Thread Closed 
Week 2: In Climbing the Income Ladder, Location Matters
09-04-2014, 10:48 AM (This post was last modified: 09-04-2014 10:57 AM by Nicki Barneck.)
Post: #1
Week 2: In Climbing the Income Ladder, Location Matters
The visualization is a choropleth map. It shows the percentage of children born in the bottom fifth of the income ladder that rose into the top fifth in areas across the US. Color is used as a visual encoding with red being the lowest percentage (4%), and blue being the highest (~35%). I think the map is quite clear and not misleading.

I think that the visualization is really good. I like that color was used to depict the percentage. I also like how the the data is displayed on a map; it makes it easy to see how certain areas are grouped together. The visualization is very clear on the percentages, what area they apply to, and what data is being analyzed. I also like that the map is interactive. On the website, a user can hover over areas and it will show the exact percentage and the name of the area that it applies to. I also like that the author put note on the edge of the graph to explain what the bottom and top fifth meant. I also like the note that made a comparison between Atlanta (in the red) and other areas.

I think that the visualization could benefit from having a note about how the different areas were divided. Were they divided by county or something else? Instead of having a legend that consists of five colors, I think it would be more helpful to have a scale. That way, viewers could more easily estimate exactly what percentage an area was. I think choosing more vibrant colors would also improve the visualization. The majority of the colors are pretty washed-out. I also think marking more cities on the map would have made the visualization more engaging.

Link: http://www.nytimes.com/2013/07/22/busine....html?_r=0


Attached File(s) Image(s)
   
Find all posts by this user
09-04-2014, 11:52 AM
Post: #2
RE: Week 2: In Climbing the Income Ladder, Location Matters
I agree with most of what Nicki says, but for the use of vibrant colors. I think it was a good decision to have just two vibrant colors (spread far apart on the color spectrum) to denote areas with lowest and highest chances of rising to higher income group, and using faded/washed out colors for everything in between as this highlights these two particular areas which in my opinion is more interesting data than knowing the exact value for the percentage of chance that children in a particular area have to rise to higher income group.
Find all posts by this user
09-04-2014, 05:48 PM
Post: #3
RE: Week 2: In Climbing the Income Ladder, Location Matters
I have noticed that some areas have white background even though there is no white color in the legend so when I hover over these areas the note says “Data not available for this area”, however data is available for other geographical areas where it might be difficult for people to live in (for example the Great Salt Lake).

So I think what Nicki said about adding a note to the article that explains how different areas were divided, is important.
Find all posts by this user
09-04-2014, 05:52 PM
Post: #4
RE: Week 2: In Climbing the Income Ladder, Location Matters
The areas are not counties. They appear to be defined by the closest major city. I think this visualization only tells part of the story. It looks like they are using "bottom fifth" and "top fifth" to refer to national income levels. This does not paint the full picture because different parts of the country have different costs of living. In an area with cheap rent, transportation, and food, a lower level of income will support a higher standard of living. I think it would have been better if this visualization had used a normalized income level that accounts for the cost of living in each area.
I agree that more colors and more unique colors would have helped with understanding. I think it is strange that the map appears to use a color scale that blends from one color to the next, but the legend only shows a few color samples. This is easier to see if you look at the shades of red and blue on the map. Most of these shades are not in the legend.
Find all posts by this user
09-05-2014, 06:11 PM
Post: #5
RE: Week 2: In Climbing the Income Ladder, Location Matters
The color choice may have been deliberate. The designer may have wanted the colors to appear to blend, a heat map type of visualization. The percent chance that a child born in the bottom fifth income level to rise to the top fifth is quantitative and continuous. Having a coloring scheme that blends from one level to the next highlights the continuous nature of the data.
Find all posts by this user
09-06-2014, 06:05 PM
Post: #6
RE: Week 2: In Climbing the Income Ladder, Location Matters
This is an interesting piece of visualization and the detailing is great.

I agree with Nicki when she says that making the map interactive is helpful. That's one of the things Tamara mentions in her book as being crucial to effective visualization as well. The interactivity of the visualization facilitates in getting accurate data.

The colors are a little confusing though. There are different shades of the same color for adjacent cities and that makes you go back & forth to refer to the legend to understand under which bracket the place belongs to. And I also noticed that in the map there are multiple shades of red where as in the legend there's only one value for red. So we have to bank on the interactivity of the visualization alone to decide the probability of a child climbing the income ladder in some cases.
Find all posts by this user
09-07-2014, 11:03 AM
Post: #7
RE: Week 2: In Climbing the Income Ladder, Location Matters
Taking only this map into context, I agree that the color visualization was an engaging part of the visualization and it was an aspect that was lacking. From a "zoomed" out perspective the colors are indeed "washed out" and they lead to some boundary detection problems. The developers could have used a better color intensity/variety to distinguish the areas. Some colors are more easily distinguished than others.

Note: Lower down in the article The New York Times provides an intriguing visualization which overcomes some of the ideas presented in this critique.
Find all posts by this user
09-07-2014, 01:33 PM
Post: #8
RE: Week 2: In Climbing the Income Ladder, Location Matters
While the map is pretty and certainly shows some trends in the south, I would like to know if the bottom 5th has been normalized or is it defined on each region? Due to this it makes me wonder how skewed the scale really is. Beyond that, the colors they chose for 10, 15 and 20 percent are hard to distinguish. This may have to do with some color blindness, but the intensity used goes from very dark red to a light yellow then a darker yellow, then yet a darker color and finally a darker blue. This makes the intensity scale for this chart form a parabola instead of using a straight scale of darker to lighter colors which makes on the fly comparisons very hard to do.
Find all posts by this user
09-07-2014, 02:32 PM
Post: #9
RE: Week 2: In Climbing the Income Ladder, Location Matters
This is an interesting example of visualization and I wanted to use the same one.

Basically, I think this map presents clear information and I agree with Nicki's opinions. Especially, user can point to one block to see detailed percentage of that area in the original website. Another good aspect is that the user is able to tell red region (lowest percentage ) and blue region (highest percentage) easily because usually users focus these two regions more. However, gradient legend might be a better choice in this visualization example since the designer is trying to apply different color intensity to present different percentage. Several points in this map visualization could be improved. For example, white blocks in the map are not explained. In addition, I would like to see abbreviation of states showing on the map so that I would get a better idea of this study of a state. Plus, the designer explains the "top fifth" under the legend but still I don't know what is "bottom fifth". Since "bottom fifth" is mentioned in the context, I think it would be more clear if adding definition of "bottom fifth" underneath of "top fifth".
Find all posts by this user
09-07-2014, 02:59 PM
Post: #10
RE: Week 2: In Climbing the Income Ladder, Location Matters
I find this one to be a good piece of visualization.

The interactive legends really make it easier for viewers to track the percentages and figure out the upward mobility in income ladder for different regions of United States. The use of colors to depict percentages make it more effective.

But on the other hand, I feel(as Nicki already mentioned) that a scale/gradient based coloring system for indicating percentages would have been more apt, where, in my opinion, color intensity could be proportional to percentage. This would reduce the user dependency on legends to understand what color indicates what percentage. Moreover the choice of darkest colors for either end of percentage spectrum can sometimes be misleading.

Also I don't see any clear cut definition for bottom fifth as it has been defined for top fifth. Demarcating regions according to a particular type would also have been more convenient.
Find all posts by this user
Thread Closed 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)