STAT 401A: Statistical Methods for Research Workers
Ag. and Bio. Sciences

Spring 2018

SYLLABUS and other useful information



 

Lectures MWF 9-9:50am, 1126 Sweeney
Laboratory: W 12:10 - 2 pm, 2272 Gilman
Instructor:  Philip Dixon
pdixon at iastate dot edu

2121 Snedecor Hall
Ames IA 50011-1210

4-2142
 

Office Hours: Thursday, 4-5 pm; Friday 3-4 pm. 2121 Snedecor
TA/grader Jingru Mu, mujingru at iastate dot edu
TA office hours: Thursday 3-4 pm. 2404 Snedecor.
Questions:: Please feel free to e-mail me ( pdixon at iastate dot edu) or Jingru mujingru at iastate dot edu) anytime with questions or comments.
On Sunday evening (the day before HW is due), PMD will check e-mail until shortly before 9pm. Anything sent later than that probably won't get answered until the next morning.
Objectives: By the end of the course, students should be able to analyze and interpret data from experimental studies using parametric and non-parametric methods. Students should be able to appropriately use statistical methods for comparisons of two groups, comparisons of multiple groups, and relating a response to one or more continuous variables. Students should understand how features of the study design influence the choice of statistical method and the type of conclusions that are appropriate. Students should recognize the conditions necessary for an appropriate statistical analysis, how to check if those conditions are met, and understand the consequences of violating those conditions.
Text Ramsey, F.L. and Schafer, D.W., 2012. The Statistical Sleuth, 3rd ed. Duxbury
Goals:
 
1) Understand variation and its consequences for drawing conclusions from data. 
2) Be familiar with some standard statistical methods:
       when and how to use them,
       how to use statistical software,
       how to interpret statistical results.
3) Be able to apply statistical principles to novel problems. 
This class emphasizes the appropriate analysis of experimental data. I presume you will be using class material within the next year. If it will be two or three years before you analyze data, I suggest you delay taking 401.
Grading: Weekly Homework: 120 pts 
Two Midterms: 100 pts each 
Final: 130 pts 
Course Outline (proposed): 
Week Dates Chapter Topic
Jan 8-12 1 Types of studies, Statistical Inference,
Data summary
Jan 15, Martin Luther King Day No class
Jan 17-19 2 Comparison of two groups:
    Hypothesis tests
Jan 22-26 2     Confidence Intervals
Jan 29-Feb 2 4     Nonparametric methods
Feb 5-9 3     Assumptions and robustness
Feb 12-16 5 Comparison of multiple groups
Feb 19-23 6 Linear combinations and multiple comparisons
Feb 21 MIDTERM I In lab
Feb 26-Mar 2 6 False Discovery Rate, Choosing a method
Mar 5-9 7, 8 Linear regression
  Mar 12-16 Spring break, no class
10  Mar 19-23 8, 9    Lack of Fit, Correlation, Multiple Regression
11  Mar 26-30 9, 10, 11    Multiple regression (cont.)
12  Apr 2-6 12 Model selection
  Apr 4 MIDTERM II in lab
13 Apr 9-13 13,14 Two-way ANOVA (intro)
14  Apr 16-20 18, 19 Contingency tables
15  Apr 23-27 20, 21 Logistic regression
Apr 25Last HW due in lab
May 2 Final Exam Weds of finals week, 7:30-9:30 am.

Details:
Sections of 401 The different sections of 401 are not interchangeable. Each is essentially a different course.
Section A focuses on the analysis of data from experimental studies, although we do briefly discuss observational studies. It will use examples relevant to the target audience (agriculture and biology). Computing will be your choice of SAS, JMP, or R.
Section B is for graduate students in social sciences and discusses both experimental and observational studies.
Section C is for graduate students in the physical sciences, math, and engineering. It includes more mathematical detail.
Section XV is an online offering for graduate students in the physical sciences, math, and engineering. It includes more mathematical detail.
Section XW is an online offering. I believe this is for the social sciences but am checking that.
Student background: Section A is intended for graduate students working in agriculture or the biological sciences, broadly interpreted. The prerequisite (Stat 101, 104, 105, or 226) is enforced for undergraduates; it is waived for graduate students.
The material I cover is intended for graduate students who will be analyzing data from their own experimental studies within a year of taking the class. Others are welcome but be aware that I have graduate-level expectations. In particular, I expect you to ask questions when you don't understand something. I use a graduate-level grading scheme (mostly A's and B's) but I reserve the right to give lower grades when appropriate.
Text: Each chapter includes two case studies, main material and a section of related issues. Please skim the case studies and read the main material in the assigned chapter(s) prior to the start of the lectures. In some chapters, parts of the related issues will also be assigned. These will be announced in class.
My lectures will cover the same concepts, but I will often use different examples and may use a different presentation. 
There is not time to lecture on all the details. I expect you to read the assigned material and ask questions on anything you don't understand.
It will probably help to reread the chapter(s) after the relevant lectures.  Through the semester, I will distribute a reading list identifying the most important parts of each chapter.
Lab: Lab time will be used for five different activities: 
Some hands-on illustrations of statistical principles.
Return HW
Discussion and Q/A on lecture material and homework problems.
Use of SAS, JMP, and/or R (most of the lab period). This part of the lab is "flipped". I provide support for three statistical computing languages. You choose which you want to work with. (And you may work with multiple if you wish). The class web site will describe how to implement statistical methods in each language. For SAS and R, this will be a file of code and a document describing what the code does and how to interpret the output. For JMP, this will be a document including screen shots that describes how to navigate the menu system to obtain the desired analysis and then how to interpret the output. You are to work through your choice of document and ask if you have any questions or if something doesn't work.
Each week, I also provide a self-assessment exercise. I recommend you work through this during the lab period, but it is optional.
You may also work on the HW during lab period.
Midterm exams will be held during the lab period in weeks 7 and 12.
Homework:
 

 
 

Weekly homework assignments will be posted on the web site and announced in class.
Goal is to provide practice using statistical concepts. 
Discussion with friends and classmates is strongly encouraged
Please write up your answers individually. Copying papers is not a good way to learn and will not be tolerated. 
No late homework accepted.
Lowest homework score will be dropped. 
Solutions will be posted on the class web page soon after the due date.
HW will be assigned no later than Wednesday noon (prior to lab), due Monday 9am (in lecture), and returned Wednesday in lab. This schedule will be modified the week of or preceding each exam.
Computing:
 
 
 
 
 

 

This class focuses on statistical concepts, not details of a specific computing package. We will rely on the computer to do most, if not all, the appropriate calculations, so most of lab time will discuss how to use statistical software.
The choice of software will be discussed in the first week of class. We will provide support for SAS, JMP, and R.

You may use another package if your lab group uses something other than SAS, JMP, or R. Please check with me to make sure that package is appropriate for this class. EXCEL is not appropriate.

If you plan on taking Stat 402, some ag/bio sections use all three languages; others use SAS. The social science and engineering sections generally use JMP.
 

Exams: Exams will be held during lab and at the designated final exam time for a Monday 9am class. You should bring a calculator. I will provide formulae and computer output. My goal is to see how well you can use class material to analyze data.

Makeup exams will be given only if you contact me and get approval prior to the scheduled exam.

Other
questions:
Please ask in class or e-mail me:  pdixon at iastate dot edu 
University
policies:
This class and its instructors follow the ISU policies on academic dishonesty, disability accomodations, dead week, harassment and discrimination, and religious accomodation. Summaries of these policies are here. For this class, the following details are relevant:
Academic honesty: I encourage you to help each other with homework and lab material. Help each other understand the problems, write code, debug code, and interpret the output. You may share code, but I encourage you to understand that code even if you didn't write it. You are required to write your answers in your own words.
Dead week:The last HW will be due during dead week.
Contact Information: If you are experiencing, or have experienced, a problem with any of the above issues, please contact Philip Dixon. If you prefer to bring a concern to the attention of university administration, please contact Dr. Max Morris, the Statistics Department Head, or academicissues@iastate.edu.