Alexander Fishkov, Ph.D. student Computer Science
In this post we will explore the ACS PUMS data on couples in which both partners hold a college degree. We want to see if pairings of certain degrees occur more commonly than others, while also investigating couples who hold the same degree. For the analysis we will use the five-year data from 2009-2013.
First, we focus our attention on couples in which both partners have at least a bachelor’s degree. The survey found nearly 12.5 million of these couples (or 25 million individuals). Among them, 4.5 million (over 18 percent) have chosen a partner with the same degree.
Both General Business and Education majors show outstanding numbers of more than 25 percent. For General Business majors, the broadness of the major and nature of the office job most likely contribute to the higher-than-average percentage. For Education majors, the situation is a bit different: college campuses and school districts are like small cities on their own, so people have a higher chance of finding a partner in the same field.
Next we present a broader view of the data. A cord diagram below represents connections between the top 20 degree majors. The width of a link connecting two majors is proportional to the number of couples holding the corresponding degrees. The width of each circle segment is proportional to the number of degree holders with this major.
The strongest relationship (and thus “most typical couple”) is between Education and Business majors: 19 percent of teachers prefer business workers, while 14 percent of business workers prefer teachers.
The next strongest relationships involve Engineering majors. They tend to marry Education graduates (14 percent/9 percent) and Medical and Health degree holders (9 percent/12 percent).
We conclude that same-degree couples are actually quite popular in many fields. Frequent degree pairings in married couples also suggest that education actually affects the choice of a future partner. Thus, deciding on a major may have even larger consequences!
About Alexander Fishkov
Alexander Fishkov, Ph.D. student Computer Science
Alexander is a Ph.D. student in Computer Science. He currently holds B.S. and M.S. degrees in Applied Math. He has experience working for industry major companies performing research in the fields of machine learning, data mining and natural language processing. In his free time, Alexander enjoys hiking, Nordic skiing and traveling.
Other posts by Alexander Fishkov:
Interesting data. Did your research take into consideration when these couples met and married? For example, did they meet as undergrads at the same school as a result of being in the same courses? Or did they meet at the workplace as unmarried singles, or were they married to someone with a different career, then divorce and marry someone of the same educational background and current career? That additional information would be really interesting to know as well.
This additional information would indeed be useful, but unfortunately it is not available in ACS. The closest questions from the survey that come to mind are the following: ‘Married in the last 12 months’ and ‘Number of times married’. No specific information is retained about the past marriages. I think that a special purpose survey is needed to address your questions.
Very nice, Alexander.
This is the first time I’ve seen a cord diagram. Looking at it, I found myself discounting the thickness of the Business cords because Business majors made up such a larger segment of the circle.
I’m wondering if, instead of the width of the cord being the number of couples, it be the percentage of those with that particular degree. The cord width for 10% of Business degree holders would be the same width for 10% of Psychology degree holders even though the latter constitutes a smaller segment arc of the circle.
This would lend to broad (“easier”) generalizations, such as, “14% of Business majors married Psychology majors”.
Hello James, thanks for the feedback!
To use your approach for the cord width, all circle segments for the degrees have to be of the same length. This way relative sizes of each degree are lost. I decided to keep this information, but I get your point about very small degrees.
You can still see the percentage equivalents of the cord size in the tool-tip.