NSF Workshop:
Integrating Computing Into the Statistics Curricula

Cheakamus Room, Fairmont Waterfront

Schedule

Friday July 30

9.00 - 9.30	Overview	An overview of the important role of computing within the different statistics programs and curricula, high-level perspective on how computing should be addressed in the syllabi and curricula, the need for teaching statistical computing in a general and intellectually rich manner rather than a sequence of "how to"'s, the changing nature of statistical computing, what we hope to achieve in this workshop.
9.30 - 11.30	Programming Fundamentals	How to teach programming. Fundamentals of R getting started, help, search path, managing objects in the workspace. Graphics high-level plotting annotating plots graphics devices and formats graphics parameters par() lattice & grid. Environment and Tools for programming. editors - Emacs & ESS, tinn-R, Eclipse Directories/Folders, paths, binary file formats. Remote login - ssh, remote graphics, ... VMWare - Unix shells on Windows (cygwin) Data types Vectorized computations Subsetting with(), subset(), local() Functions, control flow Debugging strategies & tools Efficiency - common idioms, profiling.
10.30 - 10.45	Coffee Break
11.30 - 12.00	Discussion: New Courses	Participants present what they are doing to introduce computing into their programs.
12.00 - 13.00	Lunch - on your own
11.30 - 12.00	Discussion: Course logistics	How do you organize a computing class? For example, what style of assignments to give, how to use electronic bulletin boards, chat rooms, what kinds of course materials to use.
13.30 - 15.30	Data Input & Output	Fundamentals Reading tabular data Reading subsets of large data Filtering with shell tools grep, cut, head, tail. redirection. Text manipulation Regular expressions Complex I/O Connections Non-rectangular data Other languages - sed, awk, perl, python. Binary formats
15.30 - 16.00	Reception/Snacks
16.00 - 17.00	Developing Assignments	Goals of assignments. Sources of data and existing assignments. Working in groups or individually. Grading. Our experiences creating assignments. Summer school projects.

Saturday 31 July

9.00 - 9.30	Discussion: Integrating Computing into Existing Classes	The pros and cons of adding computing aspects to more traditional statistical methodology and mathematical statistics classes. Active construction of learning by student. Strategies for integrating computing into existing classes. Sample assignments for computing in math. stat and stat. methodology classes.
9.30 - 11.30	Advanced Programming	Material for an introductory graduate class or a second, advanced undergraduate class. Efficiency and Compiled code Distributed & Parallel computing. Working with large data strategies & packages relational databases Object-oriented programming concepts. R packages, vignettes, dynamic documents & reproducible research. Software design & development. Version control
10.30 - 10.45	Coffee
11.30 - 12.00	Resources	Books Assignments, Data, & Case Studies Web sites for course & topic notes.
12.00 - 13.00	Lunch - on your own
13.00 - 13.30	Discussion: Computational Statistics	What is the role of computational statistics in a modern statistics program? What should a modern computational statistics course look like?
13.30 - 14.15	Web Technologies - presenting information	Graphics KML & Google Earth SVG JavaScript Flash Animation GUIs
14.15 - 15.00	Web Technologies - scraping and importing data	Web scraping - HTML HTML forms Web services - SOAP & REST
15.00 - 15.30	Broader Curricular Concerns	How can statistical computing improve the teaching and learning of data analysis and modern methods?
15.30 - 16.00	Reception - Snacks
16.00 - 17.00	Next Steps: Plans for Proceeding	Steps for starting a new course. Ways to disseminate materials How to foster more computing in the curriculum?