Quantitative Methodologies using Multi-Methods: Models for Social Science and Information Technology Research

✍ Scribed by Sergey V. Samoilenko and Kweku-Muata Osei-Bryson

• Publisher: Routledge
• Year: 2021
• Tongue: English
• Leaves: 311
• Category: Library

✦ Table of Contents

Cover
Half Title
Title Page
Copyright Page
Contents
Preface: Possible Uses of this Book
Introduction
SECTION I: Development of the Methodological Modules
Chapter 1: Pre-Requisite General Questions
Impact of the Assumption of Homogeneity of the Sample on Research Questions
From a Basket of Apples to a Set of Systems (Decision-Making Units)
From Systems to Systems in Context
Chapter 2: Components of Multi-Method Methodologies
Cluster Analysis (CA)
Classification Decision Trees Induction (CDTI)
Neural Networks (NNs)
Association Rules Mining (ARM)
Data Envelopment Analysis (DEA)
Multiple Regression (MR)
Chapter 3: Framework for Methodological Modules
SECTION II: Description of the Methodological Modules
Chapter 4: A1: Homogeneous Sample – DEA and DTI
Phase 1: DEA
Phase 2: DTI
Examples of Application of DEA and DTI
Chapter 5: A2: Homogeneous Sample – DEA and ARM
Phase 1: DEA
Phase 2: ARM
Examples of Application of DEA and ARM
Chapter 6: B1: Heterogeneous Sample (Groupings Are Given) – DTI and ARM
Phase 1: DTI
Phase 2: ARM
Examples of Application of DTI and ARM
Chapter 7: B2: Heterogeneous Sample (Groupings Are Given) – DTI and MR
Phase 1: DTI
Option 1: DTI Using the Data Set Comprised of a Causal Model Only
Option 2: DTI Using the Data Set without Causal Model
Option 3: DTI Using the Complete Data Set
Phase 2: MR
Option 1: MR Using the Causal Model Only
Option 2: MR Using the Adapted Causal Model – Contextual Independent Variable
Option 3: Creating a New MR Using Contextual Independent Variables
Example of Application of DTI and MR
Chapter 8: B3: Heterogeneous Sample (Groupings Are Given) – DTI, DEA, and ARM
Phase 1: DTI
Option 1: The Data Set Is Comprised of the Variables of the DEAea MODel
Option 2: The Data Set Contains Contextual Variables
Phase 2: DEA
Phase 3: ARM
Option 1: ARM to Generate “If→ (Level of the Top-Split Variable(s))”
Option 2: ARM to Generate “If→ (DEA Model’s Inputs)”
Option 3: ARM to Generate “If→ (DEA Model’ s Outputs)”
Option 4: ARM to Generate “If→ (Level of Averaged Relative Efficiency)”
Option 5: ARM to Generate “If→ (Received Categorization)”
Examples of Application of DTI, DEA, and ARM
Chapter 9: B4: Heterogeneous Sample (Groupings Are Given) – DTI, DEA, and NN
Phase 1: DTI
Phase 2: DEA
Phase 3: NN
Step 1: Generate NN Model of Transformative Capacity
Step 2: Generate Outputs of a Less Efficient Group Based on Transformative Capacity of a More Efficient Group
Step 3: Generate Outputs of a More Efficient Group Based on Transformative Capacity of a Less Efficient Group
Step 4: Compile the Generated Outputs in a New Data Set
Phase 4: DEA
Example of Application of DTI, DEA, and NN
Chapter 10: C1: Heterogeneous Sample (Groupings Are Not Known) – CA and DTI
Phase 1: CA
Phase 2: DTI
Examples of Application of CA and DTI
Chapter 11: C2: Heterogeneous Sample (Groupings Are Not Known) – CA and ARM
Phase 1: CA
Phase 2: ARM
Option 1: ARM Using Only Intrinsic Variables
Option 2: ARM Using Only Contextual Variables
Option 3: ARM Using Intrinsic and Contextual Variables
Examples of Application of CA and ARM
Chapter 12: C3: Heterogeneous Sample (Groupings Are Not Known) – CA, DTI, and MR
Phase 1: CA
Phase 2: DTI
Option 1: Data Set Is Limited to Variables of the MR Model
Option 2: Data Set Comprises Variables of the MR Model and Contextual Variables
Phase 3: MR
Example of Application of CA, DTI, and MR
Chapter 13: C4: Heterogeneous Sample (Groupings Are Not Known) – CA, DTI, and ARM
Phase 1: CA
Phase 2: DTI
Option 1: A Priori Target Variable
Option 2: CA-based Target Variable
Phase 3: ARM
Step 1
Step 2
Step 3
Step 4
Examples of Application of CA, DTI, and ARM
Chapter 14: C5: Heterogeneous Sample (Groupings Are Not Known) – CA and DEA
Phase 1: CA
Option 1: CA based on the DEA Model
Option 2: CA based on the DEA Model and Contextual Variables
Phase 2: DEA
Examples of Application of CA and DEA
Chapter 15: C6: Heterogeneous Sample (Groupings Are Not Known) – CA, DEA, and ARM
Phase 1: CA
Phase 2: DEA
Phase 3: ARM
Option 1: Complete Sample, # of Variables = the DEA Model
Option 2: Complete Sample, # of Variables = the DEA Model + Contextual Variables
Option 3: Sub-Sets of the Sample, # of Variables = the DEA Model
Option 4: Sub-sets of the Sample, # of Variables = the DEA Model + Contextual Variables
Examples of Application of CA, DEA, and ARM
Chapter 16: C7: Heterogeneous Sample (Groupings Are Not Known) – CA, DTI, and DEA
Phase 1: CA
Phase 2: DTI
Phase 3: DEA
Examples of Application of CA, DTI, and DEA
Chapter 17: C8: Heterogeneous Sample (Groupings Not Known) – CA, DTI, DEA, and NN
Phase 1: CA
Phase 2: DTI
Phase 3: DEA
Phase 4: NN
Step 1: Creating an NN Model of “Low-Level” Cluster
Step 2: Creating an NN Model of “High-Level” Cluster
Step 3: Simulation of the Outputs of “Low-Level” Cluster Using NN Model of “High-Level” Cluster
Step 4: Simulation of the Outputs of “High-Level” Cluster Using NN Model of “Low-Level” Cluster
Phase 5: DEA
Examples of Application of CA, DTI, DEA, and NN
SECTION III: Methodological Modules – Examples of Their Application
Chapter 18: A Hybrid DEA/DM-based DSS for Productivity-Driven Environments
Introduction
Description of the DSS
Externally Oriented Functionality
Internally Oriented Functionality
Architecture of the DSS
An Illustrative Application
Step 1: Is the Business Environment Homogeneous?
Step 2: What Are the Factors Responsible for Heterogeneity of the Business Environment?
Step 3: Do Groups of Competitors Differ in Terms of the Relative Efficiency?
Step 4: What Are some of the Factors Associated with the Differences in Relative Efficiency?
Step 5: Are There any Complementarities Between the Relevant Variables?
Step 6: What Is a Better Way to Improve Production of Outputs?
Conclusion
Acknowledgment
References
Chapter 19: Determining Sources of Relative Inefficiency in Heterogeneous Samples: Methodology Using Cluster Analysis, DEA, and Neural Networks
Introduction
Description of the Methodology
Description of Steps 3–5 of the Methodology
Step 3: Generate a “Black Box” Model of Transformative Capacity of Each Cluster
Step 4: Generate Simulated Sets of the Outputs for Each Cluster
Step 5: Determine the Sources of the Relative Inefficiency of the DMUs in the Sample
Motivation for Steps 3 and 5 of the Methodology
Motivation for Step 3
Motivation for Step 5
Illustrative Example
Description of the Illustrative Data Set
Application of the Methodology on the Illustrative Data Set
Results of Step 1: Evaluate the Scale Heterogeneity Status of the Data Set
Results of Step 2: Determine the Relative Efficiency Status of Each DMU
Results of Steps 3 and 4: Generate Simulated Sets of the Outputs for Each Cluster Based on Black Box Models Transformative Capacity Processes
Results of Step 5
Discussion and Conclusion
Acknowledgment
References
Chapter 20: Exploring Context Specific Micro-Economic Impacts of ICT Capabilities
Introduction
Theoretical Framework and the Research Model
The Methodology of the Study
Phase 1: Application of Data Envelopment Analysis (DEA)
Phase 1, Step 1
Phase 1, Step 2
Phase 1, Step 3
Phase 2: Decision Tree-Based Analysis
Phase 2, Step 1
Phase 2, Step 2
Description of the Data
Results of the Data Analysis
Results from Phase 1: Application of Data Envelopment Analysis (DEA)
Phase 1, Step 1
Phase 1, Step 2
Phase 1, Step 3
Results from Phase 2 – Decision Tree (DT) Based Analysis
Conclusion
Contributions to Theory
Contributions to Practice
Acknowledgment
References
Chapter 21: A Methodology for Identifying Sources of Disparities in the Socio-Economic Outcomes of ICT Capabilities in SSAs
Introduction
Research Framework
Proposed Methodology
A New Methodology: Benefits and Justifications
Phase 1: Data Envelopment Analysis (DEA)
Phase 2: Decision Tree Induction (DTI)
Phase 3: Association Rule Mining (ARM)
Research Questions and Null Hypotheses of the Study
The Data
Results of the Data Analysis
Phase 1: Data Envelopment Analysis
Phase 2: Decision Tree Induction
Phase 3: Association Rule Mining
Discussion of the Results
Conclusion
Acknowledgment
References
Chapter 22: Discovering Common Causal Structures that Describe Context-Diverse Heterogeneous Groups
Introduction
A Conceptualization of the Benchmarking Problem
Research Problem and Research Questions of the Study
The Proposed Methodology
Description of the Methodology
Justification & Benefits of the Methodology
Illustrative Example – Application to Sub-Saharan Economies
Phase 1: Define the Transformation Framework
Phase 2: Partition the Set of Decision Making Units into Meaningful Groups
Phase 3: Data Envelopment Analysis
Phase 4: Decision Tree Induction (DTI)
Phase 5: Association Rule Mining
Conclusion
Acknowledgment
References
Chapter 23: An Empirical Investigation of ICT Capabilities and the Cost of Business Start-up Procedures in Sub-Saharan African Economies
The Research Framework and Research Questions
Proposed Methodology
Phase 1: Cluster Analysis (CA)
Phase 2: Decision Tree Induction
Phase 3: Data Envelopment Analysis
Phase 4: Ordinary Least Squares Regression
Phase 5: Association Rule Mining
Data
Results of the Data Analysis
Phase 1: CA
Phase 2: DTI
Phase 3: DEA
Phase 4: OLS
Phase 5: ARM
Interpretation of the Results of the Data Analysis
Cluster Analysis
Decision Tree Induction
DEA
Ordinary Least Squares (OLS)
ARM
Discussion of the Results of the Study
Conclusion
Acknowledgment
References
Chapter 24: Exploring the Socio-Economic Impacts of ICT-Enabled Public Value in Sub-Saharan Africa
Introduction
Research Framework of the Study
Research Questions of the Study
Methodology of the Investigation
Phase 1: Cluster Analysis
Phase 2: Decision Tree Induction
Phase 3: Data Envelopment Analysis (DEA)
Phase 4: Ordinary Least Squares (OLS) Regression
Phase 5: Association Rule Mining (ARM)
Data
Results of the Data Analysis
Phase 1: CA
Phase 2: DTI
Phase 3: DEA
Phase 4: OLS
Phase 5: ARM
Discussion of the Results of the Study
Conclusion
Acknowledgment
References
Chapter 25: Contributing Factors to Information Technology Investment Utilization in Transition Economies: An Empirical Investigation
Introduction
Theoretical Framework
Growth Accounting
Theory of Complementarity
Overview on the Data
Methodology: Searching for the Determinants of the Efficiency of Utilization of Investments in Telecoms
Phase 1: Data Envelopment Analysis
Data Used to Perform DEA
Phase 2: Cluster Analysis
Data Used to Perform CA
Phase 3: Decision Tree
Data Used to Perform DT
Results
Results: DEA
Results: Cluster Analysis
Results: Decision Tree
Contribution of the Study
Summary and Conclusion
Acknowledgment
References
Appendix A
Chapter 26: Increasing the Discriminatory Power of DEA in the Presence of the Sample Heterogeneity with Cluster Analysis and Decision Trees
Introduction
The Proposed Methodology
Overview of Data Set of Illustrative Example
Description of the Methodology
Step 1: Determine the Structural Homogeneity Status of the Data Set
Step 2: Determine the Relative Efficiency Status of DMUs
Step 3: Describe the Relative Efficiency Categories
Conclusion
Acknowledgment
References
Chapter 27: An Exploration of the Intrinsic Negative Socio-Economic Implications of ICT Interventions
Introduction
Socio-Economic Impact of ICT
Tools, Machines, and ICT
Routes of Elimination and Substitution
Conditions for Elimination and Substitution
Pragmatics and Ethics of Implementation
Dimensions of Social Impact of ICT
Platform, Message, and Target
Competing with Others: Additional Implications
Competing with Others: Social Implications
Impact of Collaboration
Investigating Negative Implications of ICT: What Is the Plan?
Conclusion
References
SECTION IV: Appendix X
The Purpose and the Suggested Use of the Content in this Appendix
Appendix X1: Models of Economic Growth
References
Appendix X2: A Model of the Socio-Economic Impact of ICT
References
Index