Exam 3 - Chapter 10 - Share the story Flashcards
Data can be communicated more effectively when visualized.
What is data visualization?
Presenting results of an analysis in accesible way through charts/graphs.
What are these components of charts/graphs:
- Horizontal axis
- Vertical axis
- Scale
- Tick marks
- Data series
- Data point
Horizontal axis: Used for labels
Vertical axis: Used for values
Scale: The total values shown on vertical axis
Tick marks: Increments of the scale
Data series: raw data the chart visually represents
Data point: Individual items of the data series
Use this flaschard as a visual representation of the components of graphs/charts:
An easy and effective way to improve data visualizations is to:
Sort data in an order
What are descriptive statistics?
Brief summaries of a data set that provide a rep. of the data as a whole.
Includes:
- Mean
- Standard deviation
- Mins and max
Time series data
An effect visualization tool for displaying trends over time:
Bar charts
Line charts
Note: time series data should be sorted logically; for example, chronological order
The advantage line charts have over bar charts for visualization of trends:
Line charts lend themseviles to numerical data that extends below 0
- When the overall trend is more important to communicate rather than specific data points
The advantage bar charts have over line charts for visualization of trends:
Display specific data points easier; important when you want better descriptions of each point
Ratio data
How should this data be visualized?
Useful for trend analysis
For bar charts: Always use 0 as starting point
For line chart: Should use 0 as starting point, unless makes more sense not to; trend will stand out regardless
Pie Charts
What type of analytics do these visualizations work best with?
Useful for these descriptive analytics:
- Proportions
- Categorical data
Note: Rarely preferred # of categories > 4
What are data dashboards?
Report that contains a collection of useful visualizations and tables to check business progress/drive decisions.
Typically contain filters to make it easy to hone on a particular time, category, product, or geographic area
What are histograms?
Data visualization that displays the distribution of a particular dataset; compares numbers across intervals
Histograms vs. bar charts
What are three key differences in histograms compared to bar charts?
- Histograms represent bins/intervals: Ear bar represent subsets of data, arranged in increasing order
- Histograms use numerical data, not categorical: this is why histograms do not contain gaps between bars
- Histogram’s vertical axis shows only counts of observations: Bar charts may show any descr. statistics.
Visualizing diagnostic analysis
What are the two methods for assessing outliers?
- “Eyeball” data to determine if any observations fall outside expected distributiion
- Construct a box plot
Use this flashcard to understand box plots/outliers.