Short Course Listings
Short Course

Course Information
Course Title: (CANCELED) Statistics for the Non-Statistician with Applications to Analytical Chemistry
Categories: 1 - Pharmaceutical Sciences
2 - Data Analysis
3 - Statistics
Instructor(s): James De Muth Course Number: 20
Affiliation: University of Wisconsin
Course Date: 03/06/2017 - Monday Course Length: 1 1/2 Day Course
Start Time: 08:30 AM End Time: 05:00 PM
Course Date 2: 03/07/2017 - Tuesday    
Start Time: 08:30 AM End Time: 12:30 PM
Fee: $850 ($1175 after 2/18/17) Textbook Fee: $75

Course Description
This short course is designed to provide non-statisticians with practical information needed to apply basic statistical tests to problems typically faced in analytical chemistry. Emphasis will be placed on the most commonly utilized statistical tests. The course focuses on the efficient utilization of statistics; not abstract theory. Practical examples drawn from scientific research will be used to illustrate statistical concepts. Emphasis will be placed on the appropriate use of statistics and the interpretation of their results and presents an overview of basic elements associated with statistical tests and their practical application. This course represents a merging of two separate one-day courses that have been presented at Pittcon since 1991.

Target Audience
The course is intended for scientists and technical personnel who deal with analytical data. It is primarily for those who have never taken a formal statistics course. However, it will also serve as an excellent review for those with previous experience in statistics; especially those who have had negative experiences with statistical instruction.

Course Outline
1. Types of Research Data and Methods for their Presentation
Basic definitions used in statistics will be discussed, as well as the differences between data which represent samples and populations. Differences between discrete and continuous variables and their role in selection of appropriate statistical tests with be emphasized. Various presentation modes for handling raw data will be analyzed including histograms and descriptive statistics. Measures of central tendency will be explored including mode, median, mean, variance and standard deviation; and their importance in statistical tests.
2. Key Concepts in Making Decisions Based on Statistical Tests
The importance of appropriate sampling and various types of sampling will be addressed. The role of probability and the probability distribution on outcomes of statistical determination will be analyzed.
3. Statistical Inference: Confidence Intervals and Hypothesis Testing
Areas briefly considered will be continuous probability distributions, normal distribution, sampling distributions, standard error of the mean, and central limit theorem. The use of confidence intervals and level of significance will be discussed. Of primary interest will be the application of hypothesis testing on decision making and the methods of minimizing type I and II errors.
4. Parametric Statistics: the t Tests and One-way Analysis of Variance
The issues of standard error in confidence intervals, student t distributions and test parameters will be discussed. Statistical tests presented include: one sample t-test; two sample t test and matched pair t test. The use of the Analysis of Variance (ANOVA or F test) will be presented for situations when the independent variable presents with three or more levels. Interpretation of a significant ANOVA requires a post hoc comparison and these will be briefly presented. The complete randomized block design can be use for paired data that is measured more than twice.
5. More Applications for the Analysis of Variance
Applications of the one-way ANOVA can be expanded to other situations. The two-way ANOVA can be used when there are two or more independent variables and one dependent outcome. Measures of repeatability and reproducibility will be discussed as well as the use of the F-ratio to compare the precision of two methods.
6. Correlation, Linear Regression and Outliers
Time will not permit an in depth presentation of correlation and regression analyses. However, there will be a brief introduction to the use of statistics that involve two or more continuous variables.
7. Chi Square and Nonparametric Tests
The most commonly employed nonparametric tests are the chi square procedures. Several applications of the chi square distribution will be demonstrated including the chi square test of independence and Yates' correction for a 2x2 design. A variety of distribution free tests are available when data does not meet the criteria for the parametric tests (t-tests and F tests).

Course Instructor's Biography
James E. De Muth, Ph.D., R.Ph. is Professor Emeritus in the School of Pharmacy at the University of Wisconsin-Madison. He received the B.S. in Pharmacy degree from Drake University and the M.S. and Ph.D. degrees in Pharmacy Continuing Education from the University of Wisconsin-Madison. His primary responsibilities are the development, implementation and evaluation (relying heavily on statistical methodology) of continuing education offerings for individuals in the pharmaceutical industry and pharmacists within the United States. Dr. De Muth has authored over 40 research articles in the pharmacy and adult education literature and Basic Statistics and Pharmaceutical Statistical Applications. He has taught over 150 statistical short courses throughout the United States and internationally, including annual short courses for the Pittsburgh Conference since 1991. He served as the Chair of the Biostatistics Expert Committee (2000-2005), Chair of the General Chapters Expert Committee (2005-2010) and is currently (2010-2020) Chair of Pharmaceutical Dosage Forms Expert Committee for the United States Pharmacopeia.