Guides & Tutorials

How to build: Chart Types

  • The width, height and depth axes may each be toggled on or off. This allows you to create timelines (using only one axis), 2D scatterplots (using two axes), and full 3D charts (using all three axes). Make sure to specify a column for each axis you have toggled on!

    Some types of data tend to work better for different axes. Here are some suggestions from the Flow team on what we typically use (but please do experiment with others!):

    Width- date/time (click here for formats accepted in the csv file)

    Height- numerical (population, # of users, age, etc)

    Depth- categorical (country, part of the body, ethnicity, etc)

    Axis is timeline?

    Toggle this setting on if you're showing a data/time value for that axis. You then can apply the following different formats to show on your axis (with examples):

    • Year- 2001
    • Date (MDY)- 8/27/2001
    • Date (DMY)- 27/8/2001
    • Date (YMD)- 2001-08-27
    • Time- 12:00:00 AM
    • Month (full)- August
    • Month (short)- Aug
    • Month (short) Year- Aug 2001
    • Month (decimal) Year- 8/2001
    • Weekday- Monday
  • Categorized columns (otherwise known as Bar Charts) are used to group dots together and show counts from the dataset according to category. In other words, each dot represents a row in the csv file, and within this chart type, you can group those rows together by categorical values in a certain column. When choosing a column to use it can be either alpha or numeric, but in order to utilize the chart to convey information in an effective way, we suggest using columns with less than 25 unique values (no one wants to compare a hundred different bars in a bar chart!)

     

    Spacing between points determines the space between each of the dots, both in the horizontal and vertical. By default, this is set to 0.03.

    Using ___ points per column allows you to choose the width of each bar. By default, this is set to 5, which means the bars with be 5 dots wide. The larger the number, the more dots will be added to each row of the bar.

    Column gap determines the spacing between each of the bars themselves. By default, this is set to 0.01, increasing will spread the bars farther apart from each other.

    Arranging in a Circle

    Radius, by default, is set to zero, meaning the bar chart is oriented in a straight line. You can change this value to a number that will apply a circle orientation to the bars. The numerical input (radius) controls how large or small the circle of bars is in space.

    Radial offset allows you to change where the bars reside within the circle. Values range from 0 to 360 degrees.

  • Each dot will be mapped to its corresponding values for the latitude column and the longitude column in the dataset. Here is a useful tutorial on using Google Sheets to run a script that generates lat,lon from a list of addresses.

    Distance from map determines the space between the dot and the map.

    Stack happens automatically if dots have the same lat,lon. Stacked height controls how tall you want that stack to be, and Flow compresses or expands the stack by adding/removing space between the dots.

    Offset column takes a column with numerical values and offsets dots with highest values farthest away from the map, where smaller valued dots stick close to the surface of the map. Max offset controls how tall you want the maximum valued dot to be away from the map. Adding Connections to axis or map is usually beneficial at this point, so it can be seen where the dot is connected to on the map itself.

  • The values in the region column in the dataset must match standard ISO codes or standardized region names:

        Click here for acceptable region inputs for World maps: use Country Name, Alpha-2, Alpha-3, or Numeric codes.
        Click here for acceptable region inputs for US maps: use Subdivision Name or Code.

    Offset determines the space between the dot and the map.

    Stack can be toggled on to stack all of the dots on top of each other, which protrudes outward from the map. This represents the number of records for each region in the data. Layer height controls how tall you want that stack to be, and Flow compresses or expands the stack by adding/removing space between the dots.

    Grid can be toggled on to group all of the dots in a grid, flat on the map. This also represents the number of records for each region in the data. Grid spacing controls how large or small the grid of dots is and Flow does this by adding/removing space between the dots, as well. If regions are too small for the grid, it will begin stacking vertically.

    Offset column takes a column with numerical values and offsets dots with highest values farthest away from the map, where smaller valued dots stick close to the surface of the map. Max offset controls how tall you want the maximum valued dot to be away from the map. Adding Connections to axis or map is usually beneficial at this point, so you can show where the dot is connected to on the map itself.

  • In order to create an Area Chart, you must first create a Scatterplot with columns selected for all 3 axes. Under Global Settings, you'll find a button to create a new area chart. You'll then see a new panel where you can select the category you'd like to show the area under the curve for. This is the same column name you specified for the Depth axis in your Scatterplot. You can change the Opacity and Color settings, as well as toggle on Animate perspective change.
  • This is the default view when you create a new Swarm, which randomly assembles the dots in space. They are fit within a 2m3 cube in space, which you can alter the dimensions of by changing the width, height and depth values. Upon increasing or decreasing these numbers, you can spread or condense your dots, respectively.

    By playing with the dimensions in Random, editing Color and Size, and adding Interactions through popups and highlighting, you can add enough variation and interest to use this chart as a first Step in your Flow. This allows your audience to interact with the data in its most simple form and introduce the notion that each dot represents a data point that will move and flow throughout your data story.

  • Null is the name we give to a Swarm that's stored in a single point in space, (eg. (0,0,0)). This can be done within Random, by setting each of the dimensions to 0. This is a great way to utilize our built-in animation features. By starting with all of the dots animating from a single point and ending with any of the above charts, you can create a sense of grandeur rather than have the charts flatly appear in space.

Updated