1
Master thesis
How does big data influence the decision making tasks and how the decision making cognition modes are related to the managerial decision-making styles?
Supervisor: dr. E. Peelen
Student: Paula de Oliveira Gireli Student ID: 11696591
University of Amsterdam
Faculty of Economics and Business
Master thesis MSc Business Studies – Digital Business Date: 30th June, 2018.
2
Table of contents
ABSTRACT ... 4
1. INTRODUCTION ... 6
2. LITERATURE REVIEW ... 11
2.1 Definition of Big Data ... 11
2.1.1 Big Data Anaytics ... 14
2.1.2 Big Data Opportunities and Challenges ... 15
2.2 Decision making ... 17
2.2.1 Data driven decision making ... 20
2.2.2 Intuition decision making ... 23
2.2.3 Cognitive Continuum Theory ... 26
2.3 Research questions ... 29 2.4 Conceptual framework ... 30 3. METHODOLOGY ... 31 3.1 Research Method ... 31 3.2 Research Design ... 32 3.3 Data Collection ... 32
3.3.1 Oil and Gas Industry ... 33
3.3.2 Participants ... 34
3.3.3 Measures ... 36
3.4 Data Analysis ... 37
3.5 Validity and Reliability ... 39
4. RESULTS ... 41
4.1 Big Data ... 41
4.1.1 Group 1 ... 41
4.1.2 Group 2 ... 43
4.1.3 Key analysis of this section ... 45
4.2 Decision making process and decision style ... 46
4.2.1 Group 1 ... 47
4.2.2 Group 2 ... 49
4.2.3 Key analysis of this section ... 52
4.3 Decision making tasks ... 54
4.3.1 Group 1 ... 55
4.3.2 Group 2 ... 58
4.3.3 Key analysis of this section ... 61
4.4 Modes of Cognition ... 62 5. DISCUSSION ... 67 5.1 Implications ... 67 5.2 Limitations ... 71 5.3 Future research ... 72 6. CONCLUSION ... 74 7. REFERENCES ... 76 8. ATTACHMENTS ... 83
8.1 Attachment I - Meansuring decision style ... 83
3
List of tables, figures and Graphs Tables
Table 1. Properties of the tasks that induce intuition and analysis (Menches C., 2013) 28
Table 2. Composition of the sample 35
Table 3. Codebook 39
Figures
Figure 1. Big Data Challenges (Sivarajah U. et al, 2016) 17
Figure 2. Modes of Cognition (Dhami & Thompson, 2012) 27
Figure 3. Conceptual Framework 30
Figure 4. A typical value chain of oil & gas industry ( Pitatzis, 2016) 33
Figure 5. Group 1 – Illustrative quotes 43
Figure 6. Group 2 – Illustrative quotes 45
Figure 7. Group 1 – Tasks 57
Figure 8. Group 2 – Tasks 60
Figure 9. Mostly analysis & some intuition 63
Figure 10. equally analysis & intuition 64
Figure 11. Pure analysis 64
Graphs
Graph 1. Analysis x Intuition (LaValle S. et al, 2011) 22
Graph 2. Group 1 - Decision Style 49
Graph 3. Group 2 - Decision Style 52
Graph 4. Group 1 + Group 2 - Decision Style 54
4
Abstract
Organizations today face overwhelming amounts of data (Gordon J. et. al., 2013). Researches indicated several possibilities of the use of big data and how it can improve the decision making process. Studies also suggest that managers are relying more on data-based analysis than on intuition. Big data can be used in different ways, but this study focused on one aspect of the use of big data, which is the decision making process. Based on the literature about big data and decision making a research gap emerged, which was the need to understand the impact of big data on the tasks and cognition modes from the Cognitive Continuum Theory. Therefore the idea behind this research was to understand how big data influence the decision making tasks and how the cognition modes are related to the managerial decision-making styles. The research was exploratory in nature, and simple case was studied. The in-depth semi-structure interviews were conducted within specialists from different companies from the O&G industry. Overall, there are some empirical findings in this research that shows that data has some influence and impact in the decision tasks and cognition modes, and makes the process more analytical. This empirical study contributes to the decision making literature by considering the influence of big data in the decision making tasks and modes of cognition from the Cognitive Continuum Theory in a management context. Furthermore, this study contributes to have a better understanding of data on decisions tasks, with the aim that this insight, in future research, will help understand how big data can enhance performance /competitiveness of companies.
Keywords: Big data; Decision Making Process, Decision Making Style, Decision Tasks, Modes of Cognition.
5
Statement of originality
This document is written by Student Paula de Oliveira Gireli who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
6
1. Introduction
Big data (BD) is the oil of the new economy (Baaziz, A. & Quoniam, L. 2013). According to a report from McKinsey Global Institute (2011), big data is the next frontier for innovation, competition, and productivity (Manyika, J. et al., 2011).
Data is considered the raw material and the most strategic resource of the 21st century, similar in importance to gold and oil, and abundance is assumed with today´s 15 billion devices already connected to the internet (Elgendy, N. & Elragal A., 2016; Alharthi A. et al, 2017). Every second, more and more data is being generated from diverse sources. Gartner (2015) forecasted that 25 billion connected objects would be in use by 2020.
Big data is seen as a game changer capable of transforming the way businesses operate in many industries, it represents a new technology paradigm for data that is generated in high velocity, high volume, and with high variety (Lee I., 2017). It has the potential to transform the way organizations make their decisions (Janssen M. et al., 2016). However, as BD datasets are big and high in variety and velocity, it makes difficult to handle it using traditional tools and technique (Janssen M. et al., 2016). As mentioned by Davenport (2014), big data is too voluminous, too unstructured and too fast-moving, it demands new approaches to management and decision making that are evidence-based, fast, and support continuous decisions (Davenport T., 2014).
As Einstein famously declared “information is not knowledge”. The purpose of processing Big Data is to exploit knowledge from data to support intelligent decision (Wang H., 2016). Big data is worthless in a vacuum (Gandomi, A. & Haider M. 2014). Its potential value is unlocked only when leveraged to drive decision making. To enable such evidence-based decision making, organizations need efficient processes to turn high volumes of fast-moving and diverse data into meaningful insights (Gandomi, A. & Haider, M. 2014), and to gain valuable insights from big data it is essential to use big data analytics (Elgendy, N. & Elragal, A., 2016). Big Data analytics (BDA) is needed
7
to create value of the data (Janssen M. et al., 2016), and data needs to be accurate and decisions makers need to be able to interpret it.
Big Data analytics is the process of inspecting, cleaning, transforming, and modeling Big Data to discover and communicate useful information and patterns, suggest conclusions, and support decision making (Cao M. et al., 2015).
BDA can have a substantial influence on decision-making processes, managers need to recognize that the analytic output is useful and decide to use it to support their decisions (Thirathon U. et al., 2017). As mentioned by Provost and Fawcett (2013), data analytics is gaining increasing attention in business and consequently also Data-Driven Decision-making (DDD), which refers to the practice of basing decisions on the analysis of data, rather than on intuition (Provost & Fawcett, 2013).
Brynjolfsson and McElheran (2016) mentioned that new opportunities to collect and leverage data have led many managers relying less on intuition and more on data. The adoption of big data analytics and DDD is connected with the improvement of the performance of companies, with 5-6% higher productivity, better asset utilization, return on equity and market value (LaValle S. et al, 2011; Brynjolfsson et al., 2011). Some researches stated that in order to achieve such improvements, companies need to change their decision making process (Frisk J. & Bannister F., 2017). However, the increasing amount of information leads to a complex decision environment that may cause decision makers to lapse into using mental effort–reducing heuristics such as anchoring and adjustment (Van Bruggen, G. H. et al., 1998).
Besides the evident benefits and opportunities from implementing big data, there are some challenges related to the adoption of it, such as: data quality, data security, privacy, investment justification, data management, and shortage of qualified data scientists (Lee, 2017). However, most of the researchers agree that the challenges faced by the companies are mostly associated to managerial and cultural rather than related to data and technology
8
(LaValle S. et al, 2011; McAfee & Brynjolfsson, 2012; Kuilen, B. & Jacques, R. 2015).
As decision making process is a permanent and important element of managing any business activities, it is essential to understand the management challenges related to the adoption of big data in connection to the decision making tasks.
When we talk about big data and big data analytics, we automatically consider a rational decision making style. However, some research findings are still inconsistent in terms of what managers base their decisions on, apparently even when managers use a rational approach in their decision-making process, they may still also use intuition (Thirathon U. et al., 2017). The literature indicates that there are several situations under which intuition is more likely to be accurate.
The idea behind this research is to understand how big data influence the decision making tasks and how the decision making cognition modes are related to the managerial decision-making styles. If we understand the impact of big data in the decision making tasks and cognition modes, maybe it could help to better understand the influence of big data in the improvement of the performance of the companies.
To answer the above questions, this research will explore the impact of big data in the tasks and cognitive modes mentioned in the Cognitive Continuum Theory. The Cognitive Continuum Theory (CCT) is an adaptive theory of human judgement and suggests a continuum of cognitive modes anchored at one extreme by intuition and at the other extreme by analysis (Dunwoody P. et al., 2000; Menches C.& Saxena J. 2013). In between the two extremes are several combinations of intuition and analysis - mentioned as quasirationality – which consist of a range of modes of thinking that may be selectively used by individuals depending on the particular task being performed at the moment (Menches C.& Saxena J. 2013). The theory specifies task characteristics that are likely to induce cognitive modes at different points along the cognitive
9
continuum (Dunwoody P. et al., 2000). The CCT needs to be expanded to include an understanding of the impact of information technology on the task and cognition modes (Dhami M. & Thomson M., 2012). This gap in the existing literature will be addressed by exploring how big data impacts the tasks and cognitive modes of the Cognitive Continuum Theory.
For the aim of this study, it was chosen to collect data doing interviews with specialists from an industry where the decisions involve very large investments, are highly data dependent and has variable types of complex decisions with large amounts of uncertainty. The chosen industry is the Oil & Gas Industry (Upstream sector). Decisions in this industry determine the direction of billions of dollars every year, the complexity of a decision can range from simple such as to drill or not to drill, to more elaborate such as to determine the maximum price for a proposal, the best development process for a given asset, the drilling priority of a company´s exploration opportunities, timing to increase facilities capabilities, sign a long or short term charter contract, etc. (Coopersmith E. et al. 2001).
Therefore, the purpose of this research is to understand the impact that the adoption of big data has on decision making tasks and how the decision making cognition modes are related to the intuition and data-driven decision making styles.
Dhami and Thomson (2012) stated that few researches have applied Cognitive Continuum Theory to the management context, which is intriguing since the findings of such research can have potentially practical and policy implications for management (Dhami M. & Thomson M., 2012). Therefore, this study will contribute to the limited body of knowledge by exploring the impact of Big Data in the tasks from the Cognitive Continuum Theory in a management context.
This thesis is organized as follows: the literature review will start describing the big data and big data analytics, subsequently decision making, decision making styles and factors that influence the decision making styles will be explained. To finalize the literature review Cognitive Continuum Theory (CCT) and decision
10
making tasks will be explained. Afterward the research design will be presented. Thereafter, the research method and data collection will be given. Next, the results will be presented. The limitations of the research will be discussed and suggestions for future research will be given. To finish, a conclusion will be drawn from the findings.
11
2. Literature review
In this chapter, the main concepts of this thesis will be described. In order to understand how big data may influence the decision making tasks, first big data concept will be explained. Second big data analytics, which is a method used to analyze and acquire intelligence from big data, will be described. Decision making style (data-driven and intuition) will be described. The Cognitive Continuum Theory (CCT) and the decision making tasks will be explained. This chapter will finish with a theoretical framework.
The exploration into the existing literature of these principal concepts will help set the stage for this study and to generate the research design presented in Chapter 3.
2.1. Definition of Big Data
The era of Big Data has arrived along with large volume, complex and growing data generated by many distinct sources. Nowadays, nearly every aspect of the modern society is impacted by Big Data, involving medical, health care, business, management and government (Wanga H. et al., 2016).
While it is ubiquitous today, however, ‘big data’ as a concept is nascent and has uncertain origins (Gandomi A. & Haider M., 2014). According to some academic literature, the use of the term “big data” has been rising since late 1980s (Akerkar et. al., 2015).
As mentioned by Akerkar et al (2015), among several definitions reported in the literature, the first formal, academic definition appears in a paper submitted in July 2000 by Francis Diebold: “Big data refers to the explosion in the quantity (and sometimes, quality) of available and potentially relevant data, largely the result of recent and unprecedented advancements in data recording and storage technology (Akerkar et. al, 2015). In this new and exciting world, sample sizes are no longer fruitfully measured in “number of observations,” but rather in, say, megabytes, even data accruing at the rate of several gigabytes per day are not uncommon.” (Diebold, 2000)
12
Another good definition of Big Data is from Feblowitz (2012), where he said that “Big data technologies describe a new generation of technologies and architectures, designed to economically extract value from very large volumes of a wide variety of data, by enabling high velocity capture, discovery and/or analysis while ensuring their veracity by an automatic quality control in order to obtain a big value” (Feblowitz, 2012).
Gartner (2012) described big data as “volume, velocity and high-variety information assets that demand cost-effective, innovative forms of information processing for enhanced insight and decision making” (Laney, 2012).
Likewise, TechAmerica Foundation (2014) defined big data as “a term that describes large volumes of high velocity, complex and variable data that require advanced techniques and technologies to enable the capture, storage, distribution, management, and analysis of the information.” (TechAmerica Foundation, 2014).
The big data movement seeks to glean intelligence from data and translate that into business advantage (McAfee, A., Brynjolfsson, E., 2012).
The most popular definition in recent years uses the “Three V’s”: volume, variety and velocity (Akerkar et al., 2015). The Three V’s (volume, variety and velocity) have emerged as a common framework to describe big data (Chen, Chiang, & Storey, 2012; Kwon, Lee, & Shin, 2014). Later the Three V’s” got expanded to two additional V´s; Veracity and Value (Hashem et.al, 2015; Elgendy, N., & Elragal, A., 2104; Fadiya et.al., 2014). These 5 characteristics are explained as follows:
Volume refers to the amount of data an organization or an individual collects and/or generates (Lee, 2017). It refers to the vast quantity of structured and unstructured data that is hard to collect, manage, and analyze with the existing IT infra-structure and tools; thus, these massive data sets require new and innovative tools and approaches for capturing, storing, and analyzing data. (Alharthi A. et al, 2017). Big data volumes vary by factors, such as time and type of
13
data. What may be considered big data today may not meet the threshold in the future, as storage sizes will increase allowing even bigger data sets to be captured (Gandomi A. & Haider M., 2014).
Velocity refers to the speed at which data are generated and processed. The velocity of data increases over time. Gartner (2015) forecasted that 6.4 billion connected devices would be in use worldwide in 2016 and that the number will reach 20.8 billion by 2020. In 2016, 5.5 million new devices were estimated to be connected every day to collect, analyze, and share data. The enhanced data streaming capability of connected devices will continue to accelerate the velocity (Lee, 2017).
Variety refers to the number of data types. It refers to the fact that the data comes from various sources such as spreadsheets, traditional databases, text documents, and digital data streams. (Alharthi A. et al, 2017).
Veracity refers to credibility. IBM created Veracity as the fourth V, and it represents the unreliability and uncertainty latent in data sources. Uncertainty and unreliability arise due to incompleteness, inaccuracy, latency, inconsistency, subjectivity, and deception in data (Lee, 2017).
Value: Firms need to understand the importance of using big data to increase revenue, decrease operational costs, and serve customers better; at the same time, they must consider the investment cost of a big data project. Data would be low value in their original form, but data analytics will transform the data into a high-value strategic asset (Lee, 2017).
A data set can be called Big Data if it is formidable to perform capture, curation, analysis and visualization on it at the current technologies (Chen C. & Zhang C., 2014).
14
The speed of data growth has already exceeded Moore’s law (Wanga H. et al., 2016) and as the data sets from big data are big and high in variety and velocity, traditional tools and techniques are not able to handle it (Janssen at al. 2016). To capture the value from this enormous amount of Data, it is necessary to develop new techniques and technologies for analyzing it (Chen C. & Zhang C., 2014).
The big data has value if the companies are able to analyze it. As mentioned on the book Strategy is Digital (2016), the patterns revealed by analyzing big data, and a company’s ability to take decisions based on those insights, are what give big data its value (Cordon C. et al., 2016). Big Data is closely related to Big Data Analytics which are need to create value of the data (Elgendy & Elragal, 2014).
2.1.1. Big Data Analytics
According to Labrinidis and Jagadish (2012), Big Data Analytics (BDA) refers to methods used to examine and attain intellect from the large datasets. Thus, BDA can be regarded as a sub-process in the whole process of insight extraction from BD (Sivarajah U. et al. 2016).
Big Data analytics is the process of inspecting, cleaning, transforming, and modeling Big Data to discover and communicate useful information and patterns, suggest conclusions, and support decision making (Cao M. et al 2015). And the data analysis techniques include a wide range of disciplines such as data mining, machine learning, artificial neural networks and signal processing, most of which have shown their capabilities in processing Big Data (Wang H. et al, 2016).
Big Data analytics commonly involves combining several sources of data, some structured and others unstructured, including numbers, text, images, sound, and video, the processing of which requires a combination of different analytical methods from different disciplines (Cao M. et al 2015). An example is Ayata’s Prescriptive Analytics, which is used in oil and gas exploration to predict optimal drilling sites based on data such as images from well logs, videos of fluid flows
15
from hydraulic fractures, sounds from drilling operations, text from driller’s notes, and numbers from production reports (Cao M. et al 2015).
Many organizations are adopting Big Data Analytics with the purpose of constructing valuable information from BD. The analytics process, including the deployment and use of BDA tools, is seen by organizations as a tool to improve operational efficiency though it has strategic potential, drive new revenue streams and gain competitive advantages over business rivals (Sivarajah U et al. 2016).
The attention in Bid Data Analytics is increasing and consequently also in Data-Driven Decision-making (DDD), which refers to the practice of basing decisions on the analysis of data, rather than on intuition (Dargam F. et al 2015).
2.1.2. Big Data Opportunities and Challenges
Organizations realized that the ability to analyze and use big and complex data sets will be the most important source of competitive advantage in the 21st century (Alharthi A. et al. 2017).
Big data provides great potential for companies in creating new businesses, developing new products and services, and improving business operations (Lee I., 2017). It has the potential to deliver better customer experience, enhance internal efficiency, and, ultimately, improve profitability and competitiveness of organizations across all industries (Alharthi A. et al. 2017) .The use of big data analytics can create benefits, such as cost savings, better decision making, and higher product and service quality (Davenport, 2014). Organizations can use big data to get smarter and innovative in ways that were not possible before the advent of the ‘zettabyte era’ (LaValle et. al, 2011).
Although the benefits of BD are factual and substantial, there are some challenges that must be addressed to fully understand the potential of BD (Sivarajah U. et al. 2016).
16
In a survey conducted by Lee (2017), he highlighted six technical and management challenges: Data quality Data security Privacy Investment justification Data management
Shortage of qualified data scientists
In a study held by McAfee, A., Brynjolfsson, E., (2012) in partnership with McKinsey’s business technology office discovered that many organization were not ready to adopt the influence of big data for improving organizational performance, as it requires overcoming a number of barriers in relation to big data (McAfee, A. and Brynjolfsson, E., 2012). Alharthi et al (2017), mentioned these barriers in the article “Addressing barriers to big data”, where the barriers were divided into three types:
1. Technological barriers a. Infrastructure readiness b. Complexity of data 2. Human barriers a. Lack of skills b. Privacy 3. Organizational barriers a. culture
Also in 2017, another study related to the challenges of Big Data was held by Sivarajah U. et al. 2016, based on the data life cycle the challenges were grouped into three main categories:
Data challenges: related to the characteristics of the data itself (e.g. data, volume, variety, velocity, veracity, volatility, quality, discovery and dogmatism).
17 Process challenges: related to series of how techniques: how to capture data, how to integrate data, how to transform data, how to select the right model for analysis and how to provide the results.
Management challenges: related to privacy, security, governance and ethical aspects.
Figure 1: Big Data Challenges. Source: Sivarajah U et al. (2016). Critical analysis of Big Data challenges and analytical methods
2.2. Decision making
People are often involved in daily decision making, whether for simple or complicated problems (Hon-Tat et al 2011). Decision making can be explained as a choice between different alternatives (Hansson S., 1994). Decisions are the choices made from two or more alternatives (Judge & Robbins, 2006). Certo and Certo (2005) mentioned that decision making is the process of choosing the best alternative for reaching an objective (Certo & Certo, 2005). Bagchi N. (2010) defined decision making as a process of identification of a problem or opportunity, understanding the context in which the problem or opportunity occurs, generating alternative solutions to attack the problem or taking advantage of the opportunity and then making a choice amount the many alternatives (Bagchi N., 2010). Information is required at each stage of the decision making process (Bagchi N., 2010) and individuals are required to interpret and evaluate the information before making any decision (Hon-Tat et al 2011). Fundamentally everyone is a decision maker and everything we do consciously or unconsciously is the result of some decision (Saaty, 2008). As
18
the world grows increasingly turbulent because of the globalization of markets and rapid changes in technology, the ability to make high-quality decisions quickly in the face of complexity has become a central managerial issue and a fundamental dynamic capability (Woiceshyn J., 2009).
Henry Mintzberg argues that managers work at an unrelenting pace; work that is characterized by shortness, variety, and discontinuity, and they are strongly oriented to action (Mintzberg H., 1971). Managers are constantly switching from highly complex tasks to routine ones on a moment by moment basis (Walumbwa F. et al., 2014). Making decisions is a big part of a manager’s job since he is frequently the one with the final say (Walumbwa F. et al., 2014). And managers make strategic decisions every day, from the most usual to the most important and complex ones.
Management involves decisions and decisions can be classified into three categories based on the level at which they occur: strategic, tactical and operational (Schmidt G. & Wilhelm W., 2000). Strategic decisions have a significant influence on the degree of commitment and a significant influence on the scope of the firm (Shivakumar R., 2014). Tactical decisions are decisions about how things will get done and operational decisions refer to routine decisions that firms make each day (Shivakumar R., 2014).
The environment of the companies are rapid and fast-paced, the decisions must normally be made quickly and there is little time for reflection, therefore it is important that managers (decision makers) learn how to recognize and filter new information and to apply such information quickly in order to provide effective directions to the people they manage (Walumbwa F. et al., 2014).
Managers (or decision makers) from different organizations or even from the same organization with similar tasks and goals have big differences in the way they make decisions (Mihaela P., 2015). The differences could be explained by personality types, professional experience, managerial style, which give the managers a specific decisional making style (Mihaela P., 2015).
The decision making has different styles. Decision making styles can be
19
it is a learned habit and that the key differences among styles involve the amount of information considered during a decision and the number of alternatives identified when reaching decisions (Driver, 1979). Harren (1979) defined decision making style as individuals’ characteristic mode of perceiving and responding to decision-making tasks, he proposed a model of career decision-making style, where 3 styles were identified: dependent, rational and intuitive (Scott & Bruce, 1995). Driver, Brousseau, and Hunsaker (1993) suggested that decision-making style is defined by the amount of information gathered and the number of alternatives considered when making a decision (Scott & Bruce, 1995).
Scott and Bruce (1995), trying to integrate all earlier work on decision-making styles (Thunholm, P. 2004) defined decision-making style as the learned habitual response pattern exhibited by an individual when confronted with a decision situation. It is not a personality trait, but a habit-based propensity to react in a certain way in a specific decision context (Scott & Bruce, 1995). Five decision making styles were identified in behavioral terms (Scott & Bruce, 1995):
Rational: characterized by thorough search for and logical evaluation of alternatives;
Intuitive: characterized by reliance on hunches and feelings;
Dependent: characterized by a search for advice and directions from others;
Avoidant: characterized by attempts to avoid decision-making.
Spontaneous: characterized by a feeling of immediacy and a desire to come through the decision-making process as quickly as possible.
The rational and the intuitive styles as described by Scott and Bruce have a high similarity with the analytic and intuitive aspects of the cognitive style and are therefore unproblematic from a theoretical point of view, but the theoretical bases for the other styles are unclear (Thunholm, P. 2004). Note that cognitive style in decision-making often refers to individual ‘‘thinking practices’’ central to the understanding of decision making processes (Thunholm, P. 2004). There are several articles on individual differences in decision-making that state that
20
the terms cognitive style and decision-making style terms are often used as if they were synonym (Thunholm, P. 2004).
Considering that the definitions of the dependent and the avoidant decision-making styles cannot be referred to the analytic/intuitive classification on the information gathering or the information evaluation dimensions of cognitive style; and the spontaneous style could be considered as a kind of high-speed intuitive decision-making style (Thunholm, P. 2004). Combined with the fact that this research will focus on the decision making tasks, as mentioned in the Cognitive Continuum Theory, where the tasks characteristics are likely to induce cognitive modes of intuition and rational/analysis. It was decided, therefore, to focus only on two decision making styles: intuitive and rational (data-driven) based.
Researchers suggest that several factors can influence the decision making style, such as: environmental (stable, unstable, high-velocity) (Papadakis V. et al., 1998, Khatri & Ng, 2000, Dane & Pratt, 2007); levels of uncertainty (Agor, 1987); internal firm characteristics (such as systems, performance, size, ownership) (Papadakis V. et al., 1998); time pressure (Agor, 1987, Eisenhardt & Zbaracki, 1992; Hon-Tat et al., 2011., 2011); type of information available (Khatri & Ng, 2000); type of decision (non-routine decisions, routine decisions) (Khatri & Ng, 2000; Isenberg, 1984; Dane & Pratt, 2007); culture (Dane & Pratt, 2007, Frisk J. & Bannister F.,2017).; managerial level (Eisenhardt & Zbaracki, 1992; Agor, 1987, Papadakis V. et al., 1998).
In this study, the difference between data-driven and intuition decision making will be described.
2.2.1. Data driven decision making
Normative decision theory focuses on finding methodologies, technologies and tools (software) to identify the best decision to make based on the assumption that the decision maker is fully rational or bounded rational (Wang H. et al, 2016). Under this perspective, the data-driven decision making is considered rational.
21
As mentioned by Davenport (2013), the use of data to make decisions is not a new idea, it is as old as decision making itself (Davenport, 2013). New opportunities to collect and leverage data have led many managers to change how they make decisions, relying less on intuition and more on data (Brynjolfsson E. & McElheran K., 2016).
With enormous volumes of data now available, companies in almost every industry are focused on exploiting data for competitive advantage (Provost & Fawcett 2013). In a data-driven culture, companies approach decision making rely less in gut feeling when making major decisions (The Economist, 2013). Data-driven decision making (DDD) means to the way of basing decisions on the analysis of data rather than purely on intuition (Provost & Fawcett 2013).
Some organizational leaders are requiring analytics to exploit their growing data and computational power to get smart, and get innovative, therefore to run the businesses on data-driven decisions (LaValle S. et al., 2011). Supporters of data driven decisions suggests that data driven decisions make better predictions and better decisions than the decisions based on intuition (McAfee, A., Brynjolfsson, E., 2012). Note that, if data is not analyzed consistently and correctly, decision makers might use it to draw incorrect conclusions and if the results of data analysis are not incorporated fully into decision making, the investment on collecting and analyzing data might be for nothing (Maxwell L. N. et al, 2015).
Some consultants, such as Ram Sarvepalli from Ernst Young (2016), stated that data analytics is the key to success in the modern business environment and data is the corporate competitive advantage. Better data creates opportunities to make better decisions (Brynjolfsson & McElheran, 2016), however, more data is not always better, excessively data can make it more difficult to recognize and make sense of the data that is important (Davenport & Prusak, 2005).
There are some studies showing the direct connecting between data-driven decision making and firm performance. An article from MitSloan shows a graph where analytics trumps intuition. The tendency for top-performing organizations to apply analytics to particular activities across the organization compared with
22
lower performers. According to their study, top performers say analytics is a differentiator (LaValle S, 2011).
Graph 1: Analysis x Intuition. Source: LaValle S. et al., (2011). Big Data, Analytics and the Path From Insights to Value.
On a research held by McAfee, A., Brynjolfsson, E., (2012) partnership in McKinsey’s business technology office, they concluded that the more companies characterized themselves as data-driven, the better they performed on objective measures of financial and operational results. In particular, companies in the top third of their industry in the use of data-driven decision making were, on average, 5% more productive and 6% more profitable than their competitors.
As mentioned by Vidgen et al. (2017), becoming data-driven is not only a technical issue and requests that companies firstly organize their business analytics departments to comprise business analysts, data scientists, and IT personnel, and secondly align that business analytics capability with their
23
business strategy in order to tackle the analytics challenge in a systemic and joined- up way (Vidgen et al., 2017).
McAfee and Brynjolfsson (2012) also stated that to become data-driven is not only a technical matter, they identify five challenges for organizations in becoming data-driven: leadership, talent management, technology, decision-making, and company culture.
Actually, top level firm executives still make most of their decisions based on intuition that has been built over years of experience. McAfee and Brynjolfsson (2012) mentioned that managers and executives generally rely more on experience and intuition instead of data.
2.2.2. Intuition decision making
As mentioned by Hensman (2011), in business settings, formal techniques of rational choice may sometimes be difficult or impossible to apply (Bowman and Ambrosini, 2000). In such circumstances managers may come to rely as much upon informed intuition as they do upon rational analysis (Agor, 1989; Burke and Miller, 1999; Parikh et al., 1994; Rowan, 1986;).
The intuition research in business and management (1980s and 1990s), employed descriptive surveys of various sizes which aimed to capture managers views on the importance and the practicalities of intuition-in-use (e.g., Agor, 1989; Parikh et al., 1994; Hensman A. & Sadler-Smith E., 2011).
There are lots research focused on the role of intuition and there are several definitions from different researchers. Simon (1987) defined intuition as simply analyses frozen into habit and into the capacity for rapid response through recognition. Saddler-Smith & Shefy stated that intuition is a capacity for attaining direct knowledge or understanding without the apparent intrusion of rational thought or logical inference (Saddler-Smith & Shefy, 2004). Dane and Pratt (2007) stated that Intuitions are involuntary, affectively charged judgments arising through rapid, non-conscious and holistic associations (Dane and Pratt, 2007). It is the ‘‘insight that bypasses reasoning’’ and
24
commonly understood as an inexplicable ‘‘hunch’’ or ‘‘gut feeling’’ that tells a person what to do (Woiceshyn J., 2009).
In some business situations it is impossible to use only rational decisions (Hensman A. and Sadler-Smith E., 2011) and managers may rely as well on intuition (Agor, 1989; Burke and Miller, 1999; Parikh et al., 1994; Rowan, 1986; Hensman A. & Sadler-Smith E., 2011). Managers with good intuition can see new possibilities in any given situation; they are particularly adept at generating new ideas and providing new solutions to old problems (Agor, 1989). At a time of rapid and unprecedented change in the business environment, executives who understand how to balance their use of intuition and analytic thinking may be better prepared to lead this environment (Burke & Miller, 1999).
Agor (1989) mentioned that intuitive decisions come from a capacity to integrate and make use of information coming from both the left and right sides of the brain. He conducted a survey over 200 top executives from private and public sector in the USA, and found that the intuitive ability varied by managerial level and that executives considered intuition to function best when: there is a (a) high level of uncertainty, and (b) there is little previous precedent; (c) variables are less predictable, (d) the information are limited and the available information does not point the way to go; (e) several plausible alternative solutions exist, and (f) time is limited (Agor, 1989). Dane and Pratt (2007) suggested that intuition may be integral to successfully completing tasks that involve high complexity and short time horizons (Dane and Pratt, 2007). Dane and Pratt (2007) analyzed some studies on intuition in the management environment and identified a number of situations in which intuition is an effective cognitive mode (Dhami M. & Thomson M., 2012). Sadler-Smith and Sparrow (2008) view intuition as an ability (Dhami M., Thomson M., 2012).
Dane and Pratt (2007) reported that individuals can better rely on intuition when making a broad evaluation in a domain area where they have in-depth knowledge of the subject, or amassed intuitive expertise. They suggest that intuition is likely more constantly used by managers who have accumulated extensive experience (Sadler-Smith E. & Burke-Smalley L. 2015). And ignoring expertise can lead to failure (Sadler-Smith E. &
Burke-25
Smalley L. 2015). In 1999, Burke and Miller conducted semi-structured interviews with 60 professionals from different companies in USA and the majority of the interviewees (56%) said intuitive decisions were based on experience and 46% said intuition is based on a person's feelings or emotions when presented with information in a decision-making scenario (Burke & Miller, 1999).
Intuition does not come easily; it requires years of experience in problem solving and is founded upon a solid and complete grasp of the details of the business (Isenberg, 1984; Seebo, 1993). Intuition is central to all decisions, it is not an irrational process, it is based on a deep understanding of the situation (Khatri, N., & Ng, H. A., 2000). It is a complex phenomenon that draws from the store of knowledge in our subconscious and is rooted in past experience (Khatri, N., & Ng, H. A., 2000).
Harper (1990) mentioned that top executive may not have to use intuition very often, but, when the data do not provide a clear answer, these executives have the uncanny ability to sense what should be done and the courage of their convictions to act decisively (Khatri, N., & Ng, H. A., 2000).
Intuition is no longer considered to be a single type, several researchers in the behavioral sciences and management have identified four ‘primary types’ of intuition (Sadler-Smith E. & Burke-Smalley L. 2015): (1) expert intuition, represents an expertise-based response driven by involuntary, non-conscious processing of information; (2) social intuition, refers to the rapid and automatic evaluation of another person’s cognitive and/or affective state through the perception and non-conscious processing of verbal and/or non-verbal indicators, akin to a form of ‘‘mind-reading; (3) moral intuition and (4) creative intuition (Sadler-Smith E. & Burke-Smalley L. 2015).
Mintzberg considered the practice of management as much an art based on vision and intuition as it is a science (Sadler-Smith E. & Burke-Smalley L. 2015).
In the age of ‘‘big data,’’ Tim Lebrecht wrote, in Fortune Magazine in 2013, that although data can give us information quickly it can only serve
26
to make us ‘‘smarter not wiser’’ and that for ‘‘quick but profound decisions’’ intuition is much better (Sadler-Smith E. & Burke-Smalley L. 2015).
Although management researchers acknowledge that intuition is an important and essential cognitive tool for managing an organization, they also acknowledge that it is not sufficient (Dhami M. & Thomson M., 2012).
2.2.3. Cognitive Continuum Theory
Simon (1987) mentioned that a combination of intuition and analysis need to be employ by effective manager (Dhami M. & Thomson M., 2012). Researchers have focused on the idea that managers may move back and forth from intuitive and rational or analytic thinking (Sadler-Smith & Sparrow, 2008; Dhami M. & Thomson M., 2012).
Hammond (1996) stated that there are many situations where either analysis or intuition cannot be easily used (Dhami M. & Thomson M., 2012), he rejected the view that intuition and analysis were ‘rival’ forms of knowing, and questioned whether judgement and decision making had to be either intuitive or analytical (Parker-Tomlin M. et al. 2017). There are many difficulties and challenges in the management environment to the use of pure analysis and pure intuition (Dhami M. & Thomson M., 2012). Simon (1987) stated that successful managers must use a combination of analysis and intuition. The effective manager does not have the luxury of choosing between analytic and intuitive approaches to problems. Behaving like a manager means having command of the whole range of management skills and applying them as they become appropriate (Simon, 1987).
To try to understand the use of the combination of analytic and intuitive thought in management decision situations it is important to introduce the Cognitive Continuum Theory (CCT) and the concept of quasirationality (Dhami M. & Thomson M., 2012). The CCT was proposed by Hammond et al. in 1987; it is an adaptive theory of human judgement and suggests a continuum of cognitive modes anchored at one extreme by intuition and at the other extreme by
27
analysis (Dunwoody P. et al., 2000; Menches C.& Saxena J. 2013). In between the two extremes are several combinations of intuition and analysis - mentioned to as quasirationality – which consist of a range of modes of thinking that may be selectively used by individuals depending on the particular task being performed at the moment (Menches C.& Saxena J. 2013). Quasirationality is the combination of intuitive and analytic thought and the theory highlights the predominance of quasirationality, as opposed to pure analysis or pure intuition in human judgment and decision making (Dhami M. & Thomson M., 2012). The theory offers researchers of judgement and decision-making a framework in which the concepts of task and cognition are linked together.
The Cognitive Continuum Theory aims to improve and orient individuals and groups to their decision-making process (Parker-Tomlin M. et al. 2017). The figure 2 shows the modes of cognition along the cognitive continuum. Once cognitive processes are defined in terms of their location on a cognitive continuum, they will be found to interact in predictable ways with various task conditions located on a similar continuum (Hammond R. K. et al. 1987)
Figure 2: Modes of cognition along the cognitive continuum Source: Dhami M. & Thomson M., 2012.
The CCT specifies task characteristics that are likely to induce cognitive modes at different points along the cognitive continuum (Dunwoody P. et al., 2000), that means that Cognitive tasks along the task continuum can be measurable using some task properties that are theorized to induce intuition and analysis; they can be quantitatively differentiated from one another with regard to their
28
properties or their tendency to induce intuition, quasirationality, or analysis. (Dhami M. & Thomson M., 2012). The CCT suggests that tasks that can be decomposed into logical, sequential steps are more likely to trigger analysis, while tasks that are not easily decomposed or have ambiguous features are more likely to trigger intuition (Dane et al., 2012).
The combination of the tasks and the cognitive mode will depend on the number, nature and degree of the task properties present (Dhami M. & Thomson M., 2012). The table 1 demonstrates some properties of the task that induce intuition and analysis (Dhami M. & Thomson M., 2012) and outlines a set of predictions about how specific task characteristics will influence a decision maker’s cognition, if an individual is more likely to approach the task intuitively, analytically, or via some combination of the two modes of thought (Menches C., & Saxena J. 2013).
Table 1: Characteristics of tasks that induce intuition and analysis (Doherty and Kurz 1996; Hammond et al. 1987; Inbar et al. 2010)
Tasks can be placed on a continuum depending on their properties, with well-structured tasks (have properties that induce analysis) and ill-well-structured tasks
Analysis Intuition
29
(have task properties that induce intuition) forming the end-points of the continuum and tasks exhibiting the properties of both situated in the middle of the continuum. The properties of a task could influence the modes of cognition on the continuum (Cader R. et al. 2004).
As mentioned by Bazerman (2005), there are many important management tasks involve, limited information, information that requires processing beyond the capacity of the mind, uncertainty about available information, uncertainty about the outcome, risk, scarcity of resources, time pressure, stress and anxiety, and the need to justify decisions on grounds of legality and practicality (Dhami M. & Thomson M., 2012). These are exactly the types of situations where analysis and intuition alone would be difficult to apply; therefore they would require quasirationality (Dhami M. & Thomson M., 2012).
From a practical perspective, we know very little about how big data can influence the decision making tasks and therefore the cognitive modes; and how it is related to the managerial decision-making styles. Hammond et al. (1987) research suggests that tasks trigger the mode of thinking that ultimately leads to making an intuitive or an analytical decision, this study will explore how big data can influence this tasks that induce cognitive modes at different points along the cognitive continuum.
2.3. Research questions
The research question is: “How does big data influence the decision making tasks?
This research aims to understand how a certain phenomenon (big data) influences the tasks that induce intuition and analytical decisions in a managerial context. Additionally, other questions will be explored: “How are the modes of cognition related to the managerial decision-making styles?” and “With the implementation of big data what types of decision making tasks are likely to induce intuitive or analytics mode of cognition?”
30
2.4. Conceptual framework
All of the above comes together in the conceptual framework. Big data and big data analysis improve the decision making process, help managers to make better decisions. As decision making process is an essential element of any management context, it is important to understand the impact of the adoption of big data in the decision making tasks and how the decision making cognition modes are related to the managerial decision-making styles. Considering that the increasing amount of information leads to a complex decision environment, this study will focus in complex decision.
Figure 3: Conceptual framework. Source: Author.
Decision making style -Intuition
-Rational (Data-driven) Mostly
intuition and some analysis Modes of cognition Pure Intuition Pure Analysis Equally intuition and some analysis Mostly analysis and some intuition Big Data analysis Management Context Decision tasks Complex decisions
31
3. Methodology
3.1. Research Method
The main reason for this research is to gain in-depth understanding of how big data influence the decision making tasks. In order to answer the research question, this study will explore the impact and influence of the adoption of big data in the tasks that induce intuition and rational decisions, therefore exploratory research is most appropriate. Exploratory research is about finding general information about a topic that the researcher doesn’t clearly understand (Saunders and Lewis 2012). The purpose of an exploratory study is to seek new insights, to ask new questions and assess topics in a new light (Saunders & Lewis 2012). When exploring a phenomenon under study, the qualitative research is often used (Flick, 2009), so the exploratory nature of this study and the in-depth analysis imply the use of qualitative, therefore qualitative research method was chosen to be more suitable for this research. As mentioned by Corbin and Strauss (2015), qualitative research is a form of research in which a researcher collects and interprets data, and this method can be used to uncover and understand what lies behind any phenomenon about which little is yet know (Corbin , J. and Strauss, A., 2015)
The research philosophy will be interpretivism. Researchers who adopt an interpretative perspective wish to understand what is going on in a work organization (Saunders et al., 2016). This approach focuses on the individual perceptions of people on specific situations and experiences (Saunders et al., 2016) and questions that address the ‘how’ and ‘why’ can be better answered using an interpretative approach. For this study an interpretative approach is adopted in order to explore the influence of big data in the tasks that induce intuition and rational decisions, the conclusion of this study will be based on the interpretation of individual experience. For business and management research, the interpretivist perspective is very relevant, particularly in such fields as organizational behavior, marketing and human resource management (Saunders et al., 2016).
32
3.2. Research Design
The research design will be case study. The case study strategy can give researches a good understanding of why and how some decisions are made (Saunders & Lewis, 2012) and it is most often used in exploratory research (Saunders & Lewis, 2012). Yin (2009) stated that case studies are the preferred method when (a) "how" or "why" questions are being posed, (b) the investigator has little control over events, and (c) the focus is on a contemporary phenomenon within a real-life context, such as individual life cycles, small group behavior, organizational and managerial processes, neighborhood change, school performance, international relations, and the maturation of industries (Yin, 2009). This research aims to understand how a certain phenomenon (big data) influences the tasks that induce intuition and rational decisions in a real-life context (managerial context). The boundaries between phenomenon and context are not clearly evident. The case study method will give a good understanding of the phenomenon, therefore it was found to be the most suitable for this research.
3.3. Data collection
There are several options for getting data for qualitative research and one way to conduct an exploratory research is by interviewing specialists on a given subject. It was decided that this study will use semi-structured interviews. As mentioned by Saunders and Lewis (2012), semi-structure interviews is a method of data collection in which the interviewer asks about a set of topics using some predetermined questions, but differs the order in which the themes are covered and the questions asked (Saunders & Lewis, 2012). This type of data collection allows the researcher to collect information and knowledge from people that have some experience with the above mentioned topics.
To have the data that is necessary to answer the research question in-depth interviews were held. Some interviews were done face to face, but due to time and distance restrictions, most of the interviews were done by skype and by phone. All the interviews were audio recorded with a recording program from
33
the mobile phone and the average duration was around 35 minutes. After that, each interview was transcribed into a word document to be used for further analysis. Note that each interview had the previous approval from the interviewees for recording. To place the participant at ease, it was explained the purpose of this thesis as well as the processes for keeping the data confidential. The interviews were conducted in English and Portuguese. Participants, who are native Portuguese, were interviewed in Portuguese to be sure that no essential information is missing due to the language barrier.
It was decided that the empirical setting of the research was going to be in an industry of high relevance for big data, i.e. Oil and Gas Industry.
3.3.1. Oil and Gas Industry
The Oil & Gas (O&G) Industry is characterized by high entry barriers, such as high investments, high level of technology complexity, complex projects, highly qualified and competent personnel, etc, The Industry is technically challenging and economically risky, requires large projects and high investments in order to extract offshore oil and gas (Vega-Gorgojo G. et al., 2016). The Industry is usually divided into three sectors: upstream, midstream and downstream.
Upstream activities are related to the exploration and production (E&P) of oil and natural gas.
Midstream activities refer to the transportation, store and distribution of oil and natural gas.
Downstream activities include refining and turning hydrocarbons and natural gas into products, as well as retail and marking operations.
34
All three stages of the O&G Industry value chain are supported by service companies, such as drilling contracts, equipment providers, manpower providers, maintenance providers and all necessary services to Oil Companies (Operators). The upstream sector has the majority of the costs in the O&G Industry. As the upstream sector is more complex and with data volumes growing exponentially (Feblowitz, 2012), as an example, a modern offshore drilling platforms have about 80,000 sensors, which are forecast to generate approximately 15 petabytes (or 15 million gigabytes) of data during an asset’s lifetime (World Economic Forum, 2017). The upstream sector is considered the most interesting phase in terms of big data (Vega-Gorgojo G. et al., 2016), therefore this study is focused on the upstream phase of the oil and gas value chain.
The O&G Industry is traditionally marked by frequent boom and bust cycles. Since late 2014, it has become clear that another boom cycle is not around the corner. With the current crude price crisis situation, many large-scale projects are at risk and oil companies are looking at new ways to improve their margin costs and reduce the uncertainty of their investments – particularly through big data and data science driven solutions (Vega-Gorgojo G. et al., 2016). In this context, it was chosen to collect data doing interviews with specialists from different companies from the Upstream Oil and Gas Industry.
3.3.2. Participants
The participants will need to meet certain criteria to guarantee an accurate research design. Firstly, the interviewees have to work in the O&G Industry, at the upstream sector. Second and most important is that the participants have a managerial position and have experience in making complex and/or important decisions. Third is that the participants have some knowledge about Big Data.
For this study several companies from the O&G Industry were approached, in total 30 people were asked to participated in the interviews, the participants were contacted by phone and/or e-mail, but only 16 were available to help. The interviews were arranged with a CEO, vice president HSE, quality & maintenance, contracts & marketing director, commercial directors, commercial
35
& business development director, operational director, operational managers, contract manager, technical manager, commercial managers, technical and operational advisor and E&P contracting consultant.
Because of the short time in which this research had to be done, it was not possible to collect a larger sample of specialists. However, those who did the interview were able to contribute to this research with a valuable insight.
See below the Composition of the sample:
Group 1
Participant Company Sector Location Years of
experience Job Title Participant 1 National
Drilling Contractor
Upstream Oil & Gas
Brazil 37 years Comercial Director Participant 2 Semi-Public multinational Company (Operator) Upstream Oil & Gas
Brazil 35 years Contract Manager
Participant 3 Multinational Drilling Contractor
Upstream Oil & Gas
EUA 30 years Comercial
Director Participant 4 Multinational
Installation Services
Upstream Oil & Gas
Netherland 20 years Commercial and Business Development Director Participant 5 Multinational Drilling Contractor Upstream Oil & Gas
Brazil 15 years Marketing manager Participant 6 Multinational
Drilling Contractor
Upstream Oil & Gas
EUA 13 years Marketing and Contracts
Director Participant 7 Software
provider
Upstream Oil & Gas
Netherlands 20 years CEO
Participant 8 Semi-Public multinational
Company (Operator)
Upstream Oil & Gas
Brazil 14 years E & P contracting
consultant Group 2
Number of
Participant Company Participant Location
Years of
experience Job Title Participant 9 Multinational
Drilling Contractor
Upstream Oil & Gas
India 35 years Operations Director
Participant 10 Multinational Drilling
Upstream Oil & Gas
EUA 20 years Vice President HSE, Quality &
36 Contractor Maintenance Participant 11 International Drilling Contractor Upstream Oil & Gas
EUA 34 years Technical and Operational advisor Participant 12 Semi-Public multinational Company (Operator) Upstream Oil & Gas
Brazil 16 years Operational Manager
Participant 13 International Drilling Contractor
Upstream Oil & Gas
Brazil 16 years Technical and Operational
specialist Participant 14 Software
provider
Upstream Oil & Gas
Netherlands 7 years Technical manager Participant 15 International
Drilling Contractor
Upstream Oil & Gas
Brazil 35 years Operacional manager Participant 16 International
Drilling Contractor
Upstream Oil & Gas
France 10 years Operacional manager Table 2: Sample. Source: Author
In all the cases some related information were sent previously to the interviewees. A brief introduction of the research was given at the beginning of each the interviews. Subsequently the interviewer asked for the participant to short introduce himself/herself about the education, current position, primary responsibilities, and prior experience. Asking personal and background questions made some participants feel ease and consequently the interview were also easier to do.
After the interviews were conducted, a division became visible between the participants that work on commercial or marketing or business development functions and participants that work on the operation or maintenance functions. Therefore, the results from group 1 and group 2 will be presented separately in the results section. This allows having different viewpoints to be compared. The interview questions are described in the Attachment II.
3.3.3. Measures
Before the start of each interview, the decision making style (rational and intuitive) of each participant were measured using the Decision Making Style Inventory (DMSI) developed by Scott and Bruce (1995). Scott and Bruce’s General Decision-Making Style concept and inventory are the most
37
encompassing validated, and widely used conceptual approach (Fischer S. et al., 2015). For this study, five items were used to measure the rational decision-making style, and five items were used to measure the intuitive decision-decision-making style (see Attachment I).
Regarding the data-driven decision style, Brynjolfsson et al. (2011) developed a measure of the use of “data-driven decision making that rates how strongly firms use data across the company. The measure is based on three questions and for these research two questions were adapted to the decision making context and one question was not in line with the framework of this study (see Attachment I).
The purpose to measure the decision making styles described above is to try to understand if there is a connection between the modes of cognition and the decision making styles.
3.4. Data Analysis
The grounded theory is a qualitative research approach that was elected to be use in this study, in which theory is methodically generated from the data (Strauss & Corbin, 1990). As mentioned by Glaser and Strauss (1967), grounded theory does not aim for the “truth” (since is unknown) but it aims to conceptualize “what is going on” by using empirical data to derive hypotheses (Menches & Saxena, 2013).
Based on the grounded theory, this study employed the theoretical sampling method. As indicated by Glaser (1978), theoretical sampling is a process of data collection for generating theory where the analyst jointly collects, codes and analyses the data and decides what data to collect next and where to find it, in order to develop the theory as it emerges (Glaser, 1978). It is sampling on the basis of concepts that have proven theoretical relevance to the evolving theory (Strauss & Corbin, 1990).
As mentioned by Starks and Trinidad (2007), grounded theory relies on theoretical sampling, which involves recruiting participants with different type of experiences of the phenomenon so as to explore multiple dimensions of the
