BUS612 Study Guide
| Site: | Saylor Academy |
| Course: | BUS612: Data-Driven Communications |
| Book: | BUS612 Study Guide |
| Printed by: | Guest user |
| Date: | Monday, May 20, 2024, 3:14 AM |
Navigating this Study Guide
Study Guide Structure
In this study guide, the sections in each unit (1a., 1b., etc.) are the learning outcomes of that unit.
Beneath each learning outcome are:
- questions for you to answer independently;
- a brief summary of the learning outcome topic; and
- and resources related to the learning outcome.
At the end of each unit, there is also a list of suggested vocabulary words.
How to Use this Study Guide
- Review the entire course by reading the learning outcome summaries and suggested resources.
- Test your understanding of the course information by answering questions related to each unit learning outcome and defining and memorizing the vocabulary words at the end of each unit.
By clicking on the gear button on the top right of the screen, you can print the study guide. Then you can make notes, highlight, and underline as you work.
Through reviewing and completing the study guide, you should gain a deeper understanding of each learning outcome in the course and be better prepared for the final exam!
Unit 1: Defining the Business Objective and Sourcing Data
1a. Identify, define, and apply data analysis processes
- What is the four-step method for guiding data-driven decision-making (DDDM)?
- What are the six steps in the lifecycle of a data analysis project?
- What differences are there between the analysis and the conclusions that can be drawn from qualitative and quantitative data?
Data-driven decision-making is collecting data, extracting patterns and facts from that data, and using those facts to guide decisions. Utilizing data in decision-making is superior to using a person's, group's, or organization's intuition and can result in making better decisions, generating new business insights, and identifying new business opportunities.
To begin making data-driven decisions, the organization must start with a clear objective as to what they are trying to accomplish. It could be increased sales, reduced manufacturing costs, improved process efficiency, or any number of measurable outcomes. Once the objective(s) have been determined, the organization would gather and analyze the available data to make decisions. After the decision is implemented, it is important to determine if the results were validated by the analysis.
The objective needs to be a well-defined business challenge or query, such as determining if there is a relationship between x and y. The organization would then determine its data needs. This data has to be very robust and provide outcomes over a variety of scenarios over a consistent period of time. The results create the knowledge the organization needs to make its data-driven decision on how to proceed in the future.
Communication of analysis results with visualizations is a key component of the Data Analysis Lifecycle. Each step in the life cycle must be performed properly for the visualizations to have meaning and lead to good decisions. Understanding the steps that lead up to creating visualizations are critical to the project's success.
The 6 Steps in the Data Analysis Lifecycle are:
- Understanding the business issue(s)
- Understanding the data
- Preparing the data
- Exploratory analysis and modeling
- Validation
- Visualization and presentation
The goal of the analysis is to highlight useful information, suggest conclusions, and support decision-making. How the data is analyzed is dependent on the type of data (qualitative vs. quantitative).
To review, see 1.1: Data Analysis Processes.
1b. Apply data analysis process
- How is big data analytics used to help make better decisions?
- How do organizations use big data analysis to meet their business objectives?
Big data is data that is so large or complex that traditional data processing methods are inadequate. How it is used differs based on the situation, but in all cases, it has the following characteristics: volume, which is the amount of data; velocity, which is the speed of data delivered; and variety, which is the different types of data. When properly harnessed, big data can provide insights that a typical organization's internal data may not.
Big Data provides organizations with unprecedented opportunities to tap into their data to mine valuable business intelligence to uncover hidden patterns, unexpected relationships, and market trends or reveal preferences that may have been difficult to discover previously. With this information, organizations can make better decisions about production, financing, marketing, and sales than before.
Review how organizations use big data stream analysis to analyze consumer sentiments in Big Data Stream Analytics for Sentiment Analysis.
1c. Examine effective business analysis objectives
- What is business intelligence, and how does it affect businesses today?
- What are the five processes that comprise business intelligence architecture?
Business intelligence (BI) is the concepts and methods that support decision-making through information analysis, delivery, and processing. In many cases, BI is considered the data analysis, reporting, and query tools that help users derive valuable information.
Start-up companies also use business intelligence systems in their decision-making processes. The objectives are the same as that of an established company, which include using business intelligence, what companies need to use it successfully, and its applications in a start-up, beginning the data collection/gathering, and ending with data presentation.
The following processes comprise BI architecture:
- Data collection – the operational systems that provide the required BI data
- Data integration – involves the ETL (extract-transform-load) functions needed to transfer the data from the source into a format compatible with other data stored in the data warehouse
- Data storage – the data warehouse or data mart in which the data is stored
- Data processing – includes the concepts and tools utilized in the evaluation and analysis of the data
- Data presentation – is the process of preparing and presenting the analysis results
This figure provides an overview of the individual processes and the components of each process step.
Review the importance of business intelligence and the required process in Using BI and Decision-Making Process in Start-ups.
1d. Create effective business analysis objectives
- What are three types of objectives that can be deployed in marketing research?
- What are the five steps in the marketing research process?
How the data is collected is just as important as the output from the analysis. This is the approach taken for years for marketing research and should be the starting point for any data analysis initiative. The three types of objectives that can be deployed in marketing research are:
- Causal research - used for testing cause-and-effect relationships, typically through estimation.
- Descriptive research – used to assess a situation in the marketplace, such as product potential or consumer attitudes.
- Exploratory research – used to better define a problem or scout opportunities and includes in-depth interviews and discussion groups.
The marketing research process has six steps:
- Define the problem
- Develop an approach to the problem
- Formulate the research design
- Collect data
- Prepare and analyze data
- Prepare and present the report
1e. Evaluate effective business analysis objectives
- What are the key components of effective business analysis objectives?
- How can you evaluate the relevance and alignment of business analysis objectives with organizational goals?
- What criteria can assess the clarity and measurability of business analysis objectives?
- How can you determine the feasibility and achievability of business analysis objectives?
Evaluating effective business analysis objectives involves assessing the quality and effectiveness of objectives set for conducting business analysis within an organization. Key topics include the importance of business analysis objectives, criteria for evaluation, relevance and alignment, clarity and measurability, and feasibility and achievability.
Business analysis objectives serve as guiding principles, defining the purpose, scope, and desired outcomes of the analysis process. Certain criteria, such as specific, measurable, achievable, relevant, and time-bound (SMART), can be applied to evaluate their effectiveness. Objectives should be clear, focused, and aligned with the goals and strategic priorities of the organization. Assessing their relevance to the specific needs and challenges of the organization is essential, ensuring they contribute to overall success. Clarity and measurability are crucial aspects, necessitating clear and concise language to avoid ambiguity and identifying specific metrics or indicators to assess progress and success. Evaluating feasibility and achievability involves considering available resources, expertise, and capabilities within the organization, ensuring objectives are realistic and achievable within given constraints.
By understanding and applying these concepts, individuals can effectively evaluate business analysis objectives, ensuring they are well-defined, aligned with organizational goals, and contribute to successful business analysis outcomes.
Effective business analysis objectives should be specific, measurable, achievable, relevant, and time-bound (SMART). They should clearly define the desired outcomes, provide a clear direction for the analysis, and align with the overall purpose and goals of the organization. It is important to evaluate the relevance of business analysis objectives to ensure they address the specific needs and challenges of the organization. Objectives should align with the strategic priorities and business goals, contributing to the organization's overall success.
Clarity is essential for understanding the intended outcomes and actions required for successful analysis. Objectives should be written in clear and concise language, avoiding ambiguity. Measurability involves defining specific metrics or indicators that can be used to assess progress and success. Business analysis objectives should be evaluated for feasibility and achievability. This involves assessing available resources, expertise, and capabilities required to accomplish the objectives within the given constraints, such as time, budget, and organizational capacity.
To review, see 1.1: Data Analysis Processes.
1f. Compare and contrast the different data gathering methods and models
- What are the key differences between data-gathering methods and models?
- How do different data-gathering methods affect the quality and reliability of the collected data?
- What are the advantages and disadvantages of using specific data-gathering models?
- How do data-gathering methods and models vary across different fields or industries?
When comparing and contrasting different data collection methods, understanding the concepts below is essential to prepare for effective data collection and analysis.
Data gathering methods refer to the techniques or approaches used to collect data, while data gathering models are frameworks or structures that guide the data collection process. Data-gathering methods differ regarding the tools, procedures, and techniques employed, whereas data-gathering models differ in their underlying principles and frameworks.
The choice of data-gathering methods significantly affects the quality and reliability of collected data. Some methods may introduce biases, while others may ensure more accurate and representative data. Each data-gathering model has its strengths and weaknesses. Some models, like surveys or interviews, allow for direct participant interaction, while others, such as observational studies or experiments, offer more control over variables.
Data gathering methods and models can vary across fields or industries based on specific requirements, data types, and ethical considerations. Refer to the course resources for visual aids, examples, and case studies that illustrate the differences and applications of various data-gathering methods and models.
To review, see Data Modeling and Data Analytics.
1g. Illustrate how data collection impacts the final analysis
- How does data collection influence the outcome and accuracy of the analysis?
- What are some common challenges or biases that can arise during data collection?
- How can the choice of data collection methods affect the types of analysis that can be performed?
- How do data quality and representativeness impact the reliability of the final analysis?
Data collection plays a crucial role in shaping the outcome and accuracy of the analysis. The collected data's quality, relevance, and representativeness directly impact the conclusions and insights drawn from the analysis. Reliable and well-collected data enhance the accuracy and reliability of the final analysis. In contrast, poor data collection practices can introduce errors and biases that undermine the validity of the results.
Several challenges and biases can arise during data collection. Non-response bias occurs when certain groups are less likely to participate, leading to skewed or incomplete data. Sampling bias occurs when the sample does not accurately represent the entire population, resulting in a distorted analysis. Measurement errors, such as data recording or collection inaccuracies, can introduce noise and inaccuracies into the dataset. Awareness of these challenges is essential for implementing strategies to mitigate their impact and ensure data integrity.
The choice of data collection methods significantly affects the types of analysis that can be performed. Different methods, such as surveys, interviews, or experiments, offer strengths and limitations. Surveys provide quantitative data that can be subjected to statistical analysis, while interviews allow for capturing qualitative insights and nuanced perspectives. The selection of an appropriate data collection method should align with the research objectives and the analytical approaches planned, as different methods provide different types of data and may be better suited for specific research questions.
Data quality and representativeness are critical factors influencing the reliability of the final analysis. High-quality data, free from errors and biases, ensure accurate findings and valid conclusions. Representativeness refers to the extent to which the collected data accurately reflects the characteristics of the target population or sample. Data that is representative of the intended population enhances the generalizability and applicability of the analysis results. Ensuring data quality and representativeness is crucial for conducting reliable and robust analyses.
By understanding how data collection influences the outcome and accuracy of the analysis, recognizing common challenges and biases, considering the implications of data collection methods on analysis types, and valuing data quality and representativeness, students will develop the skills necessary to conduct rigorous and trustworthy analyses.
The data collection process directly influences the outcome and accuracy of the analysis. The collected data's quality, relevance, and representativeness shape the conclusions drawn and the insights gained. Various challenges and biases can arise during data collection, such as non-response bias, sampling bias, or measurement errors. Recognizing these issues is essential to mitigate their impact and ensure the validity of the analysis.
Different data collection methods, such as surveys, interviews, or experiments, offer distinct strengths and limitations. The choice of method should align with the research goals and the type of analysis intended to be performed. Selecting an appropriate method is crucial in generating reliable and meaningful results. The collected data's quality and representativeness significantly affect the final analysis's reliability. Data accuracy, completeness, and relevance are essential for drawing valid conclusions and making informed decisions based on the analysis.
To review, see Data Modeling and Data Analytics.
Unit 1 Vocabulary
This vocabulary list includes the terms you will need to know to successfully complete the final exam.
- achievability
- big data
- business intelligence (BI)
- causal research
- clarity
- data-driven decision-making
- data collection
- data integration
- data presentation
- data processing
- data storage
- descriptive research
- exploratory research
- feasibility
- interview
- measurability
- measurement error
- non-response bias
- objective
- sampling bias
- SMART
- survey
- variety
- velocity
- volume
Unit 2: Data Analysis
2a. Identify, define, and differentiate data analysis methods and models
- What are the differences between the four analytical models to frame tactical and strategic questions?
- What analytics domains can be deployed in an organization?
There are four primary stages of analytics development, plus cognitive analytics, for five total stages. A different question can be answered at each stage, with the early stages delivering hindsight and the later stages delivering foresight. The value of the results improves as an organization moves along the spectrum along with the complexity of implementation.
The four primary stages for tactical and strategic analytical model managers may use to frame analytics are:
- Descriptive analytics: what happened?
- Decision / diagnostic analytics: why did it happen?
- Predictive analytics: what will happen?
- Prescriptive analytics: how can we make it happen?
As the organization progresses from one stage to another, the benefits of the analysis increase while the implementation complexity increases. The focus also shifts from what happened in the past to what we can do to impact the future. Stage 5, cognitive analytics, includes artificial intelligence analytics.
Analytics can influence nearly every domain in an organization, including finance, marketing, talent, customers, risk management, transportation, and sales.
Here are some industry examples of questions that can be answered at each analytics stage.
These stages of analytics development can be implemented as an organization proceeds along the continuums of the DDDM Change Model. The less complex and past-focused stages can usually be implemented in the early phases. In contrast, organizations in the advanced implementation phases can usually perform the more complex and future-focused stages.
To review, see The Stages of Analytics Development.
2b. Evaluate, classify, and summarize data analysis findings (facts and insights)
- What are the differences between quantitative and qualitative data?
- How and when should you use quantitative versus qualitative data?
Data is at the heart of DDDM. Numerous types of data and sources must be included in a robust DDDM initiative. Every type of data must be extracted from a source, transformed into a standard format acceptable for the data warehouse, and then made available for analysis. Once data is prepared, analytics are deployed to create both hindsight and foresight analytics. You will need to understand these types of analytics and how they relate to successful DDDM initiatives.
Quantitative data is based on counting or measuring the attributes of a population. They are always numbers that specify weight, height, length, population, etc. Quantitative data can be discrete, resulting from counting with only certain numerical values, or continuous, resulting from measuring with various values.
Qualitative data is based on categories or descriptions of a population. They are usually words or letters, such as color, street name, automobile name, etc. Qualitative data includes the color of hair, year in college, month, etc.
Quantitative and qualitative data can both be used to summarize frequency distributions. Since quantitative data are always numeric, it is more often utilized for descriptive (summary) statistics than qualitative data. Quantitative data can also be used to inform a broader understanding of a population through inferential statistics.
To review, see The Difference between Qualitative and Quantitative and Qualitative and Quantitative Research.
2c. Apply data-driven knowledge, skills, and abilities to real-world datasets
- How are data-driven decisions used in everyday life?
More and more, our personal and professional daily decisions are based on data we have accessed from several sources. Businesses have been tapping into this resource for years to determine everything from who will most likely buy a product to who they should hire. A culture of data helps organizations make more informed and more accurate decisions to better utilize their resources.
Sociologists have used qualitative research methods to conduct research and obtain data to explain, predict or control an aspect of social reality. These research methods are increasingly used in the business world to examine and explain consumer behavior and other social interactions that may impact a business.
Before preparing any data analysis, you must "know" your data. This involves understanding the distribution of the data elements to help you determine what you need to analyze further. A single-variable data profile would yield a sample distribution, such as in the table below. In the profile example, the story isn't that the North region has the most sales but that the North region has more sales than the other three regions combined.
| Region | Sales |
|---|---|
| North | 60% |
| South | 22% |
| East | 13% |
| West | 5% |
| Region | Product A | Product B | Product C | Product D |
|---|---|---|---|---|
| North | 125,000 | 375,000 | 1,000,000 | 500,000 |
| South | 90,000 | 175,000 | 75,000 | 100,000 |
| East | 30,000 | 80,000 | 100,000 | 50,000 |
| West | 40,000 | 30,000 | 10,000 | 20,000 |
To review, see Research Design and Data-Driven Decisions.
Unit 2 Vocabulary
This vocabulary list includes the terms you will need to know to successfully complete the final exam.
- cognitive analytics
- decision / diagnostic analytics
- descriptive analytics
- frequency distribution
- predictive analytics
- prescriptive analytics
- qualitative data
- quantitative data
Unit 3: Data Visualization Principles and Processes
3a. Select appropriate visualizations to explore data findings
- What is the difference between presentation graphics and exploratory graphics?
Data visualizations are crucial in data analysis, including cleaning, exploring structure, outlier detection, trend identification, cluster identification, pattern recognition, evaluating modeling output, and result presentation. They are indispensable tools for exploratory data analysis and mining, ensuring data quality checks and enhancing an analyst's understanding of the data's structure and characteristics.
Presentation graphics are usually a select number of graphics created for any number of people and need to be well-designed and well-created with an effective explanatory text, either verbally or textually. They are used to convey known or summarized information.
Exploratory graphics can include several graphics created for an individual such as yourself. They don't need to be perfect but provide alternate views and additional information.
To review, see Why Is Data Visualization Important? and Presenting Data in Meaningful and Interesting Ways.
3b. Encode and annotate data visualizations
- What are the best practices for creating data-driven visualizations free of unnecessary distractions?
For visualizations to be the most effective, they must be data-driven, clear and accurate, and free of items that may distract the viewer from the message.
Interactive visualizations enable the viewer or presenter to drill down into the available data to provide additional insights.
To review, see Data Visualization.
3c. Evaluate visualization techniques
- Name the challenges in visualization methods for analyzing big data.
- Identify the purposes of big data visual representation.
- Explain the importance of rigorous analysis in the creation of data visualizations.
Visualizations are valuable tools for analyzing diverse information that is graphically formatted. Effective visualization tools consider the human brain's cognitive and perceptual properties. They aim to enhance clarity, aesthetics, and comprehension of displayed information, enabling individuals to make sense of large data sets and interact with them. In big data, visual representation serves significant purposes such as identifying hidden patterns or anomalies, facilitating flexible searches, comparing units to determine relative differences in quantities, and enabling real-time human interaction.
However, it's important to recognize that data visualization is not an ultimate objective. Instead, it is an integral part of a larger process that involves gathering, formatting, and analyzing data to generate meaningful insights. Neglecting this process is ill-advised. While using data visualization to present enticing infographics about people's beverage preferences may have minimal consequences in terms of persuasion, it becomes more critical when visualizations are employed to support or oppose public policy decisions. Misuse of data visualization for persuasive purposes can have profound implications. Without rigorous analysis, data visualization becomes, at best, mere rhetoric and, at worst, potentially harmful. A responsible approach is essential to ensure accurate interpretation and mitigate potential negative outcomes.
To review, see Describing Data.
Unit 3 Vocabulary
This vocabulary list includes the terms you will need to know to successfully complete the final exam.
- data visualizations
- exploratory graphics
- presentation graphics
Unit 4: Visualization Tools and Techniques
4a. Compare visualization tools and apply visualization techniques
- What goals can be accomplished by using visualizations?
- What are the tools and capabilities of a robust visualization tool such as PowerPoint?
Visualization tools are an important nonverbal aspect of your presentation that is completely in your control. The purpose of each visualization should be clear and almost speak for itself. Visualizations can significantly develop a presentation and should have a specific purpose the audience can easily recognize. They can also provide emphasis, effectively highlighting keywords, ideas, or relationships for the audience. They can also support a position by utilizing recognized authorities or multiple sources in graphic form. Clarity should be a key prerequisite for using visualizations. Several ways to improve clarity include limiting text on slides, using the right font size, using key images, or presenting the same data in two distinct formats, such as a line graph followed by two pie graphs. The central goal is to ensure the visualization is clear.
PowerPoint can be used to present a variety of visualizations. Its ability to integrate with numerous other applications and formats makes it an essential tool to master. Approach your visualizations with an iterative mindset, recognizing how you initially envisioned it may look completely different from the final product.
To review, see Visual Aids.
4b. Describe data findings through visual charts, graphs, and dashboards
- How would you compare the different manners to present data numerically and graphically?
- What are the differences between a data set's mean, median, and mode?
Data has been collected from surveys or experiments; it needs to be summarized and presented in a way that will be meaningful to the reader. It can be presented in numerical summaries or graphical presentations.
The center or midpoint of a data set helps describe its location. The mean and median are the two most widely used measures of the data 'center'. The mean, also called the average, is the total values divided by the number of values. The median is the middle number that splits the ordered data set into two equal parts. It is often used when the data has extreme values or outliers since precise numerical values do not affect it. Another measure of the center is the mode, which is the most frequent value. A data set can have more than one mode as long as the values have the same frequency.
Example of a Mean
Here is an example that uses the number of cans of paint bought at a home goods store, sorted by color. The total number of cans of paint bought at a home goods store for each color is recorded as follows:
Red: 12
Blue: 9
Green: 6
Black: 14
White: 18
Silver: 21
Gray: 20
Other Colors: 15
Adding these values, we obtain a total of 115 cans of paint. Dividing by the number of color categories (8), we find that the mean number of cans of paint bought is 115/8 = 14.375. Rounding this to one decimal place, the mean is 14.4.
Example of a Median
To calculate the median, we list the number of cans of paint bought in each color in ascending or descending order:
6 9 12 14 15 18 20 21
Since there is an even number of color categories (8), the median is the average of the two middle values, which are 14 and 15 in this case. Adding these two values and dividing by 2, we find that the median number of cans of paint bought by color is (14 + 15)/2 = 14.5.
In this particular example, the mean and median values are fairly close to each other, indicating a relatively balanced distribution of cans of paint bought across color categories.
A frequency table lists each item and the number of times the item appears. The level of measurement is how a data set is measured and can vary with the type of data being analyzed.
To review, see Describing Data.
4c. Organize data visualizations to convey messages and main points
- What are the best practices for creating meaningful and effective data visualizations?
- What thought process is involved in summarizing survey data into meaningful visualizations?
The human visual perception system is very powerful. When utilized properly, visualizations allow for a greater understanding of the data. Analysts need to understand how to best communicate visual information for greater understanding. It is important to keep it simple, be communicative and powerful in your data storytelling.
With better software, faster processors, and cheaper memory, it has become easier to create and iterate visualizations. With this power comes responsibility, as it is very important to create good visualizations that clearly articulate the point the analyst is trying to make. Visualizations can be effective or ineffective, which can generate very strong feelings either way.
When summarizing data for a presentation, it is important to take an iterative approach to the graphics, colors, and audience. Remove as much extraneous information as possible to enable the audience to quickly and easily reach the best conclusion from the information presented.
To review, see Best Visualization Practices.
Unit 4 Vocabulary
This vocabulary list includes the terms you will need to know to successfully complete the final exam.
- center
- frequency
- frequency table
- mean
- median
- mode
- PowerPoint
- visualization tools
Unit 5: Evaluating Data Visualizations
5a. Plan informational presentations
- What are the different sources of data for an information presentation?
- How would you compare the two ways of obtaining data? How can each one be properly used?
When you present data, start with a story: a message or a point you are trying to make. The message should be crafted for the intended audience, with a clear narrative flow and visuals that connect and convey that message.
A lot is involved in creating an informational presentation. Once the message and the point have been determined, the data required to address the query must be obtained. Using data as a source can help with all your research project's information needs. It can help to:
When planning informational presentations, it is important to evaluate various data sources. This involves comparing different methods of data acquisition and assessing their appropriate utilization.
- Acquiring background information: Data is a valuable resource for gaining a deeper understanding of the subject matter.
- Answering research questions: The evidence provided by data aids in determining the most suitable answer to your research question.
- Convincing the audience: Data often supplies evidence that supports your research question's answer, ensuring its credibility or reasonability.
- Describing the research context: Data helps provide a comprehensive picture of the circumstances surrounding your research question.
- Reporting existing perspectives: Data allows you to incorporate the opinions and findings of other researchers into your presentation.
There are two primary methods of obtaining data. The first involves utilizing pre-collected and analyzed data. The second method entails collecting data directly through environmental observations, surveys, interviews, laboratory or field measurements, or even electronically recorded data from computers or machines.
The most effective criterion for selecting a display format for your data is whether it enhances your understanding and that of your audience beyond what can be conveyed through text alone. A well-designed display should enable you to notice key points that might have been overlooked.
To review, see Data as Sources.
5b. Identify visualizations that convey data findings
- What are some key considerations when selecting visualizations for data findings?
- How do different visualizations (such as bar charts, line graphs, and scatter plots) convey specific data findings?
- How can visualizations enhance the understanding and communication of data findings?
- What are some best practices for creating effective and informative visualizations?
Identifying suitable visualizations is crucial for effectively conveying data findings. Considerations for visualization selection, types of visualizations, enhancing understanding and communication, and best practices for visualization creation are important topics to identify visualizations that convey data findings.
When selecting visualizations, it is essential to consider factors such as data type, the purpose of the analysis, the audience, and the message to be communicated. Various visualizations, including bar charts, line graphs, scatter plots, and heatmaps, have distinct strengths and limitations that align with specific data findings and relationships.
Visualizations enhance understanding and communication by providing visual cues, patterns, and trends that are more accessible than raw data alone. They facilitate concise and impactful communication of complex findings to diverse audiences.
Following best practices for visualization creation, such as the appropriate use of colors, labels, titles, axes, and legends, ensures that visualizations are informative, accurate, and visually appealing. Simplicity, clarity, and alignment with the intended message are emphasized, while clutter, misleading representations, and unnecessary complexity should be avoided.
Addressing the guiding questions, considerations for visualization selection include the data type, analysis purpose, audience, and effective communication of findings. The impact of data collection on analysis quality influences the suitability of visualizations, emphasizing the importance of accurate and representative data for meaningful and reliable findings.
Common challenges in data collection, such as non-response bias, sampling bias, and measurement errors, can introduce biases and inaccuracies that affect the validity and reliability of findings and, subsequently, the choice of visualizations.
Analytical approaches such as statistical analysis, exploratory research, and business intelligence contribute to data-driven decision-making by providing insights and patterns that can be visually represented. Statistical analyses identify relationships and patterns, exploratory research uncovers trends, and business intelligence supports strategic decision-making through data-driven visualizations.
To review, see Data Visualization to Promote Public Awareness and Creating Good Visualizaitons.
5c. Apply data narratives to visualizations
- What approaches to creating credible and effective visualizations to convey data findings are most effective?
- How can different disciplines collaborate on digital projects for educational and commercial uses?
Good visualizations don't happen by accident. The user/creator must take a systematic approach to creating credible and effective visualizations. The user must systematically think about visualizations and how to reason between different visualization design choices.
Data visualizations can be used in multiple disciplines to report data outcomes. It has been used for many cross-disciplinary applications, such as computer science, marine science, and art, to collaborate on digital projects to educate the public on ecological issues.
To review, see Creating Good Visualizaitons.
5d. Evaluate when visualizations are needed to explain a concept
- Why is reducing the need for an audience to interpret the key to create an effective presentation?
Reducing the need for the audience to interpret the findings in visualization is the key to an effective presentation. This can be accomplished by the type of chart used or by highlighting key points through color choices.
There are five distinct levels of user engagement in information visualization.
- Expose (viewing): At this level, the user comprehends how to interpret and engage with the presented data, understanding the basic mechanisms of the visualization.
- Involve (interacting): Moving beyond passive viewing, the user actively interacts with the visualization, manipulating the data to gain a deeper understanding and explore different perspectives.
- Analyze (finding trends): At this level, the user delves into the data, conducting analyses to identify patterns, trends, and outliers within the visual representation.
- Synthesize (testing hypotheses): Building upon the analysis, the user reaches a level where they can generate and evaluate hypotheses based on the data presented. This stage involves synthesizing information to formulate and test various hypotheses.
- Decide (deriving decisions): The highest level of user engagement is achieved when the user can make informed decisions and draw conclusions by critically evaluating different hypotheses and considering the implications of the presented data.
Visualization embellishment is useful for the comprehension and memorability of charts. When plain and embellished charts were compared, a user's accuracy in describing them was similar, but their recall after a two- to three-week gap was significantly better.
As an illustration, consider two distinct approaches to visually enhancing the same data set. The first example depicts a graph incorporating embellishments while preserving a bar chart's recognizable characteristics. The second example replaces the bars with a silhouette of a person positioned alongside a beverage, where the drink's height corresponds to the original bar's height. Additionally, this approach utilizes color to highlight and emphasize the data points.
To review, see Storytelling and Visualization.
5e. Evaluate the proper visualization for a given data set
- How do visualizations make large amounts of data easier to understand?
- How do visualizations reveal data at several levels of detail?
Well-crafted data visualizations present data as easily understood images. When done well, they enable the viewer to quickly perceive insights they may have missed if presented in summary tables and spreadsheets. Good data visualization does not only convert large amounts of data into images; when done well, it engages the viewer and tells a story.
Well-crafted visualizations present complex ideas or results and communicate them clearly, precisely, and efficiently. Visualizations should:
- Show the data
- Have the viewer on the substance instead of the methodology
- Avoid distorting the data
- Present many numbers in a small space
- Make large data sets easier to understand
- Present the data at several levels of detail, from a high-level overview to a deep data dive
To review, see Data Visualization Design and Principles.
5f. Present data to a non-technical audience in an ethical manner
- What are the uses of narrative charts in storytelling?
- What are the key components of an effective narrative chart?
Narrative charts are another form of data visualization. It combines the best of data visualizations, storytelling, and some of our favorite historical or fictional tales. They describe an experience, event, or a sequence of events in the form of a story. Often, they are used to depict the structure of a movie or book.
The key is that these narratives are usually created for non-technical audiences, and they must follow the sequence of events from the original story source.
To review, see Narrative Charts Tell the Tale.
Unit 5 Vocabulary
This vocabulary list includes the terms you will need to know to successfully complete the final exam.
- analyze
- decide
- direct data collection
- expose
- involve
- narrative
- narrative chart
- pre-collected data
- story
- synthesize
- systematic approach
Unit 6: Storytelling with Data
6a. Identify, define, and apply basic principles and methods of graphical design
- Why are data visualizations essential for exploratory data analysis and data mining?
Data visualizations play a crucial role in data analysis, including data cleaning, exploration of data structure, outlier and anomaly detection, trend and cluster identification, identification of local patterns, evaluation of modeling output, and effective presentation of results. They are particularly essential for exploratory data analysis and mining, as they allow analysts to assess data quality, enhance their understanding of the data's structure and characteristics, and gain valuable insights before delving deeper into the analysis process.
To review, see Using Data Visualization.
6b. Identify and define visual types, data encoding, and textual annotations for visual communications and storytelling
- What is the difference between linear, user-directed, parallel, and random-access storytelling?
The sequence or order of events makes a big difference in storytelling and refers to the path the viewer takes in the visualization. Stories can be presented in several ways, including:
- Linear, where the story sequence path is linear in order and is prescribed by the author
- User-directed, where the user selects a path from alternatives or creates their own path
- Parallel, where several paths can be visualized or followed at the same time
- Random access, where there is no prescribed path. This is more commonly referred to as an "overview" path.
To review, see How to Present Data and Presenting Using Storytelling.
6c. Evaluate how the use of visual communications improves the conveyance of ideas, information, and interpretation of data
- What basic storytelling principles help the audience better understand analytics results?
- What are the important components that differentiate a dashboard from other data visualizations?
Presenting the results of an analysis as part of a story could help the audience better understand the results. The best stories utilize the basic storytelling principles, which are:
- Change is more important than chronology
- "Twists" make the story interesting
- Figure out the important points before you start
- Intrigue and delight
- No data dumps
- Create an immersive experience for your audience
With the explosion in visualizations, dashboards have become a very popular tool for organizations to manage their business. Dashboards typically focus on high-level performance metrics that provide key indicators that can alert managers when action is needed. There are several basics of creating effective dashboards and what should and should not be included.
To review, see What’s in a Dashboard?.
Unit 6 Vocabulary
This vocabulary list includes the terms you will need to know to successfully complete the final exam.
- dashboard
- linear
- parallel
- random access
- storytelling principles
- user-directed