Understanding Mindshift Learning: The Transition to ObjectOriented Development

1
Understanding Mindshift Learning The Transition
to Object-Oriented Development

• Deborah J. Armstrong and Bill C. Hardgrave
• MIS Quarterly (in press)

2
Motivation

• IT professionals are repeatedly asked to learn
  new tools, techniques and processes
• Transitions often require a shift in mindset
  (mindshift)
• Examples
• Shift from mainframe to client-server
• Move to object-oriented software development
• Without mindshift advantages may be lost
• Why is learning during a mindshift so difficult?

3
Context

• Iivari, Hirschheim and Kleins (1998 2000-2001)
  Information Systems (IS) development framework
• Four hierarchical levels of framework
• Paradigm
• Approach
• Methodology
• Technique

4
More Context

• Differences in Traditional and OO development
  occur at the approach level. Iivari et al
  (2000-2001)
• Approach components
• Set of goals
• Guiding principles and beliefs
• Fundamental concepts
• Principles for the ISD process
• Learning process may begin with individuals being
  introduced to the fundamental concepts of the new
  approach.

5
Literature Review

• Three themes in software development literature
  from learning / knowledge structures perspective
• Successful IS development education focused on
  semantics first, then syntax (e.g. Spohrer
  Soloway, 1986 Hardgrave and Doke, 2000)
• Experts create abstract (semantically focused)
  knowledge structures, novices have more concrete
  (syntactically focused) knowledge structures
  (e.g. Adelson, 1981 1984)
• Experienced developers have difficulty moving
  from traditional to OO approach (e.g. Rosson
  Alpert, 1990)

6
Learning Process

• Knowledge Structure
• A representation of an individuals knowledge
  which includes domain-specific concepts and the
  relations among those concepts (Dorsey, Campbell,
  Foster Miles, 1999)
• Concept Knowledge
• Ideas and information embodied in the knowledge
  (Ausubel, 1963)
• Incremental learning
• Mindshift learning
• Proactive Interference (Underwood, 1957)

7
Base Theory
Existing Knowledge Structure
Modification (incremental)
Introduce Concepts
Concept Knowledge
New Knowledge Structure
Creation (mindshift)
8
Refining the Theory

• Motivation
• Strengthen theory
• Context specific
• Identified OO concepts
• Inductive Approach
• Gathered insights from experts

9
Revised Theory
Traditional Software Development Knowledge
Structure
OO Software Development Concepts
OO Software Development Concept Knowledge

• Learning
• Novel
• Changed
• Carryover

10
Hypotheses Development
High
OO Concept Knowledge Score
Low
High
Degree of Novelty
11
Hypotheses

• H1. A developers OO concept knowledge score will
  have a U-shaped (curvilinear) relationship with
  the degree of perceived novelty.
• H2. A developers carryover concept knowledge
  score will be greater than his or her changed
  concept knowledge score.
• H3. A developers carryover concept knowledge
  score will be greater than his or her novel
  concept knowledge score.
• H4. A developers novel concept knowledge score
  will be greater than his or her changed concept
  knowledge score.

12
Method

• Subjects
• Sample criteria both traditional and OO
  experience
• 81 software developers (response rate 39)
• 16 organizations
• Instrument Development
• Degree of perceived novelty (9 items)
• OO concept knowledge (27 items)
• Level of perceived difficulty (9 items)
• Validation

13
Hypothesis Testing H1

• OO Concept Knowledge Score a ß1Novelty
  ß2Novelty2

14
Regression
Quadratic Model
Linear Model
Thus H1 is supported
15
Data Preparation

• Categorize concepts (based on degree of perceived
  novelty)
• Carryover 0-24
• Changed 25-75
• Novel 76-100
• Placed scores for each concept into categories

16
Object Concept Categorization
17
Hypothesis Testing
Category Means Novel 2.29 Changed
1.62 Carryover 2.24
t Tests
18
Results

• Why no significant difference between Novel and
  Carryover ?
• Rival hypothesis analysis
• Years of OO experience
• category
• concept
• Difficulty

19
Data Analysis Years of OO Experience
20
Data AnalysisYears of OO Experience/Category
21
Data AnalysisYears of OO Experience/Concept
22
Data AnalysisLevel of Concept Difficulty

23
Bottom Line

• It is not years of OO experience
• It is not difficulty of the OO concept
• It is the perceived novelty of the OO concept
  that impacts learning

24
Implications

• Individual
• Increase productivity, increase employee
  satisfaction, increase employee retention
• Organizational
• Decrease training costs, increase software
  quality, encourage wider adoption of OO
• Transitions
• Training decision aid, change management
  initiatives, increase commitment

25
Future Research

• Goal
• Understand IS professionals mental models and
  changes in mental models across transitions
• Projects
• Test full theory
• Test various antecedent conditions (e.g.,
  multiple existing mental models)
• Test various contexts (e.g., ERP, SOA)

26
(No Transcript)