Lessons Learned from a
PLTL-CS Program
Christian Murphy, University of Pennsylvania
Rita Powell, National Center for Women & IT
Adam Cannon, Columbia University
Kristen Parton, Columbia University
March 10, 2011
Emerging Scholars Program (ESP)
A Peer Led Team Learning Program (PLTL)
ESP concept – Uri Treisman at UC Berkeley (Math Workshop)
ESP/PLTL long applied in STEM disciplines
• 1 or 2 hour long weekly workshops adjunct to an
academic course
• Workshops led by a undergraduate “peer leader”
• 6 to 15 student participants per workshop
• Students work together to solve problems without pressure of
exams or grades
• Weekly meetings between peer leaders and program
coordinators
• Coordinators evaluate program at end of semester to
measure success, make adjustments
ESP-PLTL in Computer Science
• NSF funded eight-university study in 2004
• Developed, implemented and evaluated an ESP-PLTL
program in CS1 over three years
• Published detailed findings, results, course modules
• Horowitz ( Wisconsin-Madison) and Rodger (Duke)
SIGCSE paper in 2009
• Found that active recruiting combined with ESP-PLTL
is an effective approach to attracting and retaining
under-represented students in an intro CS class
• ESP-PLTL increased grades, retention in CS1
Columbia Emerging
Scholars Program
Background: CS1 at Columbia

Large, lecture-style class of about 150-200
students with no lab or recitation

Required for many other Engineering majors

Most students have not yet declared a major


Students are admitted to a School and declare in
their second year
Approximately 40% female enrollment

Compared to less than 10% in the major in 2007
Columbia Emerging Scholars Program

Pilot program in Spring 2008

One section, six participants

Adjunct to the CS 1 class

Coordinated by 2 PhD students and CS1 instructor

Seed grant from NCWIT in summer 2008

Has continued to grow

Currently two sections, 8-10 participants each

Over 80 students have completed the program
Workshop Topics


Focus on collaborative problem solving and
algorithmic thinking

No programming whatsoever

No homework or preparation
Topics from AI, graph theory, robotics, graphics,
etc.

Most material created from scratch

Some material from pltlcs.org website
1. “Mä hach’a challwawa challwataxa.”
2. “Kimsa hach’a challwawa challwataxa.”
3. “Mä challwa mä hach’a challwampiwa challwataxa.”
4. “Mä hach’a challwa kimsa challwallampiwa
challwataxa.”
5. “Paya challwallawa challwataxa.”
6. “Mä challwalla paya challwampiwa challwataxa.”
7. “Kimsa challwa paya challwallampiwa challwataxa.”
Coed vs. Women-Only

Originally CESP was only for female students

“How much do you like all-female workshops?”



46%: I prefer it that way

42%: it doesn't matter
Fall '09 and Spring '10: one all-female section
and one coed section
Fully coed as of Fall 2010
Lessons Learned
Recruiting and Selection




Possibly the most important ingredient to a
successful program
Know what you're looking for... and what you're
willing to accept
In-class announcements from an undergrad
work better than unsolicited emails
Institutionalization would help
Setting Expectations



Regardless of how you explain it, students may
still not understand how a PLTL program works
Have former participants describe their
experiences
“Open house workshop” for all applicants
Scheduling

Friday afternoons tend to be the best



One section in early afternoon, one a bit later
Get a commitment from peer leaders
Be aware of conflicts with CS1 activities,
religious holidays, Grace Hopper Conf., etc.
Training Peer Leaders

Choose peer leaders from past participants



CESP uses a “workshop assistant” position to
groom potential peer leaders
Training should be ongoing, even if the peer
leader is already familiar with the material
Peer leaders should present material at staff
meeting earlier in the week
Discussion vs. Problem Solving



Students tend to prefer activities that have clear
“right” answers
Discussing the merits of different approaches/
algorithms not as appealing (but still important!)
Open-ended discussions about ethical/legal
issues in computing have been lowest-rated
Evaluating the Program



Crucial to know the facts:

How many students go on to CS2?

How many major/minor in CS?
Be nice to administration and make sure they
know you'll want this data
Stay in touch with past participants so they're
not surprised when they hear from you
Initial Results
Women’s Representation in the Computer
Science Major at Columbia 2005-2010
25
% of women CS majors
20
15
Start of CESP
10
5
0
2005-06
2006-07
2007-08
2008-09
2009-10
Other Results


45% of CESP participants who have declared a
major have chosen Computer Science
Peer leaders are the ones who benefit the most



Two have participated in CRA-W Distributed Research
Experience for Undergraduates
One CRA Outstanding Undergraduate Research
honorable mention
One currently working at Microsoft
Ongoing Evaluation


What is the quantitative impact of CESP on
recruiting and retention?

Number of students majoring in CS

Number of students who take CS2
What is the qualitative impact of CESP?

Attitudes toward CS

Effect of coed vs. all-female workshops
Links
National Center for Women & IT
http://www.ncwit.org
Columbia Emerging Scholars Program
http://www.cs.columbia.edu/esp
Peer-led Team Learning in CS
http://www.pltlcs.org
Revolutionizing the face
of technology
The NCWIT Alliances
To Join the K-12 Alliance, email info@ncwit.org