Michel Baudin‘s comments: This morning’s New York Times contains an article with data visualizations at varying levels of detail that are far more sophisticated than the usual pie charts and stacked bar charts commonly found in the American press as well as in business presentations and shop floor performance dashboards.
The exact meaning of the above chart between the title and the lead of the article is not immediately obvious. After looking at it for a minute or two, you realize that it has a high data-to-ink ratio: it makes a non-trivial point in a flourish-free format that I think Edward Tufte would approve.
The article is about the relative representation of different groups in the student population of 101 institutions, including the Ivy League, University of California campuses, “top liberal arts colleges,” “other top universities,” and “public flagship universities.” The study compares the proportion of freshmen enrolled from each group to their proportion in the college-age population as a whole.
The body of the article explains the metric used, primarily through charts like the following:
It follows through with facet charts of the evolution of the mix of the freshman population over 35 years in each institution, in the following format, which lets you see striking differences between the Ivy League and the University of California campuses:
Note that the label of each curve is attached to the curve and matches its color, making each chart easier to read than if the labels were off in a legend block, as is common in Excel charts.
Giving us the facts is the primary job of a newspaper; interpreting them for us, not so much. Readers should draw their own conclusions. This article only has a few unnecessary paragraphs assigning differences in the quality of K-12 education as the root cause. I can think of a myriad of other causes and many more questions. For example, in each group, what is the proportion of applicants who are admitted? And what part is played by the cost of attending these top colleges?…
You can argue with the study equating “college-age population” with the number of 18-year olds, or with the definition of the groups, but all these choices are stated at the end of the article, along with the data sources.
Aug 24 2017
Sophisticated Graphics In The New York Times
“Even after decades of affirmative action, black and Hispanic students are more underrepresented at the nation’s top colleges and universities than they were 35 years ago, according to a New York Times analysis. The share of black freshmen at elite schools is virtually unchanged since 1980. Black students are just 6 percent of freshmen but 15 percent of college-age Americans, as the chart below shows.”
Sourced through the New York Times
Michel Baudin‘s comments: This morning’s New York Times contains an article with data visualizations at varying levels of detail that are far more sophisticated than the usual pie charts and stacked bar charts commonly found in the American press as well as in business presentations and shop floor performance dashboards.
The exact meaning of the above chart between the title and the lead of the article is not immediately obvious. After looking at it for a minute or two, you realize that it has a high data-to-ink ratio: it makes a non-trivial point in a flourish-free format that I think Edward Tufte would approve.
The article is about the relative representation of different groups in the student population of 101 institutions, including the Ivy League, University of California campuses, “top liberal arts colleges,” “other top universities,” and “public flagship universities.” The study compares the proportion of freshmen enrolled from each group to their proportion in the college-age population as a whole.
The body of the article explains the metric used, primarily through charts like the following:
It follows through with facet charts of the evolution of the mix of the freshman population over 35 years in each institution, in the following format, which lets you see striking differences between the Ivy League and the University of California campuses:
Note that the label of each curve is attached to the curve and matches its color, making each chart easier to read than if the labels were off in a legend block, as is common in Excel charts.
Giving us the facts is the primary job of a newspaper; interpreting them for us, not so much. Readers should draw their own conclusions. This article only has a few unnecessary paragraphs assigning differences in the quality of K-12 education as the root cause. I can think of a myriad of other causes and many more questions. For example, in each group, what is the proportion of applicants who are admitted? And what part is played by the cost of attending these top colleges?…
You can argue with the study equating “college-age population” with the number of 18-year olds, or with the definition of the groups, but all these choices are stated at the end of the article, along with the data sources.
#datavidualization, #chartjunk, #datatoinkratio
Contents
Share this:
Like this:
Related
By Michel Baudin • Press clippings • 0 • Tags: Chart Junk, Data visualization, Data-to-ink ratio