Post on 12-Jul-2015
Creative Expression to Motivate Interest in
Computing
Mark Guzdial & Barbara Ericson
School of Interactive Computing
Story
• Programming is hard:
Need motivation to get there
• Rigor, Jobs, Creative Expression
• Media Computation
• Example Lesson with Media Computation
• Along the Pipeline:
From Elementary Students to Teachers
The Rainfall Problem
• Problem: Read in integers that represent
daily rainfall, and printout the average daily
rainfall.
• If the input value of rainfall is less than zero,
prompt the user for a new rainfall amount.
• When you read in 99999, print out the
average of the positive integers that were
input other than 99999.
Results at Yale in Pascal in 1983
% of Students who got
it right
Novices (3/4 through first
course)
14%
Intermediates (3/4
through second course)
36%
Advanced (Jrs and Srs in
Systems Programming)
69%
Programming is hard
• Elliot Soloway and his students replicated this
study several times.
• Others have used this same problem with similar
results (Most recently: Venable, Tan, and Lister,
2009)
• 2001: McCracken Working Group Study
• The first of several Multi-Institutional, Multi-National
(MIMN) studies of CS education
• Out of a possible 110 points, average score was 22.89.
• 2004 Lister Group; 2010 Allison Elliot Tew
"There are three things to
emphasize in teaching: The first is
motivation,
the second is motivation,
and the third is (you guessed it)
motivation."
- Terrel H. Bell, U.S. Secretary of Education,
1981–1985 (Bell, 1995)
So why study programming?
• In Europe and Australasia, Computer
Science is a discipline worthy of rigorous
study.
• In The United States, Computer Science is
about jobs.
Simon Peyton-Jones,
CAS
Emphasis in US: Jobs
1.4M IT jobs in US400K graduates
OCTOBER 5, 2012
Code.org
Why another motivation?
• Who is not motivated by the current
approach?
• Whose talents are we missing out on
because of this motivation?
Thanks
to Brian
Danielak
Media Computation:
A Context to Motivate Learning Computing
• Fall 1999:
All students at Georgia Tech must take a course
in computer science.
• Considered part of General Education, like
mathematics, social science, humanities…
• 1999-2003: Only one course met the requirement.
• Shackelford’s pseudocode approach in 1999
• Later Scheme: How to Design Programs (MIT Press)
• Less than half of students in Liberal Arts, Architecture,
or Business passed this course.
Contextualized Computing Education
• What’s going on?
• Research results: Computing is
“tedious, boring, irrelevant”
• Since Spring 2003, Georgia Tech
teaches three introductory CS
courses.
• Based on Margolis and Fisher’s
“alternative paths”
• Each course introduces computing
using a context (examples,
homework assignments, lecture
discussion) relevant to majors.
• Make computing relevant by teaching it
in terms of what computers are good for
(from the students’ perspective)
Media Computation:
A Context of Digital Expression
• Programming across data abstractions• Iteration as creating
negative and grayscaleimages
• Indexing in a range as removing redeye
• Algorithms for blending both images and sounds
• Information encodings as sound visualizations
15
def negative(picture):
for px in getPixels(picture):
red=getRed(px)
green=getGreen(px)
blue=getBlue(px)
negColor=makeColor(255-red,255-green,255-blue)
setColor(px,negColor)
def clearRed(picture):
for pixel in getPixels(picture):
setRed(pixel,0)
def greyscale(picture):
for p in getPixels(picture):
redness=getRed(p)
greenness=getGreen(p)
blueness=getBlue(p)
luminance=(redness+blueness+greenness)/3
setColor(p, makeColor(luminance,luminance,luminance))
17
Open-ended, contextualized homework
in Media Computation CS1 & CS2
Sound collage
Linked list
as canon
Flags
Your assignment is to write a function that will create a
collage of pictures. Your collage will be made by
copying at least 3 pictures onto a blank canvas. You
must use at least 3 different pictures. When finished,
your collage should look like the flag of a country.
Study-Abroad
CS
Results:CS1“Media Computation”
Change in Success rates in CS1 “Media
Computation” from Spring 2003 to Fall 2005
(Overall 85%)
Architecture 46.7% 85.7%
Biology 64.4% 90.4%
Economics 54.5% 92.0%
History 46.5% 67.6%
Management 48.5% 87.8%
Public Policy 47.9% 85.4%
We have sustained that pace
Table 1: R et ent ion dat a for G eor gia Tech ’s M edia-Com p cour se fr om Fal l 2006–Fal l 2012
Fem ale M ale Tot al
Passing grades 2102 1659 3761
Failing grades 208 235 443
Withdraw 30 46 76
DFW Total 238 281 519
% DFW-total 5.5% 6.5% 12%% DFW-set 10.1% 14.5% 12%
t hat we were wrong. Whatever incites plagiarism, Media-Comp does not seem to impact plagiarism.
3. THE RETENTION HYPOTHESISThe retent ion hypothesis is the most explored and best
supported MediaComp hypothesis. We originally hypoth-esized that we would have a failure2 rate less than 15%.Our results in the pilot offering of the course showed thatless than 12% of students withdrew from the course or had
a failing grade (Unpublished internal report [23]3 ). In fact ,the retent ion rate rose compared to our t radit ional course
in both our MediaComp course and in our Engineering-oriented MATLAB course that started thesamesemester[14].The average failure rate for students from the Colleges ofArchitecture, Management , and Liberal Arts had been over
50% in the two years before MediaComp started, and lessthan 15% in the following two years [27].
A review of the last six year ’s of ferings of the Georgia Tech
MediaComp course (4,292 total students) show the failurerate cont inues to be below 15% (Table 1). We use 4,280
students in our calculat ions because students took an in-complete and finished the course the following semester, andare not considered to be passed or failed. We compute %DFW-total in terms of DFW for men, women, and totalover 4280, then %-DFW in terms of DFW over the number
within that set . The course at Georgia Tech was major-
ity female because the majors who were required to take it
(from the Colleges of Liberal Arts, Management , and Archi-tecture) are more female (than the other colleges at GeorgiaTech – Engineering, Comput ing, and Sciences).
In published accountsof other schools’ MediaComp courses,the retent ion hypothesis has been well-supported. We stud-
ied the first adopt ion of MediaComp at Gainesville StateCollege, and found an increase in retent ion over the t radi-
t ional course [43]. TheUniversity of I llinois-Chicago adoptedMediaComp for their “ CS 0.5” course, and they found a sig-nificant improvement in retent ion [38]. The University ofCalifornia at San Diego adopted MediaComp in 2008 and
document an improvement in retent ion [37]. Five yearslat er, Porter & Simon presented a longer term analysis onhow their adopt ion of peer inst ruct ion, pair programming,
and MediaComp in 2008 led to greater retent ion with thecomputer science major measured into the second year of
undergraduate [34].
2Students who withdraw from the course or earn a failinggrade are considered to have ‘failed’ (DFW) in this analysis.The students who earn a passing grade (A, B, or C) are saidto have ‘succeeded’ and were ‘retained.’3ht t p: / / coweb. cc. gat ech. edu/ medi aComp- pl an/upl oads/ 45/ FI NAL- r epor t - pi l ot - of f er i ng. 1. pdf
3.1 Explaining the retention effectWhy did retent ion improve at mult iple schools over sev-
eral years? There may be mult iple causes, and they may
be different at different schools. We can imagine differences
in grading pract ices between different MediaComp courseswhich might explain retent ion differences, for example. Giventhat the result was similar across different MediaComp in-stant iat ions, and that our goal is to build a theoret ical ex-planat ion, we will focus on the common curriculum between
the classes. How might the MediaComp curriculum haveinfluenced retent ion?
In our first year of implement ing MediaComp, we con-ducted several interview studies to understand how studentssaw the course. Students told us that the course was “mot i-
vat ing” and “ fun.” Why? Several explanatory themes arosefrom those studies:
• Students told us that they appreciated that the course
was “ tailored” t o their major, in both the MediaComp
and Engineering courses [14].
• Students told us that they found the course to be awelcome opportunity to be creat ive [24][35].
• The theme of relevance appeared in prominent ly inour findings. Female students told us that they un-
derstood what they would do with the content of theMediaComp, but those in the t radit ional course saidthat they did not know what they would do with whatthey were learning [35]. In a follow-on study a year af-
ter we started the MediaComp course at Georgia Tech,we found that 19% of the respondents to a survey hadactually done some programming outside any course
context [24].
Here is a quote that captures the relevance point well [24],where a student describes coding beyond the requirementsof the course:
I just wish I had more t ime to play around with
that and make neat effects. But JES (thePythonIDE for the MediaComp course) will be on mycomputer forever, so... that ’s thenice thing aboutthis class is that you could go as deep into the
homework as you wanted. So, I ’d turn it in andthen me and my roommate would do more afterto see what we could do with it .
We found student react ion to the course surprising whenviewed from a perspect ive of “ thick authent icity” [36]. Stu-dents found MediaComp relevant and something that theycould use for later programming. Yet the course is quiteinauthentic [27]. Real programmers who manipulat e me-
dia do not write code like what students do in MediaComp.MediaComp is far too slow to be usable in pract ical set -
t ings. However, students clearly saw that they could douseful tasks with code in MediaComp [24], even if thosetasks would not be performed in that way and with thosetools by professionals.
We made a theoret ical choice at this point in our explo-rat ion of MediaComp, in part driven by the finding that theretent ion effect occurred in both the MediaComp course andthe Engineering-focused course. The Engineering course did
not make an effort to provide a creat ive out let . We decidedto focus on the sense of relevance. Creat ivity may be an im-
portant part of why MediaComp is mot ivat ing to students.
UCSD’s PI+PP+MediaComp Experiment (SIGCSE 2013)
• UCSD changed CS1
(quarter system) in
2008 to:
• Peer Instruction
• Pair Programming
• Media Computation
• Tracking students
since 2001.
• Increase retention of CS
majors into second year
by 30% (from 51% to
81%)
BS in Computational Media
• Joint between School
of Literature, Media,
and Communications
(Liberal Arts) and
Computing.
• 45% Female
• Most gender-balanced
ABET-accredited
computing program in
US
A Sample Lesson
How sound works:
Acoustics, the physics of sound
Digitizing Sound: How do
we get that into bytes?
• Remember in calculus, estimating the curve by creating rectangles?
• We can do the same to estimate the sound curvewith samples.
Now let’s talk about information
• Get a sound and make an empty picture.
• For each pixel in empty picture,
• Get the next sample from the sound.
• If the value is less than -2000,
make the pixel blue.
• if the value is greater than 2000,
make the pixel red.
• Otherwise, make the pixel green
Beyond Undergrad
• Using creative expression to motivate
elementary and secondary students.
• And teachers.
AP CS APicture Lab - Lots of Fun
We just completed the Picture Lab and I wanted to take a moment to thank
Barbara Ericson for all her hard work on such an engaging unit. We had a lot of
fun working through the exercises. And there were plenty of interesting
peripheral concepts that could be explored to whatever depth felt right.
We worked in groups of 3 students. For each set of exercises, students could
choose one problem to tackle. When they were done, they shared their solutions
with the other 2. For an added challenge at the very end, I had them choose
between the Green Screen, Hidden Message, and Pesky Tourist problems.
Has anyone else worked though the unit and have any ideas to share?
--Tonya Herron
Anderson High School
Cincinnati, Ohio
Summer Camp Collages
Summer Camp Collages
Summer Camp Collages
Summer Camp Green Screen
Summer Camp Green Screen
Teacher Collages
Teacher Collages
Teacher Collages
Teacher Green Screen
Other Creative Tools for K-12
• Scratch
• Alice
• LEGO Robots
• Artbotics
• EarSketch
• Writing programs to remix music
Scratch – Elementary School
Scratch – Middle School
Alice – High School
Alice – High School
WeDo Spin Art
Drawing with Robots
ArtBotics
• Use LEGO NXT and EV3 to make art
• http://artbotics.cs.uml.edu/wordpress/?page_id
=177
• https://www.youtube.com/user/Artbotics
EarSketch
• Write programs to remix music
• http://earsketch.gatech.edu
With thanks to our supporters
• US National Science Foundation
• Statewide BPC Alliance: Project “Georgia Computes!”
http://www.gacomputes.org
• Expanding Computing Education Pathways Alliance,
http://expandingcomputing.org
• CCLI and CPATH Grants, and now CE21 to produce new media
• Georgia’s Department of Education
• GVU Center, and Institute for People and
Technology (iPaT) at Georgia Tech
Thank you!
• http://www.cc.gatech.edu/~mark.guzdial
http://home.cc.gatech.edu/csl
• http://coweb.cc.gatech.edu/ice-gt
For more on MediaComp:• http://www.mediacomputation.org