Me

• Luther Tychonievich (Tie CON of Itch)
• CS instructor since 2013 (UVA, then UIUC)
• Have taught many subjects:
  programming, computer architecture, mathematics, computational theory, data structures, computer graphics, teaching methods, web programming, systems programming, security, computational geometry, computer-aided geometric design
• Currently teach
  • CS 340 Introduction to Computer Systems
  • CS 418 Interactive Computer Graphics
• Scholarship on teaching practice, inclusivity, data standards

Research

Moves beyond current limits of knowledge

• Novel, never done before (except replication studies)
• Assumes you already know up to current limits of knowledge

If you knew it would work, it wouldn’t be research

• Sometimes “didn’t work” is new, sharable knowledge
• Experienced researchers can usually pick things that work

Hard to scope

• Estimating difficulty, time needed is challenging when it’s novel

Varies significantly by subfield

Undergraduate Research

All faculty like ugrad research, but not all find it easy to support.

Obstacle: getting you to the edge of knowledge

Solution: take the 400-level course first

Solution: read research papers

Obstacle: insufficient time to mentor

Solution: only take research students when other work is lower than usual

Obstacle: too much noise in requests

Solution: ignore long, indirect emails

Obstacle: research takes time to yield results

Solution: prefer students with several semesters remaining

Some prefer students who can join existing projects;
Others prefer students who come with a research idea of their own.

Two Common Faculty Roles

• Teaching-focused
  • ∼25 CS faculty (“teaching” or “lecturer” in job title)
  • Teach most required CS classes
  • 80% of time on teaching, helping school operate
  • 20% (8hr/week) on scholarship
• Tenure-stream
  • ∼110 CS faculty
  • Supervise most research
  • 70% of time on research, mentoring researchers, aiding research community
  • 30% (12hr/week) on teaching, helping school operate
• Significant individual variation in time allocation

My Scholarship

• I rarely engage in research
  • Sometimes study some aspect of student behavior
  • Sometimes mentor students going beyond CS 418
• I regularly engage in scholarship
  • read, digest, and disseminate practices in inclusive teaching
  • explore alternative ways to educate
  • write, maintain, and edit data standards for family history software
  • develop tools and curricula for CS courses
  • implement teaching research and report my experience
  • review and metareview CS Education experience reports

Example: explore ways to educate

In grading, I’ve tried

• A pool of points totaling 100%
• A pool of points totalling 250%, pick any 100
• Two pools of points, one totalling 200% (pick any 50%) and one totalling 50%
• Six topics, each tested pass-fail, each with 2 retakes, with mappings like “pass 1–4 = B” and “pass all but 2 = C+”

and asked questions like

• What do students like?

• Which seem to me to

  • best match grade to knowledge?
  • best match grade to effort?
  • motivate the most learning?

  (if I measured this, not just “seems to me”, it would become research)

Example: mentoring beyond 418

1. Seventy students took my Computer Graphics

2. Two came to my office and asked “how can we render a cartoon scene with shadows and reflections?”

3. I recommended several papers, which they read

4. I helped them understand those papers, had them read more

5. We brainstormed five possible approaches

6. They tried three of them, all building off of code from class

7. One worked well enough

8. They presented a poster at SIGGRAPH (15 months after the course ended)

Summary

• Research is going beyond what is currently known

  1. Take 400-level course
  2. Brief targeted contact
  3. 2+ semesters of work after course

• I do some research

  • CS Education
  • Student-designed post-elective

• I do much scholarship

  • Dissemination
  • Exploration
  • Standards
  • Tools and curricula
  • Experience reports