Understanding Mindshift Learning: The Transition to ObjectOriented Development

1
Understanding Mindshift LearningThe Transition
to Object-Oriented Development

• Authors Deborah J. Armstrong
• Florida State University
• Bill C. Hardgrave
• University of Arkansas
• Presented By DJ Susko
• Cleveland State University

2
Introduction

• IT professionals are constantly facing changes in
  their in their work environment.
• When concepts change, we witness a shift in
  mindset (mindshift).
• Examples
• Mainframe to Client-Server Computing
• Traditional to Object-Oriented Development
• People employ new technologies with old ideas,
  losing advantages of new mindshift.

3
Introduction

• Learning during a mindshift is difficult
• Why is learning difficult?
• Study explores software developers transitioning
  to O-O development.
• Seeks to understand difficulties encountered
  during the acquisition of fundamental O-O concept
  knowledge.

4
Context

• Framework for classifying IS development.
• Developed in 1998, 2000-01 by Iivari, Hirscheim,
  and Klein.
• Four hierarchal levels for framework
• Paradigm
• Functionalism, Social Relativism
  Neohumanism, and Radical Structuralism
• Approach
• Methodology
• Techniques

5
Definitions

• O-O Software Development
• Developing software that is centered on the
  concepts of cooperating objects and classes.
• Traditional Software Development
• Represents any non-object-oriented software
  development approach.

6
More Context

• Using IS framework, shift occurs at the approach
  level.
• Learning process begins with individuals being
  introduced to fundamental concepts that define
  the approach.
• Objective Understand the difficulty while
  learning fundamental concepts that underlay the
  new approach.

7
Past Research

• Three major research themes emerged
• Semantics first, then Syntax.
• Semantically vs. Syntactically
  Focused Knowledge Structures.
• Transition from Traditional to O-O
  Approach.

8
The Learning Process

• Knowledge Structure
• A representation of person's knowledge that
  includes both a set of domain-specific concepts
  and relations among those concepts.
• Concept Knowledge
• Actual ideas and information embodied in the
  knowledge of events or objects designated by a
  label.

9
The Learning Process

• Incremental Learning
• Knowledge structure most closely matching the
  concept is activated.
• As knowledge increases, a person revises existing
  knowledge structures to organize knowledge.
• Example Java learner, new JComboBox concept

10
The Learning Process

• Mindshift Learning
• Learner activates existing knowledge structure
  that is only partially appropriate.
• Proactive Interference
• Knowledge that is inappropriate in new domain.
• Existing knowledge interferes with learning
  process.
• Makes it difficult to understand concept within
  concept of the new mindset.

11
Cognitive Processing

• Novel
• Encounters situation unfamiliar or unknown.
• Promotions or career changes trigger active
  thinking.
• Discrepant
• Unexpected failure or disruption between
  expectations and reality.
• Difference in performance review between manager
  and employee.

12
Cognitive Processing

• Deliberate Initiative
• Response when asked to think or while being
  explicitly questioned.
• Career planning triggers active thinking
• Reflect on goals, resources, opportunities

13
Base Theory
14
Refining the Theory

• Determine concepts fundamental to O-O
• Concept's origin is important to understanding it
• Some are borrowed from Traditional (attribute)
• Some are new to O-O (inheritance)
• Some are contradictory (encapsulation)

15
Inductive Approach to Refine Theory

• 3 OO experts asked to assist with learning
  portion.
• Sort the 9 OO concepts into 3 categories.
• Each expert experienced difficulty with sorting.
• Each understood novel category.
• Failed sorting led to new categories
• Novel, Changed, and Carryover.

16
New Categories

• NovelSame as before.
• Changed
• Any concept that had an existing meaning in
  traditional development, but new meaning in the
  O-O development context.
• Carryover
• Concepts originally defined in traditional and
  hold the same meaning in O-O development.

17
Revised Theory
18
Hypotheses Development

• Novel Concepts (high novelty)
• Integrate directly into OO concept knowledge.
• Traditional knowledge will have no influence.
• Changed Concepts (low novelty)
• Proactive interference may occur.
• Existing traditional knowledge will negatively
  influence learner's knowledge of concept.
• Carried Over Concepts (somewhere in between)
• Positive influence on learner's knowledge.

19
Hypotheses Development
High
Carryover
O-O Concept Knowledge
Novel
Changed
Low (0)
High (100)
Degree of Novelty
20
Hypotheses

• H1 A developer's OO concept knowledge score will
  have a U-shaped relationship with the degree of
  perceived novelty.
• H2 A developer's carryover concept knowledge
  score will be greater than his/her changed
  concept knowledge score.
• H3 A developer's carryover concept knowledge
  score will be greater than his/her novel concept
  knowledge score.
• H4 A developer's novel concept knowledge score
  will be greater than his/her changed concept
  knowledge score.

21
Method

• Survey sent via mail or e-mail to organizations.
• Criteria Both traditional and OO development.
• Sample
• 81 software developers from 16 companies and
  various industries.
• 39 Response Rate.

22
Sample Table
23
Instrument Development

• First Section
• 9 items measuring degree of perceived novelty.
• Second Section
• 27 items measuring OO concept knowledge.
• Third Section
• 9 items measuring level of perceived learning
  difficulty for each concept.
• Validation

24
Hypothesis Testing- H1
25
Hypothesis Testing- H1

• H1 is supported. OO concept knowledge score does
  have a U-shaped relationship with the degree of
  perceived novelty.

26
Data Prep for Testing H2-H4

• Necessary to categorize the concepts by subject.
• Based on a degree of perceived novelty
• Carryover 0 24
• Changed 25 75
• Novel 76 100

27
Hypothesis Testing H2-H4

• H2 Supported.
• H3 Not Supported.
• H4 Supported.

28
Looking Deeper into H3

• Why no significance between Carryover and Novel?
• Study believes that novel concept knowledge
  improved over time making it equal to his/her
  knowledge of carryover concepts.
• Looked at years of OO experience for answers.
• Very small, but significant correlation found.

29
Years of 00 Experience
30
Do Scores Improve Over Time?
Knowledge improves over time in novel and
carryover. Problems caused by proactive
interference tend to persist over time.
31
Experience vs. Perceived Novelty
32
Concept's Difficulty
33
Limitations/Future Research

• Wasn't possible to directly test the impact of
  experience over time on OO concept knowledge.
• Test an individual's knowledge at various points
  in time.
• Theory has the potential to generalize to other
  mindshift learning situations.
• Principles identified in the MLT may improve
  understanding.

34
Conclusion

• Previous research stated mindshift learning was
  more difficult than incremental learning, but
  didn't answer why.
• Our findings indicate lower scores in areas
  perceived as changed.
• We can use results for organizations to change
  the training process for individuals.

35

• Questions??