Computer Science Education Research

1
Computer Science Education Research a
smattering of results

• Carol Zander
• Threshold Concepts
• Debugging
• Software Design
• Fixed vs. Growth Mindset

2
Threshold Concepts

• Threshold concepts (Meyer and Land, 2005)
• Transformative
• Integrative
• Irreversible
• Troublesome
• Characteristic of the discipline

3
Threshold Concepts

• For example
• Math limit
• Literature irony
• Physics heat transfer
• Economics opportunity cost
• Medicine pain

4
Threshold Concepts

• Good evidence that two specific concepts satisfy
  the criteria
• - Object-oriented programming
• - Pointers/memory

5
Threshold Concepts other work

• Searching for threshold concepts and
  investigation overview (Eckerdal et al., 2006
  Boustedt et al., 2007 Zander et al., 2008
  Thomas et al., 2009)
• Learning troublesome concepts (McCartney et al.,
  2007)
• Liminal spaces (Eckerdal et al., 2007, McCartney
  et al., 2007)
• Integration associated with object-oriented
  programming (Sanders et al., 2008)
• Transformations related to abstraction (Moström
  et al., 2008)
• Transformations (Moström et al., 2009, Zander et
  al., 2009, Thomas et al., to appear)

6
Threshold Concepts recent work

• Harnessing Surprise Tales from Students'
  Transformational Biographies  
• Special Interest Group on Computer Science
  Education (SIGCSE), 2010, to appear
• Based on philosophical position that surprise
  causes learning
• (MRI scans give evidence that memory and the
  learning of concepts is enhanced by surprise.)
• as educators, we should try and set up
  situations in which our students experience it
  more often.

7
Debugging

• Debugging Finding, Fixing and Flailing, A
  Multi-Institutional Study of Novice Debuggers
• Computer Science Education Journal, 2008
• Figure shows quality of debugging strategies
  versus bugs fixed.
• Student self-assessment of debugging ability
  shown in error bars.

8
Debugging

• Journal, 2008

9
Debugging

• Debugging The Good, the Bad, and the Quirky
• a Qualitative Analysis of Novices Strategies
• Special Interest Group on Computer Science
  Education (SIGCSE), 2008

10
Debugging strategies
Gain domain knowledge Isolating the problem
Tracing, including Mental, Print Debugger Pattern matching
Tracing, including Mental, Print Debugger Consider alternatives
Testing Environmental
Understanding code Work around problem
Using resources Just in Case
Using Tools Tinkering
11
Software Design

• Categorizing Student Software Designs Methods,
  results, and implications
• Computer Science Education Journal, 2005
• Can Graduating Students Design Software?
• Special Interest Group on Computer Science
  Education (SIGCSE), 2006

12
Software Design

• Large study 22 researchers from 22
  institutions, 4 countries (150 graduating
  senior designs)
• Design task Given a challenging design problem
  Analyze, decompose and organize a task into
  meaningful and manageable parts

13
Software design categories

• Nothing little or no content
• Restatement restates design requirements
• Skumtomte small amount past restatement
• First step some significant work past
  restatement
• Partial design provides a description of the
  parts, and overview showing relationships, but
  incomplete
• Complete design well-developed solutions which
  include an understandable overview, part
  descriptions with explicit communication between
  the parts

14
Can graduating students design software? NO
15
Fixed vs. Growth Mindset

• Carol Dweck studied mindsets about intelligence
  and academic success
• Fixed mindset (fixed intelligence) avoid
  challenges and perform worse academically
• Grow mindset intelligence can grow with effort
  and hard work
• Take-away message Dont tell students they are
  successful because they are smart emphasize
  hard work leads to success
• Research done Saying is believing exercise