NYC Swipe Map
To create a swipe map you need 2 layers: 1) a base map with a background and transparent marks/polygons and 2) a duplicate map with colored marks/polygons and no base map. The 2nd map needs to be overlaid on top of the base map with a transparent background.
To create the swipe effect add a parameter based on the longitude. The color is and IF THEN calx tied to the parameter. If the map longitude is less than the parameter longitude (left of line) then set a negative normalized value. And if the map longitude is higher than the parameter longitude (right of the line) then set a positive, second normalized value. The two normalized values in the above case are median household income and health insurance coverage.
Finally, add a reference line to 2nd map tied to parameter. Then just line up the two maps with the parameter and bing bango: swipe map.
Download my version to reverse engineer. Let me know if you have any questions.
Tadpole Scatter
The tadpole scatter incorporates a tertiary encoding of change vis-a-vis the direction of the tadpole mark. The direction shows change in the x and y coordinates over time. For example, up and right (45° angle) shows positive sales (x-axis) and profit (y-axis) increases. This is similar to a comet chart but doesn't overwhelm the user with busy lines and twice as many data points so they can focus on the primary message.
The primary message in the above example is sales vs profit by product sub-category. The color indicates product category and the direction or angle of the tadpole is the change from one year to the next. With the tadpole you can easily see which product sub-categories are improving up and/or right versus getting worse down and/or left. You can hover over to see the comet pattern which mirrors the direction of the tadpole.
Here is a comparison with another example. This chart is showing employee performance for 2 variables: impact (x-axis) and quality (y-axis). The tadpole scatter shows the x and y position and the direction or angle of the tadpole shows the change from year to the next. It is easier to see the difference between Eli and Saquon. These two have similar scores but Eli is trending up and Saquon is trending down.
These patterns are harder to see in a traditional scatter. You can see 2019 performance clearly but no difference between how these employees have changed over time. So it is harder to contextualize their annual performance as a sign of improvement or not.
The comet plot below shows a line for each employee. The thicker part of the line is 2019 performance. You are able to see change over time. But it is much busier and it is harder to immediately see 2019 performance.
The primary message in the above example is sales vs profit by product sub-category. The color indicates product category and the direction or angle of the tadpole is the change from one year to the next. With the tadpole you can easily see which product sub-categories are improving up and/or right versus getting worse down and/or left. You can hover over to see the comet pattern which mirrors the direction of the tadpole.
Here is a comparison with another example. This chart is showing employee performance for 2 variables: impact (x-axis) and quality (y-axis). The tadpole scatter shows the x and y position and the direction or angle of the tadpole shows the change from year to the next. It is easier to see the difference between Eli and Saquon. These two have similar scores but Eli is trending up and Saquon is trending down.
These patterns are harder to see in a traditional scatter. You can see 2019 performance clearly but no difference between how these employees have changed over time. So it is harder to contextualize their annual performance as a sign of improvement or not.
