Thread Closed 
Week 4 : Visualizing a century of meat consumption
09-18-2014, 01:53 AM (This post was last modified: 09-18-2014 10:19 AM by Shravanthi.)
Post: #1
Week 4 : Visualizing a century of meat consumption
[Image: 15food_graphic.html]
http://www.nytimes.com/imagepages/2011/0...aphic.html

Description :
This is an infographic visualization on increased consumption of meat in the United States over the century especially after the world war 2 with the growth of fast food industry and intensive farming methods. The other factor that contributed towards this was the adoption of modern refrigeration techniques in food storage. The dietary guidelines published by the health department also dictates the consumption measures.

Pros:
The graph narrates the story of meat consumption by documenting the events that either led to the increase or decrease very neatly.
The causal narrative is easy to understand.
I would keep the white grid lines as they give a good sense of the amount of consumption in pounds.
The color coding to highlight each of the periods like the Korean war and the vietnam war works well.

Cons:
I would add an element of interactivity to this chart and a better color palette could be more suitable.


An interesting thing to mull over would be to analyze how this contributed to national health and obesity in the US.
This visualization is taken from the link http://www.nytimes.com/imagepages/2011/0...aphic.html
Find all posts by this user
09-18-2014, 06:29 PM
Post: #2
RE: Week 4 : Visualizing a century of meat consumption
While the narrative infographic visualization does a good job of describing "A Century of Meat", I feel the visualization could have been even better if the following changes were there:

1. Separate legend describing the what each colored line indicates. This could make it much easier than having to look out for the bold characters on the graphs where there are lot of lines and other categories (eggs, turkeys, etc) which might cause confusion.

2. The choice of colors as Shravanthi has rightly mentioned could be definitely improved. The color coding used for "Pork" and the background graphic are shade variations of the same color. I felt a stronger color could have been used for "pork" for easy identification.

3. Personally I would like to see the solid lines being replaced by dotted lines as that would make it relatively easier to follow the path, when we have a lot of lines originating/overlapping/branching out from a common point.
Find all posts by this user
09-19-2014, 08:41 PM
Post: #3
RE: Week 4 : Visualizing a century of meat consumption
Quote:The color coding to highlight each of the periods like the Korean war and the vietnam war works well.
Agreed, but I would make the color slightly darker and not have it dissolve into the background.

Quote:I would add an element of interactivity to this chart...

I like the idea of making this an interactive graph. I think if I were to do this, I would make the interactive portion display hyperlinks to historical articles that helped back up certain assumptions made by this graph. For example, did dietary guidelines really cut consumption of red meats? Or was the reduction in the consumption of red meat due to the fact doctors were telling their patients around this time period that they should reduce red meat consumption due it's "relationship" to an increased chance for heart problems?

In addition to adding links to articles that supported assumptions made by this graph, I also think they should have added rollover that would highlight a given data set when someone had their cursor over a particular meat.

Quote:and a better color palette could be more suitable.
Yeah, they should have spent a lot more time picking out a more distinguishable color pallet.


In response to Anirudh
Quote:1.Separate legend describing the what each colored line indicates.
This would definitely make this graph much more readable.

Quote:3. Personally I would like to see the solid lines being replaced by dotted lines as that would make it relatively easier to follow the path, when we have a lot of lines originating/overlapping/branching out from a common point.
I don't think this would make the graph much more readable, as it would create a sense of disjointedness.
Find all posts by this user
09-20-2014, 11:29 AM
Post: #4
RE: Week 4 : Visualizing a century of meat consumption
Shravanthi gave an excellent analysis of the visualization.

I agree with previous posts on the issue of color and labeling.

Some of the colors chosen by the author representing different kinds of meat are similar, especially for veal, turkey and pork. We may change to colors with sharper difference.

The labels make the graph messy. There three types of labels: one for the time series, like "Beef"; one for events, like "First McDonald's restaurant opens"; and one for era, like "World War I". The events labels are identified by a connector intersecting the time series with a circle. However, the era label and the time series label could be confusing. For example, for a careless read, the yellowish curve could mean either "Egg" or "Vietnam War".

A better practice is to use legend to indicate the representation of the time series (removing time series labels), to put the era label in the same row, for example, between 50 and 60, and move the paragraph of the description in the upper-left corner out of the graph.
Find all posts by this user
09-21-2014, 12:10 PM
Post: #5
RE: Week 4 : Visualizing a century of meat consumption
I agree with many of the points that have already been made, especially about how the labels could be adjusted to make this visualization less confusing. Incorporating a legend and adjusting the location of the labels for spans of time so that they are just below the axis would be a great start. It would have the advantage of providing consistent locations in which to find similar types of data, instead of placing labels wherever they happen to fit, regardless of what category those labels belong to.

As far as the color scheme goes, I don't feel like this example demonstrates anything flagrantly wrong. For instance, it isn't overly difficult to tell different each series apart in the regions where they overlap (though it wouldn't be a bad idea to adjust the hues that are similar to the background).

In spite of not doing anything wrong, though, the use of color could certainly be enhanced. I notice that the labels for chicken, pork, and beef are in bold, indicating that those particular lines are supposed to be emphasized to get some of this visualization's main points across. I think changing the properties of those particular lines to make them pop out would be a better choice than a bold label. Heavily saturating them in comparison to the other categories could be one possible option.
Find all posts by this user
09-21-2014, 02:50 PM
Post: #6
RE: Week 4 : Visualizing a century of meat consumption
The visualization is quite simple and easy to understand. I liked the way the color intensity is used to highlight the war periods.

However, I think there could have been couple of improvements in this visualization.

1. Mouse hover on any line should display the meat name it corresponds to.
2. The color used for lamb, veal and turkey is very similar. Its difficult to differentiate among them. Better color combination should have been used.
3. Option to select a meat trend which highlights its graph.
Find all posts by this user
09-21-2014, 02:54 PM
Post: #7
RE: Week 4 : Visualizing a century of meat consumption
This visualization is very well summarized by Shravanthi and I agree to most of the points mentioned by others however I feel color selection if just fine and it goes well with the background. However they could have used different colors for Turkey, Veal and Pork. At the same time they have used Bold fonts for Chicken, Pork and Beef which are the top three meat being consumed. The color shade describing the time of war is good. The key events mentioned in the graph provides good idea why the consumption of a particular meat has gone up or down. But I don't get these 2 events
1. Why was "First McDonald's restaurant opens" placed on Beef line? Does that mean McDonald's were selling just Beef at that time?
2. Why was "The other white meat ad campaign" placed on Pork line? Was it to do something with Pork?

May be I can get answer to this if the graph was interactive and as mentioned by ThatGuyCraig display links to related articles.
Find all posts by this user
09-21-2014, 03:20 PM
Post: #8
RE: Week 4 : Visualizing a century of meat consumption
I think this visualization is simple, and easy to understand, i can easy see the change for each kind of meat by years, and it listed some important events, which i can clearly see how each kind of meat changed by this event, such as World War II, Korean War, and so on.
But i also think there would be some way which can improve this visualization.
The color of Lamb, Turkey and Veal are too similiar to clearly see how the line changed. For me, it is easily mix up sometimes.
But, in general, this visualization, which i thought, is a goog visualization.
Find all posts by this user
09-21-2014, 04:05 PM
Post: #9
RE: Week 4 : Visualizing a century of meat consumption
For the most part I think the graph does a pretty good job of relaying a lot of information. If you take a step back and look at how many details are contained with this graph it is pretty amazing.

Obviously because of the amount of information, like has been stated before, the graph becomes a little congested at certain points. And yes the colors could be improved upon, adding more contrast and distinction.

Improvements aside, I think there a few elements that while appear simple, provide some great insight. The dark areas of the background signifying wars and the great depression add a lot. It was even decided to allow a gradual color change for the great depressions, symbolizing its slow recovery. Also interesting, is the fact that the campaigns to try and get people to eat more of a specific type of meat really brought on no change.

Good analysis and great choice for a visualization.
Find all posts by this user
09-21-2014, 08:06 PM
Post: #10
RE: Week 4 : Visualizing a century of meat consumption
I think this visualization is very effective and easy to understand.
As Shravanthi already pointed out, this graph nicely demonstrates the events that led to the fluctuations in various kinds of meat consumption due to different turn of events in the United States.

What I strongly feel this graphic lacks is a legend for indicating which color indicates which meat. While following some lines one can often confuse what kind of meat line they are following and there the lack of a legend makes things more difficult. It is even more confusing in case of eggs, veal, turkey and fish because the writing is not in bold.

It is also difficult to follow the veal line because of the more intense lamb line over it. So I totally agree with Anirudh that dotted lines with more distinct and sharper colors would have been a better fit for this visualization.
Find all posts by this user
Thread Closed 


Forum Jump:


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