statistical data. Some are simple observation, while others necessitate the use of sophisticated and highly mathematical tools.

 

5. Interpretation. After the completion of the analysis, the findings must be interpreted. Correct interpretation will lead to a valid conclusion of the study.

 

 Reference:

 

 Davis, D., Simon, M., & Utts, J. (2002). Statistics and Research Methods for Managerial Decisions. Mason: South-Western College Publishing.

 

All of the material here has been developed from the following texts. For more information please buy and read the latest versions of these texts.

  • W. J. Conover, Practical Nonparametric Statistics 2nd ed, 1980, John Wiley and Sons, Inc.
  • Bowerman, Bruce L., Richard T. O'Connell, Business Statistics in Practice, 3rd edition, 2003, McGraw-Hill Companies, Inc. 
  • Cooper and Schindler. (2003). Business Research Methods, Eighth Edition. New York: McGraw-Hill.Lind and Mason. (2003). 
  • Lind. (2004). Statistical techniques in business & economics (11th ed). New York: McGraw-Hill.
  • Lind. (2005). Statistical techniques in business & economics (12th ed). New York: McGraw-Hill.
  • More to come...

Step One: Understand What is Needed:

  1. Define and Write Out Goals - Under line Information Being Requested, You may place in terms of Ho: and H1: (What are you trying to decide or test?)
  2. Understand Statistical Definitions, symbols, and Methods - Look up the terms you do not understand. 
  3. Find Data -- The following link will take you to possible research links.  http://www.westbrookstevens.com/links_to_others.htm 

Step Two: Decide on Method

  1. Descriptive: Start as simply as possible.  Simple Descriptive Statistics are methods of organizing, summarizing, and presenting data in informative ways (includes simple statistical calculations such as Mean, Mode, etc and pure graphical methods) 
  2. Inferential:  Inferential Statistics allows you to make a decision, estimate values, predict outcomes, or generalize about a population, based on a sample.
    1. To Determine the Probability of something happening (for classes MHR 3040, QNT 554, and  QNT 530).
    2. When estimating population parameters, one may use Confidence Intervals. (for classes MHR 3040, QNT 554, and  QNT 531)
    3. Looking to provide evidence for making a decision, you may Test a Hypothesis. (for classes MHR 3040, QNT 554, and  QNT 531)
    4. To Test or Compare Two of More Sets of Data
    5. Regression Models
    6. Non-parametric (See Statistical Definitions)
    7. Design of Experiments (DOE)

    Here is are a couple of links to find the appropriate method.

    1. http://www.graphpad.com/www/Book/Choose.htm 
    2. http://www.ats.ucla.edu/stat/stata/whatstat/default.htm
    3. KEY TO SELECTING CORRECT TEST

Step Three: Use Method: (Follow the links to the left to find methods and steps in using methods.)

Step Four: Testing Ho:

Step Five: Testing Residuals:

Plotting is a way of detecting correlation between residuals. A tendency to have runs of positive and negative residuals indicates positive correlation, implying the independence has been violated. Randomization is important in obtaining independence. If errors are normally and independently distributed with mean zero and constant but unknown variance s^2, then ANOVA procedure is exact test of Ho. 

  1. Calculate eij = Yij - Average Yi. 
  2. Place in the order as listed in data. k = 1,2,…N (3) Calculate Pk = (K-1/2)/N

Example Problems:  You may find example problems next to the subject when appropriate.  This site is always under construction, so it may take a while to put the information in the right places.