Data Analytics - Exam 1

data

raw figures and captured facts, such as categories, measures, and calculations

information

knowledge gained from data that is relevant for analysis purposes

data analytics

the process of analyzing raw data to answer questions or provide insights

self-service business intelligence (SSBI) software

easy-to-use, accessible software that can prepare data, analyze data, and report results

• provides extended data processing capabilities for preparing data, analyzing data, and reporting data analysis results

• easy to use

two key features of SSBI software

null value

an unknown or missing value in a data set

data visualization

graphical representation of information and data to provide meaning and insights during the data analysis process

dashboard

a graphical user interface that shows key performance indicators for an organization; useful tool to communication key information to everyone who needs it

1. plan

2. analyze

3. report

three stages in the data analysis process

1. understand motivation

2. determine the objective

3. design the data and analysis strategy

three steps in the planning stage of the data analysis process

motivation

the reason or stimulus for performing data analysis; the “why” behind a project

external motivation

the project originates from a request or requirement by another party, such as external stakeholders (investors, creditors, etc.)

internal motivation

the project originates by a desire to better serve a client, better understand phenomena to gain business intelligence, or to perform job responsibilities; the incremental information gained is believed to outweigh potential costs involved with performing the data analyses

objective

the goal of a data analytics project; a statement that details what the project will accomplish

• determine the data necessary to answer questions

• decide what type of analysis is appropriate considering both the data and those questions

two aspects involved in developing a strategy for the planning stage

• internal data

• external data

two categories of data

internal data

data generated within an organization, such as sales and customer data; more easily controlled and verified by an organization

external data

data that are acquired from outside an organization, such as weather or social media data; somewhat riskier to use since we often cannot know if the data are accurate or complete; can provide more insights that internal data alone cannot provide

• descriptive

• diagnostic

• predictive

• prescriptive

four types of data analysis methods

descriptive analytics

data analysis method designed to understand and investigate what is currently happening or what has happened in the past; most common and easily understood analytics method; first analytics method performed to help understand data; examples include - sum, count, average, median, standard deviation

diagnostic analytics

data analysis method designed to understand and reveal why something has happened; examples include - anomaly and outlier detection, trend analysis, pattern recognition

predictive analysis

data analysis method that helps understand and predict what is likely to happen in the future; method uses data, statistical algorithms, and machine learning to identify the likelihood of future outcomes based on historical data; examples include - forecasting, regression analysis, time-series analysis

prescriptive analysis

data analysis method that determines the best course of action to achieve a goal in a specific scenario; helps understand what should happen to meet goals and objectives; these analyses recommend one or more possible courses of action; examples include optimization and what-if analyses

1. prepare data (ETL)

2. build information models

3. explore the data

three steps in the analysis stage of the data analysis process

extract-transform-load (ETL)

the process of retrieving raw data from a source, cleaning, restructuring, and/or integrating them with other data, and then loading the data into software for analysis purposes

the process by which data is retrieved from a source - could be downloading an Excel file or obtaining data from a database

what does extracting in the ETL process refer to?

the process by which data are cleaned, restructured, and/or integrated with other data prior to using it for analysis

what does transforming in the ETL process refer to?

data profiling

process of reviewing the data for possible issues

the process of importing transformed data into the software used to perform analyses; many types of analysis software include Excel and Power BI

what does loading in the ETL process refer to?

building information models

the creation of information needed for analysis purposes, starting from the data collected; examples include - calculations for net income, profit margins, total assets, etc.

identify patterns, trends, or unusual observations; lets one discover, question, and investigate data relationships to successfully execute data analysis objectives

what does exploring data in the analysis stage refer to?

1. interpret results

2. communicate results

two steps in the reporting stage of the data analysis process

the process of reviewing analyses to be sure they make sense based on the project’s objective and that the results are valid and reliable

what does data analysis interpretation in the reporting stage refer to?

• oral presentations

• written reports

• dashboards

• data visualizations

different ways in which the results of a data analysis project can be communicated

data analytics mindset

the professional habit of critically thinking through the planning, analysis, and reporting of data analysis results before making and communicating a professional choice or decision; minimizes the risk of biased or subjective thoughts; businesses can make decisions based on evidence rather than assumptions; individuals must focus on developing skills such as critical thinking, data literacy, technological agility, and communication skills

• ask the right questions

• extract, transform, and load relevant data

• apply appropriate data analytics techniques

• interpret and share the results with stakeholders

a data analytics mindset includes the abilities to …

• critical thinking

• data literacy

• technological agility

• communication skills

four skills in developing a data analytics mindset

critical thinking

disciplined reasoning used to investigate, understand, and evaluate an event, opportunity, or issue; the foundation of a data analytics mindset

reasoning

the human process of logically forming conclusions, judgements, or inferences from facts

data literacy

the ability to read, write, and communicate data in context, including an understanding of data sources and constructs, analytical methods and techniques applied, and the ability to describe the use of case application and resulting value

technological agility

an awareness of the latest technological developments and a willingness to work with new tools and try new things; the ability to quickly and smoothy adapt to or integrate current technologies with newer, different, disruptive, expansive, or convergent technologies

communication skills

being able to concisely, coherently share results to multiple audiences; involves writing clear and effective memos and reports, preparing successful presentations, creating meaningful data visualizations for analysis, and telling compelling data stories

• stakeholders

• purpose

• alternatives

• risks

• knowledge

• self-reflection

six elements of critical thinking when performing data analytics

stakeholders

individuals or groups who may be impacted by and/or have an interest in the outcome of a data analysis project

internal stakeholders

individuals or groups involved in a business’ operations who may be impact by and/or have an interest in the outcome of a data analysis project; includes an organization’s managers and employees

external stakeholders

individuals or groups outside a company who may be impacted by and/or have an interest in the outcome of a data analysis project; includes investors, creditors, regulators, etc.

the main reason for an analysis; knowing the reason for data analyses and articulating specific questions maintains focus on the data and analyses steps that will achieve objectives; clarifies questions, goals, or issues

identifying the purpose in critical thinking refers to

• single system

• multiple system

• no system

three types of questions

single system questions

type of question that employs evidence and reasoning within a specific domain; you can reach a correct answer

multiple systems

type of question that employs evidence and reasoning across several domains; you can reach an informed judgment, but there are better and worse answers

no system

type of question that requires stating a subjective preference; cannot assess the answer

alternatives

different options that are considered and ranked based on the objectives and goals of the data analysis

• obtaining information to inform your reasoning

  • seek out diverse or opposing views

• identify the system(s), concepts, and theories that may apply

what does considering the alternatives in critical thinking involve?

risk

obstacles and challenges to our thinking or analyses

• data risk

• analysis risk

• assumptions risk

• biases risk

four types of risk

data risk

risk that involves choosing inappropriate data, or data that are incomplete or incorrect

analysis risk

risk that involves choosing an inappropriate type of analysis, or apply a data analysis method incorrectly

assumptions risk

risk that involves not understanding or evaluating assumptions about the data or results

biases risk

risk that involves unconscious and mental shortcuts that can affect decisions; closed-minded

knowledge

the concepts that add meaning to the data and analyses

• recognizing when more knowledge is necessary to complete the analysis

• acquiring knowledge from a reliable source

• learning how to correctly apply it to data analysis

identifying knowledge gaps in critical thinking includes

self-reflection

a review of what worked and what did not

• asking “did you have control and resources needed to make progress”

• identifying assumptions

  • degree of randomness vs predictability

• refining questions, goals, or issues

• developing findings or recommendations

• considering the implications

performing self-reflection in critical thinking includes

• clear

• fair

• logical

• actionable

• relevant

critical thinking - decisions that are:

• volume

• variety

• velocity

• veracity

• value

five v’s of data

data volume

the amount of data selected for an analysis project

cloud data services

a popular solution for storing high volume data; these services provide secure data storage on large servers using the organization’s own resources or third-party resources

• easily expandible data storage capacities

• expert data management

• secure data, hardware, and software

• IT expertise and advice

benefits of cloud data services

• private

• public

• hybrid

three types of cloud data services

private clouds

data clouds with restricted access that store the data for one organization or shared between an agreed-upon group of organizations; often include a single large company, a group of supply chain partners, or non-competing independent companies from different industries

• offers high data privacy and data security

• offers high data interactivity

• more customization and adaptability are possible

benefits of data storage for a single organization

• can be the most expensive option of cloud data services

• may be difficult to manage if developed in-house without hiring staff with cloud expertise

costs of data storage for a single organization

• offers high data privacy and data security

• offers high data interactivity between supply chain partners

• more customization and adaptability are possible

• sharing data storage makes it a less expensive option for each company

benefits of data storage that is shared by a set group of non-competing organizations

• can have some restrictions on customization and adaptability

• may involve shared contracts and joint revisions

costs of data storage that is shared by a set group of non-competing organizations

public clouds

data clouds that securely store data from multiple companies on shared servers using virtual server data separators; shared by many organizations who can easily come and go

• lowest individual company cost

• fastest implementation

• most appropriate for small businesses

benefits of public clouds

* least adaptable or customizable cloud data service

cost of public clouds

hybrid clouds

data clouds that offer both private and public cloud data storage, serving the needs of the security and use characteristics of the data involved; allow organizations to choose which data subset is best stored in a private cloud, while storing the rest in public clouds for cost efficiencies

data variety

the diversity of data structures and measurement scales in data that are useful for analysis

data velocity

the speed at which new data points are generated

data veracity

the reliability, or the integrity of the data used for analysis

• whether the data will likely provide insights for the analysis objective, either by itself or as a component in an informational model, such as a net income calculation

• how accurate the data must be to generate useful insights for the objective

• how complete the data must be to generate useful insights of the objective

three things data veracity depends upon

data value

the benefits of certain data given the objective of a data analysis project

data mining

the systematic process/practice of looking for issues or patterns occurring within or between data fields; lets us better understand data behavior and test expectations about data values, patterns, or relationships; most common form of data analytics

blockchain technologies

new online technology that creates a single, shared record of transactions between parties

smart contracts

secure, online shared digital contracts; specific and popular application of blockchain technology; allow business contracts to be securely negotiated, settled, and signed online in a fraction of the traditional time

robotic process automation (RPA)

set of software technologies that record a specific order of human keystrokes and mouse activity steps within or across digital applications; used for stable processes where the steps of the process do not change over time; once saved or recorded, these routines can be repeated by simply rerunning the tool

process mining tools

tools that measure the timing and flow of data captured as business events occur; help businesses better evaluate process efficiency and effectiveness

continuous auditing technologies

programmed modules embedded in an organization’s AIS that evaluate transactions as they occur throughout the year; routine transactions are typically captured and tested in automated routines for consistent processing; results in more accurate assessment of risk and reduces audit workload at year end

textual analysis

the process of analyzing word choice in footnotes, investor communications, blogs, social media postings, and other documentation

textual mining

the process of transforming text into a format that can be used for analysis; evaluates word choice and usage to reveal insights about honesty, transparency, intent, and sentiments in communications by management, elected officials, and customers

cognitive technologies

artificial intelligence technologies that use algorithms that mimic the human thought process; can range from simplistic decision models to dynamic, complex models

data set

a collection of data columns and rows available for analysis

relational database

a collection of logically related data that can be retrieved, manipulated, and updated to meet users’ needs; structured tables composed of rows and columns

table (files)

how data related to an object of interest are stored and linked in a relational database; can be linked together with relationships

files

a database is a collection of …

rows and columns

tables are comprised of…

one record or instance of the entity

what does each row in a table represent?

records

data in the rows of the data set representing instances of the phenomena being captured; represent the collection of columns that hold the descriptions of a single occurrence of the data set’s purpose

instance

a specific, unique representation of the entity; rows of a data set