Hackathons as an Informal Learning Platform

Presented at SIGCSE 2016

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

Video at https://youtu.be/akjFxR4sShc

Transcript

1. Hackathons as an 
 Informal Learning Platform Arnab Nandi / Meris Mandernach
 Computer Science & Engineering / University Libraries http://hack.osu.edu
2. Outline •  What is a Hackathon? •  Learning Environments •  The OHI/O Hackathon •  Findings •  Future Work
3. Outline •  What is a Hackathon? •  Learning Environments •  The OHI/O Hackathon •  Findings •  Future Work
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)
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)
6. Motivation •  Fostering a tech culture
 at Ohio State •  Long-term Investment 
 towards Students •  Ecosystem •  Industry •  Research •  Education
7. Needs •  Gap between 
 university education and
 industry needs •  Holistic learning •  End-to-end skills •  Working in teams
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
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
10. Outline •  What is a Hackathon? •  Learning Environments •  The OHI/O Hackathon •  Findings •  Future Work “Input / Output”
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
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
13. 2013 2014 History 2015 2016 •  1st University-wide
 Hackathon •  Focus: writing working code: 
 demo or don’t •  103 participants! 200+ 500+
15. Outline •  What is a Hackathon? •  Learning Environments •  The OHI/O Hackathon •  Findings •  Future Work
16. 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)
17. 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
18. 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
19. Demographics: Insights • 2x year-on-year growth • Wider participation outside CSE / OSU • Almost 3x as many women • Participation across grad & undergrad
20. 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.
21. 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. ”
22. 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)
23. 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&
24. 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&
25. 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)
26. 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."
27. Peer Learning •  "Ability to create something useful 
 and learn from others." •  "It allowed me to participate and gain
 experience in a group programming setting”
28. 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.”
29. 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
30. Outline •  What is a Hackathon? •  Learning Environments •  The OHI/O Hackathon •  Findings •  Future Work
31. 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
32. 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
33. Thank you!
 Paper, slides, videos, and more at 
 http://go.osu.edu/hacksigcse