MAA 520 Data Science Fall 2021

PROFESSOR:        Dale Lehman 
Email:            Dale.Lehman@loras.edu
Class meets Tuesdays, 6:30-8:30 PM online and Nov 13-14 in person.
Grading:    Homework 50%
                   Group Cases 50%
                 
You are permitted to resubmit each homework once.  Resubmitted homework will receive half the credit that it would receive if submitted on time.  You can resubmit parts of the homework that you did not do or did not adequately understand.  Thus, if you were to hand in no homework on its due date, but redo the assignments perfectly after we go over it, you would receive a grade of C for the homework portion of the course.  My standards for grading are:

A    demonstrate a command of the tools
B    control of the tools, but too many omissions or errors for it to be command
C    effort is clear, but the tools are controlling you rather than the reverse
D    effort is not clear
F    you have convinced me that you did not really try

Textbooks:  
None required, but any basic introductory statistics book will serve as a useful reference.
Brief Outline (complete outline on eLearn)

Data Manipulation, Cleaning, and Visualization

Statistical Inference:  one or two variables

Statistical Inference:  more than two variables

Multiple Regression Modeling

Classification Modeling

Complete Syllabus on eLearn


Course Objectives

"If the statistics are boring, then you've got the wrong numbers."

(Edward Tufte, The Visual Display of Quantitative Information, Graphics Press).  The purpose of this course is to  develop your abilities to manipulate, analyze, and present data.  We will explore a wide variety of data in our efforts to find interesting stories and to develop an appreciation for both the limits and power of data analysis.  This course relies extensively on the use of JMP, a powerful statistical package.  The software will be provided for you.

While this course is not demanding mathematically, it does presume a basic understanding of algebra, geometry, and a previous course in statistics. You are not assumed to remember much from your prior statistics course, but the basic concepts (e.g., standard deviation) will be reviewed rather than presented as completely new ideas.   The computer will do most of the tedious work, but you need to provide intelligent instructions and be able to decipher the computer's graphical, algebraic, and numerical output.  At the conclusion of the course, you should be able to take almost any data set, conduct exploratory analysis of the data, perform somewhat appropriate analytical methods, and present your findings.  This course demands time and effort, but the payoff is large in terms of usable skills and enjoyment (in my opinion).