Is this statistic about how many Latinos/Hispanics have a PhD in the USA correct?
This information came from a tweet I shared to express my excitement for finishing my PhD and, in the process, increase by 1 the number of underrepresented minorities in the US with a doctoral degree. I did not expect for this tweet to get the attention it has, but I'm glad it's opened the doors for more in-depth discussions about diversity and representation in academia.
To clarify, I got my information from the following sources:
< 2% of people in the US have a #phd
–United Stated Census Bureau, Educational Attainment of the Population 18 Years and Over, by Age, Sex, Race, and Hispanic Origin: 2014
Of those, 6.5% are Hispanic
–National Science Foundation, Survey of Earned Doctorates 2016 (this number only includes Hispanic/Latinos that are also US citizens)
These were back-of-the-envelope calculations based on limited information I had at the time. I appreciate that others in this thread have provided further sources of information regarding this statistic. It seems like my back-of-the-envelope calculations were not far from the most recent figures of Hispanic/Latino representation in doctoral degrees.
My goal was not political or inflammatory in nature, I merely wanted to bring attention to the low numbers of Hispanic/Latinos with doctoral degrees, share my excitement for finishing my degree, and connect with others in a similar position. I hope this response can bring some clarity.
The US Census Bureau gathers data on educational attainment of people living in the US, broken down by race, sex, age, and other categories. They specifically include "Hispanic origin" (though they do not consider this to be a "race"). The data come from the American Community Survey and are widely considered to be authoritative.
Here is their data for 2017.
If you look in Table 1, "Educational Attainment of the Population 18 Years and Over, by Age, Sex, Race, and Hispanic Origin: 2017", in the file labeled "Hispanic (of any race)", you'll find they estimate there are 235,000 people over age 18 of Hispanic origin who have a doctoral degree.
In the "All Races" file you can find that the total population over age 18 in 2017 was 246,325,000. So of the total US adult population, about 0.095% are people of Hispanic origin with a doctoral degree; even less than Freytes-Ortiz's figure. Also note that "Doctoral degree" may include degrees other than PhD, such as EdD, DD, etc (though not MD, DDS, JD, etc, which would be under "Professional Degrees").
This file also estimates 4,096,000 people of all races with doctoral degrees. From this, I get that 1.6% of the adult US population has a PhD, and of those, 5.7% are of Hispanic origin. Pretty close to Freytes-Ortiz. She may be using data from a different year, or a different source.
So, I did a run using Census data, but a few clarifications (which some people have pointed out, but just to bring it all together): Census asks separate questions about a person's race (e.g., white, black, Asian, etc) and a person's ethnicity (that they are of Hispanic/Latino origin, or not).
As someone mentioned earlier, Census does not distinguish between Hispanic, Latino, or Spanish. You can find a CPS questionnaire here. I do think it is a little strange that she alternated between words (though Census does treat them interchangeably, for whatever that's worth).
Okay, so with those caveats -- When reporting poverty by education, Census usually limits to adults over 25, presumably because many 18 year olds will receive college degrees but are to young to do have done so (for example, see Table 3 from their annual report here).
Using Census microdata from the March 2017 Current Population Survey, we find
- 1.88% of the population 25+ have a PhD
- of these, 5.7% are Hispanic or Latino (compared to 15% of the entire population 25+)
- 0.7% of the Hispanic/Latino population 25+ have a PhD
- 0.11% of the population 25+ are both Hispanic/Latino and have a PhD.
As Nate mentioned, she could simply be using data from an earlier year or from a different source like the American Community Survey.
This is pretty consistent with my priors; however, I'm glad you're surprised -- it means your program must be doing something right in terms of engaging diverse communities. Hopefully that spreads!