Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Hackathons as an Informal Learning Platform

Hackathons as an Informal Learning Platform

Presented at SIGCSE 2016

More details at http://go.osu.edu/hacksigcse

Video at https://youtu.be/akjFxR4sShc

Arnab Nandi

March 04, 2016
Tweet

More Decks by Arnab Nandi

Other Decks in Technology

Transcript

  1. Hackathons as an 

    Informal Learning Platform
    Arnab Nandi / Meris Mandernach

    Computer Science & Engineering / University Libraries
    http://hack.osu.edu

    View full-size slide

  2. Outline
    •  What is a Hackathon?
    •  Learning Environments
    •  The OHI/O Hackathon
    •  Findings
    •  Future Work

    View full-size slide

  3. Outline
    •  What is a Hackathon?
    •  Learning Environments
    •  The OHI/O Hackathon
    •  Findings
    •  Future Work

    View full-size slide

  4. hackathon
    hack·a·thon /ˈhakəˌTHän/
    An event, typically lasting several days, in which
    a large number of people meet to engage in
    collaborative computer programming —OXFORD DICTIONARIES
    • hack (cut with rough or heavy blows, cobble together)
    • marathon
    (note: has nothing to do with security / breaking into things)

    View full-size slide

  5. What is a Hackathon?
    “figure it out” /
    learn along the way!
    idea
    working software
    (or hardware)
    short span of time
    (24-36 hours)
    Needs to work, but
    doesn’t need to be
    perfect
    Make new friends
    (possibly co-founders)
    (peer learning)

    View full-size slide

  6. Motivation
    •  Fostering a tech culture

    at Ohio State
    •  Long-term Investment 

    towards Students

    •  Ecosystem
    •  Industry
    •  Research
    •  Education

    View full-size slide

  7. Needs
    •  Gap between 

    university education and

    industry needs
    •  Holistic learning
    •  End-to-end skills
    •  Working in teams

    View full-size slide

  8. Student Learning Experiences
    •  Project based learning
    • Real-world applications Irani, 2015


    • Rapid prototyping

    with industry focus Sigfridsson, 2007; Sousa, 2013; Briscoe & Mulligan 2014
    • Hackathons supporting learning Skirpan, 2015

    View full-size slide

  9. Student Learning Experiences
    • Informal Educational Opportunities


    •  Gamification of education Wolz et al, 2006


    • Group and Peer Learning / Teaching


    •  Competition, Cooperation 

    and Individualistic Learning Johnson and Johnson, 1987


    •  Competition to spur peer learning Topping, 2005


    View full-size slide

  10. Outline
    •  What is a Hackathon?
    •  Learning Environments
    •  The OHI/O Hackathon
    •  Findings
    •  Future Work

    “Input / Output”

    View full-size slide

  11. OHI/O Hackathon
    • Annual 24—36 hr Hackathon Event
    • Open to University Students
    • Industry Sponsors
    • Team-based
    • There’s a theme, 

    but you can build anything you want
    • Focus: making a workable project

    View full-size slide

  12. Hackathon Format
    • Pre-event Talks, Team Formation
    • Event:
    • Tech Talks
    • Hacking
    • Mentors available
    • 4-hour shifts
    • Industry & Alumni
    • Judging (Staff, Faculty & Industry)
    • Showcase & Awards

    View full-size slide

  13. 2013 2014
    History
    2015 2016
    •  1st University-wide

    Hackathon
    •  Focus: writing
    working code: 

    demo or don’t
    •  103 participants!
    200+
    500+

    View full-size slide

  14. Outline
    •  What is a Hackathon?
    •  Learning Environments
    •  The OHI/O Hackathon
    •  Findings
    •  Future Work

    View full-size slide

  15. Data Collected
    •  Cross-reference
    •  Registration information (OSU students only)
    •  Academic records
    •  GPA, Major, Year of study, Gender
    •  Github commits
    •  2014 only
    •  Analysis for 5 teams
    •  Post-event surveys
    •  Response rates: 20% (2013), 36% (2014)

    View full-size slide

  16. Demographics: Majors
    Major 2013 (%age). 2014 (%age).
    Computer Science & Engineering 79 71
    Electrical Engineering 8 12
    Physics & Engineering Physics 6 3.5
    Mechanical Engineering 1 3
    Exploration / Undeclared 1 3
    Business 0 2
    Chemistry & Chemical Engineering 0 1
    Biomedical 0 1
    Architechture & Planning 0 1
    Geography 2 0.5
    Aero, Astronomy, Astrophysics 2 0.5
    Mathematics 0 0.5
    Civil Engineering 0 0.5
    Industrial Systems Engineering 0 0.5
    Psychology 1 0

    View full-size slide

  17. Demographics: Year
    6
    12
    19
    33
    22
    37
    33
    78
    23
    24
    0 10 20 30 40 50 60 70 80 90
    2013
    2014
    Number of Students
    Year of Event
    Grad
    4th
    3rd
    2nd
    1st

    View full-size slide

  18. Demographics: Insights
    • 2x year-on-year growth
    • Wider participation outside CSE / OSU
    • Almost 3x as many women
    • Participation across grad & undergrad

    View full-size slide

  19. Academic Performance
    •  Possible negative impact?
    •  Considered non-academic activity
    •  High time involvement
    •  Discovery
    •  Hackathon Students have 

    2—5% higher GPA!
    •  (note: not significant)
    •  Our hypothesis:
    •  Does not cause / impact GPA
    •  Self-selects a high achieving group of students
    time participant noted:“While working on a project with
    was fun, and I learned a lot from it, the chance to talk
    r like-minded people and meeting mentors was definitely
    the most invaluable experiences from the event.”
    ome mentors - they really made it for me.”
    CADEMIC PERFORMANCE
    ble 3, we present a term-by-term analysis of Cu-
    and Term GPAs for Undergraduate students dur-
    existence of our program. (Some numbers / mea-
    e withheld for institutional privacy and administra-
    sons.) While we do not expect a single weekend
    ch year to have a measurable impact on a student’s
    ance, does the long-term impact on student morale,
    ity, self-confidence and peer-learning have a bear-
    academic performance? One concern is that the
    d time involvement of such extracurricular events
    with homework and class projects, and are a time-
    a student’s already busy academic schedule: they
    ve a negative impact on in-class performance. In
    Category / Measure Au 13 Sp 14 Au 14 Sp 15
    Hackathon Undergrads
    Count 129 132 160 148
    Term GPA 3.136 3.073 3.155 3.131
    Cumulative GPA 3.243 3.218 3.177 3.193
    Degree-Seeking Undergrads
    Term GPA 3.031 3.066 3.073 3.121
    Cumulative GPA 3.071 3.088 3.088 3.112
    Hackathon CSE-BS
    Count 70 71 71 64
    Term GPA 3.192 3.124 3.204 3.141
    Cumulative GPA 3.254 3.242 3.224 3.231
    Non-Hackathon CSE-BS
    Count 621
    Term GPA 3.025
    Cumulative GPA 3.123
    Table 3: Average Cumulative and Term GPAs for Un-
    dergraduate students. Hackathon-participating students
    have small (2–5%) but consistently higher GPAs than non-
    participating students.

    View full-size slide

  20. Anecdotal Evidence: 

    Impact on In-class Participation
    • Post-event emails from CSE instructors (2015):


    •  “…Typically at the end of the semester the excitement level is low, but I’ve been
    getting a number of comments from students about how the Hackathon
    recharged their interest in their CS coursework even as the slog of the
    semester had been getting them down on their classes overall.”

    •  “Talking to my students about their experience this week, they were unanimously
    positive and still excited about have participated. 

    The community feeling has even carried over into closed labs this week, with
    groups spontaneously forming to solve problems, and students moving
    around the lab to meet up with others and talk about the assignments. ”

    View full-size slide

  21. Data Collected
    •  Cross-reference
    •  Registration information (OSU students only)
    •  Academic records
    •  GPA, Major, Year of study, Gender
    •  Github commits
    •  2014 only
    •  Analysis for 5 teams
    •  Post-event surveys
    •  Response rates: 20% (2013), 36% (2014)

    View full-size slide

  22. Findings: How Teams Work
    “Relay” Model
    “Waves” Model
    10/3/14&3:36&PM& 10/3/14&9:36&PM& 10/4/14&3:36&AM& 10/4/14&9:36&AM& 10/4/14&3:36&PM& 10/4/14&9:36&PM& 10/5/14&3:36&AM& 10/5/14&9:36&AM&
    10/3/14&12:00&PM& 10/3/14&6:00&PM& 10/4/14&12:00&AM& 10/4/14&6:00&AM& 10/4/14&12:00&PM& 10/4/14&6:00&PM& 10/5/14&12:00&AM& 10/5/14&6:00&AM&

    View full-size slide

  23. Findings: Progress over Time
    0"
    5"
    10"
    15"
    20"
    18" 19" 20" 21" 22" 23" 0" 1" 2" 3" 4" 5" 6" 7" 8" 9" 10" 11" 12" 13" 14" 15" 16" 17" 18" 19" 20" 21" 22" 23" 0" 1" 2" 3" 4" 5" 6" 7"
    Sleep%/%Rest%Period%
    Final%push%before%deadline%
    Number'of'commits'per'hour,'5'teams'
    0"
    500"
    1000"
    1500"
    2000"
    2500"
    3000"
    3500"
    4000"
    4500"
    5000"
    10/3/14"12:48" 10/3/14"17:36" 10/3/14"22:24" 10/4/14"3:12" 10/4/14"8:00" 10/4/14"12:48" 10/4/14"17:36" 10/4/14"22:24" 10/5/14"3:12" 10/5/14"8:00"
    Each%dot%=%source%code%commit%
    Final%push%before%deadline%
    Lines&of&code&over&-me,&5&teams&

    View full-size slide

  24. Data Collected
    •  Cross-reference
    •  Registration information (OSU students only)
    •  Academic records
    •  GPA, Major, Year of study, Gender
    •  Github commits
    •  2014 only
    •  Analysis for 5 teams
    •  Post-event surveys
    •  Response rates: 20% (2013), 36% (2014)

    View full-size slide

  25. A sense of accomplishment
    •  “I loved the fact that this event gave me time to learn
    how to implement real world applications 

    of Computer Science”

    •  "The ability to sit down for an extended period of time
    and just work on a project allowed me to accomplish
    more in 24 hours than I have since the start of the
    school year and it made me feel like I actually did
    something useful for once."

    View full-size slide

  26. Peer Learning
    •  "Ability to create something useful 

    and learn from others."


    •  "It allowed me to participate and gain

    experience in a group programming setting”

    View full-size slide

  27. Peer Learning / Mentorship
    “While working on a project with friends was fun, and
    I learned a lot from it, the chance to talk to other like-
    minded people and meeting mentors was definitely
    one of the most invaluable experiences from the
    event.”

    “It was so cool that mentors were available through
    the night to provide guidance and talk about how the
    projects we were working on would apply in the real
    world.”

    View full-size slide

  28. Impact: Example Projects
    • Team Building: Suicide prevention app
    • Psychology PhD student + CSE undergraduate student
    • Met at event
    • Semester-long research: 3D Robot Arm controlled by gestures
    • Went on to become an URO-funded project
    • Commercialization: Retail discount app
    • Recently launched on the app store

    View full-size slide

  29. Outline
    •  What is a Hackathon?
    •  Learning Environments
    •  The OHI/O Hackathon
    •  Findings
    •  Future Work

    View full-size slide

  30. Future Work
    •  Demonstrating long-term learning
    •  Student involvement & ownership
    •  The OHI/O Program
    •  Extend beyond CSE
    •  Travel to other hackathons
    •  Year-round talks & events
    •  Industry engagement
    •  Integration with existing curriculum
    •  Capstones
    •  Course enrollment

    View full-size slide

  31. Summary: 

    A platform for informal learning


    •  Increased Engagement & Enthusiasm
    •  Peer learning despite competitive setting
    •  Mentorship opportunities
    •  Requires students to demonstrate comprehensive knowledge
    •  Opportunity for integrating educational elements 

    into hackathons

    View full-size slide

  32. Thank you!

    Paper, slides, videos, and more at 

    http://go.osu.edu/hacksigcse

    View full-size slide