Most of the familiar statistical packages social scientists work with are not well-equipped for analysis of text. Python is one tool often used with text data.
Here is a series of Python tutorials posted on Neal Caren’s Github site. Notice the wide-prevalence of code sharing. That is a feature of much of the folks who work in this field.
You can follow his tutorials on Python or take a Coursera course by a UM professor in February. Another option is the Coursera Data Science specialization offered via Johns Hopkins. This set of courses skips Python but includes a snapshot of the variety of concentrations in this field.
Learning Python for Social Scientists [list curated by Neal Caren]
Programming for Everybody (Python) [University of Michigan via Coursera]
Data Science Specialization [Johns Hopkins via Coursera]
Here’s a rendering of that specialization from a student in the Data Toolbox course:
Source: Uri Grodzinski
Who knew that the name Violet was such a good example of a bi-modal distribution?
This was drawn from a very fun post:
How to Tell Someone’s Age When All You Know is Her Name
Nate Silver and Allison McCann | FiveThirtyEight blog
May 29, 2014
We had a previous post on fun with the Social Security names database.
This age of names example is a great applied demography exercise – calculating the median age of names. For that you’ll need a link to the full names database and cohort life tables:
Beyond the Top 1000 Names
Cohort Life Tables for the Social Security Areas by Calendar Year
Here’s also a nice link to some Big Data exercises via Python. There is a lot of code sharing in this GitHub repository.
Both the Pew Research Center and the FiveThirtyEight blog have done write up about the trouble the U.S. Census has counting same-sex couples.
Pew’s story (which came out in September) discusses the way gender reporting on the census confounds the data.
The story in FiveThirtyEight reports on how the Census Bureau is working to make it’s questions gather more accurate data.
The Antidote for “Anecdata”: A Little Science Can Separate Data Privacy Facts from Folklore
Daniel Barth-Jones | Info/Law Blog [Harvard]
November 21, 2014
This is a great piece that shows again that most of the publicity about re-identification in data are overblown:
The 11 in 173 million risk demonstrated for this celebrity ride re-identification (or 1 in 15,743,614) is truly infinitesimal. To put this in perspective, this risk is over 1,000 times smaller than one’s lifetime risk of being hit by lighting. With proper de-identification applied and the cryptographic hash problem fixed in any future data releases, this spooky specter of celebrity cyber-stalking using TLC taxi data is likely to vanish as soon as one turns on the lights.
This blog post is in reaction to the release of NYC taxi medallion data, which were improperly anonymized. A previous blog post described the data.
Here is the piece that sensationalizes the possibility of re-identification, based on famous people who ride cabs.
Riding with the Stars: Passenger Privacy in the NYC Taxicab Dataset
Anthony Tockar | Neustar Blog
September 15, 2014
This is a big data resource, and more. Check out the reaction to the bad anonymization here.
20GB of uncompressed data comprising more than 173 million individual trips. Each trip record includes the pickup and dropoff location and time, anonymized hack licence number and medallion number (i.e. the taxi’s unique id number, 3F38, in my photo above), and other metadata.
Before the link to the data, here’s an analysis based on similar data:
Why New Yorkers Can’t Find a Taxi When It Rains
Eric Jaffe | City Lab Blog
October 20, 2014
Provides a nice synopsis of some research using taxi cab rides. Read it for the links to the formal research papers.
New York City Taxi Cab Trips [in small chunks]
FOILing NYC’s Taxi Trip Data
Chris Whong | personal website of an Urbanist, Mapmaker, Data Junkie
March 18, 2014
a synopsis of how he got the data via a FOIA request & a link to the data on rides/fares as single files, instead of the chunked version above.
and the story about how the taxicab medallion IDs were improperly anonymized:
Poorly anonymized logs reveal NYC cab drivers’ detailed whereabouts
Dan Goodin | ars technica
June 23, 2014
On Taxis and Rainbows: Lessons from NYC’s improperly anonymized taxi logs
Vijay Pandurangan | Medium blog
The PSC Infoblog has reported on this earlier here, but this is still of interest.
The U.S. Census Is Trying To Get A More Accurate Count Of Arab Americans
Ben Casselman | fivethirtyeight.com Blog
November 24, 2014
Note that this article mentions that the Census Bureau did a special tabulation for Homeland Security to provide counts of Arab populations by geography (place and zip code).
Some Arabs have expressed reluctance to identify themselves on a government form, especially after the Census Bureau shared detailed data on the Arab-American population with the Department of Homeland Security in the early 2000s
These “detailed tabulations” referenced above, were public use tables from American FactFinder. Here’s the original FOIA request from the Electronic Privacy Information Center:
FOIA request: Department of Homeland Security Obtained Data on Arab Americans From Census Bureau [Source: EPIC]
Here is the example for Places drawn from DP-2. Here’s the example for Zip Codes drawn from (Tables PCT16 and PCT17).
Researchers at the Census Bureau and the Minnesota Population Center matched census responses across the 2000 and 2010 Census and examined changes in the Race/Hispanic origin responses. They found that the biggest change was Hispanics changing their “Other Race” choice in 2000 to “White” in 2010. The groups least likely to change responses were those who identified themselves as non-Hispanic white, black or Asian in 2000.
America’s Churning Races: Race and Ethnic Response Changes between Census 2000 and the 2010 Census
Center for Administrative Records Research and Applications & Minnesota Population Center
CARRA Working Paper #2014-09
August 8, 2014
This paper is nicely summarized here – based on the PAA version:
Millions of Americans changed their racial or ethnic identity from one census to the next
D’Vera Cohn | Pew Research Center
May 5, 2014
As a reminder, the Census Bureau has spent the better part of the past year, looking to change how it collects data on race and ethnicity.
U.S. Census looking at big changes in how it asks about race and ethnicity
D’Vera Cohn | Pew
March 14, 2014
2010 Census Race and Hispanic Origin Alternative Questionnaire Experiment[website]
2010 Census Race and Hispanic Origin Alternative Questionnaire Experiment[report]
By: Ben Wieder
A new federal initiative that could provide millions of students with a free lunch might have an unexpected cost for researchers and state educational agencies.
“It’s obviously good for kids, but from a pure data perspective it provides some weaknesses,” said Brandon LeBeau, an assistant professor at the University of Iowa’s College of Education who has studied the use of free lunch eligibility in education research.
Read the full story
Recent research based on an original question in the Health and Retirement Study (HRS) shows how few folks felt they would live to 75 (at age 50). There is also an association with the low probability folks dying earlier than the high probability folks.
Popular press coverage and a Brookings publication below:
You’ll probably live much longer than you think you will
Christopher Ingraham | Washington Post [Wonkblog]
November 10, 2014
Better Financial Security in Retirement? Realizing the Promise of Longevity Annuities
Katharine Abraham and Benjamin Harris | Brookings
November 6, 2014
Abstract | Full Paper
You can comment on the Census Bureau’s plans to remove some questions from the American Community Survey (marriage history and field of study in college) via the Federal Register:
link to Federal Register Notice
A working paper by Kennedy and Ruggles provides some talking points on the marriage history question: “Breaking up is Hard to Count . . . ” And, quite a number of researchers of the STEM population, including the migration of STEM folks, ought to be interested in the field of study question.
It is interesting to note that questions that were thought to be vulnerable (flush toilet, leaving time for work, income, and mental/emotional disability) were unscathed. For historical purposes (e.g., April 2014) it is interesting to review a summary of these touchy questions.
Here is a summary of how the Census Bureau came up with the questions to be eliminated. It comes down to a grid of mandated/required questions x user burden/cost:
American Community Survey (ACS) Content Review
Gary Chappell |Census Bureau
October 9, 2014
Other helpful links are on the ACS Content Review website.