<%@ Language=VBScript EnableSessionState=False %> <% on error resume next %> Salisbury University WAC Program

Writing Across the Curriculum

 

Holloway Hall

Writing Intensive Course Designs
Mathematics

Modern Statistics with Computer Analysis
Developed by Bob Tardiff

Modern Statistics with Computer Analysis is an introduction to using widely accepted statistical techniques to extract meaningful information from data.     There are three major concepts that students must understand in order to extract meaningful information from data.   First, students need to develop the ability to recognize how and why the data were gathered.   Students learn how to summarize or describe the data accurately using standard techniques.   Finally, if the method used to acquire the data meets certain standards, then students learn how a statistical inference can be made and how to measure the reliability of the method used to make that inference.  

Students completing Modern Statistics with Computer Analysis are expected to have an understanding of how

  • To understand how methods for acquiring data affect the data’s usefulness;

  • To use standard descriptive techniques to describe data; and

  • To use widely used inferential techniques to infer characteristics of a larger group.

Measuring the reliability of inferential statistical techniques relies on probability theory and consequently, inferential arguments are delicate.   The classical approach to measuring reliability, which is the primary focus of this course, is to measure the reliability of the method used for making the inference as opposed to measuring the reliability of the inference itself.   The Bayesian approach to inference addresses the reliability of the inference itself but is computationally much more difficult to implement, and more importantly, the Bayesian approach must assign probabilities to quantities that many view as non-random. 

Cheap computing is giving rise to some profound changes in statistics.   Classical statistics evolved in the early to mid 20th century focused on normal distribution theory because the requisite computations were straight forward and readily done on a mechanical calculator.  However many of these classical techniques, though widely accepted, lack robustness; i.e., they are sensitive to departures from underlying mathematical/probabilistic assumptions.    Cheap computing has given statisticians a tool for developing and continuing to develop robust techniques.  These techniques which are   computationally intense are increasingly becoming part of mainstream statistics.   

Cheap computing is also giving rise to massive data sets, data sets with millions of data points each of which has several dimensions.    Techniques for describing, summarizing and extracting meaningful information from these massive data sets or what is now known as data mining, is at the forefront of today’s statistical research.   

Formal Writing Assignments
Evaluating Data Sources

Evaluating Statistical Inference

Informal Writing Assignment
A Short E-mail on Today's Class