Learn the skill of makings sense out of data in sexy storylines and communicating them to your audience. The importance of data cleaning the most crucial process to reveal relevant information lies in this chapter. Explain the phases of Exploratory Data Analysis, which include the power of quickly obtaining insights from the raw data. Learn a wide variety of ways to visualize data, ensuring that critical insights are shared clearly and interestingly in the end result.
Learning Objectives:
- Learn the importance of ethics and accessibility in data visualization.
- Explore in some depth the part EDA has for the data practitioner in developing and sharing stories from raw data.
- Sensibly, ethics and accessibility regarding data stories should be explored closer.
- Gain knowledge about exploratory data analysis (EDA) and get to know its various advantages for understanding data.
- Appreciate the need for aligning EDA methods with business objectives within PACE.
- Identify the six main stages: discovering, structuring, cleaning, joining, validating, and presenting, as part of the EDA process.
- Explain how data analysis creates narratives for data professionals from raw sources of data.
PRACTICE QUIZ: TEST YOUR KNOWLEDGE: TELL STORIES WITH DATA
1. Fill in the blank: The presenting stage of exploratory data analysis involves sharing _____, which can include graphs, charts, diagrams, or dashboards.
- databases
- data visualizations (CORRECT)
- data frames
- datasets
Correct: The preliminary stage of exploratory data analysis involves the presentation of data visualizations-such as graphs, charts, or diagrams or even dashboards-to promote the insights gleaned from such visualizations. It thus entails showing major findings from the data to an audience clearly, compellingly, and accessibly.
2. During which exploratory data analysis practice might a data professional familiarize themself with the meaning of column headers in a dataset?
- Discovering (CORRECT)
- Validating
- Joining
- Structuring
Correct: A data professional considers what the names of the columns mean in a dataset during the discovering phase of explorative data analysis (EDA). At this stage, data professionals get acquainted with the dataset: its organization and identification of the kinds of data it has.
3. If sampled data is organized in such a way that it does not accurately represent its population as a whole, what problem will occur?
- Biased data (CORRECT)
- Unclean data
- Disorganized data
- Unfiltered data
Correct: Biased sampled data can be defined as organizing sampled data in one way and not representing the population correctly as an entire entity. Data bias occurs when preference for or against particular person or group or thing causes systematic distortion of data analytical results leading to inaccurate conclusions.
PRACTICE QUIZ: TEST YOUR KNOWLEDGE: HOW PACE INFORMS EDA AND DATA VISUALIZATIONS
1. What are the primary drivers of a data-driven story? Select all that apply.
- Sales predictions
- Stakeholder theories
- Project goals (CORRECT)
- Project purpose (CORRECT)
Correct: In essence, the insights extracted from data and the stories they weave are powered and developed by the purpose and goals of the project. In fact, the objectives direct the analysis in that they ensure the extracted insights from the data align with the intended outcomes. Consequently, such outcomes become logical and meaningful, taking on what may be considered action-oriented.
2. Fill in the blank: In order to help avoid _____ in the workplace, data professionals share the PACE plan with stakeholders and team members.
- unnecessary meetings
- competition
- miscommunication (CORRECT)
- unintentional bias
Correct: Data professionals would also be inclined to share with their stakeholders and team members the PACE plan, which is meant to avoid possible miscommunications at the workplace. PACE allows teams to communicate, problem-solve, and make decisions in an efficient and expeditious manner on their way towards project objectives.
3. Why is it important to maintain proper scale of a graph’s axes in a data visualization?
- To take advantage of white space
- To tell a more interesting data story
- To change stakeholders’ minds
- To avoid misrepresenting the data (CORRECT)
Correct: To ensure accurate representation of data in a visualization, one should maintain the right scale of axes in the graph. Badly scaled axes may result in the skewing of data that may mislead the viewer in interpreting the data, causing incorrect assumptions and possibly missing the actual view of relationships and patterns in the data.
QUIZ: MODULE 1 CHALLENGE
1. Fill in the blank: The type of data being studied and the _____ guide the order of the six practices of exploratory data analysis.
- size of the dataset
- company mission
- needs of the data team (CORRECT)
- available hardware and software
Correct!
2. A data team leader at a clothing manufacturer reviews a dataset that will be used to decide where to open new retail stores. They conceptualize how their analytics team can most effectively use the dataset. Which exploratory data analysis process does this scenario describe?
- Validating
- Joining
- Discovering (CORRECT)
- Cleaning
Correct!
3. What are the goals of the structuring exploratory data analysis step? Select all that apply.
- Correcting misspellings or other errors
- Prepare data to be effectively modeled (CORRECT)
- Make data easier to visualize and explain (CORRECT)
- Group data in such a way that it accurately represents the dataset as a whole (CORRECT)
Correct!
4. Which of the following statements correctly compare data cleaning to data validation during exploratory data analysis? Select all that apply.
- Cleaning is the process of confirming that no errors were introduced during validation.
- Both data cleaning and data validation involve eliminating any misspellings in the data. (CORRECT)
- Cleaning involves ensuring the data is useful. (CORRECT)
- Validating involves verifying the data is of high quality. (CORRECT)
Correct!
5. Fill in the blank: A data professional discovers that their dataset does not have enough data. Therefore, they choose to add more data during the _____ process.
- Joining (CORRECT)
- structuring
- validating
- cleaning
Correct!
6. What steps may be involved with presenting data insights to others during exploratory data analysis? Select all that apply.
- Remove written descriptions to save people time when viewing the visualizations
- Share a cleaned dataset for additional analysis (CORRECT)
- Ask team members or stakeholders for feedback (CORRECT)
- Make the visualizations available to others for further modeling (CORRECT)
Correct!
7. What are some strategies that a data professional might use to help avoid miscommunication in the workplace? Select all that apply.
- Provide audiences with raw data for their own exploration.
- Share the PACE plan with all stakeholders. (CORRECT)
- Present primary analysis with a working group to get feedback. (CORRECT)
- Understand stakeholders’ most important goals before presenting to them. (CORRECT)
Correct!
8. A data professional works on a project that uses data from a study about farming in Africa. They consider how to use the PACE framework to perform exploratory data analysis practices effectively. Which of the following objectives will this help them achieve? Select all that apply.
- Conform to client expectations by misrepresenting the data
- Confirm that the data represents an appropriate number of African geographical regions (CORRECT)
- Ensure ethical depictions of the farmers represented in the study (CORRECT)
- Maintain focus on the project purpose (CORRECT)
Correct!
9. Fill in the blank: The exploratory data analysis process is_____, which means data professionals often work through the six practices multiple times.
- Supplementary
- Immutable
- transitory
- iterative (CORRECT)
Correct!
10. Fill in the blank: A data professional might add more context to the data during the _____ process by adding information from other data sources.
- Joining (CORRECT)
- structuring
- cleaning
- validating
Correct!
11. Fill in the blank: To avoid _____ in the workplace, data professionals can share initial data findings with a working group to get feedback before providing analyses to all stakeholders.
- competition
- silos
- favoritism
- miscommunication (CORRECT)
Correct!
12. A data professional at a financial investment company familiarizes themselves with a dataset for a new investment project. They consider the meaning of the column headers and how many total data points exist. Which exploratory data analysis process does this scenario describe?
- Validating
- Joining
- Cleaning
- Discovering (CORRECT)
Correct!
13. Fill in the blank: In exploratory data analysis, _____ is the process of augmenting a dataset by adding values from other sources.
- cleaning
- structuring
- validating
- joining (CORRECT)
Correct!
14. A data professional works on a project that uses data from a study about mental health in Europe. They consider how to use the PACE framework to perform exploratory data analysis practices effectively. Which of the following objectives will this help them achieve? Select all that apply.
- Modify the data in order to meet all project deadlines
- Ensure ethical depictions of the mental health subjects represented in the study (CORRECT)
- Maintain focus on key priorities and project purpose (CORRECT)
- Confirm that the data represents an appropriate number of European geographical regions (CORRECT)
Correct!
15. A data professional works in the research and development department of a high-tech firm. They receive a dataset that will be used when creating next year’s products. They review the data and consider key questions about it. Which exploratory data analysis process does this scenario describe?
- Joining
- Discovering (CORRECT)
- Cleaning
- Validating
Correct!
16. What procedures take place during the structuring exploratory data analysis step? Select all that apply.
- Share data with stakeholders
- Organize data columns based on the data within the dataset (CORRECT)
- Transform raw data from the dataset (CORRECT)
- Group data into categories that represent the dataset (CORRECT)
Correct!
17. What processes do data professionals perform during the structuring exploratory data analysis step? Select all that apply.
- Create data visualizations
- Transform raw data. (CORRECT)
- Categorize data into categories representing the dataset (CORRECT)
- Organize raw data (CORRECT)
Correct!
18. What are some best practices associated with visualizing data during exploratory data analysis? Select all that apply.
- Design visualizations specifically to support your personal hypotheses.
- Create visualizations that are ethical, accessible, and representative of the data. (CORRECT)
- Ensure visualizations are guided by the story uncovered by the data. (CORRECT)
- Use data visualizations throughout exploratory data analysis to better understand the data. (CORRECT)
Correct!
19. Fill in the blank: Exploratory data analysis is the process of investigating, organizing, and analyzing datasets and _____ their main characteristics.
- Modifying
- Summarizing (CORRECT)
- preparing
- augmenting
Correct!
20. A data professional works on a project that uses data from a study about teachers in Australia. They apply the PACE framework to perform exploratory data analysis practices effectively. Which of the following objectives will this help them achieve? Select all that apply.
- Conform to stakeholder expectations by misrepresenting the data
- Ensure ethical depictions of the teachers represented in the study (CORRECT)
- Keep project priorities in order (CORRECT)
- Confirm that the data represents an appropriate number of Australian geographical regions (CORRECT)
Correct!
21. Fill in the blank: A data professional discovers that their dataset does not have enough data. Therefore, they choose to add more data during the _____ process.
- Validating
- Structuring
- cleaning
- joining (CORRECT)
Correct!
22. Fill in the blank: To avoid miscommunication in the workplace, data professionals can share _____ with a working group to get early feedback.
- metadata
- changelogs
- initial data findings (CORRECT)
- raw data
Correct!
23. What may be involved with visualizing data during exploratory data analysis? Select all that apply.
- Asking stakeholders to hold their comments until the final official presentation
- Making data visualizations available to team members for further analysis or modeling (CORRECT)
- Considering people with auditory impairments by providing captioned descriptions about the data (CORRECT)
- Considering people with visual impairments by describing the data in detail (CORRECT)
Correct!
24. What is the process data professionals use to investigate, organize, and analyze datasets in order to summarize the data’s main characteristics?
- Storytelling with data
- Exploratory data analysis (CORRECT)
- Data strategy
- Data visualization
Correct: Exploratory Data Analysis (EDA) is the process that data professionals use to expose, organize, and analyze data sets in order to understand their prominent characteristics. It also serves to expose various patterns, associations, and other insights inherent within the data before proceeding to more sophisticated modeling procedures.
25. When a data professional discusses a project plan and company goals with stakeholders, which element of the PACE model are they engaged in?
- Analyze
- Construct
- Plan (CORRECT)
- Execute
Correct: When a data professional discusses a project plan and company goals with stakeholders, they are engaged in the plan element of the PACE model. During planning, data professionals define the scope of the project and identify the informational needs of the organization.