Thread Closed 
Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
09-25-2014, 04:48 PM
Post: #1
Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
[Image: CA_Migration_v2_101-01.png]

This visualization shows the 10 largest state-to-state migration between California and other states in the time frames of 1955 to 1960 and 1995 to 2000. Since this information is about state migration the information pertaining to location has been encoded using a map of the US. The direction of migration has been encoded using arrows. These arrows perform double duty by using the width of an arrow to show the amount of population migration. Finally, this data covers two time periods which is encoded using color, red for 1995 to 1960 and blue for 1995 to 2000.

There are two major insights that can be obtained from this visualization. The first is that their was a large amount of immigration to California in the 50's, and a secondary wave of migration out of the state in the 90's. The second is that these two waves are related to two different geographic areas. The bulk of migrants came from the northern Midwest and the Northeast in the 50's, where as the migrants leaving California chose to settle all along the southern United States and Pacific Northwest.

For the most part I find this visualization to be effective but there are some minor problems with it. At first glance it's not clear if Colorado, Utah and Nevada are part of the large red arrow, it's only clear if you back track to the beginning of the arrow and realize that the states the tails of the arrow are in are where it originates from.

Source: http://www.census.gov/dataviz/visualizations/051/
Find all posts by this user
09-27-2014, 12:49 PM
Post: #2
RE: Week 5: Net Migration Between California and Other States: 1955-1960 & 1995-2000
I agree that this visualization has issues, which make it difficult to interpret.

The largest downfall in my opinion is the encoding of the net migration within the width of the arrow. Due to the spatial location of each state, it makes it difficult to draw comparisons between the net migration values for individual states without referencing the associated text. This is only complicated by the number of values that this size encoding must differentiate between. For example, referring to the 1995 to 2000 values for Arizona and Texas, their arrow widths look near the same while their values differ greatly. Given this, I find it difficult to draw conclusions about the cumulative migration for a given time period as arrow width inconsistencies would be compounded when merging arrow paths.

I also find their use of layering to be ineffective as it makes for a cluttered visualization due to the number of arrows, making it more difficult to draw conclusions.

One improvement that I would suggest is to split this into a small multiples view with a map pertaining to each time period. Rather than arrows, one could encode the data with more effective channels as they no longer have to differentiate between time period. For example, direction of migration could be encoded by hue rather than an arrow, eliminating the need to trace a path to determine direction. This could be implemented by filling a given state with a particular hue. Luminance could then be used to encode the value of net migration. While this may still pose discriminability issues due to the number of states represented, I feel it would be a vast improvement over the current arrow width method.
Find all posts by this user
09-27-2014, 01:30 PM
Post: #3
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
I agree this visualization has a few issues. The large red arrow is really hard to decipher, it is not clear what portions of the immigration to California came from Utah, Colorado or Nevada (maybe there was none?). Also, the blue and red arrows that cross over Oklahoma look to be about the same width, but the values are very different; 94,900 for the blue arrow and 83,700 for the red.

I do like the idea of having the width reflect the number of people in or out of each state, but I think it could have been done in a better way. Mapping each state to a different color would be a huge improvement.
Find all posts by this user
09-27-2014, 03:16 PM
Post: #4
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
This visualization is a nice representation of population immigration from one state to another. Different color coding used for different time frames is also looking fine and makes it clear to understand the population flow to and from California. It looks visually appealing and easy to understand.

However, the width of arrows doesn't seem to be a good option to show amount of population migrated. To understand the exact number of people migrated, we have to look at the number associated with each arrow. Arrow width for North Carolina seems more as compared to that for Florida, for the period 1995 to 2000. But, when you see at the exact number its rather reverse.
Find all posts by this user
09-27-2014, 06:17 PM
Post: #5
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
The representation of thick arrow for cumulative immigration was a good idea.
Red arrow represents the migration between 1955-1960, however it only shows migration "to" California and not "from" California.
Blue represents migration "from" California to various states between 1995-2000, but the graph also shows the migration "to" California from NY. I am pretty sure there was decent amount of migration to California between 1995-2000.
A quick look at the graph would have led us to assume that red arrow also represents the migration "to" California between 1955-1960 and that blue arrow represents the migration "from" California between 1995-2000. But the presence blue arrow from NY crosses out our assumption.
Overall, it was a decent graph!
Find all posts by this user
09-28-2014, 10:26 PM
Post: #6
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
This visualization seems simple and effective in conveying its information about migration patterns from and to California and other states in the USA with emphasis on California. Unlike some of the others, I did not really have any confusion with regard to the space the arrows spanned.
I did however have a problem with making comparisons between states. On the other hand, I guess that wasn't the point of the visualization. It was intended to depict the migration with emphasis on California which, in my view, was shown well.
Find all posts by this user
09-28-2014, 11:08 PM
Post: #7
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
I liked the visualization, but I had some minor quibbles:

1) Since the legend already states that the target state is California, and the only stat being traced is people migrating in/out from there, there didn't seem any information encoding in the arrows originating/ending there.

2) At times, the "area" of the arrow (which encodes no information in this graphic) overpowered the signal of the width somewhat. For example, although NC had fewer people migrating than GA (29k vs 37k), the greater length of the former arrow makes it more prominent despite being narrower.

3) At some states (such as Colorado), the blue and red arrows could get mixed up. For instance, Colorado is a net inflow state, as shown by the blue arrow entering it from CA. However, the red arrow passing through from Texas obscures this message on first viewing.

It's a pretty graphic, but perhaps simply coloring the states red or blue depending on whether they were inflow/outflow states might have presented a clearer image.
Find all posts by this user
09-28-2014, 11:16 PM
Post: #8
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
Very nice piece of visualization and nicely reviewed by Greg Berrett.
Over all it gives a very good picture of people migrating into and out of California. The use of arrow with different color and thickness to indicate the flow and the number of people involved in migration works for me. This works fine because the designer has considered only few states. If many states were considered then this visualization might fail. Also the designer has considered north east states for incoming migrants and south east sates for outgoing migrants and there is no overlap between any arrows which is clean but doesn't give the overall picture.
Find all posts by this user
09-29-2014, 08:52 AM
Post: #9
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
The visualization is effective in communicating the net migration between California and other states. Though there were no indication of people migrating out of California during 1955-1960(may be there was none?), it is effective in the information of people migrating to California. The color difference between two timeline helps to easily identify the data intended to be communicated. Overall, I think this is an effective visualization.
Find all posts by this user
10-05-2014, 10:36 PM
Post: #10
RE: Week 5: Net Migration Between California and Other States: 1955-1960 and 1995-2000
I liked this visualization and I feel that you did a good job on critiquing it. It was a very simple visualization. There were only two colors, which were used to distinguish in-going and out-going migration. At first I wasn't sure what kind of migration but I am assuming that it is migration of people. I would agree that the position of the arrows might be misleading. For example it can be hard to see where the arrows are coming from but it can also mean this is the actual path they chose to take. I don't know. Does the arrow point to where they actually ended up in that state the arrow is point to? I am again not sure. I would think that they should tell you if it is literally the path or not but it doesn't tell you. I also felt like the numbers were hard to read, the creators were mostly wanting the consumer to see the flows and not really the numbers, I guess. However, overall I think this is a very well done visualization, because it is simple and straight forward.
Find all posts by this user
Thread Closed 


Forum Jump:


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