Week 7 - Chart visualizations
Characteristics of Area Charts
- Shaded in below the trend line - good for comparisons
Characteristics of stacked area
- show changes in composition over time - categorizes are stacked together
Characteristics of timeline charts
- used to present continuous data - time should always rune from left to right - do not skip values for consistent data
Characteristics of stacked column chart
- used to show comparison
Main characteristic of line charts
- used to show trends
________ data would be considered the least sophisticated type of data. A. Ratio B. Nominal C. Ordinal D. Interval
B. Nominal
Charts for Qualitative Data when you want to show proportion
Bar charts pie charts stacked bar chart tree maps heat maps symbol maps word clouds
(Data Driven: Quantitative) What type of chart is used for Outlier detection?
Box and Whisker plot
What is the connection between statistics and modeling and optimization? A. Data mining B. What-if? C. Simulation and risk
C. simulation and risk
In the textbook, exhibit 4-8 gives chart suggestions for data. These options include all of the following except: A. Geographic data B. Outlier detection C. Relationship between variables D. Normal distribution curves
D. Normal distribution curves
Gold, silver and bronze medals would be examples of: A. Test data B. Nominal data C. Structured data D. Ordinal data
D. Ordinal data
Which of the following is NOT a typical example of nominal data? A. Gender B. Ethnic group C. Hair color D. SAT scores
D. SAT scores
What is the more appropriate chart type when showing a relationship between two variables? A. bar chart B. Histogram C. Pie Chart D. Scatter chart
D. Scatter chart
(Data Driven: Quantitative) What type of chart is used for geographic data?
Filled map
Explain the "P and A"
Give an overview of your model and limitations you faced
Explain the "M"
If appropriate, describe issues you encountered in the ETL process
Which chart type is better for working with dates over time? Line graph or bar graph
Line graph
Tools for creating visualizations
Tableau Microsoft BI
Word clouds
formed by counting the frequency of each word mentioned in a dataset
Normal Distribution
median, mean, and mode are all equal, so half of all the observations fall below the mean and the other half fall above the mean
Improving your charts comes down to choosing appropriate _______ and choosing your ________ effectively
scale; colors
(Data Driven: Quantitative) What type of chart is used for relationship between two variables?
scatter plot
what type of charts can be used for geographic (Qualitative) data?
symbol map
Declarative visualizations
- "declare" or present your findings - made after the data analysis has been completed and are meant to exhibit was was found in the analysis steps
Don'ts of writing a report
- Don't ramble like in essay question on an exam - Don't use flowery descriptive language - Don't be condescending "as you can clearly see...."
What are the two questions you need to ask yourself when determining the purpose of your data visualization?
- Is your purpose declarative or exploratory? - Is your data qualitative or quantitative?
Standard normal distribution
- a special case of the normal distribution used for standardizing data - its mean is 0 - its standard deviation is 1
What type of charts can be used for comparison?
- bar chart - pie chart - stacked bar chart - tree map - heat map
Which charts are appropriate for qualitative data?
- bar charts - pie charts - stacked bar charts
(Conceptual: Qualitative) What type of charts are used to show comparisons between variables?
- bar charts - pie charts - stacked bar charts - tree map - heat map
What are the three categories of data that represent conceptual (qualitative) data?
- comparison - geographic data - text data
Explain the "T"
- discuss what's next in your analysis - how frequently will it be updated? - are there trends or outliers that should be paid attention?
Dos of writing a report
- get to the point quickly - be clear, unambiguous, correct, interesting, and direct - separate facts and data from opinions - consider your audience - use plain language
Characteristics of bar charts
- horizontal column charts - used if number of categories is greater than seven - good for displaying set with negative numbers
Use charts when:
- is used to convey a message that is contained in the shape of the data - is used to show a relationship between many values
What types of charts are appropriate for quantitative data?
- line charts - box and whisker plots - scatter plots - filled geographic maps
What are the 4 category types of data that represent Data-Driven (Quantitative) data?
- outlier detection - relationship between two variables - trend over time - geographic data
Things to do to consider your audience and tone
- place the focus on your audience - craft different versions for different audiences - use an appropriate tone - provide the right content - avoid too much detail
Explain the "C"
- provide an explanation of the visual you chose - describe any items that stand out of that are interesting
types of quantitive data
- ratio - interval - discrete - continuous - distributions of mean, median, and standard deviation
Characteristic of Scatter charts
- used for correlation and distribution analysis - can help in spotting anomalies or outliers
Characteristics of column charts
- used if the numbers of categories are small (up to five) - used if one of the data dimensions is time - show trends only if there are a reasonably-low number of data points
Characteristics of Pie charts
- used to visualize a part to whole relationship or a composition - sum of all segments should be 100% - do not used if they're segments are almost identical
Proportion
- used with quantitative data - calculated by counting the number of items in a particular category, then dividing that number by the total number of observations
When should you use Excel to communicate your results?
- when your data analysis project is more declarative than exploratory
Exploratory Visualizations
- you are performing the test plan directly in visualization softwares
Use tables when:
- you need to compare or look up individual values - you require precise values - values involve multiple units of measure - the data has to communicate quantitative information, but not trends
What is the connection between statistics and business intelligence/information systems? A. Data mining B. What if? C. Simulation and risk
A. Data mining
Line charts are not recommended for what types of data? A. Qualitative data B. Quantitative data C. Normalized data D. Continuous data
A. Qualitative data
The Fahrenheit scale of temperature measurement would best be described as an example of: A. Discrete data B. Interval data C. Ratio data D. Nominal data
B. interval data
Justin Zobel suggests that revising your writing requires you to "be egoless-ready to dislike anything you have previously written", suggesting that it is _______ you need to please. A. the customer B. the reader C. your boss D. yourself
B. the reader
What is the connection between business intelligence/information systems and modeling and optimization? A. Data mining? B. What if? C. Simulation and risk
B. what if?
__________ data would be considered the most sophisticated type of data A. ordinal B. interval C. Ratio D. Nominal
C. Ratio
In the late 1960's, Ed Altman developed a model to predict if a company was at severe risk of going bankrupt. He called his statistic Altman's Z-score, now a widely used score in finance. Based on the name of the statistic, which statistical distribution would you guess this came from? A. Uniform distribution B. Standardized normal distribution C. Poisson distribution D. Normal distribution
D. standardized normal distribution
Explain the "I"
Explain what was being researched and the purpose of the project
Scale and increments to consider
How much data do you need to show? What do you do with outliers? What is the baseline? Would context or reference lines make the scale more meaningful?
(Data Driven: Quantitative) What type of chart is used to show trend over time?
Line chart
Which tool is used for basic declarative charts?
Microsoft Excel
3 types of of Qualitative data
Nominal Ordinal proportion
(Conceptual: Qualitative) what type of chart is used to show text data?
Word cloud
What type of charts can be used for text data?
Word cloud
The way you write your report is dependent on your __________.
audience
What type of charts can be used for outlier detection?
box and whisker plot
Symbol maps are best used to:
express qualitative data proportions across geographic areas
What type of chart can be used to show geographic (quantitative) data?
filled map
What type of chart can be used to show trend over time?
line chart
Charts appropriate for quantitative data when you want to show complex data
line chart box and whisker plots scatter plots filled geographic maps
What type of charts can be used for relationship between two variables?
scatter plot
(Conceptual: Qualitative) What type of chart is used to show geographic data?
symbol data
Exploratory visualizations
used to gain insights while you are interacting with data; you do not have a clue about what information lies within your data
Declarative visualizations
used to present findings
True/False The x-axis of quantitive chart is supposed to start at 0
True
True/False When using time in charts, set it on the horizontal axis
True
True/False Human eyes are much better at telling differences between lengths than determining differences between areas
True Ex: A stacked bar chart is easier for people to understand the size difference compared to a pie chart