Description:
This graph uses tuition vs graduation rate as a measurement to determine the education quality. In this graph, X-axis is the 6-year graduation rate and Y-axis is the tuition. Also, public institutions are represented as blue, private non-profit institutions are in green and private for profit institutions are in brown. The size of the bubble indicates the full-time-equivalent.
What I like:
The author did a good job on telling the audiences why he creates this graph. His research starts from a story that a student who has 3.9 GPA drops out of school because he thinks the tuition he paid won’t be paid back in the future. Therefore, his graph focused on tuition vs graduation rate. [1]
He also separated all the education institutions into three main kinds, which makes the analysis more accurate. Besides that, size of the bubble is very functional. This meets the requirement of aesthetics. It’s very easy for the audience to read the message directly from the graph. For example, the audience can easily figure out that the brown bubble on the left is the largest one. Also, most of the tuition of public school is below $10000, as well as, for the private non-profit institutions, the graduation rate has the positive relationship with tuition.
What I dislike:
The graph doesn’t meet the requirement of validation. The data of the graph is not accurate. According to the data from NCES, public institutions have the lowest graduation rate, which is below 30%. [2]
Also, the data showed in this graph is very subjective. There exist many other external factors, which will result in the difference in the final result. For example, some student many have double major. This will also increase the tuition and duration of graduation years. Moreover, the data focus on the 6-year graduation rate. However, some majors, such as medical science or , may require more than 6 years, and there majors may have a lower graduation rate with higher tuition.
Therefore, the author has a good approach to the question. However, the accuracy of the graph still needs to be concerned.
Reference: