Dataset: Food
Introduction
The dataset that our group pick was from CORGIS talked about Food. There are over 4000 datas in the dataset that contains majority nutirients and components in each kind of food.
Brianstorming
By portraying our groups data, we would like to lay out the contents in each of the food and the components. During the brainstorm, we were all thinking that we should designing something for healthy. So here are the pictures for our brainstorming. Creative group were thinking that they would map all the contents in the fast food, cookies and candies that we normally eat. The purpose was to “gross” the people who will be looking at the graph, and telling them you might stop eating these.
For clear graph brainstorm, we were thinking that the bar chart or box plot on analyzing the minerals, vitamins, and Nutrients in food.
Clear communication and reasoning
In the clear communication style, we did three graphs in this. The goal of these graphs was trying to illustrate which meat, fruit or cereals that people should choose during workout or on a diet or for the people who wants to keep healthy.
First thinking on designing the first graph, we would like to show the audience that the data of proteins and fat in 6 big categories of meat — Beef, Chicken, Fish, Lamb, Pork, Turkey. First time we would like to do the comparison histogram graph to visualize the percentage of these two contents. By looking this graph, people can balance between protein and fat from their needs. Later on, we thought that for visualizing, we thought that the graph should be in one bar for the ratio. So for example, fat is 5 grams out of 100 grams of the meat and 30 grams of protein out of 100 grams of meat. Thus, the ratio would be 30/5 which is 6. So in the graph, we can just see which bat is higher that the ratio is larger. For those who want to benefit themselves on gain more protein, they could choose the right meat for them in the graph.
Here is the design for prototype testing, while people were testing it through, they thought that the ratio is harder to read than the compare histogram graph. However, while we designing the graph, this graph was the one that we thought should be the easiest to read. Probably it’s because that the ratio might mislead the people on way of thinking this graph.
For second graph on clear graph, we chose to do box plot on four cereals brands — Kellogg, Kraft, Quaker, and General Mill’s. The design concept of this graph is that we wanted to lay out all the cereals kind in each brand and find out the mean sugar percentage of each brand. By looking at the box plot, we can easily find out that which brand has an overall lower mean of sugar. Or we can see the whisker of the box plot to see which brand has the highest sugar percentage. Due to the definition of boxplot, we can see the skewness of the graph and see how outliers affect the graph. If people who has special favorite on certain brand of cereals, they can search by sugar contains in that brand, because most of them all have the same lowest sugar percentage.
While we do prototyping of the graph, people thought that the line for mean wasn’t clear enough to read. However, the way of plot out all the data on the graph gives an concrete outlook of each brand.
For the third graph we had the graph on vitamins and sugars in fruit. We were thinking on doing the graph by horizontal stack. So that readers can visualize the graph by seeing how big the block is on the graph. However, we found out that the units of each contents were different, Vitamin A’s unit is mcg, Vitamin B and Vitamins C’s units are mg, and sugar’s unit is gram. Within these different units, if we put on the same scale, the graph would be too hard to visualize. Meanwhile, if we just leave them as different unit, the graph would be totally wrong. Thus, we thought to do the standard deviation. So that each content will compare with themselves.
Here is the prototype test that we had. For the first time, we didn’t have the scale at the side so it’s hard for people to read the graph. They didn’t really understand about how to read the graph. For example, when we look at the Apricots distribution, we can tell that the vitamin A is way above the standard deviation than any other. Thus if people who want to take more Vitamin A in their fruit, Apricots would be the top choice for them to choose among these 10 fruits.
Persuasive communication and storytelling
For their designing, they used the fat person to resemble the people who eat those unhealthy food for their whole life.
For the sugar content in the snacks, the pie chart plot out on the actual food. So it’s more harsh to the people who eat them. For example, when I was looking at the chocolate chip cookie, i was shocked by that half of that cookie was made by sugar.
Final Testing
While people were looking at our graph, it took normally 1 min for them to understand the graph and absorbed the information that provided from the graph. The feedback that they gave us were “I know fish have more protein than any other meat” or “I am lucky I always have Kraft for my breakfast.” These feedback definitely shown that people did understand our graph, which means that we achieved our goal.
Contrast on these two parts
For the clear graph, other than it’s clear. I think it contains more data than the storytelling one. It deals more data than the other one. Plus, understanding of the data would be a little bit harder too. However, for the creative group, it’s more persuasive than the clear graph one. Thus, I think the trade of between these two groups are the high level information or the big picture of certain dataset.
Conclusion
By sketching the 5 design sheets, it made the design and the goal more clear than before. Overall the whole process went well. If there were more time on doing this project. I would like to make the graph looks more fun in color, or the shape. Although for clear group, we don’t need to focus on attracting people’s attention, making the graph more fun or easier to read should be improved.