Web & Social Media Analytics Previous Year Question Paper.pdf
PhD Thesis Igor Barahona July 26th of 2013
1. The level of adoption
of analytical tools.
Igor BARAHONA
Barcelona, Spain. July 26th of 2013
2. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
Contents.
2
3. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
3
4. Processor
Dhrystone
MIPS
Cost Year
UNIVAC I
0.002 MIPS at
2.25 MHz
$11.500.000.
USD
1951UNIVAC I
Computers performance.
1951
IBM
System/370 mod
el 158-3
IBM
System/370 mo
del 158-3
1 MIPS at
8.69 MHz
$2,248,550.
USD
1971 1971
INTEL 286 Intel 286
2.66 MIPS at
12.5 MHz
$3,500. USD 1982 1982
Intel
Pentium III Intel Pentium III
2,054 MIPS at
600 MHz
$2,000. USD 1999 1999
Intel Core i7
2600K
128,300 MIPS
at 3.4 GHz
$1,000. USD 2011 2011
Intel
Core i7
2600K
5. If computers are more powerful...........
Lots of data but not information
Powerful computers but
unstructured problems
Difficulties of getting fast and
accurate information.
Make sense of “data tsunami”
that is hitting modern industries
5
What does it
imply?
Burby & Atchison (2007)
Kaushilk (2011)
6. The business environment...................
More complexity requires analyzing
real-time-data and for making
better decisions.
Reduction on differentiation
points due to the globalization of
markets.
Customers better informed with
more alternatives.
Data has to be converted into
“Information” that triggers
managerial action.
6
How is it
changing?
Stubbs (2011)
McDonough (2009)
7. Extensive utilization of
data, information and
quantitative models.
Understand past / present
performance
Reduce uncertainty
Predict future results
Making better decisions based on
quantitative evidence
OUTPUTS
ADDED VALUE
INPUTS
7
It can be defined in terms of inputs and outputs
What is an analytical tool?
Davenport & Harris (2007)
8. • Using analytics
– Finding the best customers, and charging
them the right price
– Minimizing inventory in supply chains
– Allocating costs accurately and
understanding how financial performance is
driven.
Using analytical tools is good.......
8
But It is better competing with them...
• Competing with analytics.
– Making analytics and fact-
based decisions a key
element of strategy and
competition
Davenport, Harris & Morrison (2010)
9. Thesis Objectives
1. Propose a theoretical scale to measure the level of adoption of
analytical tools in companies.
2. Design a reliable and valid instrument to collect data from a
sample of companies located in Barcelona, Spain.
3. Analyze data collected from the surveyed companies, in order to
draw conclusions about the level of adoption of analytical tools in
Barcelona by applying the Statistical Engineering approach.
4. Rank the sampled companies in the scale by applying the
Evidential Reasoning approach.
5. Conduct in-depth interviews with managers, consultants and
academics with the purpose of finding out soft and unstructured
aspects about the level of adoption of analytical tools in
Barcelona by applying the Laddering Methodology.
6. Based on results generated, provide practical guidelines to
stakeholders who are interested in expanding the use of analytical
tools in companies and creating competitive advantages.
The level of
adoption of
analytical
tools.
9
10. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
10
11. Systemic
thinking
Management
support
Removing obstacles
Human, technical and
financial resources.
.
Encouraging staff
involved in the project
Emergency
Hierarchy
Communication
Control
1
4 key-drivers for LAAT expansion.
2
Deming, (2000)
Deming, (2000)
Hahn et al (2000)
Yeo (1993)
Tort-Martorell et al
(2011)
Hoerl & Snee (2010)
Checkland (1999)
11
12. DB. Competitive
advantage
Communication
outside the
company.
Lower price / cost
Market niche
Differentiation
Privileged location.
Customer Relationship Managers (CRM)
Trust
Long term relationships
3
4
Supply chain Managers (SCM)
4 key-drivers for LAAT expansion.
Langfield-Smith &
Greenwood (1998)
Davenport, Harris &
Morrison (2010)
Porter (1990)
Poon &Wagner
(2001)
Blanchard (2010)
12
13. 13
Theoretical model and 4 key drivers.
LAAT
MANAGEMENT
SUPPORT
COMMUNICATION
OUTSIDE
SYSTEMIC
THINKING
DB. COMPETITIVE
ADVANTAJE
13
14. 1. We proposed a 5 level
scale
2. At level 1 we find
companies that do not use
any analytical tool.
3. At level 5 we find
companies that use
analytical tools as a
strategic support for their
competitive advantage.
4. At levels 2, 3 and 4 we find
companies that are
improving on using
analytical tools
Five level scale
14
15. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
15
16. Analytical tools on finance.
S – Strength
W – Weaknesses
O – Opportunities
T – Threats.
Score Cards
Financial benchmarking
Predictive analytics
applied to Ratios analysis,
Balance sheets and
income statements .
Janis (2008) Xu DL (2012) Morris et al. (2002)
16
17. Analytical tools on manufacturing
Design of experiments
(DOE)
Six Sigma.
Statistical Process Control.
(SPC)
The seven management
tools
Surface response
Hoerl et al (1993) Ishikawa (1988)
Futami (1986) Deming (2000)
17
18. Analytical tools on R&D
Clinical trials
Control groups
Survival analysis
Stochastic process
Multivariate analysis
Davenport, Harris &
Morison (2010)
Liu et al (2008)
18
19. Analytical tools on Human Resources
Multivariate regression
Assess intangible assets
MCDA methods
Correspondence analysis
Decisions trees
Harris, Craig & Egan (2009)
Lewis (2003)
Armstrong (2012)
19
20. Analytical tools on Marketing
Time series
General linear models
MCDA methods
Multivariate analysis
Survey research methods
Customer relationships
management
Armstrong (2012)
Deming (2002)
Burby & Atchison (2007)
20
21. Analytical tools with suppliers
Decisions trees
MCDA methods
Multivariate analysis
Supply chain management
Petroni & Braglia (2000)
Verma & Pullman (1998)
Nydick & Hill (1992)
21
22. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
22
23. Questionnaire design.
• A 7-step methodology was
adapted to design and
validate the scale.
• In the first part the issue is
to provide valid and
reliable items
• The second part is focused
on the validity and
reliability of the scales
23
Menor & Roth (2007)
24. Theoretical domain (1/7)
• Four theoretical constructs
were investigated
• A total of 17 items were
derived from the
theoretical constructs
• Each item was associated
to a five level Likert scale.
24
Bryman (2012)
Menor & Roth (2007)
Michie et al (2005)
25. Item generation (2/7)
The “degree of
understanding” was
calculated in order to
ensure each item is
understandable and easy
to read. (Kappa index for
multiple raters)
ITEM Judge1 Judge2 Judge3 Judge4 Judge5 Judge6 Judge7 Judge8
DB-CA1 4 4 4 4 4 4 4 4
DB-CA2 4 4 4 4 4 4 4 4
DB-CA3 4 5 4 5 4 5 4 5
DB-CA4 5 5 4 4 4 4 4 4
DB-CA5 4 4 4 4 4 4 4 4
MS-DA1 4 4 4 4 4 4 4 4
MS-DA2 4 4 4 4 4 4 4 4
MS-DA3 4 4 4 5 4 5 4 4
MS-DA4 4 4 4 4 4 4 4 4
MS-DA5 4 4 4 4 4 4 4 4
MS-DA6 5 5 5 5 5 5 5 5
SYS1 5 5 5 5 5 5 5 5
SYS2 5 5 5 5 5 5 5 5
SYS3 4 5 5 5 5 5 4 5
SYS4 5 5 5 5 5 5 5 5
SYS5 5 5 5 5 5 5 5 5
COMOUT 5 5 5 5 5 5 5 5
A total of four constructs
were operationalized in 17
items. All the questions were
designed in a Likert scale
from 1 to 5
Grade Kappa
Standard
Error
z Prob>Z
4 0.77980 0.045835 170132 <.0001
5 0.77980 0.045835 170132 <.0001
Overall 0.77980 0.045835 170132 <.0001
Item refining (3/7)
25
Good level of
understanding.
Fleiss (1971) Cohen (1960)
26. Questionnaire development (4/7)
Section
Number
of items
Categorical questions 3
Data Based Competitive Advantage 5
Management Support Data Analysis 6
Systemic Thinking 5
Communication outside the company 1
Total 20
Structure of the first draft
of questionnaire
1
Two steps on the pilot test
It was shared in social
networks
2 It was sent to 300
companies members of
the UPC-Alumni
31 responses
were obtained
Improving the
order of questions
Reviewing
features of the
cover letter
Final writing of
questions
26
28. Survey data collection (5/7)
602.161
41.152
86.094,00
474.915
Total Indústria Construcció Serveis
6,064 companies were invited by
sending it electronically
255 responses were obtained.
Analytics diagnostic free of charge.
Open to share results.
28
IDESCAT (2013)
30. Reliability (6/7)
Cronbach Alpha
ITEM Response
We apply analytical tools in all decisions we
make
strongly agree
completely agreeWe exploited and analyzed plenty of data
during the last year
The use of statistics is useless to build
competitive advantages in our company
completely agree
Alphas are
helpful to
identify these
type of
incoherence
Formulation and outputs
K= Items of the section
Si= standard deviation of the item
St= standard deviation of the section
Subsection Alpha
Data Based Competitive Advantage. (DB-
CA)
0.8884
Management Support in Data Analysis.
(MS-DA)
0.8025
Systemic Thinking (SYS) 0.7761
Communication Outside the Company
(COM-OUT)
1.0000
30
Cronbach (1951) Streiner (2003)
31. Reliability (6/7)
Interclass correlation coefficient
row /company-effect
column/ item-effect
Source of variation
Sum of
Sq
D.F
Mean of
Sq
F-
Value
Pr(>F)
Between Companies (row-effect) 1734.138 153 11.334
Within
Companies
(item-effect)
Within
Companies
817.168 15 54.478 55.768 .000
Residuals 2241.894 2295 .977
Total 3059.063 2310 1.324
Formulation and outputs
Intra- class Correlation Coefficient (ICC)
Two-way
Random Effect
Model
ICC
95.00% C.I
Lower Upper
Average
Measure
(Within effect)
.887 .851 .915
Shrout & Fleiss (1979) Tian (2005)31
32. Item and scale refinement (7/7)
Explanatory factor analysis (EFA)
Quetionnaire ITEM Factor1 Factor2 Factor3 Factor4
Understanding benetifs DB_CA1 0.757
Product Improvement DB_CA2 0.756
Statistics Support DB_CA3 0.831
Statistics Importance DB_CA4 0.806
Statistics Encouragement DB_CA5 0.659
Staticstics Training MS_DA1 0.826
New knowledge implementation MS_DA2 0.723
Data collection process MS_DA3 0.527
Budget for projects MS_DA4 0.837
Technological resources MS_DA5 0.622
Competitor's Investigation MS_DA6 0.561
Efforts recognition SYS1 0.595
Mission understanding SYS2 0.693
Communication openness SYS3 0.571
Teamwork culture SYS4 0.764
Reinforcement on data usage SYS5 0.534
Communication suppliers/customers COM_OUT 0.852
DB-CA. Data-Based
Competitive
Advantage
MS-DA. Management
Support on Data
Analysis
SYS. Systemic Vision
of the business
COM-OUT.
Communication
Outside company.
(clients and suppliers)
In order to validate our questionnaire, the 17 items were clustered on the first 4 factors
using the loadings as classification criteria
32
Krzanowski (2000) Long (1983) Kaiser (1958)
33. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
33
34. Statistical Engineering case of study
Don't focus on the introduction of new theories but rather
how they might be best utilized for practical benefit
Strategic:
Statistical
Thinking.
Tactical:
Statistical
Engineering.
Operational:
Statistical Methods
and Tools.
Statistical
Theory.
Statistical
Practice.
How to utilize the principles and techniques of statistical
science for benefit of humankind.
How to best utilize statistical concepts, methods, and tools
and integrate them with information technology and other
relevant sciences to generate improved results.
Systemic thinking
Variance reduction
Holistic approach
Statistical Methods
Specialized software
Skilled staff
34
Hoerl & Snee (2010) Anderson-Cook et al (2012) Hoerl & Snee (2012)
35. Statistical Engineering case of study
Data collection2
Confirmatory analysis3
Relationship
between companies
4
Relationship between key
drivers
5
Conclusions
Understanding project’s
scope
1
Flowchart
Questionnaire
Survey
Datasheet
Correspondence analysis (CA)
Factor Analysis
Logistic Regression (LR)
Correlation Matrix (CM)
illustrates the relation between statistical thinking
and statistical methods.
35
Seven statistical tools were wisely integrated in a
five step process to accomplish a unique objective.
36. Data collection (2/5)
6,064 companies
were invited by sent
it electronically
255 responses
were obtained.
Confirmatory analysis (3/5)
Quetionnaire ITEM Factor1 Factor2 Factor3 Factor4
Understanding benetifs DB_CA1 0.757
Product Improvement DB_CA2 0.756
Statistics Support DB_CA3 0.831
Statistics Importance DB_CA4 0.806
Statistics Encouragement DB_CA5 0.659
Staticstics Training MS_DA1 0.826
New know ledge implementation MS_DA2 0.723
Data collection process MS_DA3 0.527
Budget for projects MS_DA4 0.837
Technological resources MS_DA5 0.622
Competitor's Investigation MS_DA6 0.561
Efforts recognition SYS1 0.595
Mission understanding SYS2 0.693
Communication openness SYS3 0.571
Teamw ork culture SYS4 0.764
Reinforcement on data usage SYS5 0.534
Communication suppliers/customers COM_OUT 0.852
ITEMS
36
37. 37
The 255 responses were discomposed
and represented at the 2 biggest factors
Correspondence analysis (4/5)
39. 39
Level 1 is close from Micro Size.
Level 4 is close from Middle Size
40. 40
Services Companies are more suitable to be
analytical oriented
Products Companies are more related with level 1
and Micro size
41. 41
Middle size companies are closer to “better and
different” strategies.
There is a group for Micro-size, Products, Level 1
and No Competitive Advantage
41
42. COMMUNICATION
OUTSIDE
COMPANY
DB.
COMPETITIVE
ADVANTAGE
SYSTEMATIC
THINKING
MANAGEMENT
SUPPORT. DA
C.M allows us to understand and quantify relationships
between the AVERAGES of the Key Drivers
0.702
0.648
0.300
Pearson Correlation Coefficients
DBCA MSDA SYS COMOUT
DBCA. Data Based
Competitive Advantage
1.000 0.70243 0.69484 0.05246
MSDA. Management
support data analysis
1.000 0.64852 -0.03397
SYS. Systematic Thinking 1.000 0.30036
COMOUT. Communication
Outside Company
1.000
0.695 These
correlations
were
calculates with
the AVERAGES
of ITEMS.
Correlation Matrix (4/5)
42
Hair, et al (2006)
Krzanowski (2000)
43. To predict if on a set of 255 Spanish companies, either a company has analytics aspirations
or not. (Level=>4)
Level 4 is the starting point of the use of analytical tools as a distinctive competence in the
industry
RESPONSE VARIABLE:
0Y
1Y
If the company does not has analytical aspirations. (Level<4)
If the company has analytical aspirations. (Level>=4)
NOANALYTICAL
ASPIRATIONS. (LEVEL
1 , 2AND 3)
ANALYTICAL
ASPIRATIONS
(LEVEL 4 AND 5)
TOTAL
186 69 255
73% 27% 100%
PREDICTORS
G1 Understanding the benefits of Statistics
G2 Statistics builds the Comp. Adv
G3 There is one mission and vision
G4 Communication with clients and suppliers
The predictors were
taken from the
questionnaire ITEMS
Logistic regression (5/5)
43
Philip and Teachman (1998)
44. )(43210
1
ijkllkji GGGG
P
P
Ln
THE MODEL
have p-values less than 0.05,
indicating that there is
sufficient evidence that the
coefficients are not zero using
an alfa level of 95%
The goodness-of-fit tests,
with p-value equal to 1.000.
Indicate that there is
insufficient evidence to
claim that the model does
not fit the data adequately.
1. UNDERSTANDING THE BENEFITS OF APPLIED STATISTICS BUSINESS.
2. BUILDING A COMPETITIVE ADVANTAGES BY DATA ANALYSIS.
3. ESTABLISHING A MISSION AND VISION STATEMENTS ON THE COMPANY
4. STIMULATING COMMUNICATION OUTSIDE COMPANY.
Coefficients for
these variables
are not cero.
Logistic Regression Table
Odds 95% CI
Predictor Coef SE Coef Z P Ratio Lower Upper
Constant -17.8045 3.13596 -5.68 0.000
DB_CA1 1.65439 0.313537 5.28 0.000 5.23 2.83 9.67
DB_CA3 0.723906 0.271505 2.67 0.008 2.06 1.21 3.51
SYS2 1.12321 0.273354 4.11 0.000 3.07 1.80 5.25
COM_OUT 1.54055 0.382019 4.03 0.000 4.67 2.21 9.87
Goodness-of-Fit Tests
Method Chi-Square DF P
Pearson 105.652 111 0.625
Deviance 72.350 111 0.998
Hosmer-Lemeshow 4.405 8 0.819
44
45. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
45
46. Evidential Reasoning case of study
A generic evidence-based multi-criteria decision analysis (MCDA) approach
for dealing with problems having both quantitative and qualitative criteria
under various uncertainties including ignorance and randomness.
Main documented applications:
• Environmental impact assessment,
• Organizational self-assessment
• Portfolio investments
• Prioritizing voices of customers.
The ER approach is implemented in a software called Intelligent
Decision System.
The Belief Decision Matrix allows us more realistic assessments than
traditional Decisions Matrix.
Accepts data of different formats with various types of uncertainties as
inputs, such as single numerical values, probability distribution, and
subjective judgments with belief degrees.
Main features in
the ER approach
Xu & Yang (2001) Yang & Singh (1994)
46
47. Main ER assumptions
Main
assumptions.
Exclusiveness of grades: A grade Hn is assumed to be mutually exclusive of
another Hn+1
Completeness of criteria: Suppose an overall criterion is assessed through
m sub-criteria. These m sub-criteria are said to be collectively exhaustive
or complete.
Weight as degree of importance: The weight of a sub-criterion yi,
denoted by wi, is a degree of importance of yi in the assessment of the
overall criterion.
Distribution assessment.
Generation of the overall belief
Linear algorithm Non -linear algorithm
Xu & Yang (2002) Yang (2001)
Yang & Singh
(1994)
47
48. Model definition2
Relate father and bottom
attributes.
3
Assigning weights.4
Assigning belief degree.
5
Calculate assessments.
6
Data collection1
Implement the set of rules
Pair-wise comparison
Define uncertainty
Sensitivity test
Overall performance
Compare alternatives
Transform means-values to degree
of belief
Conclusions
Evidential reasoning case of study
A six-step methodology was adapted for this case of
study48
Apply ER algorithm to extract relevant conclusions
about which attributes clearly contribute to the
expansion of LAAT and therefore to reach
competitive advantages.
49. Model definition (2/6)
Model summary Grades
Numer of parent attributes: 4 u(H1):= u (Analytic ignorance) =0.00
u(H2):= u (Local applications) = 0.25
Numer of bottom atributes: 17 u(H3):= u (Analytical aspirations) =0.50
Selected method for relating parent and
bottom attributes: RULE-BASED
APPROACH (Yang 2001)
u(H4):= u (Analytics as a systems) = 0.75
u(H5):= u(Analytics as competitive advantage) = 1.00
49
50. Relate father and bottom attributes. (3/6)
Relating attributes is defined as how the assigned
grades are converted to the ones of their parents
If MSDA is worst =0.00
Then Overall Performance is Analytical
Ignorance=100%
If MSDA is poor=0.25
Then Overall Performance is Local
Focus=100%
If MSDA is average=0.50
Then Overall Performance is Analytical
aspirations=100%
If MSDA is good=0.75
Then Overall Performance is Analytics
as System=100%
If MSDA is excellent=1.00
Then Overall Performance is Analytics
as Comp. Advantage=100%
In similar way, the attributes
• SYS
• COM-OUT
• DBCA
were related to their parent attributes50
Yang (2001)
51. Assigning weights . (4/6)
0.22
0.18
0.21 0.21
0.17
0.14 0.15 0.16
0.12
0.16
0.28
0.16
0.20 0.18
0.23 0.22
1.00
Data-Based competitive
advantage
Management support on data
analysis
Systemic Thinking Communication outside the
company
MS-DA6 was the most important. It refers to whether the top
management promotes the use of data to evaluate how the
competitors are evolving
SYS4 refers to whether there is a teamwork culture in the
company.
51
52. Assigning belief degree. (5/6)
The degree of belief represents the extent to which an answer is believed to be true.
The following expression was utilized for
assigning belief degrees
For SYS4 µ= 3.80
The belief
structure is
{(“Worst” with β=0.00),
(“Poor” with β =0.00),
(“Average” with β =0.20),
(“Good” with β =0.80),
(“Best” with β =0.00)}.
In this way the mean-value was transformed into 5
values, which describe the scenario more accurately
The transformation was applied to the 17 attributes
of the model
52
53. Calculate assessments. (6/6)
The overall performance
Middle companies are slightly more
analytical oriented than big
This result is coherent with CA of the
slide 42.
In the CA, middle companies
are closest to the Level 5.
Distance among Big and
Middles is also small.
53
55. Sensitivity test
Calculate assessments. (6/6)
Micro Company Small Company
Middle company Big Company
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Averagescore
Average scores for the Overall PerformanceAverage scores for the Overall Performance
Weight of Systematic Thinking
Givenweight
Micro Company Small Company
Middle company Big Company
0%
20%
40%
60%
80%
100%
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Averagescore
Average scores for the Overall PerformanceAverage scores for the Overall Performance
Weight of Communication Outside the company
GivenweightSensitivity to changes in the weight of
SYS
For lower weights on SYS sensitivity
increases
For higher weights, the sensitivity
decreases
Sensitivity to changes in the weight of COM-
OUT
For higher weights, the sensitivity
increases
55
For lower weights on COM-OUT, the
sensitivity decreases
56. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
56
57. 57
How customers translate the attributes of products
into meaningful associations with respect to
self, following Means-End Theory
Understand how customers underlying personal
motivations with respect to any given product or
service
Higher level distinctions provides a perspective of
how the attributes are processed from a motivational
perspective
Investigate about the connections between the
attributes and personal motivations.
Reynolds & Gutman 1988, Gutman 1982
The Laddering technique
57
58. How ladders are built?
ATTRIBUTES
CONSEQUENCES
VALUES
Reynolds & Gutman 1988; Gutman 1982
The Laddering technique
Full-bodied taste/
less alcohol
Avoid getting drunk
(wasted) / Socialize
Sense of belonging /
Responsibility to family
59. Respondent Industry Length
1 Project Manager Consultancy Services 50 min
2 CEO Entertainment Services 55 min
3 Head of department Government 1:15 hr
4 Owner / Founder Marketing Research 56 min
5 CEO
Plastic packages
manufacturer
45 min
6 Professor / Researcher Academy 36 min
7 Professor / Researcher Academy 35 min
8 Business consultant Consultancy Services 50 min
9 Analytics consultant Consultancy Services 48 min
10 Professor / Researcher Academy 52 min
The persons who responded the interview.
The designed script was used
in all interviews
The interviews took place on
respondent ’s office
There are digital records for
each one interview
The script’s structure follows
the 5 key drivers
The Laddering technique
59
60. The Laddering technique
60
One example of ladder
ATTRIBUTES
CONSEQUENCES
VALUES
A Improve the knowledge of data
A Goal setting
Ladder taken from the
interview with a CEO of
Packaging
Manufacturer
C Lower cost
C Continuous learning
60
V Serving the society
V Add value to stakeholders
61. Implication Matrix
Data is accessible and
supports decisions (1)
Improve D. Analysis
(21) 17 times
Improve D. Analysis
(21) add value to
stakeholders (29)
18 times
62. Hierarchy Value Map (HMV)
HVM is a way to graphically represent the most
dominant connections. It is a representation of the
linkages across levels of abstraction, starting with
attributes and finishing with values
It should include ladders with 4
or more direct relations. (A total
of 84 in this case)
The main purpose is to highlight
meaningful connections between
(A)-(C)-(V)
Obtained by the cumulative
frequency of direct relations.
Reynolds & Gutman 1988; Gutman 1982
62
63. (1)
Data is
accessible
and supports
decisions
(2)
Data
online
(5)
High
skilled
staff
(6)
Enough
support
(7)
High
tech
(4)
standar
dized
proced
ures
(12)
the most
efficient
structure
(15)
innovate
products
and
services
(8)
Commun
ication
with C&S
(10)
informati
on
outside
(11)
Market
research
(9)
Creativit
y to new
ideas
(14)
Respond
more
quickly
(13)
Flexibility
(3)
Goal
Setting
17 7 13115 6 3 58 7 46 563
(21)
Improve
data
analysis
(28)
staff
efficiency
and
motivation
(16)
Analyze
data from
market
(24)
Knowledge
of data
(19)
Exceeding
customer
exp
(20)
Good
image of
the
company
10
8
(29)
add value
to stake
holders
(30)
Being a
leader
(31)
Communic
ation and
trust
(33)
Passion, Qu
ality and
Excellence
(25)
Long term
relationships
7
8
11
14
12
(22)
improvin
g process
(23)
Improvin
g results
4 7
(27)
More
money (17)
Continuous
learning
(18)
Distinctive
competence
(26)
Lower
cost 4
65
14
11
4
4
6
13
12
(34)
serving
the
society
(32)
honesty
and
credibility
6 7 9 6
14
10
18
8
Hierarchy Value Map (HMV)
63
64. Summary table
The 10 attributes which have the biggest number of relations.
They concentrate the 80% of the total relations .
This table allow us to identify the attributes which have the biggest
impact on values.
64
65. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
65
66. LEVEL 1
A small interest on using
analytical tools shows up.
LEVEL
2
Small and local success brings the
attention of the Senior
Management.
Final point. Analytical
initiatives did not reach
expectations from the senior
management.
Having the leadership in the
market though the use of
analytical tools.
“Prove-it”
Strategy
LEVEL 3
LEVEL 4
LEVEL 5
The first attempt to broad an
analytical project. Define an
analytical vision “Plan your work” .
Embedding the strategic and
critical process with the Analytical
Vision. “Work your plan”
Analytical
Ignorance
Analytical
Focus
Analytical
Aspirations
Analytical
Engineering
Analytics as
competitive
advantage
Road Map for upgrading the scale
Keep working on:
Diagnostic Actions Diagnostic.
Until the highest level in the scale is
reached.
66
68. 1. Introduction.
2. The level of adoption of analytical tools. A theoretical
perspective.
3. The analytical tools in different areas of the company.
4. The questionnaire design.
5. The cases of study.
• Statistical Engineering.
• Evidential Reasoning.
• The laddering technique.
6. Conclusions and guidelines to stakeholders.
7. Further lines of research.
68
69. Scale aggregation
Overall assessment. Level of
Adoption of Analytical Tools.
(LAAT)
Questionnaire
Data-Based
competitive
advantages
In-depth
interviews
Operative
attributes
Systematic
thinking
Management
support
Communication
outside
Tactical
features
Organizational
values
17 attributes on a five-level scale 33 concepts on a three-level scale
How aggregate them
while losing or
distorting information
is prevented?
69
70. Scale aggregation
The research presented on:
Yang, J. B., Xu, D. L., Xie, X., & Maddulapalli, A. K. (2011). Multicriteria
evidential reasoning decision modelling and analysis-prioritizing voices of
customer.Journal of the Operational Research Society, 62(9), 1638-1654.
Will be adapted to investigate this aggregation and
accomplish the following objectives:
• Investigate the scales from questionnaires and in-depth
interviews in order to aggregate them into a unique framework.
• Apply the evidential reasoning approach for calculating the
overall performance of the level of adoption of analytical tools.
• Offer relevant guidelines to organizations that are interesting
in improving their analytical capabilities.
70
71. Publications
Conference papers
Submitted for publication
Barahona, I., & Riba, A. (2012). Applied Statistics on Business at Spain: A Case of
Statistical Engineering. In ASA (Ed.). In 2012 Joint Statistical Meetings. Vol. Book of
abstracts, pp. p 246). San Diego CA:(ASA).
Barahona Igor, & Alex, R. (2013). The level of adoption of analytical tools in
Barcelona, Spain. In JIPI (Ed.). In Jornada d'Investigadors Predoctorals
Interdisciplinària[February 7th of 2013]. Vol. Book of abstracts, pp. Page 7.).
Barcelona, Spain:(Universitat de Barcelona).
Igor, B., & Alex, R. (2011a). Applied statistics as competitive advantage. In ENBIS (Ed.).
In11th Annual ENBIS Conference. Vol. Book of abstracts, pp. P. 67).
Coimbra, Portugal:(ENBIS).
Igor, B., & Alex, R. (2011b). La estadística aplicada a la gestión como una ventaja
competitiva. In S. d. e. aplicada (Ed.). In I Jornades de Consultoria Estadística i Software.
pp. p. 16-17). Barcelona, Spain:(Servei d'estadística aplicada).
Barahona, Igor., Riba Alex & Yang, Jian-Bo. “The level of adoption of analytical tools in
Spain. An empirical study based on the evidential reasoning approach”. Decisions Support
Systems. Ref. No: DECSUP-D-13-00247
Barahona, Igor & Riba Alex. “The level of use of statistical tools. A case of statistical
Engineering”. Quality Engineering. Ref. No: LQEN-2013-0088.