Designing good experiments is essential to being able to answer research questions. Knowing how to control variation among experimental units and account for it in the design process is crucial to obtaining data which can be meaningfully analyzed. Agronomic experiments present some unique challenges that influence both their design and analysis. Field-plot experiments often have a spatial component that may be addressed in the design using blocking and other techniques, but sometimes requires advanced analytical methods. There is often a temporal component that may require special consideration in the design and analysis phases of experimentation. Sometimes data collected in agronomic experiments do not conform to common assumptions and need special treatment in the analysis.
This course is designed to help you learn about statistical methods commonly used in agronomic research, learn how to use software for applying these methods, and to provide an opportunity for you to practice your analytical skills. The goal is to help you develop your ability to apply experimental statistics to agronomic research. Fundamental principles of experimental design will be covered in the context of field-plot experimentation. Experimental designs commonly used in agronomic research including complete and incomplete block designs, split-plot designs, Latin squares and other spatial designs will be described and analytical approaches to evaluating data from them presented. Additional emphasis will be placed on the development of linear additive models and calculation of expected mean squares for various experimental designs. Application of these principles to statistical computing using SASŪ will be covered in a weekly laboratory.
The class will meet for lectures twice a week at 9:30 a.m. in 1022 Agronomy Hall. We also meet each Thursday for a lab in the same room from 2:10–4:00 p.m. The lab is designed for you to apply the techniques discussed in lecture to problem sets using the SAS System for statistical analysis.
There will be three exams each covering only the material presented since the previous one. Each exam is worth 25% of your grade and homework accounts for the other 25%. You can always figure out how you are doing in the course by calculating your average performance on exams and homework on a percentage basis and then multiplying by 0.75 and 0.25, respectively. Sum these figures and you will have your score on a percentage basis. I grade on the standard ISU +/- scale with some adjustments if necessary.
The ability to design and analyze experiments is fundamental to a successful research career. Statistical methods are tools of the trade and your ability to function as a researcher will pretty much depend on your knowledge and skill in applying them. Hopefully, when you finish this course you will feel comfortable using the most common tools and know where to find help when you come up against a problem requiring an advanced approach.
Because our goal is to learn how to use these tools rather than just about them, the approach may be a little different than what you are used to in other courses. Designing and analyzing experiments takes skill. You would not expect to learn how to play a musical instrument or excel in a sport by reading and listening to someone lecture about it. Becoming good at these activities takes practice and it’s the same with experimental design. You have to practice designing and analyzing experiments to get good at it.
This course has three fundamental components that work together to help you develop your knowledge and skills: 1) theory, 2) technique and 3) practice. In lecture you will learn about statistical methods that are used in agronomic research and some of the theory behind them. In lab you will learn techniques used to analyze experiments using the methods discussed in lecture. Finally, you will have an opportunity to test your knowledge and practice your skills by completing homework assignments related to what you have learned in lecture and lab. My philosophy on homework is that it should be a learning experience. In this class it is not really used to assess your performance. It is designed for you to practice what you are learning. It will be graded, but not nearly as rigorously as an exam. In fact, as long as your turn your homework assignments in on time and have made a fair effort in solving the problems, you will be allowed to fix any errors and resubmit them for full credit.
There are two books you can use for the course readings:
Glaz, Barry and Kathleen M. Yeater. 2018. Applied Statistics in Agricultural, Biological, and Environmental Sciences. American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America, Inc., Madison WI.
doi:10.2134/appliedstatistics. (Required, but free online to ISU students)
Bowley, Stephen R. 2015. A Hitchhiker's Guide to Statistics in Biology, Generalized Linear Mixed Model Ed. Any Old Subject Books, Guelph ON. ISBN 978-0-9685500-4-5.
Each lecture topic will have assigned and additional readings associated with it. Some of the readings will be from other sources and you will be provided with links to read these materials online. The book edited by Glaz and Yeater is available to you for free through the ACSESS Digital Library which Parks Library subscribes to. It is a recent publication of the American Society of Agronomy that focuses on statistical methods used in the crop, soil, and environmental sciences. It covers the subject matter quite thoroughly and sometimes at a depth beyond the scope of this class. However, it will be a very useful reference for you as you develop statistical skills appropriate to your research in the future. The book by Bowley is more of an overview of statistical methods used in the plant sciences. You do not need to purchase it, but you may find it useful for providing some context for the methods you will be learning.
Recognizing that your time is valuable, readings are separated into assigned and additional readings. The assigned readings provide important background information and reinforce the material covered in lecture. You will get more out of the course if you do the assigned readings. The additional readings are generally more in depth and are there to provide reference material in case you want to delve further into a particular topic now or in the future.
WEBPAGE AND CALENDAR
All of the lecture notes, reading assignments, homework assignments and labs can be accessed through the menus on this course homepage. Some of the materials on the website are password protected. This is required to comply with the Fair Use standard for copyright protection. Once the class is up and running you will be authenticated from your ISU username and password. Please let me know if you have any trouble accessing course materials.
When you are trying to connect from off campus, you may need to establish a virtual private network (VPN) connection to gain permission to access the server. If you have not previously loaded the VPN software, download and install it following the instructions at https://www.it.iastate.edu/services/vpn. Each subsequent use only requires you to login.
The most important area of the webpage is the course calendar which you can access by clicking on CALENDAR in the header. The lecture topics for each class meeting are listed in the calendar. Click on the topic and you will be presented with reading assignments, links to notes, supporting materials, and homework assignments. You should do the assigned reading to get the most out of the course. The additional reading is for those of you who want to learn more about a particular topic. It is not required, but I encourage you to print out all the articles that are listed for reading. You may find them useful at a later date even if you do not need them now.
A handout of the slides presented in each lecture is available through the assignments webpage. Just click on the title of the lecture to open the notes in Adobe Acrobat. Even though all of the lecture topics are listed with the assignments, links to the notes are usually made active just a few days before the lecture is given. This gives me time to revise the notes if necessary and helps ensure that everyone is on the same page (literally). You are encouraged to print the notes and bring them to class with you so that you can make annotations on them.
Homework is due on the day under which it is listed. You need to complete the assignment and hand it in during class on the day it is due to get full credit. It you turn it in late, you will not have the option of having it re-graded later and you will lose a point for each class period it is late. So what’s the big deal with homework and why the concern about when it’s done? I usually discuss the homework assignment in class on the day it is due. If you have made an attempt to do it before then, you will be better able to follow the discussion and learn from it. If you wait until after the discussion to do the homework, you will likely do better but learn less; hence the grading penalty for those who put it off until after the discussion.
An iCalendar file that contains lecture topics and assignments is available for downloading and importing into your desktop calendar. See the downloads page above for the file and instructions on how to import it into Outlook. You can import it into any calendar program that can read the *.ics format. The calendar entries contain links to the course assignments webpage where you will find the notes, readings and other information.
The lab meets every Thursday at 2:10. The lab will start with a brief intro on the day’s topic. You will be given handouts with SAS code to type in and run. Data for the lab exercises can be downloaded from the course webpage. In the calendar, click on the lab link for that date and you will be taken to a page where you can view reading materials and data for the topic that is being covered. I make you type the code because I believe it helps you learn the process better and how to detect and deal with errors that invariably occur. SAS uses a legacy programming interface that will probably seem unfamiliar to you. However, it’s easy to learn and by a few labs in you will likely have no difficulty in using it.
You are strongly encouraged to bring your own laptop for use in the lab. You can access SAS (and other statistical software) from your laptop by linking to the campus virtual machine server at: https://vdi.iastate.edu. The initial screen gives you the choice of installing client software for accessing the VM or accessing it through HTML (web browser). Most of you will want to use the latter. Once you intall the client or initiate a web session select the Statistics application server. On the virtual desktop you will find a folder labelled Statistics. Click on the folder and select SAS 9.4 from the program listing to intitiate a SAS session. Anything you create during a session on the VM is lost when you end it. For that reason you should link to your CyBox folder from within the session so that you can save SAS programs and other files you create. A better alternative for running SAS on your own laptop is installing the program on it and executing it locally. This has the advantages of better performance and easy use of resources local to your laptop such as disk drives, file folders, and printers. The SAS license is free to students. It takes a little effort to download and install it, but it is usually worth it especially if you intend to keep using SAS after the class is over. See the next section for details about obtaining a free SAS license.
No worries if you do not have a laptop to bring to lab. One will be made available to you for use during lab. They will be located in a cart in front of the room. Your access to SAS from these computers will be through the campus virtual machine using the same link as above.
SAS AND EXCEL
There are two software programs which you will be using frequently throughout the course: SAS and Excel. If you are not already familiar with them you should become quite comfortable using them by midway through the semester.
Excel is a very good spreadsheet program familiar to most agronomists and students, but it is inadequate for many statistical analyses. Excel is an excellent program for data entry, ease in copying and pasting, page and print set up, and formatting. Thus, it is an excellent tool to enter data and format tables of results. However, SAS is superior for most statistical analyses. It is fairly easy to import data from Excel into SAS and to copy results into Word documents or Excel sheets, and your homework assignments will give plenty of practice in this. The latest version of SAS installs a SAS add-in to Excel that provides you point-and-click access to a host of analysis features many of which will be covered in this course. There is also a version of SAS that runs under a powerful shell called SAS Enterprise Guide. In both the Excel and Enterprise versions, these programs provide an interface to the same SAS engine that the programming interface you will be learning in lab uses. There are two primary reasons that we will be working with SAS at the programming level: 1) greater flexibility and access to features not available in the point-and-click programs, and 2) believe it or not, greater simplicity in use.
SAS recently created a University Edition of its software that is free for those learning and teaching statistics and quantitative methods. It uses a new programming interface called SAS Studio which is a full-featured programming environment with many familiar features found in other advanced programming environments. It has right-click access to context sensitive syntax that is very useful when coding procedures that are new and unfamiliar to you. All of the programs developed for the lab will run under SAS Studio and you may find the documentation and support more user friendly than traditional SAS manuals. You can download and install SAS Studio directly from SAS at: https://www.sas.com/en_in/software/university-edition/download-software.html. The program runs under Linux and the installation process will likely require you to create a virtual machine on your computer. Consequently, installing and running SAS Studio is a little more complicated than using programs that are designed to run under the Windows or Mac operating systems. Some of you, however, may find it to be worth the effort.
As the complexity of the analyses you learn increases, certain SAS procedures have options that enable you to analyze your results while accounting for departures from the usual assumptions. The SAS program can handle missing data correctly and performs calculations at a higher precision than some more elementary statistical programs. SAS is a very standard statistical analysis program. It has capabilities far in excess to what will be needed in this class, but which may be useful to you later as you progress with your quantitative skills. SAS has been around a long time and the SAS interface reflects this legacy. It originally ran on mainframe computers and received instructions and data from stacks of punch cards. We will be using the interactive programming interface, which may seem foreign to you at first. However, you will quickly master it and find it to be a very powerful tool for analyzing data. Once you get used to interpreting SAS output, you should be better able to interpret output from the many other statistical computing programs, such as R, Minitab, Genstat, and JMP. If you feel strongly about using some other program, you are encouraged to do so on your homework assignments. As long as you achieve a correct answer you will get full credit. However, be aware that there may be some exam questions specific to SAS code. These questions work two ways: they test your knowledge of SAS and how it can be used but more importantly, from the code you write your understanding of the problem can be inferred.
As described above, you can access both Excel and SAS and bring them to your desktop using the VM server. Both programs are also available to students at very steep discounts through licensing agreements with the manufacturers. You can obtain an annual license for SAS for free at: https://www.it.iastate.edu/services/software-students. This is a great opportunity and will come in very handy when you begin to analyze data from your own research. You can also learn about obtaining a license for Microsoft Office (and Excel) at the same website.
STUDENTS WITH DISSABILITIES
Iowa State University is committed to assuring that all educational activities are free from discrimination and harassment based on disability status. All students requesting accommodations are required to meet with staff in Student Disability Resources (SDR) to establish eligibility. A Student Academic Accommodation Request (SAAR) form will be provided to eligible students. The provision of reasonable accommodations in this course will be arranged after timely delivery of the SAAR form to the instructor. Students are encouraged to deliver completed SAAR forms as early in the semester as possible. SDR, a unit in the Dean of Students Office, is located in room 1076, Student Services Building or online at www.dso.iastate.edu/dr/. Contact SDR by e-mail at firstname.lastname@example.org or by phone at 515-294-7220 for additional information.
Office: 1571 Agronomy Hall
Mailbox: 2101 Agronomy Hall
Office hours: Tuesday / Thursday 8:30-9:30 a.m.
195A Seed Science
Mailbox: 1126 Agronomy Hall
Office hours: Monday / Wednesday 4:00-5:00 p.m.
1126K Agronomy Hall