Learn R
A Free Software Environment for Statistical Computing and Graphics
The course is designed for natural resource managers and is open to all who are interested without charge. R (the open-source/free version of S) provides a comprehensive environment for statistical analysis and graphics and is unrivaled in the availability of new, cutting-edge applications. It runs under most operating systems. Audio of the presentations is available either using Voice of IP through your computer speakers and optional microphone or headset or by calling a phone bridge long distance. Live video of the presenter's computer screen is available over the web. You can also share your computer's screen with other participants when asking a question or making a point. An audio and video recording of the presentations and discussion will be available on our FTP site after the presentations. There are no specific prerequisites but some knowledge of statistics would be helpful. A basic knowledge of computers and the internet will be assumed.
The course is presented by the US Geological Survey, Status and Trends Program (Paul Geissler, Paul_Geissler@usgs.gov) and the National Park Service, Inventory and Monitoring Program (Tom Philippi, Tom_Philippi@nps.gov). Please contact us for more information. Please also see Tom's companion site.
The course will start Monday, November 9. There will be presentations on Mondays and Wednesdays, and a lab on Tuesdays for two hours. The times will be: Hawaii 9:00-11:00, Alaska 10:00-12:00, Pacific 11:00-1:00, Mountain 12:00-2:00, Central 1:00-3:00, Eastern 2:00-4:00, UTC 7:00-9:00. The course will continue until we finish the outline.
Last year's course website: http://www.fort.usgs.gov/brdscience/learnR08.htm#schedule
Please register for both
• the course at http://www.fort.usgs.gov/brdscience/courseRegister.aspx and
• the webinar at https://www1.gotomeeting.com/register/125974937
We ask that everyone register individually for the course, so we can email updates to you and so you can participate in the email discussion.
We also ask that person that will be watching the webinar individually and one person from each group that will be watching it together register for the webinar. We can have up to 1,000 people participate in the webinar, so spaces probably not be a problem, but often people like to participate in groups to discuss the presentation.
US Department of the Interior employees can get credit for this course on their record by registering with DOI LEARN https://doilearn.doi.gov/ Web registration has closed, but Paul will still register people through the first week of class. A certificate of participation is also available.
R is a very powerful system for statistical computations and graphics, which runs on Windows, UNIX and Mac computers. You can think of it as a combination of a statistics package and a programming language. It can be downloaded for free from http://www.r-project.org/ . The R Wiki provides an online forum http://wiki.r-project.org/rwiki/doku.php and documentation.
ADVANTAGES
• With the increasing cost of commercial statistical package, a free package is very attractive. However, free does not imply second rate. R is a high quality package that is better than commercial package in many respects.
• There are over 2,000 contributed packages (extensions) available for R to perform a great variety of statistical and graphical procedures.
• R includes a powerful programming language for selecting, manipulating and transforming data.
• R is interactive and supports data analysis, which should be interactive and exploratory.
• Although SAS is the most common statistical package in general use, R (or S) is the most popular with statistical researchers (Faraway 2005: x).
• New statistical methods often are available first in R. For example, GRTS analyses are only available in R at this time to my knowledge.
• R can easily import and export data to and from Microsoft Access and Excel as well as text files.
PRESENTATIONS
Paul Geissler (USGS Status & Trends Program, Paul_Geissler@usgs.gov) lecture on Mondays and Wednesdays and Tom Philippi (NPS Inventory and Monitoring Program, Tom_Philippi@nps.gov) will conduct a lab on Tuedsays. We will start by showing how to download and install R and the R Commander and Rattle user interfaces. The first part of the course will address the needs of the casual or occasioinal user, who uses the menus to do common analyses and graphics. For the casual user, the free online
documentation
will probably be sufficient. Later we will introduce the R language, and provide an introduction to statistical procedures, using Michael Crawley (2007, The R Book, Wiley, 942 pages, $67). We will also demonstrate the extended statistical analyses of selected data sets using R. We plan on following-up the introduction with modules on special topics on interest to natural resource managers. Suggestions of data sets to analyze and of additional topics will be appreciated.
Minimum computer requirements for Webinar are:
For PC-based Users
•
Required: Windows® 2000, XP, 2003 Server or Vista
•
Required: Internet Explorer® 6.0 or newer, or Mozilla® Firefox® 2.0 or newer (JavaScript™ and Java™ enabled)
• Internet Connection Required: cable modem, DSL or better recommended
•
Recommended: Minimum of Pentium® class 1GHz CPU with 512 MB of RAM (2 GB of RAM for Windows Vista)
For Mac®-based Users
•
Required: Mac OS® X 10.4 (Tiger®) or later
•
Required: Safari™ 3.0 or newer, Firefox® 2.0 or newer; (JavaScript™ and Java™ enabled)
• Internet Connection Required: cable modem, DSL or better recommended
• Required: PowerPC G4/G5 or Intel processor, 512 MB of RAM or better
To Use VoIP
• Required: Fast broadband connection (384 kbps or more recommended)
• Required: Microphone and speakers (USB headset recommended)
If you are having problems connection to the webinar, see the GoToWebinar FAQs, especially the one about your firewall at https://www1.gotowebinar.com/en_US/webinar/pre/support.tmpl
You can also contact their help at http://support.gotomeeting.com/ics/support/default.asp?deptID=5641
or call them at:
| Country | Toll-Free | Direct Dial |
|---|---|---|
| United States | 1-800-263-6317 | +1-805-617-7000 |
| United Kingdom | 0800-404-9279 | +1-805-617-7000 |
| Europe | 00-800-7770-0011 | +1-805-617-7000 |
| Mexico | 001-888-840-6244 | +1-805-617-7000 |
| India | 000-800-100-3359 | +1-805-617-7000 |
| Australia | 0011-800-7770-0011 | +1-805-617-7000 |
| Asia-Pacific | 0011-805-617-7000 | +1-805-617-7000 |
SCHEDULE:
There will be lectures on Mondays and Wednesdays, and labs on Tuesdays, starting November 9, and continuing until we complete the outline.
| Hawaii | Alaska | Pacific | Mountain | Central | Eastern | UTC |
| 9:00-11:00 | 10:00-12:00 | 11:00-1:00 | 12:00-2:00 | 1:00-3:00 | 2:00-4:00 | 7:00-9:00 pm |
The date for later topics, and links will provided as the course progresses. Some topics will take several sessions to cover. We will update this schedule and the topic pages (listed in the link column) shortly before each session. The recordings will be available shortly after the presentations at ftp://ftpext.usgs.gov/pub/cr/co/fort.collins/Geissler/LearnR.
I thought a long time about how to organize this course and decided to make take two approaches to R. First, I will demonstrate the use of menus in R Commander and Rattle. I agree with John Fox, the author of R Commander when he said "I must confess that I'm not terribly enamored of menu/dialog box interfaces to statistical software, but I do feel that these interfaces have a role for introductory and occasional use. The Commander interface is not innovative, but I hope that it's simple and familiar. One of my design goals was to wean users from the GUI to writing commands, which is one motivation for the script window. It is relatively easy for me to add statistical functionality to the R Commander, and I'd appreciate suggestions for what you'd like to see implemented. Please remember, however, that my intention is to keep things simple and basic. In particular, I don't like extensive menu/dialog-box interfaces to large statistical systems that attempt to provide access to every option and procedure. In R, of course, which relies on hundreds of contributed packages, this is not feasible anyway."
The second approach is use written commands. Menus are convenient, but you can only access the full power of R by using written commands. I know that commands are difficult to remember, and I often use the menus instead of looking up the command. However, if you use R very much, you will soon reach the limits of menus and have to use the commands. For that reason, books on R concentrate on the command approach. One contribution of this course is to demonstrate the menus. After covering the R Commander and Rattle menus for occasional use, I will circle back and cover the R language and commands. Last year, I focused on the command appraoch, and you can see those notes for the command approach.
I will focus on using R. Although I will briefly discuss the underlying statistics, the course will not teach you statistics. The course follows Crawley (2007) The R Book, and I expect that you will be familiar with the concepts presented in that text.
I will develop web pages for each topic as the course progresses and try to stay about a week ahead of the class. There will be some breaks in the presentations. Please check here for the current schedule.
Essentially all the informaton will be available either on the topic web pages or in the text book(s), so you should not have to take notes.
| Link | Topic | Date | Recording (topics) * |
|---|---|---|---|
| Topic 1 Tom's Notes |
1. Getting started • Making sure everyone can connect to the webinar • Downloading and setting up R, Packages, R Commander and Rattle • Setting defaults in Rprofile.site We will start R during the next session. |
Nov. 9 | None |
| Menu Approach | |||
| Topic 2 |
R Commander Menus 2. File, Edit and Data menus • Importing data |
Nov. 16 |
R10-topic02.wmv |
| Lab 1 | Lab: Understanding data in R, simple data access in Rcmdr (and possibly rattle), simple data access via the command line. | Nov. 17 | R10-lab01.wmv |
| Topic 3 Topic 4 |
R Commander & Rattle Menus 3. Graphics 4. Classical tests, Crawley (2007) Chapter 8 |
Nov. 18 | |
| No sessions Thanksgiving week - November 23-25 | |||
| Topic 4 Topic 5 |
R Commander & Rattle Menus |
Nov. 30 Dec. 2 |
|
| Lab: Data manipulation (long v. wide format, reshape, merge, subset); advanced data i/o (SQL databases & IRMA services?) |
Dec. 1 | ||
| Lab: Graphics | |||
| No sessions December 7-9, Paul and Tom have a meeting. | |||
| Linear Models Overview and Regression, Crawley (2007) Chapters 9 and 10 | |||
| Regression Diagnostics, Fox (2002), Chapter 6 | |||
| Analysis of variance, Crawley (2007) Chapter 11 | |||
| Analysis of covariance, Crawley (2007) Chapter 12 | |||
| Generalized Linear Models - Count Data, Crawley (2007) Chapters 13 and 14. | |||
| Count Data in Tables, Crawley (2007) Chapter 15 | |||
| Proportion Data, Crawley (2007) Chapter 16 | |||
| Binary Response Variables, Crawley (2007) Chapter 17 | |||
| Generalized Additive Models, Crawley (2007) Chapter 18 | |||
| Mixed-Effects Models, Crawley (2007) Chapter 19 | |||
| Non-linear Regression, Crawley (2007) Chapter 20 | |||
| Tree Methods, Crawley (2007) Chapter 21 | |||
| Multivariate Statistics, Crawley (2007) Chapter 23 | |||
| Spatial Statistics, Crawley (2007) Chapter 24 | |||
| Time Series, Crawley (2007) Chapter 22 | |||
| Command Approach | |||
| Introduction to the R language for data manipulations | |||
| Introduction to the R Language | |||
| Classical tests, Crawley (2007) Chapters 8 and 15 |
* Files are available at ftp://ftpext.usgs.gov/pub/cr/co/fort.collins/Geissler/LearnR/. Be sure to download the recordings before you open them. Files will be available on the FTP site, before this page can be updated. I am waiting for an editing progem to improve the raw recordings.
Special Topics
| Topic | Presenter | Date | Registration | Recording | Handouts |
|---|---|---|---|---|---|
Frequently Asked Questions
Question: I will have to miss a class. Are there make up classes?
Answer: Each class is is being recorded and is available on the FTP site. The topic web pages will present the same information.
Question: Who do I contact if I have problems or have a question.
Answer: Please contact Paul Geissler (Paul_Geissler@usgs.gov, 970-226-9482) and Tom Philippi (Tom_Philippi@nps.gov, 970-225-3586) if you have problems or do not understand something. If we can't resolve it by e-mail or over the phone, we can set up a one-on-one or small group interactive web session so we can see the problem.
Question: I'm enjoying your class. I'm wondering how many R classes you'll teach, and whether you'll do this again. The answers to that will dictate how important it is to keep up with what's going on, to read supplementary materials, etc.
Answer: We don't know how many classes. We plan to keep going as long as there is interest, as it is a big subject. We are now going over the basics, but we hope to get to more advanced topics and to invite others to present on areas they are working with. This is the second year for the course, and we plan to offer it annually.
References
To access the online help system enter the command help.start() . To search, click on "Search Engine & Keywords" .
You can also enter the command help(lm) where lm is the name of the command you want help on.
Many excellent, free references and information are available at the R-Project site (http://www.r-project.org/).
• Manuals Provides a list of free downloadable references. Click on the link for "contributed documentation" for a more extensive list.
• Search - extremely useful
• FAQ - Frequently Asked Questions.
• CRAN - Comprehensive R Archive Network lists the mirror servers where you can download R and packages.
• Books - annotated list
• Mailing lists - You can join e-mailing lists for updates. The R-help list has 75-100 posts a day, which are batched and sent daily.
Some free documentation.
• An Introduction to R (Venables, Smith & R Core Development team) http://cran.r-project.org/doc/manuals/R-intro.pdf provides a quick overview.
• The R FAQ (http://cran.r-project.org/doc/FAQ/R-FAQ.html) and Windows FAQ (http://cran.r-project.org/bin/windows/base/rw-FAQ.html) provides more information.
• Data Import/Export (http://cran.r-project.org/doc/manuals/R-data.pdf)
Several good introductions to R statistics and graphics are available in the contributed library
• Using R for Data Analysis and Graphics - Introduction, Examples and Commentary by John Maindonald (http://cran.r-project.org/doc/contrib/usingR-2.pdf data http://wwwmaths.anu.edu.au/~johnm/r/dsets/)
• Simple R by John Verzani (http://cran.r-project.org/doc/contrib/Verzani-SimpleR.pdf)
• Practical Regression and Anova using R by Julian Faraway (http://cran.r-project.org/doc/contrib/Faraway-PRA.pdf) with examples and scripts available at http://www.stat.lsa.umich.edu/~faraway/book/
• An Introduction to R: Software for Statistical Modeling& Computing by Petra Kuhnert and Bill Venables (http://cran.r-project.org/doc/contrib/Kuhnert+Venables-R_Course_Notes.zip)
R can most easily be accessed using the R Commander graphical user interface.
• Getting Started With the R Commander: A Basic-Statistics Graphical User Interface to R by John Fox http://socserv.mcmaster.ca/jfox/Getting-Started-with-the-Rcmdr.pdf
With so many commands, reference cards are very handy
• R reference card by Jonathan Baron - 1 page (http://cran.r-project.org/doc/contrib/refcard.pdf)
• R reference card by Tom Short - 4 pages (http://www.rpad.org/Rpad/R-refcard.pdf)
Text Books
Crawley, M. J. 2007. The R Book, Wiley, 942 pages, about $67. I will use Crawley as the primary text for the course because it provides a relatively comprehensive introduction to R. I suggest purchasing this book to get the most out for the course and will not repeat the explainations of statistical procedures in the text. "An excellent, comprehensive and very readable text. Building on the success of the author’s best selling Statistics: An Introduction using R, The R Book is packed with worked examples, providing an all inclusive guide to R, ideal for novice and more accomplished users alike. The book assumes no background in statistics or computing and introduces the advantages of the R environment, detailing its applications in a wide range of disciplines.
• Provides the first comprehensive reference manual for the R language, including practical guidance and full coverage of the graphics facilities.
• Introduces all the statistical models covered by R, beginning with simple classical tests such as chi-square and t-test.
• Proceeds to examine more advance methods, from regression and analysis of variance, through to generalized linear models, generalized mixed models, time series, spatial statistics, multivariate statistics and much more."
Data are available at http://www.bio.ic.ac.uk/research/mjcraw/therbook/data/ The author's page.
A search found the following prices. We do not recommend any bookseller.
Amazon, $67, http://www.amazon.com/R-Book-Michael-J-Crawley/dp/0470510242/ref=sr_1_1?ie=UTF8&s=books&qid=1255451055&sr=1-1
Alibris, $68,
http://www.alibris.com/booksearch?S=R&qisbn=9780470510247&qsort=p&siteID=uTKYNncJNRY-O1bFaP1tUnwqUX_ejfgKIA
Barns & Noble, $75 http://search.barnesandnoble.com/The-R-Book/Michael-J-Crawley/e/9780470510247/?itm=1
Borders, $110, http://www.borders.com/online/store/TitleDetail?sku=0470510242
Wiley, $110, http://www.wiley.com/WileyCDA/WileyTitle/productCd-0470510242.html
Maindonald, J. and J. Braun. 2007. Data analysis and graphics using R - an example-based approach (Second Edition), Cambridge University Press, 502 pages, about $80. DAAG package. This my favorate R text, but it focuses on statistics at an advanced level and often puts the R code in footnotes. They are the authors of the DAAG (Data Analysis and Graphics) package. They "begin this example-based introduction to data analysis with a tutorial in R, and this allows them to demonstrate elementary concepts and methodologies for data analysis with real world examples drawn from their experience as teachers and consultants. The detailed discussion of regression methods that makes up the core of the book leads on to more advanced statistical concepts. As these are explained, the facilities that allow them to be implemented in the R system are illustrated. R code and data sets for all examples are available on the Internet. This allows mathematical content to be kept to a minimum while statistical and scientific issues are still covered in depth. This book is therefore suitable for either the researcher requiring practical skills in data analysis, or the student looking for examples of applications to complement a more theoretically oriented course. Only basic statistical knowledge is assumed, approximately that for a first undergraduate course, and the methods demonstrated are suitable for use in fields as diverse as biology, social science, medicine and engineering." See the book web sites for supplemental information and scripts http://www.stats.uwo.ca/DAAG/ http://wwwmaths.anu.edu.au/~johnm/r-book.html
Everutt B. S. and T. Hothorn. 2010. A handbook of statistical analyses using R, second edition. Chapman & Hall, 355 pages, about $37, provides an excellent gentle intorduction to statistics using R. "It presents straightforward, self-
contained descriptions of how to perform a variety of statistical analyses in the R
environment. From simple inference to recursive partitioning and cluster analysis, the authors lead you methodically through the steps, commands, and
interpretation of the results, addressing theory and statistical background only when
useful or necessary. They begin with an introduction to R, discussing the syntax, general
operators, and basic data manipulation while summarizing the most important features.
Numerous figures highlight R's strong graphical capabilities, and exercises at the end of
each chapter reinforce the techniques and concepts presented." All data sets and code
used in the book are available as a downloadable package from CRAN, the R online
archive.
Venables, W, N, and B. D. Ripley. 2002. Modern applied statistics with S, Springer. 495 pages, about $90. This excellent text is more advanced and detailed than Crawley 2007 and Maindonald, J. and J. Braun 2007, which applies equally to R and S. MASS package.
"Readers are assumed to have a basic grounding in statistics, and so the book in intended for would-be users of S-PLUS and both students and researchers using statistics. Throughout, the emphasis is on presenting practical problems and full analyses of real data sets. Many of the methods discussed are state-of-the-art approaches to topics such as linear, nonlinear, and smooth regression models, tree-based methods, multivariate analysis and pattern recognition, survival analysis, time series and spatial statistics. Throughout, modern techniques such as robust methods, non-parametric smoothing, and bootstrapping are used where appropriate. This third edition is intended for users of S-PLUS 4.5, 5.0, 2000 or later, although S-PLUS 3.3/4 are also considered. The major change from the second edition is coverage of the current versions of S-PLUS. The material has been extensively rewritten using new examples and the latest computationally intensive methods. The companion volume on S Programming will provide an in-depth guide for those writing software in the S language."
Fox, J. 2002. An R and S-Plus companion to applied regression. Sage, 275 pages, about $63. car package. A companion to a statistical text, which provides many easy-to-follow examples of using R (and S). John Fox is the author of R Commander.
The book provides a valuable supplement to texts on regression analysis and linear models by showing readers how to put into practice the strategies and techniques involved in modern statistical methodology. It explains clearly the use of a very sophisticated and powerful statistical software system. And, while the examples and objectives are focused closely on regression and related techniques, the discussion successfully conveys general advice and principles for statistical computing with the S system. http://socserv.mcmaster.ca/jfox/Books/Companion/index.html
Venables, W. N. and B. D. Ripley. 2000. S Programming. Springer, 264 pages, about $95. It has an excellent explanation of the S language implemented in R, which is used for data selection and manipulation. I will not use this book directly, but it is a good source when you can't understand some of the free documentation. S is a high-level language for manipulating, analyzing, and displaying data. It forms the basis of two highly acclaimed and widely used data analysis software systems, the commercial S-PLUS and the Open Source R. This book provides an in-depth guide to writing software in the S language under either or both of those systems. It is intended for readers who have some acquaintance with the S language and want to know how to use it more effectively, for example, to build reusable tools for streamlining routine data analysis or to implement new statistical methods.
Spector, P. 2008. Data Manipulation with R. Springer, 152 pages, about $55. Many users, especially those with experience in other languages, do not take advantage of the full power of R. Because of the nature of R, solutions that make sense in other languages may not be very efficient in R. This book presents a wide array of methods applicable for reading data into R, and efficiently manipulating that data. Most experienced R users discover that, especially when working with large data sets, it may be helpful to use other programs, notably databases, in conjunction with R. Accordingly, the use of databases in R is covered in detail, along with methods for extracting data from spreadsheets and datasets created by other programs. Character manipulation, while sometimes overlooked within R, is also covered in detail, allowing problems that are traditionally solved by scripting languages to be carried out entirely within R. For users with experience in other languages, guidelines for the effective use of programming constructs like loops are provided. Since many statistical modeling and graphics functions need their data presented in a data frame, techniques for converting the output of commonly used functions to data frames are provided throughout the book.
Murrell, P. 2006. R Graphics. Chapman & Hall. 301 pages about $79. R Graphics presents the first complete, authoritative exposition on the R graphical system. After an introductory overview of R graphics facilities, the presentation first focuses on the traditional graphics system, showing how to work the traditional functions, describing functions that are available to produce complete plots, and how to customize the details of plots. The second part of the book describes the grid graphics system - a system unique to R and much more powerful than the traditional system. The author, who was integral in the development of the grid system, shows, starting from a blank page, how it can be used to produce graphical scenes. He also describes how to develop new graphical functions that are easy for others to use and build on. Appendices contain a brief introduction to the R system in general and discuss how the traditional and grid graphics systems can be combined. Much of the information presented in this book cannot be found anywhere else.
Faraway, J. J. 2005. Linear models with R. Chapman & Hall, 229 pages, about $76. A more in-depth treatment of regression and analysis of variance. (http://www.stat.lsa.umich.edu/~faraway/LMR) Install the R package faraway. This more advanced text focuses on the practice of regression and analysis of variance. It clearly demonstrates the different methods available and more importantly, in which situations each one applies. It covers all of the standard topics, from the basics of estimation to missing data, factorial designs, and block designs, but it also includes discussions on topics, such as model uncertainty, rarely addressed in books of this type. The presentation incorporates an abundance of examples that clarify both the use of each technique and the conclusions one can draw from the results. A brief introduction to R is presented in an appendix, and all of the datasets used in the book are available for download from the Internet. The author assumes that readers know the essentials of statistical inference and have a basic knowledge of data analysis, linear algebra, and calculus. The treatment reflects his view of statistical theory and his belief that qualitative statistical concepts, while somewhat more difficult to learn, are just as important because they enable us to practice statistics rather than just talk about it.
Faraway, J. J. 2006. Expanding the linear model with R. Chapman & Hall, 301 pages, about $76. "This book surveys the techniques that grow from the regression model, presenting three extension to that framework: generalized linear models (GLS,), mixed effect models, and nonparametric regression models, including "GLM diagnostics, generalized linear mixed models, trees, and neural netwworks. To demonstarate the inter play of theory and practice, throughout the book the author weaves the use of the R software environment to analyze the data of real examples, providing all the R commands necessary to reporduce the analyses. A supporting Web iste holds all of the data descried in the book and will host future updaes to the material."
Williams, G. (In preporation) Data mining desktop survival guide. http://datamining.togaware.com/survivor/index.html Graham Williams is the author of the Rattle graphical interface. "Data mining is about building models from data. We build models to gain insights into the world and how the world works. A data miner, in building models, deploys many different data analysis and model building techniques. Our choices depend on the business problems to be solved. Although data mining is not the only approach it is becoming very widely used because it is well suited to the data environments we find in today's enterprises. This is characterised by the volume of data available, commonly in the gigabytes and fast approaching the terabytes, and the complexity of that data, both in terms of the relationships that are awaiting discovery in the data and the data types available today, including text, image, audio, and video."
Crawley, M. J. 2005. Statistics: an introduction using R. Wiley, 327 pages, about $43 paperback. An easy introduction, focused on undergraduates, postgraduates without statistics and those who wish to switch to R. (http://www.imperial.ac.uk/bio/research/crawley/statistics) Features step-by-step instructions that assume no mathematics, statistics or
programming background, helping the non-statistician to fully understand
the methodology. Uses a series of realistic examples, developing step-wise from the simplest
cases, with the emphasis on checking the assumptions (e.g. constancy of
variance and normality of errors) and the adequacy of the model chosen
to fit the data. The emphasis throughout is on estimation of effect sizes and confidence
intervals, rather than on hypothesis testing. Covers the full range of statistical techniques likely to be needed to analyze
the data from research projects, including elementary material like t-tests
and chi-squared tests, intermediate methods like regression and analysis of
variance, and more advanced techniques like generalized linear modeling.
Websites
R Labs for Vegetation Ecologists.
This section of the Laboratory for Dynamic Synthetic Vegephenonenology (LabDSV) includes tutorials and lab exercises for a course in quantitative analysis and multivariate statistics in vegetation ecology.
NCEAS Scientific Computing: R Language Short Course
An Introduction to the R Programming Environment, K. A. Garrett, P. D. Esker, and A. H. Sparks
Vegan: R functions for vegetation ecologists
The Journal of Statistical Software issue on R for ecology
In-person Courses
National Conservation Training Center Data Analysis courses now use R. Free to DOI personnel; $1150 for others. These are statistics courses for natural resource and alnd management.
Data Analysis I: Jan 4-8, 2010.
Data Analysis II: Ecological Modeling in R
Data Analysis IIIA: Species Distribution Modeling using R
Discussion
Please send comments and suggestions to Paul and Tom. We will batch comments them and email them to participants who wish to receive them. Archive copies will be posted here.
Some participants have reported that the course emails are being diverted to spam folders.
Current - These issues are being batched and will be emailed to participants who choose to receive them.
An active subgroup of Mac users has formed, helping each other with problems on a Mac. Please email Paul if you want to goin that list.
Alison: I have been unsuccessful at getting onto the webinar from my home
computer. Instead I have been downloading the Topics you have made
available.
Must I be at a government computer to participate?
Paul: There should not be a problem in viewing the webinar from home. There may be a connection speed issue if you are on a dial-up line. See GoToWebinar FAQs, especially the one about your firewall at https://www1.gotowebinar.com/en_US/webinar/pre/support.tmpl
You can also contact their help at http://support.gotomeeting.com/ics/support/default.asp?deptID=5641
or call them at:
| Country | Toll-Free | Direct Dial |
|---|---|---|
| United States | 1-800-263-6317 | +1-805-617-7000 |
| United Kingdom | 0800-404-9279 | +1-805-617-7000 |
| Europe | 00-800-7770-0011 | +1-805-617-7000 |
| Mexico | 001-888-840-6244 | +1-805-617-7000 |
| India | 000-800-100-3359 | +1-805-617-7000 |
| Australia | 0011-800-7770-0011 | +1-805-617-7000 |
| Asia-Pacific | 0011-805-617-7000 | +1-805-617-7000 |
Sarah:
1. When I load R Commander, I receive this warning. [2] WARNING: The Windows version of the R Commander works best under RGui
with the single-document interface (SDI); see ?Commander. Is this something to worry about?
2. I could not find "DAAG" or "possum" under data packages, did I miss out on installing something?
3. Under Tools, "Load R cmdr plug-ins" is greyed out for me.
Paul:
1. I run R Commander with the MDI without problems. See Tom's site (http://science.nature.nps.gov/im/monitor/stats/R/Installation.cfm#Installation) for a discussion.
2. You need to install the DAAG package. From the R console, select Packages > Install Package(s) >
select a mirror and the packages you want to install.
3. It is greyed out because you have not installed plug ins that are not loaded. See above to load other packages.
Jamie: Based on the drop-down menus, it doesn't look like R-Commander or Rattle offers much in the way of either multivariate statistics or resampling-based nonparametric statistics. Am I right about this? If so, can you recommend other GUI packages that are good for these branches of statistics?
Paul: For the full power of R, you need to use the command language. Only a small part is available through menus. R Commander does offer principal component, factor and cluster analysis, but beyond that I recommend using the command language. I suggest searching on Google for the "CRAN Task View: Multivariate Statistics".
Tom:
Rcmdr.FactoMineR is the plugin for FactoMineR, which performs many
multivariate analyses in the PCA, CA, etc. family.
I wouldn't want to trust any complex analyses such as resampling that was
run via a GUI: there are too many options that don't have universal
defaults.
Ginny: One thing I was wondering about though--why don't you include in your
stats references Helsel and Hirsch's Statistical Methods in Water Resources
http://pubs.er.usgs.gov/usgspubs/twri/twri04A3?
The report is now available for free as a USGS report. Possibly you did not
include it in the past because (1) only the printed version of the report was available
for a charge from the publisher or (2) your primary focus is on statistics related to biological
studies--versus water quality (chemical concentration) type studies--a primary focus in
the Helsel and HIrsch book. But, I wonder if some of your listeners might find Helsel & Hirsch
useful because they do water quality studies and Helsel and HIrsch's book is a well-written free source.
Paul: The focus of the course is on using the R statistical package. I looked quickly their publication, and I did not see any examples using R.
Tom: I find that book more valuable than the NPS wq monitoring guidelines.
However, if I recall correctly it doesn't provide R code. I'm pretty sure
I stuck it in as a general monitoring reference somewhere in the NPS I&M
website.
I may have a lab session on NADA and R-Hydro, although that's a bit far
afield from my duties or expertise, and R-Hydro isn't very far along yet.
Joe: I am a Course Leader/Wildlife Biologist at the National Conservation Training Center (NCTC) in West Virginia. I am responsible for the analytical training courses (data analysis and population viability analysis) at NCTC. The first two courses the data analysis curriculum were recently change to include R as the main tool to teach statistics. Previously we used MINTAB and SYSTAT in our courses, but changed over about a year and half ago. I am, again, taking your on-line seminar on using R, and I over heard that there were a few (or at least one student) with issues. For those students in need of a basic introduction in statistics and want to learn R, I would suggest DAI here at NCTC. The course starts with basics and ends up learning and using t-tests in R. They may find taking a week-long course here at NCTC advantageous. The two courses that might be beneficial to the students, who are in need individual attention are:
Data Analysis I: Concepts and Procedures using R, CSP4200
Data Analysis II: Ecological Modeling using R, CSP4210
My purpose here is provide an avenue to students that find or need a concentrated/enclosed format to learning helpful, and for some it may be necessary. So whenever you see the need, point them in our direction, and we will try to service them and do the best we can.
Joe W. Witt, Ph.D.
Course Leader / Wildlife Biologist
Branch of Conservation Science & Policy
National Conservation Training Center
698 Conservation Way
Shepherdstown, WV 25443
304-876-7447
304-876-7234 (Fax)
Lisa:
1) Is there a way to reverse the "set case names" action? For example, once I get rid of the row ID number in the "weather" data set and replace the ID by the data, how do I go back to having the row ID number as it was originally?
2) I attempted to use the "aggregate variables in active data set" action on my own dataset. For some reason the option was grayed out. I compared my dataset to the "possum" dataset and could not figure out the reason that this menu option was grayed out.
Paul: Thanks, these are excellent questions.
1) You can use the command weather$Date = row.names(weather) to copy the row name to the variable Date. You can then use the menu (Data > Active data set > Set case names) to set another row name if you wish. You can use the menu (Data > Manage variables in active data set > Add observation number to data set) to number the observations, which then could be set at row names.
2) There many not be any factor in the data set that you could use as the index for aggregation. Use the menu (Data > Manage variables in active data set > Convert numeric variables to factors) to define a factor over which you want to aggregate.
Morgan:
1) Is there a way to perform a match merge between two datasets based on a unique identifier?
2) In R commander, can you use non-active datasets as you would in the typical RGUI (for example if I wanted to use values from Pam1 while I am working with possum in R commander)?
Paul:
1) Consider the following data modified from Phil Spector (2008, Data Manipulation with R, Springer, page 143), who has an excellent discussion of R merge.Copy these lines into the R Commander script window and submit them. To go beyond the common operations, we need to use commands. You can get more information on merge with the commend ?merge.
x=data.frame(a=c(1,1,2,3,4,5),x=c(8,9,12,14,21,8))
x
y=data.frame(a=c(1,3,4,6,6),y=c(8,14,19,2,3))
y
merge(x,y)
merge(x,y,all=TRUE)
merge(x,y,all.x=TRUE)
merge(x,y,all.y=TRUE)
Note that there are duplicate a values in both x and y. Here are the results from the output window.
> x=data.frame(a=c(1,1,2,3,4,5),x=c(8,9,12,14,21,8)) > x a x 1 1 8 2 1 9 3 2 12 4 3 14 5 4 21 6 5 8 > y=data.frame(a=c(1,3,4,6,6),y=c(8,14,19,2,3)) > y a y 1 1 8 2 3 14 3 4 19 4 6 2 5 6 3 > merge(x,y) a x y 1 1 8 8 2 1 9 8 3 3 14 14 4 4 21 19 > merge(x,y,all=TRUE) a x y 1 1 8 8 2 1 9 8 3 2 12 NA 4 3 14 14 5 4 21 19 6 5 8 NA 7 6 NA 2 8 6 NA 3 > merge(x,y,all.x=TRUE) a x y 1 1 8 8 2 1 9 8 3 2 12 NA 4 3 14 14 5 4 21 19 6 5 8 NA > merge(x,y,all.y=TRUE) a x y 1 1 8 8 2 1 9 8 3 3 14 14 4 4 21 19 5 6 NA 2 6 6 NA 32) R Commander is designed to operate on one active data set. Usually one would want to merge data sets in some logical manner in order to work on them. In the command language you can use the syntax ddtat.set$variable as in cbind(x$x,y$y), which yields the following. Here cbind is column bind. Because the vectors are of different length, R repeats the y values to fill out the matrix.
[,1] [,2] [1,] 8 8 [2,] 9 14 [3,] 12 19 [4,] 14 2 [5,] 21 3 [6,] 8 8Pam: Had another problem importing her Excel spreadsheet. After copying to the clipboard and importing from the clipboard, one of her numeric variables was interpreted as a factor, so she could not compute a mean or other statistica with it. To remove a factor, use submit the command Pam$af=as.numeric(as.character(Pam$af)) in the R Commander script window, where Pam is the data set name and af is the variable name.
Erik got the following error message: Error in rattle() : RGtk2 package is not available but is required for the GUI.
Tom: Other folks with that problem had installed tcltk with administrator permissions, but somehow only for use by administrator, not by all users. I recommend either downloading & installing tcltk from the ggobi website, or uninstalling and reinstalling it, trying to make sure that it is available for all users. I don't know how permissions are set up on USGS computers.
Doug: I’d like to confirm what happens in the labs. Will labs ONLY cover what’s already been addressed in lecture?
Paul: The labs will cover new material, related to the lectures.
Ken: I a brief communication with John Fox he said:
Unless the problem has been fixed, the data editor on the Mac crashes R when
it is called with an empty data frame as an argument. The work around is to
enter the data into a text file and read them via "Data -> Import data -> from text file or clipboard".
The text file and clipboard solution work fine.
Carlos: I have followed in detail the instructions in
http://datamining.togaware.com/survivor/Install_Macintosh.html
to install rattle but I have not been sucessful
The problem seems to be that RGtk2 does not get installed no matter what I try:
> install.packages("RGtk2")
Aviso en install.packages("RGtk2") :
argument 'lib' is missing: using '/Users/nalu/Library/R/2.10/library'
Mensajes de aviso perdidos
In getDependencies(pkgs, dependencies, available, lib) :
package 'RGtk2' is not available
Any suggestions would be much appreciated
Mike: I had the same problem with Rattle. I found it a bit tricky to fix.
I downloaded the RGtk package from http://cran.r-project.org/, and expanded it.
Unfortunately, iy didn't expand itself into the R libraries like the other packages; it wound up in my Downloads folder.
I tried several downloads and expansions, with the same result.
Then I opened the R menu Packages and Data/Package Installer. From the first scrolling list, I selected Local Package Directory, and clicked Install.
This brought up a standard OSX directory window. I navigated to RGtk2, and selected "libs", then clicked "Open".
That installed the package in the R library. However, you still need to go to Packages and Data/Package Manager, and select RGtk -- this allows R to load the package into memory. I suppose that with a properly-set-up Rprofile.site, this would be done automatically when R starts.
At this point you can type
install.packages("rattle", dependencies=TRUE)
then
library(rattle)
rattle()
Make sure you have X11 open -- one way is to click the X button in the R menus --or R will crash.
The Rattle GUI opens, but complains about cairoDevice. I searched for cairoDevice in the CRAN package library, and eventually got a window stating
* using R version 2.10.0 Patched (2009-11-05 r50313)
* using session charset: ASCII
* checking for file 'cairoDevice/DESCRIPTION' ... OK
* this is package 'cairoDevice' version '2.10'
* checking package name space information ... OK
* checking package dependencies ... OK
* checking if this is a source package ... OK
* checking for executable files ... OK
* checking whether package 'cairoDevice' can be installed ... ERROR
Installation failed.
See '/Volumes/Data/Builds/packages/leopard-universal/results/2.10/cairoDevice.Rcheck/00install.out' for details.
See http://www.r-project.org/nosvn/R.check/r-release-macosx-ix86/cairoDevice-00install.html for details.
So it looks like cairoDevice is not available for OSX. However, when I typed
library(rattle)
and
rattle(),
I got the Rattle GUI -- but with many warnings in the R command line.
Sorry I couldn't give you more, but maybe someone can pick this up and run with it.
Carlos: Michael, thank you so so very much!!
This has been a wonderful help for me and I got rattle installed. Yet it did not work forme exactly as you so clearly described, so I am adding what I did to your instructions in case it can be useful to somebody else.
I had the same problem with Rattle. I found it a bit tricky to fix. I tried all this running R64.app which -I think- is the version for 64 bites which I was told to use, but I got an error message that hinted at some 64 bite problem. So I tried running R.app and it worked well. I downloaded the RGtk package from http://cran.r-project.org/, and expanded it. Unfortunately, iy didn't expand itself into the R libraries like the other packages; it wound up in my Downloads folder. I tried several downloads and expansions, with the same result. Then I opened the R menu Packages and Data/Package Installer. From the first scrolling list, I selected Local Package Directory, and clicked Install. This brought up a standard OSX directory window. I navigated to RGtk2, and selected "libs", then clicked "Open". This did not work for me. The error mesage said that some files were missing. Then I selected the whole folder "RGtk2" and it worked That installed the package in the R library. However, you still need to go to Packages and Data/Package Manager, and select RGtk -- this allows R to load the package into memory. I suppose that with a properly-set-up Rprofile.site, this would be done automatically when R starts. the packages seem in alfabetical order, but they are not, so keep looking down until you see RGtk At this point you can type install.packages("rattle", dependencies=TRUE) then library(rattle) rattle() Make sure you have X11 open -- one way is to click the X button in the R menus --or R will crash. The Rattle GUI opens, but complains about cairoDevice. I searched for cairoDevice in the CRAN package library, and eventually got a window stating " * using R version 2.10.0 Patched (2009-11-05 r50313) * using session charset: ASCII * checking for file 'cairoDevice/DESCRIPTION' ... OK * this is package 'cairoDevice' version '2.10' * checking package name space information ... OK * checking package dependencies ... OK * checking if this is a source package ... OK * checking for executable files ... OK * checking whether package 'cairoDevice' can be installed ... ERROR Installation failed. See '/Volumes/Data/Builds/packages/leopard-universal/results/2.10/cairoDevice.Rcheck/00install.out' for details. See http://www.r-project.org/nosvn/R.check/r-release-macosx-ix86/cairoDevice-00install.html for details. " So it looks like cairoDevice is not available for OSX. However, when I typed library(rattle) and rattle(), I got the Rattle GUI -- but with many warnings in the R command line. SAme for me, but the rattle window aldo opened.
Katie: Thanks for the message. I think the problem that I'm having is that I need to have Mac OS10.5 (leopard) and I only have tiger. Do you know if anyone has had success loading on tiger? I've followed all of the instructions in this e-mail and have also done a lot of exploring with trying to load rattle and there are certain packages/dependencies that won't load. Do you have any insight into this or will I have to figure out a way to get leopard?
Mike: There's a version for 10.4.4 at
http://cran.r-project.org/bin/macosx/R-2.10.0.dmg .
I haven't tried it, but maybe it will work for you.
Katie: Thanks Mike, I can get R up and running, I'm just having problems with Rcmdr and
rattle. Not sure if anyone is still on the ancient tiger as I am,
but has anybody with tiger gotten Rcmdr and rattle to work?
Tom: Given that tiger is 10.4 and mach kernel BSDunix based, you should be in
much better shape. Again, you should be able to get help at UC Davis. However, because the
binary for Mac os 10.4 is R 2.10.0, that part should be fine for Rcmdr.
The major difference I am aware of is that 10.4 didn't necessarily install
X11 or tcltk, but you should be able to install them from the original dvd.
Katie: In case anyone else is having trouble still, I was finally able to load Rcmdr on my Mac (tiger) following the instructions on this website: http://socserv.mcmaster.ca/jfox/Misc/Rcmdr/installation-notes.html
Carlos: Anybody knows where is the file "Rprofile.site" on a mac, or if it
has a different name? (Spotlight cannot find it)
Sudhit: But for profile in Mac, you can go to R and preferences where you can set your library locations etc.
Mike: I think the file needs to be created. I did not find it in my install. From what I found earlier, I think the file needs to be stored in /Library/Frameworks/R.framework/Resources/etc
Since this is a text file, it should be easy to create, and the text in
the Rprofile.site from Monday should work.
However, I'd be cautious about copying the Windows text file, due to the
extraneous line-ending symbol which Windows uses. I would create a fresh
Mac file, and copy and paste (or type) the text into it. I haven't has a
chance to try this yet.
Tim: Can not start rattle on Mac. Receive the following error when I try to
install package rattle with dependencies -
Warning: dependencies ‘RGtk2’, ‘cairoDevice’, ‘rggobi’ are not available
trying URL
'http://watson.nci.nih.gov/cran_mirror/bin/macosx/leopard/contrib/2.10/rattle_2.4.90.tgz'
Content type 'application/x-gzip' length 658639 bytes (643 Kb)
opened URL
==================================================
downloaded 643 Kb
I am guessing there is something that needs to be installed in a certain
order that has not been done correctly. Maybe I need to uninstall the
previous installs of rattle?? Everything else works at this point
although I don't understand having ggobi as a stand alone package but
will wait till we are using the program to understand that.
---------------- Batch sent 11/8/9 ---------------------------
I think I have been able to shut off the confusing messages from webinar about sessions that have been canceled. When setting up the webinar, I specified a repeating webinar every Monday, Wednesday and Friday, and no, we will not have sessions on Veteran's day or Thanksgiving week. The real schedule is at http://www.fort.usgs.gov/brdscience/LearnR.htm#schedule
I think the session on Monday went well, but there are still an number of issues. I will try to resolve some of them below and have updated Topic 1 to reflect these corrections and clarifications. However, if you are still having problems with the webinar, or installing and loading R, R Commander and Rattle, I have scheduled another webinar on Thursday at the same time to resolve any remaining problems.
There have been problems downloading the webinar quick reference guide, so I have copied it to our FTP (ftp://ftpext.usgs.gov/pub/cr/co/fort.collins/Geissler/GoToWebinar_Attendee_QuickRef_Guide.pdf) site.
Participant List
People often would like a list of course participants, so they can contact others. I will prepare a list of participants and contact information for distribution to the group. Please let me know if you would NOT like to be included in the list. If you would like to update your information, you can do so at
http://www.fort.usgs.gov/brdscience/CourseEdit.aspx
R Commander:
Bronwen Wang found that you can load both the biodiversity and HH plugins if you load them in the following order. Thank you very much, this is a very helpful suggestion. I will update the suggestion on topic 1 to reflect this change.
library(BiodiversityR)
BiodiversityRGUI()
library(RcmdrPlugin.HH)
library(rattle)
rattle()
I had left out the library statement for BiodiversityR, see above.
All mirrors do not have the BiodiversityR package. If yours does not, try installing it manually from the packages menu of the R console from the USA (CA 1) mirror. In general, if you can't find a package, try another mirror. Note the R is VERY case sensitive.
Rattle:
The link to download Glade is missing the .exe even though you can see it. It should be
http://downloads.sourceforge.net/gladewin32/gtk-2.12.9-win32-2.exe
Links:
Installing rattle
on MS Windows
on Mac
on GNU/Linux
General
Data Mining Desktop Survival Guide
Tom's instructions are:
I didn't have rattle on my main computer, so I'm working through the
install right now. Unfortunately, I already have GTK and ggobi.
If you go to the ggobi site,
http://www.ggobi.org/downloads/#windows
You need to first download gtk:
http://downloads.sourceforge.net/gladewin32/gtk-2.12.9-win32-2.exe
then install it. [I have it download into c:\R, but it doesn't matter
where as long as you can find it to run it.]
Then, download and run the ggobi installer
http://www.ggobi.org/downloads/ggobi-2.1.8.exe
I install to c:/R/ggobi, just because I only use ggobi with R and I want to
keep my machine easy for our IT folks to inspect.
Then, start R, and paste the following into the command window:
install.packages("rattle")
You should get a message that looks like:
-----------------------------------------------------------------
also installing the dependency ‘pmml’
trying URL
'http://cran.stat.ucla.edu/bin/windows/contrib/2.10/pmml_1.2.20.zip'
Content type 'application/zip' length 78028 bytes (76 Kb)
opened URL
downloaded 76 Kb
trying URL
'http://cran.stat.ucla.edu/bin/windows/contrib/2.10/rattle_2.5.2.zip'
Content type 'application/zip' length 671781 bytes (656 Kb)
opened URL
downloaded 656 Kb
package 'pmml' successfully unpacked and MD5 sums checked
package 'rattle' successfully unpacked and MD5 sums checked
The downloaded packages are in
C:\Documents and Settings\Tom Philippi\Local
Settings\Temp\Rtmp1NPy9f\downloaded_packages
------------------------------------------------------------------------------
Then, type in the following 2 commands:
library(rattle)
rattle()
and a window entitled "R Data Miner" should appear.
If you get errors at the GTK or ggobi stage, you may have file permissions
issues.
Mac:
There were a number of issues about installing on a Mac. We do not have any experience with a Mac. Please send any suggestions to me, and I will forward them to the group with the batched comments.
See http://science.nature.nps.gov/im/monitor/stats/R/InstallationMac.cfm
To install the Mac and Linux version, see CRAN (http://cran.r-project.org/). Also click on the FAQs on the left for both general and Mac specific frequently asked question.
Bob Ourso writes: Here is a website with Tcl/Tk Binaries for the Mac for those that may need it...
http://www.categorifiedcoder.info/tcltk/
Mike Schenk (NOAA/NWSI) writes: I just got Rcmdr to run on OSX 10.5, and I kept some notes of the steps: first blow-by-blow, then summary. I didn't tackle setting up the Rprofile file. I list the steps below.
There were binaries for 10.4 and 10.5 at http://cran.r-project.org/.
Under my install of 10.5.7:
I searched for tcltk, and did not find it in the factory install of 10.5.7. I assume this is an option on the install disk. It was easy to download the install for tcltk from cran at
http://cran.r-project.org/bin/macosx/tools/ .
I also downloaded R for Mac OS from http://cran.r-project.org/ , and installed it.
At this point, I have not edited the Rprofile file.
I launched R, and following the instructions from http://socserv.mcmaster.ca/jfox/Misc/Rcmdr/installation-notes.html , I typed
install.packages("Rcmdr", dependencies=TRUE)
at the command line in order to install Rcmdr.
This took me to the mirror site dialog, and I selected a mirror.
This took quite a while to download, due to multiple dependencies from Rcmdr.
When it was finished, I clicked the X button on the R console to launch X11, and closed the xterm window.
Then I typed library(Rcmdr), and voila -- the R commander window appeared.
The key seems to be that tcltk is not a default factory install from Apple. Also, X11 is needed to run R Commander. In earlier versions of OSX, X11 was an optional install also, but it seems that OSX 10.5 has it as default.
With a DSL connection, the entire process took well under an hour. The longest part was downloading Rcmdr and the dependent packages.
So...
1. Download and install OSX version of R from http://cran.r-project.org/ .
2. Download and install OSX tcltk from http://cran.r-project.org/bin/macosx/tools/ .
3. Launch R and install the Rcmdr package by typing install.packages("Rcmdr", dependencies=TRUE) at the command line .
4. Launch X11 from the R console.
5. Type library(Rcmdr) at the command line.
The Startup initialization file should be located in
/Library/Frameworks/R.framework/Resources/etc
From the OSX Help file,
"Then R searches for the site-wide startup profile unless the command line option --no-site-file was given. The name of this file is taken from the value of the R_PROFILE environment variable. If this variable is unset, the default is ‘R_HOME/etc/Rprofile.site’, which is used if it exists (which it does not in a ‘factory-fresh’ installation)."
Thank you very much. - Paul
Comments
Jason: I have been pulled into the field this week. Will it be possible for me to view recordings of the webinars?
Paul: Normally, I will record the sessions and post them on an FTP site, but I will not do so for today's session because it is devoted to helping people with the webinar and installing R. There are instruction at http://www.fort.usgs.gov/brdscience/LearnR10-01.htm and http://science.nature.nps.gov/im/monitor/stats/R/Installation.cfm. Please contact me if you have a problem.
Tami: I am having trouble registering for the Learn R course through DOI Learn.
I followed your suggestions for the search.... However, I can't find the class. Any suggestions?
Paul: Registration has closed because the class starts today. However, I have registered you for the course. Please contact me if any other DOI employees would like to register with DOI Learn.
Rich: I successfully installed R on my Win32 XP box using the instructions at:http://www.fort.usgs.gov/brdscience/LearnR10-01.htm
Here are a few comments from the install:
1. It would be helpful to note that installing the packages can take
some time because of all the dependencies. I fired this up just
before I was going to head into the office, and then had to wait 10-15
min for everything to download. So you might want to add the standard "go get a cup of coffee" kind of instruction there. ;-) [Paul: Yes indeed!]
2. The link which is supposed to go to <http://downloads.sourceforge.net/gladewin32/gtk-2.12.9-win32-2.exe>
actually points to <http://downloads.sourceforge.net/gladewin32/gtk-2.12.9-win32-2.> (the .exe is missing) and that lands you on a page prompting you to
download Glade3-3.6.1.exe which may work, for all I know, but if you
want folks to download gtk-2.12.9-win32-2.exe, I guess you should fix
the link. [Paul: I will fix the link and check. Rattle uses Glade for interactive graphics. If it doesn't work this is probably the problem.]
3. For CRAN mirror, the notes have you selecting: <http://cran.cnr.Berkeley.edu> but then the copy-and-paste lines for
Rprofile.site use <http://cran.stat.ucla.edu>. I used <http://cran.cnr.Berkeley.edu> for Rprofile.site file as well,
figuring it might be better to use the same site for both (not knowing
if all the mirrors are identical). Might be nice to explain this or
use the same site for both. [Paul: Mirrors are not identical, see #4.]
4. After cut-and-pasting the suggested lines for the Rprofile.site
file (and noticing that they don't exactly match the screen grab
example of Rprofile.site) and restarting "R", I got these messages:Warning in library(package, lib.loc = lib.loc, character.only = TRUE,
logical.return = TRUE, :
there is no package called 'BiodiversityR'
Warning: package BiodiversityR in options("defaultPackages") was not found. [Paul: BiodiversityR is not available on all mirrors. Try installing it from the R console, selecting the USA(CA 1) mirror.]
Other than that, all went smoothly! [Paul: Thank you very much for your helpful comments. I will pass them on to the group with the next batch.
John: Are you going to offer your class in "R" a second time this year??
Paul: The class will continue for a few months and will not be repeated until next year. However, I hope the web pages and recording will help those who cannot participate in the webinars.
David: I get the message: WARNING: The Windows version of the R Commander works best under RGui with the single-document interface (SDI); see ?Commander.
Paul: I get the same message and do not have a problem with Rcmdr.
Tom's page says:
If you will use the R Commander GUI (we will for the first several weeks of the Learn R course) or the Tinn-R editor/GUI, you must choose SDI. Graphics packages built with tcltk also require SDI. Otherwise, I prefer MDI to keep all of my R windows within a single large window to keep them organized on my moderate-sized monitor (e.g., I can minimize the containing window to clear all of them off the desktop temporarily). When I had a huge monitor I preferred SDI, where each R window was a completely independent window on the monitor. This is an option you can easily change later by editing the etc/Console file and switching the commenting (#) between the MDI = yes and MDI = no lines.
Excerpts from webinar question log:
Q: Hi, I'm on muted & can't figure out how to un-mute
A: To avoid line noise that will make it impossible for anyone to hear, I need to unmute participants one or two at a time.
Q: when I logged in it said something like I am entering on listening only mode even though I have a headset.
A: To avoid line noise that will make it impossible for anyone to hear, I need to unmute participants one or two at a time.
Q:I will not be using a microphone. If I have a question, should I also click the hand icon and start typing my question when you recognize me.
A: `If you do not have a microphone, just type your question. I will check periodically. If you have a microphone or are on the phone, raise your hand by clicking the hand icon to speak.
Q: Is there just a pdf in the webinar viewer right now? If we can see this then everything is working okay, right?\
A: Yes, you are seeing my screen.
Q: Can we scroll through the pdf file or is it simply a screen shot?
A: No, unless Paul gives us explicit control of his screen, only he can scroll or otherwise manipulate his screen.
Q: Should I have my R application running?
A: For this session only, it will be good to run R, as we are getting set up. During lectures, I will be probably be going too fast for you to run the procedures - just watch. With the labs, there will be time to try things in R.
Q: What version R will we be using?
A: I am running R 10.0.0, but previous versions will probably also work.
Q: Can't call in. "Access code is invalid"
A: You may be looking at the PDF. When you click on the webinar link you should see a control panel like the one below with the actual call in number and access code for this session.
Q: i'm seeing an audio not connected message and I don't hear anything.
A: I you are calling the phone bridge, you need to change the audio mode to telephone. See the above image.
Q: I noticed when I went through the download procedures, that one of the modules, lattice, was built under R, version 2.10.0. I wasn't sure if you wanted to be using it during the course, so I included it anyway. And I was also wondering if we should we include all of the packages you listed in the code sample?
A: It doesn't matter too much, as you can always download them on the fly when you need them. I've updated my installation to 2.10.0, although Paul might still be using 2.9.2. There's more changes between those 2 versions than even between 2.7 and 2.9.9: compiled html help is no longer an option, etc.
Q: is this course appropriate for people without programming experience?
A: We will start by discussing the menu approach which will not require programming, other than to get R set up. Later we will discuss the command approach, which will require extensive programming in the R language. Some background in statistics will be very helpful but not essential as you can read the background in the referenced books.
Q: I'm currently getting "chatter" in my audio, is this a bandwidth problem for my machine, or are others hearing this as well?
A: I'm getting reasonable audio via webinar right now (I'm not where Paul is), but last year we had very choppy audio because of bandwidth issues.
A: Some of the problem is line noise when I unmute some participants to let them talk. I will try to keep those who are not talking muted to minimize the problem.
Q: Ok. I wonder if I need to update R. I've been using 2.8.1...?
A: You don't need to, but I recommend updating to 2.10.0. If you have a number of packages downloaded, look at my (NPS) installation page for how to update the packages at the same time.
http://science.nature.nps.gov/im/monitor/stats/R/Installation.cfm
Q: I installed R but package "Foreign" did not installed.
A: Try "foreign" all lower case. R is (very) case sensitive.
Q: But I can't seem to speak with either my web camera mic or my headset
A: When you raise your hand by clicking on the hand icon, I will recognize you and unmute you so you can speak. If too many people are unmated at one time the line noise makes it vary hard for people to hear. A web camera mic is not recommended.
Q: I'm having trouble installing R. Following Tom's instructions I just uninstalled R version 8. Now trying to install V10. Downloaded the zipped file. Now don't see install setup.
A: The R executable version should be an .exe executable, not a zip file. The zip file might be the source code. Try the link at the top of my installation page.
Q: I am trying to understand the how people are able to "raise a hand" and ask one of you a question.
A: Look for the hand icon on the webinar control panel a shown below.
Q: What is "rattle"?
A: Rattle is a graphical user interface for data mining, which is a very nice complement to R Commander.
Q: Is there a way to find out which packages are loaded?
A: The way I usually find out is try to use the package, and if it is not loaded or installed, I get a not found message.
Q: Is it there any other way to install packages without the warnings.
A: Most warning are normal and can be ignored. Try it to see if it works.
Q: Is there a good place to look for my local CRAN mirror URL (to add to the Rprofile.site file?) I'm sure I can find it eventually, so if you don't have the info handy, that's okay.
A: At http://cran.us.r-project.org/ click on Mirror in the left column
Q: When I added the commander and rattle packages, I end up with three windows, the console, commander and rattle is this normal?
A: Yes, that is what you want. You now have three ways of interacting with R. (1) type commands into the R console. (2) Use the general menus in R commander, (3) Use the more specialized menu in rattle for data mining and exploration.
Q: When I load R commander, it tells me that there are several missing packages that it uses, and asks me if I want to load them. Do I need to? If I click no, R commander opens fine, so I think I am all set without the other packages for now...?[Kr
A: Allow R to install those other packages (dependencies). Those other packages are for the functionality behind Rcmdr. For example, when you ask for a plot in Rcmdr, it actually calls those other packages.
Q: Will we only be covering the packages you listed? I was interested in the NADA package in particular and would be like that to include that in the discussion, if possible. Thanks.
A: The NADA package contains methods described by Dennis R. Helsel in his book “Nondetects and Data Analysis: Statistics for Censored Environmental Data”. That is a specialized topic, which I am not planning on covering unless there is an expression of interest by a number of people. At the end of the course, you may wish to lead a session on it so we can all learn together.
Q: When I try to download packages it goes to my Temp files. Can you tell me how to change the directory for downloaded packages?
A: I think temporary ZIP files are downloaded to temporary directory, and the packages goes to the directory you specified (or default) when you installed R.
Q: When I click yes to install the other packages, and select the option to do it from CRAN, it loads A LOT more than the 5 or 6 packages it said were missing that RCDM uses. Should I let it do them all? It went for quite some time and I just decided to stop it after a couple of minutes.
A. Yes there are a LOT of packages and it takes a long time. R commander and Rattle are interfaces which use many other packages. Enjoy a coffee break.
Q: I am running the script with a limited number of packages, but I see many dozens more being downloaded. Is this expected behavior?
A: Yes, R commander and rattle are interfaces which use many other packages.
Q: I only see two questions from the audience in my questions log. I expect that there have been more questions than that, and I'm wondering if I'm not seeing all of the questions and, if not, why not?
A: I think you only see your questions and the answers that the presenter sent to the group.
Q: Not sure what you mean by "mirror" What is the best mirror for Alaska?
A: Mirrors are web servers around the world that keep the complete R distribution available, rather than everyone hitting the same server. You might want Fred Hutchinson at UW or UC Berkeley as your mirror:
http://cran.fhcrc.org/
http://cran.cnr.berkeley.edu/
Q: I can't find BiodiversityR.
A: BiodiversityR is a package on the main CRAN set of packages.
install.packages("BiodiversityR")
worked for me. If it fails, try pointing to a different mirror:
http://cran.stat.ucla.edu/
is a good one...
Q: echo in the audio
A; When I unmute someone, the echo can come from a speaker phone or from a microphone too close to the speakers.
Q: Which folder should I unzip the downloaded packages?
A: Don't unzip the files. The installation process unzips them into the correct directories. Each package puts some files in library, so in docs, etc.
Q: Has anyone had any experience using R with program DISTANCE? The latest Version of Distance says that it's compatible with R v. 2.9.1. Are you aware of any problems using R v. 2.10?
A: It should work fine with 2.10.0. It won't work with much older versions of R.
Q: how do we save the Rprofile.site file in notepad, after adding your settings? Notepad doesn't recognize it as a text file and won't let me save it.
A: You can always save it with .txt., then in file explorer rename it to the correct name.
A: It may be a problem of not having write permission to program files. If you can, run notepad as an administrator by right clicking on the icon. Tom suggests installing R in another directory to avoid this problem.
---------------- Batch sent 11/8/9 ---------------------------
Julia: I’m probably going to sign up to Learn R (great opportunity). However, I looked at the 2008 course material and find that I need a Windows Media Player codec (G2M3 codec) to play some of the video files, but some I have no problem with (a bit odd). I’m looking into getting some updated version with that codec through my computer administrator because I’m on a managed system. Just wondering if this will be a problem for me during the course and if there is something else I might be able to do to get around the problem?
Paul: There have been continuing problems with the recordings. I am not sure what can be done about it. I do not have the problem after downloading the files, so I cannot reproduce the problem. Downloading the files before opening them helps. In any case, all that I am going to say is either on the web pages or in the The R Book.
Annett: With the help of our IT administrator I installed R. The only error I'm getting is about the biodiversity package. The error from the initial R Console is pasted below. In the installation instructions, the command to download packages doesn't include a biodiversity package as far as I can tell, but profile options that we're asked to paste into our own, does call a biodiversity package, so I'm wondering if this is the problem. Is the biodiversity package needed for the course?
Paul: The BiodiversityR package is not an essential part of the course, but I plan on discussing it because I think it will be of general interest. Try to install it by selecting Packages > Install package(s) from the R Console and select BiodiversityR. Not all mirrors have all packages. If you can't find it, select Packages > Set CRAN mirror > USA (CA1) [UCLA].
Annett: Thank you - that worked. The error is now gone.
Rachel: I'm considering registering for your R course which starts this Monday. My computer does not have a microphone. Is there a way to phone into the course instead of using the computer's speaker system? Thanks,
Paul: Yes, you have the options:
* Call our phone bridge long-distance. You will receive login information when you register for the webinar.
* Listen on you computer speakers and type in your questions or comments.
* Purchase a USB headset for about $25
Tom: You can use your computer's speakers to hear the presentations, or you can use the dial-in line for your phone. Given that we have over 850 registered participants, the phone lines will all be on mute by default, and we will unmute the line of the specific participant asking a question. However the Webinar software also has an im-like facility built into it that lets you type questions, and that is how the vast majority of questions will be taken. While Paul is presenting, I try to answer specific or unique questions typed into the program (only the 2 of us can see the questions), and Paul pauses every 10 minutes or so to publicly answer questions of more general interest. I believe that you receive instructions for the dial-in line when you register for the webinar.
Abby: I am looking forward to your class but I have run into a problem. I thought I only needed a MasOS 10.4 but the R-2.10.0.dmg says it is for 10.4.4 and higher (I have a 10.4.11). Is there another version for my apparently massively outdated Mac, or am I at a total impasse? Thanks for your help,
Paul: Looking at http://cran.us.r-project.org/ and clicking on MacOS X, I found:
"This directory contains binaries for a base distribution and packages to run on Mac OS X (release 10.2 and above). Mac OS 8.6 to 9.2 (and Mac OS X 10.1) are no longer supported but you can find the last supported release of R for these systems (which is R 1.7.1) here"
I read that as release 10.2 and above is supported, but that you can also download earlier releases of R that support previous Mac OS releases.
Mark: I have enrolled in your R course and have worked through the notes for Monday's class. I am receiving a warning about package 'lmer' when I open R. That warning is pasted below. I cannot find information about this package on the internet. Package 'lmer' is included in a list of default packages in some code suggested for the file 'Rprofile.site'. That code is immediately below. I wonder whether 'lmer' should actually be 'lme4'? Thank you in advance for any advice. Mark
Paul: Yes, you should install and load package lme4, which contains lmer for mixed models. I think that is a change from last year. Thank you very much for pointing out the problem.
Tom: I have both nlme (which has the older, stable code for fitting generalized linear mixed models) and lme4 (which can fit complex crossed & nested random effects, but sometimes still doesn't converge).
Thanks for pointing out the problem on my web page; I've corrected it.
Jayne: Which R version are you using for the webinar? I see on the R home page they have recently released 2.10.0.
Paul: I am using version 2.10.0, but to my knowledge, there are not major changes.
Gloria: Has the audio quality problem that occurred during the last course offering been corrected?
Paul: The audio problems we had for the first two sessions last year was corrected, and we did not have a problem for the rest of the course.
Scott: Please clarify the schedule: Is it Monday through Wednesday starting at 12 noon (Mountain Time)? I think Wednesday is a holiday (Veteran's Day).
Paul: You are correct that we are not having a class on Wednesday. It is easier to schedule a block in Webinar.
See http://www.fort.usgs.gov/brdscience/LearnR.htm#schedule for the real schedule.
---------------- Batch sent 11/4/9 ---------------------------
The R class starts, next Monday, November 9.
See the course website for updates (http://www.fort.usgs.gov/brdscience/learnR.htm).
So far we have 744 people registered.
Please be sure you have also registered for the webinar at https://www1.gotomeeting.com/register/125974937 if you plan on watching the live presentation. If you are watching in a group, only one person needs to register and sign on to the webinar.
US Department of the Interior employees can get credit for this course on their record by registering with DOI LEARN https://doilearn.doi.gov/ Select course catalog and search for "R statistical package". A certificate of participation is also available.
Note that there are three separate registrations:
* the course to get on the emailing list and participate in discussions - http://www.fort.usgs.gov/brdscience/courseRegister.aspx
* the webinar to view live presentations - https://www1.gotomeeting.com/register/125974937
* DOI LEARN to have the course on the record of DOI employees - https://doilearn.doi.gov/
I regret the necessary duplication.
The first session is an introduction to the course and helping people with any problems with the webinar or with setting up R. You may want to try setting up the webinar and R before the session (see http://www.fort.usgs.gov/brdscience/LearnR10-01.htm and http://science.nature.nps.gov/im/monitor/stats/R/Installation.cfm). The introduction to R starts on November 16.
There will not be a lab on November 10 or a session on Veteran's Day November 11. There will also not be sessions during Thanksgiving week, November 23-25. I will try to keep the schedule at http://www.fort.usgs.gov/brdscience/LearnR.htm#schedule current.
Questions and Comments are emailed to the group in batches and archived at http://www.fort.usgs.gov/brdscience/LearnR.htm#Discussion
Paul: Several people have had a problem with verifying their registration because of a problem with the program. It has been fixed now. I apologize for the problem. This step verifies that we have your email address correct and that you want to receive course emails.
Ben: I won't be able to attend all sessions on the new times. Will the webinar be recorded and available for viewing afterwards?
Paul: Yes, after the sessions the recordings will be available on ftp://ftpext.usgs.gov/pub/cr/co/fort.collins/Geissler/LearnR. You can download last year's recordings using the links on http://www.fort.usgs.gov/brdscience/learnR08.htm Be sure to download the files before trying to open them.
Sean: I am interested in joining the R stats courses. I have one question though. What type of software will I need to run the VoIP? For most of the courses I will be in the field but will have internet access at the time of the course via hotel internet connection and my laptop. I just want to make sure I can access everything so I am not overwhelmed during the first presentation trying to get set up.
Paul: We are using GoToWebinar (https://www1.gotowebinar.com/). There is no cost to participants.
To attend a Webinar on a PC, the following is required: * Internet Explorer® 6.0 or newer, Mozilla® Firefox® 2.0 or newer (JavaScript™ and Java™ enabled) * Windows® 2000, XP, 2003 Server or Vista * Cable modem, DSL or better Internet connection * Minimum of Pentium® class 1GHz CPU with 512 MB of RAM (Recommended) (2 GB of RAM for Windows® Vista) Participants wishing to connect to audio using VoIP will need a fast Internet connection, a microphone and speakers (a USB headset is recommended).
Mac® computer users can attend Webinars. System requirements: * Mac OS X 10.4 (Tiger®) or newer * Safari 3.0 or newer, Firefox 2.0 or newer (JavaScript™ and Java™ enabled) * Cable modem, DSL or better Internet connection * Power PC G4/G5 or Intel processor (521 MB of RAM or better recommended) Participants wishing to connect to audio using VoIP will need a fast Internet connection, a microphone and speakers (a USB headset is recommended).
Linux is not supported.
The first session will be devoted to helping people get set up using both R, R commander and webinar.
Jock: Thanks for the R course confirmation & schedule. Quick question
about the schedule though... are you really going to have classes
between Christmas & New Years? I will be out of the office then,
as I suppose most other folks will be as well.
Paul: No, we will be skipping a number of sessions that fall on holidays, starting with Veterans Day (November 11). Tom and I will also have to cancel some sessions because of other commitments. The schedule will be flexible, but we will let you know about a week ahead of time. Thanks for clarifying.
Marcus: Your initial e-mail notification suggested "6-8 weeks including more advanced statistical analyses", but the below schedule shows 16 weeks. Is this correct? Is there a course outline for each day/week? My colleges and I may have to pick and choose days given our busy work schedules.
Paul: I really do not know how long the class will last, as it depends on the interest of the participants and how much detail people want. We are putting the general information first, thinking that people will participate as long as it meets their needs.
Diane: Can you give us the course number for DOI learn? I can't find it using the search engine which only allows one search term.
Paul:
The course code for DOI Learn is: USGS-BRD-B-ST-RStat-C1576 You should also be able to find it searching for "R Statistical Package "
Eric: When I registered for the course I was not aware that a webinar would also be offered. Is participation in the webinar essential, or is it intended as a supplement to the course?
Paul: There was an error with the webinar times. The course and the webinar are the same thing, with the webinar being to web broadcast system to watch the course. I apologize for the confusion.
Seth: Thanks for offering this very helpful class. Perhaps people would appreciate a heads up when the "how to install" page is updated, that you promised Regina below, so we could work out any bugs regarding installation before class begins. I have current version of R but not the commander program.
Paul: That page has been updated, although we may make additional changes as we think of them. R version 10.0 is now available, and you will get that version if you click on the link for the first session (http://www.fort.usgs.gov/brdscience/LearnR1.htm). I have updated that page to include Rattle.
Kam: Am curious about the other courses that appear to be available when registering for the R Course. I’d like to narrow it down to 1 or 2, but honestly they all look interesting, applicable and useful. Are the other courses similarly open to non-USGS participants? Thanks, looking forward to the start of the R course.
Paul: We hope to offer several other courses, but the R course is the only one we have scheduled so far.
Sam: I initially had trouble opening the .wmv files from last year's course. Once I upgraded to Windows Media Player version 11 everything worked fine. Not sure why!
Paul: Thank you very much for letting us know. I will pass it on to the group.
Claire: I am interested in the R statistical package course scheduled for November,
but I have some major scheduling conflicts including travel and attending a
workshop during the entire second week of class. This leads me to ask two
questions of you: 1) would I be able to learn something useful from just
the first week of class (two presentations and one lab)? and 2) do you
expect to present the course again in the near future?
Paul: The first week will be devoted to getting set-up and started, so I think it will be very useful if you plan to use R. You should be able to catch-up on the second week by following the web page and the audio/video recordings. We hope to offer the course again next fall.
Kurt: Will each webinar for the three day periods (e.g., 9-11 Nov.) be on
unique topics or will some be repeats? I ask this because the dates on the
webinar schedule that was emailed to me do not match those on the website.
Also, the ftp site you listed as carrying the webinar recordings does not
appear when I click the link. Is that because there isn't any content yet?
Paul: Each webinar will be unique, but often more than one two-hour session will be required to complete a topic. The FTP site is empty now because we have not had any sessions yet to record.
Hallie: I signed up for your R seminar course, but just realized that I have class on Monday and Wednesdays at 1 p.m. (central time). Will I be able to view lectures if I cannot watch them live? Thank you.
Paul: You should be able to follow from the web pages for each topic. Audio/visual recordings will also be available for download. Registering for the class will allow you to receive updates and participate in email discussions.
Ivan: What is the time and duration of the R-course webinar? At first I had 1-3 pm CST, and the course was “6-8 weeks”. I then received an email specifying the time as 10-11 am CST, and running from Nov 9, 2009 to Feb 24, 2010, which is more like 16 weeks (going from 8 weeks for a two hour course to 16 weeks for a one hour course makes sense). The text from the email seems to have changed, now I have 13 pm CST, from Nov 9, 2009 to Feb 24, 2010.
Paul: Sorry about the confusion. The times are:
Hawaii 9:00-11:00 Alaska 10:00-12:00 Pacific 11:00-1:00 Mountain 12:00-2:00 Central 1:00-3:00 Eastern 2:00-4:00 UTC 7:00-9:00 pm. Hawaii and UTC have been corrected for standard time.
I really do not know how many weeks the course will go. We are adding material to last year's course. We will cover the menus first and then circle back and cover the R command language in more depth. Some casual or occasional users may want to only cover the menu approach.
Tim: I have a quick question about the upcoming R class (November 9). DOI Learn lists the course as three days from 1200-1400? So it is 3 days, 2 hours each day, if needed? I assume this is Mountain Time since you are in CO?
Tom: Paul will make presentations on Mondays & Wednesdays 2-4pm EST (12-2 MST), working through the material. I will hold lab sessions on Tuesdays at the same time. My labs will attempt to be a bit more hands-on. Last year we strung together lab sessions on getting complex data into R, data manipulation, data exploration (primarily graphical), and then some statistical tests, all related to downscaled climate change projections for participants parks or areas of interest. We won't do climate change projections again this year.
You can participate in Paul's presentations without participating in my labs. Its a bit harder to participate in my labs without either doing Paul's presentations or a fair bit of outside work, as I'm focusing more on applying topics that Paul has introduced.
I suppose that participation in Paul's M-W sessions is sufficient for obtaining DOI Learn credit; Paul and I haven't discussed how the credit works. I doubt that we'll give a final exam, but for all I know we may have to have some test or project to demonstrate that you learned something in order for DOI Learn credit.
Paul: I agree with Tom. For DOI LEARN credit or a certificate of participation, we will expect participation in a large proportion of the lectures or your certification that you have viewed the recordings, at least for the menu portion of the course. If you are participating with a group, please let us know who signed on to the webinar for the group.
Bob: As I mentioned in my initial email to you, the webinar is schedule too late in the day (Eastern Time) for me to participate. I have family commitments that necessitate me working a 0600-1420 schedule.
Paul: I regret that the course is at an inconvenient time for you, but hope that the web pages and recordings will be helpful.
I am automatically generating reminders, understanding that there are good reasons why some people have not registered for the webinar. I apologize for the inconvenience, but it is important to contact those who have neglected to register for the webinar. You will receive another reminder later in the week. Please ignore it. There are 744 people registered for the course so far, making it difficult to follow each registration individually.
Gary: I noticed that some of the times of the webinar conflict with some of my teaching, so there will be times that I cannot attend or will log in late. Is that a problem?
Paul: It is not a problem. I hope that the web pages and recordings will help you catch up on the parts that you have to miss.
Liviu: And thank you for the reminder. I vaguely remember having registered
for the webinar, but the software apparently requires that I attend
using either a Win or Mac OS. Is it possible to connect using Linux, my main operating system?
Paul: I regret that webinar does not support Linux. If you do not have access to a PC or Mac, I am afraid your only option would be to follow the web pages and recordings, although people using Linux have had problems with the recordings.
Ivan: I wanted to let you know, I just downloaded R 2.10.0 and was getting the packages for the course and I did not find “foreign” . Do you think this is because I downloaded R 2.10.0 instead of R 2.9.2? It has been my experience that R is pretty good about being backward compatible from older versions.
Paul: The foreign package has functions for reading and writing data stored by statistical packages such as Minitab, S, SAS, SPSS, Stata, Systat, ..., and for reading and writing .dbf (dBase) files.
I think it is installed as part of the BiodiversityR package [correction - rattle package]. I have foreign on my PC but I do not see it on the CRAN mirrors and am not sure how I got it. A Windows binary and more information is available at http://cran.r-project.org/web/packages/foreign/index.html
Lindsey: I have an extensive climate data set and wondering if you would be
interested in using some of it for the "Intro to R class." I have hourly
climate data (wind, solar, temp, RH...) from 15 meteorological sites in
Glacier National Park over various lengths of period throughout the past
fifteen years.
I could provide one or more metrics from any number of these stations for a
specific time period. Please let me know if you are interested and, if so,
how you would like the data prepared.
I am taking a Wilderness First Responder Class this week and will have
access to email during the evenings only. However, I will have cell phone
coverage throughout the day and am able to chat during 12-1 pm Mountain
time. You may call me at (406) 314-3113 and leave a message, or email me
with specific instructions if you would like me to prepare the data.
Tom: I would be happy to use those data in my lab sessions as a case study of
dealing with time series data (simple decomposition into 24hr and 1year
periodic components, plus a smoothed trend and residuals), and from a more
ecological/resource scientist perspective, how to use the climatological packages (at least 3 exist) to make climate diagrams, and how to think
about your data and extract meaningful attributes from instrumentation
records (e.g., frost-free days, dates of first & last freeze, etc.).
What format are they in? The folks at NCAR use R so there are good tools
for grabbing complex data in netCDF files. At the other extreme, .csv
files are great, and we can probably handle anything in between (including
SQL relational databases, even MSaccess).
Thanks for the suggestion.
Lou: May I register for the certificate later once I know I can complete the course? Thanks
Paul: Yes, please send me an email requesting a certificate of participation after we have completed the menu part of the course.
Katie: I'll be traveling quite a bit the first 2 weeks of the course with limited access to the internet. I know the lectures will be posted online after they are given, will the labs be posted as well?
Tom: If Paul can set up the recording, the Tuesday 12-1:30 labs will be recorded, as that's his Webinar license. I will try to generate a lab-specific web page on my NPS R site each week.
Unless disaster ensues on Monday and more folks need help with installation, there won't be a lab Tuesday Nov 10th. The lab Nov 17th will be on getting data from Access,other relational databases, and other sources in and out of R. Frankly, you might do better with the lab page or my R data i/o page on my NPS R website than the recording of the lab.
---------------- Batch sent 10/26/9 ---------------------------
Please register for both
• the webinar at https://www1.gotomeeting.com/register/125974937and
• the course at http://www.fort.usgs.gov/brdscience/courseRegister.aspx
We ask that everyone register individually for the course, so we can email updates to you and so you can participate in the email discussion.
We also ask that person that will be watching the webinar individually and one person from each group that will be watching it together register for the webinar. After you register for the webinar, the system will email you logon information so you can watch the presentations. We can have up to 1,000 people participate in the webinar, so spaces probably not be a problem, but often people like to participate in groups to discuss the presentation.
US Department of the Interior employees can get credit for this course on their record by registering with DOI LEARN https://doilearn.doi.gov/ Select course catalog and search for "R statistical package". A certificate of participation is also available.
Kirsten: If I don't want to use a microphone (since I have a shared office, and will be using headphones so as not to disturb others), will it be possible to submit questions by IM instead? Also, have there been any problems in the past with company firewalls restricting VoIP?
Paul: You can always type questions into the webinar question box. IM is not supported - I cannot do too many things at the same time. We have not had any reported problems using VoIP.
Jennifer: I am interested in your R Stats Package online course that is beginning on Nov 9, however, the times are not good for me. I would like to be able to view the recordings afterward. Do I need to sign up for the course or will they be available for anyone to view?
Paul: It is not necessary for you to register in order to view the recordings, but we suggest that you do so. Registering will put you on the email list for announcements and allow you to participate in email discussions. It will also help us get a count of the participants.
Ed: I am interested in the course you are offering. I tried to view last year’s course and got a message that a G2M3 codec was required. I could not find this anywhere. Is there special software I need to use other than media player or is the course proprietary to USGS staff?
Paul: I am very surprised that you got that message, because I have tried to convert all the recordings to windows media format (.wmv). G2M3 codec is GoToWebinar's proprietary format https://www.gotomeeting.com/codec
Satindra: I just wanted to confirm that this course is in fact available to anyone; even those not affiliated with the USGS and/or NPS. I don't see any mention of it being restricted to USGS/NPS affiliates either in your e-mail or on the registration page but since the offer seems so generous, I wanted to double-check.
Paul: Yes, the course is open to all who are interested without charge. The USGS has the mission to provide science support to DOI and other land management agencies. The R course is part of that support. However, as there is little extra cost incurred by opening the course to everyone, we did so. We will probably not approach the maximum number of participants (1,000). R is an open source (free) system, and this is our way of giving a little back to the community.
Al: Are you going to be using Tinn-R as the interface w/ R? Or, do you
recommend using it?
Paul: Although many people like the Tinn-R interface, I prefer R Commander because it also includes a menu for basic statistics and graphics and because I prefer the interface. Both work well and Tinn-R has the advantage of syntax highlighting. I will present the R Commander and Rattle GUIs.
Brian: This course could be very valuable to me. I have only limited exposure to R, though I have programmed in S. But I would like to make the switch from SAS to R - and this could make that transition much easier. Timing for me is terrible - I will be out of town the week of Nov. 9 and again the week of Nov. 30. So, I am wondering whether it would make sense to register and take the course during the time I am available. If there are resources I could use to catch up (e.g., powerpoints or something similar) then this could work.
Paul: Unfortunately, the timing will always be bad for some people. Notes and an audio/video of the presentation will be available on the website. You can view last year's notes and recordings at http://www.fort.usgs.gov/brdscience/learnR08.htm. I suggest registering for the course, so you will receive updates and can participate in the email discussions.
Susan: I’ve registered for your course, but am not sure I will have the time this fall to make it work. Will you be offering the class again in the near future?
Paul: We will probably offer the course again next year. However, web pages with the information and audio/video recordings will be available when your schedule permits. The first few sessions should get you started, and then you can view sections as you need them.
Peter: I have registered for this year's R Training and have started to have a look
at last year's recordings.. The first session starts by discussing initial set. I am interested in this years Webinar set up as I want to find out if I can
use my laptop which runs Linux. Is there a link or instructions on the
website? I have had a quick look - did I miss something?
Paul: I have not posted the Webinar link yet. Unfortunately, Webinar does not support Linux. The recordings are in Windows Media Format (wmv).
Peter: Well... linux could probably support Webinar if they didn't use M$ propriety encryption.
Peter: I am looking through the last year's LEARN R recordings and have hit a snag with the LearnR2.wmv file.
I have tried to view this file on both my windows and linux PCs there seems to be a codec issue? Not even the latest MS Media Player knows what to do with it. My VLC media player at least produced this message on failure: No suitable decoder module: VLC does not support the audio or video format "G2M3". Unfortunately there is no way for you to fix this.
Paul: Try downloading the file to your hard disk by right clicking on the link and selecting "save link as". I just downloaded the file and was able to listen to it.
Peter: I had already grabbed all the 2008 wmv files to play with. I spent a hour trying to convert them to something useful ( to me ) but I have yet to find the G2M3 codec. Looks like it was initially as part of M$_Media_Player_9 but is now incompatible with what is on my second (windows)PC - M$_Media_Player_11. The issue is that G2M3 is a special video codec used by GoToMeeting and GoToWebinar for encrypted recording and streaming (Just typical of M$_windows to make it hard for anyone who chooses not to use their software). The G2M3 codec is used by the meeting software as a default format. It is possible to record standard WMV files without the encryption but one must make a conscience choice to do so. Anyway looks like you have not run across this issue before so I am on my own. Thanks for your suggestion.
Paul: I hear your frustration. As you say, I have not run into this problem before. GoToWebinar seems to be the best option for offering these webinars, and using that approach, I do not have any other recording option other than their own proprietary format. I commend your independent spirit, but it does come with frustrations. I will send your comments to the group with the next batch, and perhaps someone will have a solution. If you find a solution, I am sure others will be interested.
Peter: Thanks for your email.
I have eschewed my Linux laptop and have a Windows PC ready. I still can't view many of last year's webinar movies even with the latest Media Player 11. I hope this isn't a big problem to solve for this year's training sessions.
Paul: Thanks for hanging in there. I will have to work on the recording problem.
Mary: I'm in the same boat as 'Peter' regarding Linux. I was just able to play
one of your archived wmv files by using:
mplayer -vo gl LearnR1.wmv
(R4 played fine, but for R2 I just get the audio, and I haven't checked
out any others). I haven't tried this solution yet: http://www.sugarcrm.com/forums/showthread.php?p=146189#post146189 If you forward this message to Peter, maybe between the two of us we can exchange ideas and figure out how to view your all of the files, and if
we're lucky, maybe even the live sessions. No need to drag everyone into the discussion, but I'd be happy to report back on any successes
that we have. One thing that I did read is that the G2M3 codec is the default for webinar, but that there is an option to set it to record with a more readily available codec.
Paul: Thanks for working on this problem. I will have to work on it. With webinar, I convert the files from their proprietary format to windows wmv format. This is their option, and I do not have control over it. I will copy Peter.
Eileen: I am very interested in taking the course on the R statistical package that
you are offering beginning in November. Please let me know what I will
need to do to sign up.
Tom: Welcome! I didn't realize that Fort Collins had tropical forestry: clearly
I'm stuck on the cold side of town.
If you go to the USGS course website
there is a link for registering and information on the course, including some information for downloading
and attempting to install R beforehand. I don't know FS IT rules; I have a
guide on R and NPS IT rules on my NPS website:
http://science.nature.nps.gov/im/monitor/stats/R/Installation.cfm
Regina: Thanks for teaching this class; I am looking forward to it. At this point
the, the only class I know I will miss is the first class. I guess since
you have the getting started page, the best way for me to "make up " for
this is to try to install the package myself? prior to class?
Any other suggestions?
Thanks
Tom: You should have no problems if you have local admin rights on your
computer. I have a page of installation instructions tailored to NPS on my
NPS pages:
http://science.nature.nps.gov/im/monitor/stats/R/Installation.cfm
I'll be updating that page soon to include information on easy ways to
update versions of R and all installed packages. You can email me if you
have problems.
Rebecca: I had a question about the schedule for the R course. I am interested in R for the occasional user. Looking at the schedule, which classes should I be a part of and which ones would best fit my occasional use needs?
Paul: I suggest the first few sessions on how to set up R and use the menus. For general use the R Commander menus are useful. For data mining, you may like the Rattle package. One strategy is to start the course and drop out when it gets too deep. We will try to start slow.
Joseph: Several coworkers and I are interested in taking your course on the R statistical package and we were trying to decide the best way to register. Would it be beneficial for us to register individually? We were thinking of setting up the Monday and Wednesday sessions in a conference room so we could openly discuss the topics presented, and then we would participate in the Tuesday labs at our individual work stations. I assume doing it this way we would need to individually register? Several of us have R on our local computers already too. Thanks for your time!
Paul: We will appreciate each of you registering individually. That way each will get updates and be able to participate in the email discussions. It will also allow us to get a count of the number of participants. There is a separate registration for the webinar, and only one person needs to register for the webinar if you plan to watch in a conference room. If you will be participating in the labs at your individual work stations, then each person would need to register for both the course at http://www.fort.usgs.gov/brdscience/courseRegister.aspx and the webinar at https://www1.gotomeeting.com/register/125974937.
---------------- Batch sent 10/14/9 ---------------------------
Amy: I just signed up for the R class in hopes to refresh what I learned from a 2 week class this summer. I am planning on at least sitting in on specific lectures (or catching them later) but will likely not catch all the lectures. I just wanted to check that this would be agreeable to you and that I would not be taking the place of someone able to take the entire course.
Paul: The webinar has a capacity of 1,000 participants, and we are unlikely to come close to that number, so the number of participants is not limited. I encourage you to participate in the sessions that interest you.
Patrick:Thanks for offering this, I'm definitely interested. I'll be out of town, and away from the internet, for a couple weeks during this course, will the presentations on the FTP site be kept there for a while so that I can get caught up?
Paul: We intend to keep the recordings up permanently. However, they are automatically deleted after a while, and I need to remember to copy the files back to the FTP site. If the recordings are deleted, please let me know and I will copy them back to the FTP site.
Rita: I just saw the announcement for the upcoming R course. I am interested in learning R and in addition, I plan to use RMark at some point to analyze some nest survival data. You asked for input on potential data sets. I have data on White-headed Woodpecker nest and adult survival and am currently in the process of analyzing it using the MARK nest survival model. But am also interested in learning RMark. Is that something that would be of interest to you and applicable to the MARK course?
Paul: I have not used RMark, but this is clearly a topic of interest to natural resource managers. It will be an excellent special topic, after the regular course. Thanks for volunteering your data.
Mike: This class sounds excellent and I have a PhD student who is
interested but did not know if this was open only to USGS or DOI employees?
Paul: Our primary audience is DOI and State land management agencies, but there is no extra cost to opening it up to everyone. The webinar has a capacity of 1,000 participants, and we are unlikely to come close to that number, so the number of participants is not limited.
Roy: The confounding problem is that the documentation often seems a bit or
sometimes very thin. sometimes it just says we did like Book X, but book X
doesn't really explain it well either.
Just recently I asked an author of recent JAWRA water quality paper what
equations she used for spatial autocorrelation and partial autocorrelation or
just partial correlation and the answer was "spatial autocorrelation generated in R using ade4
and ncf packages" but then she didn't seem to know exactly what R was doing, what
assumptions, what actual equations were in use, what limitations were on the analyses, were etc., etc.
Actually, R appears to have many related autocorrelation or partial
autocorrelation functions, although which one to use in a specific
situation is often not clear, nor is it clear how the results of these
various functions would be different.... So it is hard to know which R function is best for a particular task, how
it differs from others, which ones apply only to certain types of data and
not others, what equations they are actually using.
Paul: R has over 2,000 contributed packages, many of which have overlapping capability. Although the duplication can lead to confusion, it is also a considerable strength to have multiple options. All the packages on the CRAN website have been peer reviewed, but often somewhat different approaches or different defaults are used. Specifying which package with which options, clearly specifies what analysis was done, although as you point out, it may not be easy to learn the details. However, a published reference is generally cited. The R documentation does not attempt to teach statistics, one must consult text books or the primary literature.
Jennifer: I am interested in learning about the online R course that you are
offering. What type of connection do I need to participate?
How many hours would you estimate would be needed outside of the class time per week?
Paul: To attend a Webinar on a PC, the following is required:
*
Internet Explorer® 6.0 or newer, Mozilla® Firefox® 2.0 or newer (JavaScript™ and Java™ enabled)
* Windows® 2000, XP, 2003 Server or Vista
* Cable modem, DSL or better Internet connection
* Minimum of Pentium® class 1GHz CPU with 512 MB of RAM (Recommended) (2 GB of RAM for Windows® Vista)
The number of hours outside class would depend on your interest. You could just listen to the lectures and obtain an overview of what is available in R. It may take ten hours a week to work through the examples, and a shorter time if you only worked a few examples.
Stevan: Thank you for the update, I very much look forward to your course. I assume you will send instructions as to how we access the video as we draw nearer?
Paul: Yes, I will sent out the link and post it on the website shortly.
Adelaide: Could you possibly give me an outline itinerary of the topics and the days they would be covered?
Paul: An outline is on the website, but we will have to see how long each topic takes. I will post the schedule a week ahead of time.
Jill: I am currently running a graduate stats class that uses R and would like to advertise your short course to my
students as a possible supplement to their course work here. Are you open to
having non-US students in your online class?
Paul: Yes, they are very welcome to participate. Any suggestions on the course will be very welcome.
Greg: Will the presentations on R be available on a FTP site and how do I access it? I can't attend during the scheduled Mondays and Wednesdays.
Paul: The recordings will be available after the presentations, and the links will be posted on the website http://www.fort.usgs.gov/brdscience/learnR.htm.
Last year's course and recordings are available at http://www.fort.usgs.gov/brdscience/learnR08.htm
I suggest that you register for the free course at http://www.fort.usgs.gov/brdscience/courseRegister.aspx so we will have your email address and can send you updates and so you can participate in the online discussions.
Kurt: I intend to take the course and in looking up "The R Book" I found
it's price on Amazon.com to be substantially lower than the one you quoted
(i.e., $67.23). I thought I'd let you know so that you might widely
disseminate this information to reach those in your audience who might be
daunted by the $110.00 price you quoted in your first email.
Paul: Thanks. A search found the following prices. We do not recommend any bookseller.
Amazon, $67, http://www.amazon.com/R-Book-Michael-J-Crawley/dp/0470510242/ref=sr_1_1?ie=UTF8&s=books&qid=1255451055&sr=1-1
Alibris, $68, http://www.alibris.com/booksearch?S=R&qisbn=9780470510247&qsort=p&siteID=uTKYNncJNRY-O1bFaP1tUnwqUX_ejfgKIA
Barns & Noble, $75 http://search.barnesandnoble.com/The-R-Book/Michael-J-Crawley/e/9780470510247/?itm=1
Borders, $110, http://www.borders.com/online/store/TitleDetail?sku=0470510242
Wiley, $110, http://www.wiley.com/WileyCDA/WileyTitle/productCd-0470510242.html
Dolores: I registered for your R course and I was wondering what software besides R will we be using for the class. I need to get FWS approval prior to installing software to my computer. Will we be using WebEx or some other conferencing software, etc.?
Paul: We will be using GoToWebinar (https://www1.gotowebinar.com/en_US/webinar/entry/entry.tmpl) conferencing software. It does not require administrator access. There is no cost for participants. For interactive graphics, the rattle package uses GTK+ 2.12.9 (11MB) http://downloads.sourceforge.net/gladewin32/gtk-2.12.9-win32-2.exe and GGobi (807KB) http://www.ggobi.org/downloads/ggobi-2.1.8.exe, but they are not essential.
