Fundamentals of Data Visualization Coursera Quiz Answers

Get All Weeks Fundamentals of Data Visualization Coursera Quiz Answers

Data is everywhere. Charts, graphs, and other types of information visualizations help people to make sense of this data. This course explores the design, development, and evaluation of such information visualizations. By combining aspects of design, computer graphics, HCI, and data science, you will gain hands-on experience with creating visualizations, using exploratory tools, and architecting data narratives. Topics include user-centered design, web-based visualization, data cognition and perception, and design evaluation.

This course can be taken for academic credit as part of CU Boulder’s Master of Science in Data Science (MS-DS) degree offered on the Coursera platform. The MS-DS is an interdisciplinary degree that brings together faculty from CU Boulder’s departments of Applied Mathematics, Computer Science, Information Science, and others.

With performance-based admissions and no application process, the MS-DS is ideal for individuals with a broad range of undergraduate education and/or professional experience in computer science, information science, mathematics, and statistics.

Enroll on Coursera

Week 01: Fundamentals of Data Visualization Coursera Quiz Answers

What is the primary function of a mark?

  • To indicate the most critical data
  • To represent data attributes
  • To overcome the limitations of statistics

The three basic questions to ask when deconstructing a visualization are:

  1. What data is being shown?
  2. ???
  3. How has the visualization been designed?
  • Why is the data presented?
  • Where did the data come from?
  • When was the data last updated?

What allows us to deconstruct a chart into its constituent pieces?

  • Mappings
  • Grammar of Graphics
  • Visual Channels

What defines the range of values mapped to a given set of axes?

  • Statistics
  • Data
  • Scales

What defines the shape of marks in a visualization?

  • Aesthetics
  • Facets
  • Geometric Objects

I have a dataset that I’d like to visualize that describes major world events over the last year. The dataset has an attribute for the month the event occurred in, the number of countries involved in the event (e.g., January, February), and the genre of event (e.g., political, social, environmental). How would you categorize the data types of these three attributes respectively?

  • Ordinal, Continuous, Categorical
  • Ordinal, Ordinal, Categorical
  • Categorical, Continuous, Categorical

If you divide continuous data into three intervals and represent each interval using either a square, circle, or triangle, the visualization violates the principle of

  • Mapping
  • Effectiveness
  • Expressiveness

What are the inputs and outputs of a visualization mapping?

  • Perception & Visual Channel
  • Data Value & Statistic
  • Data Value & Visual Encoding

When might it be useful to use a log scale?

  • When you need to make your visualization more intuitive
  • When the data has a heavily skewed distribution
  • Log scales are never useful

Why is it misleading to use the diameter of a circle to encode data?

  • Diameter shows a smaller effect than using the radius
  • Diameter is less intuitive to reason about
  • Diameter significantly increases the area of a mark

Which of the following color encoding types should be used to show how far temperature readings are above or below freezing?

  • Diverging
  • Sequential
  • Categorical

In which of the following situations would a rainbow colormap be useful?

  • When you need to ensure your visualization is as intuitive as possible
  • When you need to emphasize the differences between ranges of values
  • When you are not representing continuous data

Why might interaction be undesirable?

  • It doesn’t let us work with datasets that are too large to show everything at once
  • Some interactions may not be intuitive
  • Analysts might have many different questions about their data.

What workflow should interactive visualizations typically follow?

  • *A: Overview first. Zoom and filter. Details on demand.
  • Show as much data as possible. Collapse data on demand.
  • Look at the data. Highlight the data. Show a tooltip.

Which of the following techniques allows a visualization to prevent its readers from getting lost as they interact with their data?

  • Selection
  • Brushing
  • Overview+Detail

What should visualizations typically do in response to a selection interaction?

  • Remove the selected mark from the visualization
  • Make relevant data easier to find
  • Show that part of the graph in more detail

What is the primary reason a visualization may want to arrange data alphabetically along an axis?

  • To make it easier to find data
  • To highlight statistical patterns
  • To cluster similar data

What technique should I use if I want to select interesting data in one chart and filter for only that data in a second chart?

  • Multiple View Coordination
  • Semantic Zoom
  • Seriation

I have a dataset with the following composition:

  • Sales Totals between $10k and $15k per week
  • Date between January 2000 and December 2020

I create a scatterplot of this data with the date on the x-axis and the sales totals on the y-axis. Which of the following y-axis scales is most appropriate to analyze the fluctuations in sales overtime?

  • Sales Totals from $0 – $15k
  • Sales Totals from $10k – $15k
  • Difference from mean Sales Totals from $-5k to $5k

Which of the following is *not* a reason that 3D charts can mislead?

  • Reading data on 3D axes is harder than on 2D axes
  • 3D charts are harder to build than 2D charts
  • 3D charts can occlude important values

Week 02: Fundamentals of Data Visualization Coursera Quiz Answers

Why are tasks important to characterize in visualizations?

  • They allow us to separate the goals of an analysis from the specifics of a domain
  • They allow us to readily define the set of interactions we need to implement
  • They define the knowledge that people build through a given visualization design

How do people typically build knowledge from visualizations?

  • Linearly
  • Collaboratively
  • Iteratively

What aspect of a task defines why someone is performing it?

  • Workflows
  • Purpose
  • Roles

What should come first: task characterization or preliminary design?

  • Task Characterization
  • Preliminary Design
  • You cannot complete one without the other

Why is provenance important to consider when designing a visualization?

  • It defines who your audience is
  • It defines the means by which our visualizations should respond to a given input
  • It defines the knowledge that people have before and after conducting a task

Why are low-fidelity prototypes important for visualization design projects?

  • The foster a bias towards action
  • They closely resemble the target functionality
  • They present a near-perfect vision of the final design to stakeholders

What is the primary advantage of the Five Design Sheets approach compared to a Design study?

  • FDS produces better shared expectations of a project
  • FDS offers more time for iteration
  • FDS is a significantly faster way to start a project

What happens in the Design phase of a Design Study?

  • You characterize the problem space you are working in
  • You outline representations or interactions for your visualization
  • You build a functional prototype

Which of the following projects is best suited for a Design Study approach?

  • Working with your team to visualize how sales volumes for a given product have changed over time.
  • Working with a marketing team to understand what historical strategies have been most effective for a company.
  • Working with a client with limited availability to create a basic mock-up of an internal dashboard of key performance indicators.

Which of the following is a benefit of visualizations over statistics?

  • Visualizations typically communicate patterns in data more intuitively
  • Visualizations typically communicate data more concisely
  • Visualizations typically communicate data more precisely

How might we use knowledge of visual perception to drive visualization design?

  • Perception tells us the best way to visualize our data
  • Perception allows us to make predictions about what people see in data
  • Perception allows us to compute statistics more effectively than other methods

Which of the following information do we not receive information about during our first glance at a visualization (i.e., before attention kicks in)?

  • The largest value
  • The distribution of values
  • The most salient data

How might we best indicate that a set of datapoints are related?

  • Place them all in the spatial envelope.
  • Make them all salient.
  • Make them all the same hue.

I have a dataset that has Temperatures, Elevations, and Snowfall Depths for various mountain tops around the world. If my stakeholders care most about comparing temperatures, what visual channel should I use to communicate temperature?

  • Position
  • The amount of red or blue.
  • Area.

I have a dataset that has Temperatures, Elevations, and Snowfall Depths for various mountain tops around the world. Which combination of visual channels will allow people to best analyze that data?

  • Position, color, and size
  • Position, height, and width
  • Width, color, and area

I have a node-link diagram that has too much visual clutter to read effectively. Which of the following is not likely to reduce the amount of clutter in the graph?

  • Aggregate groups of connected nodes into single, larger nodes
  • Make the graph smaller
  • Organize the nodes such that groups of highly related nodes are close together

Why is uncertainty important in visualization?

  • It provides information about how likely our conclusions are to be correct
  • It limits the number of available visualization designs
  • It allows us to aggregate data

What is a key disadvantage of bar charts with error bars compared to violin plots?

  • Bar charts are more complex than violin plots
  • Bar charts can only represent small datasets
  • People think points below the mean are more likely in bar charts

True or False: Uncertainty visualization can only be used when the data is aggregated by category?

  • True
  • False
  • It depends on the type of visualization used.

Why might it be useful to not explicitly encode uncertainty?

  • Uncertainty visualizations make decision making harder
  • People can infer statistics about a dataset robust to outliers and other potential errors
  • People aren’t good at reasoning under uncertainty

Week 03: Fundamentals of Data Visualization Coursera Quiz Answers

Why might we prefer to evaluate individual components of a visualization (e.g., the algorithms used, performance on specific tasks, intuitiveness of interactions) over evaluating its holistic utility?

  • It isn’t possible to evaluate a tool’s holistic utility
  • To understand the kinds of knowledge someone gains with the visualization
  • To identify specific areas for improvement

Why might we choose to evaluate the holistic utility of a visualization over that of its specific components (e.g., the algorithms used, performance on specific tasks, intuitiveness of interactions)?

  • It isn’t possible to evaluate a tool’s individual components
  • To understand the kinds of knowledge someone gains with the visualization
  • To identify specific areas for improvement

The primary purpose of a visualization is:

  • Transparency
  • Insight
  • Communication

Which of the following is *not* a characteristic of a good insight?

  • Quantitative
  • Deep
  • Unexpected

What metric for assessing insights can best characterize a tool’s value in snowballing (i.e., leading to new, complex discoveries)?

  • Number of Insights
  • Time to Insight
  • Depth of Insights

Identify the best qualitative evaluation method for the following scenario:

  • You want to develop a formative understanding of the kinds of tasks energy analysts conduct with their data sampled from several hundred companies.
  • Journaling Studies
  • Systematic Survey
  • Thinkalouds

Identify the best qualitative evaluation method for the following scenario:

  • You want to develop a summative understanding of specific trade-offs in a dashboard design for web security monitoring for cybersecurity experts as they use the tool.
  • Journaling Studies
  • Semistructured Interviews
  • Thinkalouds

Identify the best qualitative evaluation method for the following scenario:

  • You want to develop a formative understanding of the most pressing needs are for tools for biological sequence analysis.
  • Journaling Studies
  • Semistructured Interviews
  • Thinkalouds

Identify the best qualitative evaluation method for the following scenario:

  • You want to develop a summative understanding of how well a tool for communicating global migration patterns works for biologists.
  • Journaling Studies
  • Semistructured Interviews
  • Systematic Surveys

Which of the following is not a measure collected in qualitative evaluations?

  • Accuracy
  • Insight
  • Anecdotes

For this series of questions, you will design an experiment that tests what visualization type you should use to communicate correlation between two data attributes.

  • Which of the following is a decision making task for this experiment?
  • The graph below shows the relationship between two attributes. Estimate the correlation coefficient of the attributes.
  • The two graphs below show the relationship between different attributes in a dataset. Choose the visualization that shows the highest correlation between the x- and y-attributes.
  • These graphs show the relationship between a target attribute and a proxy attribute that is easier to collect data about. Choose the proxy attribute that best approximates the target attribute.
  • For this series of questions, you will design an experiment that tests what visualization type you should use to communicate correlation between two data attributes.

What would your dependent variable be?

  • The distribution of the visualized data
  • Visualization type
  • Accuracy

For this series of questions, you will design an experiment that tests what visualization type you should use to communicate correlation between two data attributes.

Why might you want to test the type of visualization used as a between-subjects variable?

  • You want greater statistical power
  • You want to test a wider range of designs
  • You want to reduce the variance created by testing different users

What is one advantage of using real data instead of synthetic data in your visualizations?

  • It reflects the distributions of data that people are likely to see when they use your tool
  • It captures a wider range of statistical patterns.
  • It is easier to get real data

Why is formative evaluation important for visualization design?

  • It proves that our solution works
  • It offers grounded feedback into the insight people build with a tool
  • It allows us to test our assumptions as we build a tool

Which of the following is an example of summative evaluation?

  • An experiment comparing how well different visual channels communicate variance to decide which to use in our visualization
  • A journaling study that demonstrates several insights developed using a tool
  • A semistructured interview study that identifies common themes in geophysical communication tools
  • Determine whether the following scenario would best be served by a qualitative evaluation or an experiment:

Determine whether a new solution provides less biased trend estimates than a prior solution.

  • Qualitative/Insight-Based Evaluation
  • Experimental Evaluation

Determine whether the following scenario would best be served by a qualitative evaluation or an experiment:

  • Determine how effectively a target solution communicates patterns in a dataset for which there are no previous solutions.
  • Qualitative/Insight-Based Evaluation
  • Experimental Evaluation

Determine whether the following scenario would best be served by a qualitative evaluation or an experiment:

  • Determine the depth to which a visualization tool allows scientists to build complex knowledge.
  • Qualitative/Insight-Based Evaluation
  • Experimental Evaluation

Why are the tools created through design studies best evaluated using qualitative evaluation?

  • Design studies solve complex problems with the aim of building knowledge
  • There are too many steps in a design study, making experimental evaluation infeasible
  • Design studies are only used for solutions where we don’t care about objective performance measures
Conclusion:

I hope this Fundamentals of Data Visualization Coursera Quiz Answers would be useful for you to learn something new from the Course. If it helped you, don’t forget to bookmark our site for more Quiz Answers.

This course is intended for audiences of all experiences who are interested in learning about new skills in a business context; there are no prerequisite courses.

Keep Learning!

Get All Weeks Vital Skills for Data Science Specialization

Data Science as a Field Coursera Quiz Answers

Ethical Issues in Data Science Coursera Quiz Answers

Cybersecurity for Data Science Coursera Quiz Answers

Fundamentals of Data Visualization Coursera Quiz Answers

Leave a Reply

Your email address will not be published.

error: Content is protected !!