SC 07

Chad Thompson
Computational literacy – doing science the modern way
We used to do
A theory
B experiment
Now the third element is computational models
Car engineers did car design and crash test
Capacity to understand the relationship between domain knowledge of science and knowledge required to understand and apply computational models
Study by Krell Institute
Measure student performance on computational literacy and curriculum content
Treatment – taught usinig interactive simulation (scaffolded computational model)
Control 0 students using traditional

Predator/prey
Spread of disease
Carbon cycle
Rock cycle

Computationallabotroy.org

Pre test to 600 students
Then removed the ones who had problems to control variables like lack of computers, access to alb etc
Also between control and treatment, verified the variable of race and income

Evidence of improvement is there, but there is no proof
What did Chad learn from research?
Teachers matter
Type of simulation matters
The story behind the simulation matters
Technical environment matters

www.krellinst.org
www.computationallaboratory.org

Look here for the simulations

Saturday, November 10, 2007
Chad Thompson of Krell Institute presented his research on computational literacy in four states.

Steve Gordon
What minimum computational science skills should undergraduates have?
What minor?
Simulation and modeling
programming and algorithms
Differential equations and discrete dynamical systems
numerical methods
optimization
parallel programming
scientific visualization
Should all students have the skills?
Based in feedback from industries, they have now created modules
Example:
MODSIM:
1. Explain the role of modeling in science and engineering
2. Analyze modsim in computational science
create a model
examine mathematical functions
etc
Programming and algorithm competencies
fundamentals of problem solving
understand and write pseudo code
code in programmng lan
use of fundamentals of algorithm
DQ and discrete dynamical systems
---first order linear differential and difference eequeations
--- Laplace transforms
Numerical methods
-- Monte carlo methods
-- ordinary and partial DQ
Optimization
-- use
-- linear and nonlinear systems
Parallel programming
-- related architectures
-- pp using MPI
-- parallel scalability
Scientific visualization (SV)
--- SciVis needs
-- computer graphics concepts
-- ise surface
-- volumetric rendering
Integrated approach
Project based learning approach
capstone experiences
-- computationally oriented course in major disciplines
-- research or internship experience
www.rrscs.org/minor/competencyfinal.pdf
minor/courses.shtml





Research: Ekiga