Judul : I Believe In Astrology
link : I Believe In Astrology
On my train ride in this morning, I saw an article posted on Twitter about Pell Graduation rates at the 80 largest universities in America. If you want to look at a boring table of static data, just click here.
But I wanted to see if there were any patterns, so I copied the table, pasted it into Excel and then opened in Tableau to visualize it. I think it tells an interesting story, although the data set is unfortunately limited, and with no key to merge the data with another set, it loses some potential.
Start by looking at the first view. For each institution, there are three columns: The overall six-year graduation rate; the six-year graduation rate of Pell recipients, and the spread, with the values on spread sorted from low to high. In this instance, a negative number means Pell students graduate at a higher rate than the student body overall, and a positive number means just the opposite. As you scroll down the list from top to bottom, ask yourself what makes the pattern make sense? There are dozens, but all I could see was, "football," but you might see "big public research university." Or something else all together.
If you want to sort by another column, hover over the axis until the little icon pops up and click away. The "reset" at lower left does just what it says it does.
The second view (on the tabs across the top) shows the Pell graduation rate scattered against the percentage of freshmen with Pell. The bubbles are colored and sized by spread (blue and large are good for Pell students; red and small, not so much.) Right away you see the pattern: If you enroll fewer Pell students, your Pell graduation rate is higher. My hypothesis would be that more selective institutions (who have higher graduation rates overall) a) simply select the most capable from among the poor students they admit, and b) have more resources to fund the smaller percentage of low-income students.
What do you see?
Women have made tremendous strides in educational attainment of bachelor's degrees in the last half of 20th century and the first decade of the 21st. And even though doctoral degrees have lagged behind, we can see dramatic changes there as well.
Take a look at this visualization using National Science Foundation Data (this link downloads the data for you in Excel as Table 14). What you see over time is a dramatic increase in the number of women who earned doctorates since 1983, but also a shift in the percentage distributions. Women are now the majority in Life Sciences, Education, and Social Sciences, and close to dead even with men in all fields except Physical Sciences and Engineering.
The second view (using the tabs across the top) shows doctorate by broad discipline over time. Use the filter at the top to compare men and women, or to see the totals. Note the tremendous percentage growth in women in engineering since 1983: From 124 to 2,051, an increase of over 1,500%.
While it's not necessarily true that most doctoral recipients work in higher education, it's true that higher education gets most of its instructional faculty from doctoral recipients; the long, slow trend (assuming it will continue, or even just stabilize) means there are some interesting changes in store in the higher education labor force in the coming decades. It's possible college faculty will look very different 20 years from now
What do you think?
P.S. You might also be interested in this, showing bachelor's attainment over time.