A questionnaire change: politics or not?

The Census Bureau has changed how it will measure insurance coverage in the Current Population Survey – Census Bureau announcement.

Many on both the left and right are unhappy with this change because of the discontinuity with the insurance measure over time – in other words, before Obamacare and after Obamacare.

Many of the comments reflect ignorance on how much research goes into changing, adding, deleting questions or that OMB always reviews questionnaire changes (even under Republican presidents). Also this change is to the questionnaire in the CPS, not the American Community Survey (or long-form census).

But, this quote probably reflects the view of many health analysts: “Getting worked up into an increasingly heated health nerd rage about the Census changes. We’re losing our best data source on Obamacare.” via Twitter. [See tweet and comments]

Here is a compilation of articles in the order they were published:

Census Bureau: Sorry, We’re Going to Have a Hard Time Measuring . .
Patrick Brennan | National Review Online
April 15, 2014

Obama Hijacks Census to Whitewash ObamaCare’s Failure
B. Christopher Agee | Western Journalism
April 15, 2014

Both Sides in Obamacare Fight Slam Census Bureau
Rob Garver | Fiscal Times
April 15, 2014

Census Survey Revisions Mask Health Law Effects
Robert Pear | New York Times
April 15, 2014

Want Useful Data on Obamacare? The Census Bureau Won’t Help
Peter Suderman | Reason (blog)
April 15, 2014

Manipulating Obamacare Stats: The Census Bureau’s Suspect Timing
Guy Benson | Town Hall (blog)
April 15, 2014

Obama’s cooking the Census for Obamacare?
Megan McArdle | Newsday
April 15, 2014

“You Had One Job” [Federal Health Insurance Statistics Edition]
Ben Mathis-Lilley | Slate
April 15, 2014

Republicans Accuse Census Bureau of Trying to ‘Hide the Effects of Obamacare’
Michael McAuliff | Huffington Post
April 15, 2014

Finding “war brides” in the ACS

Several years ago the Census Bureau added a “what year were you married” question to the American Community Survey. This was an uncontroversial change to the questionnaire because it helped shore up data on marriages.

See link from the IPUMS for all the new marriage/divorce timing variables

The CDC used to collate marriage and divorce certificate data from state vital statistics offices, but ceased this operation in the mid-1990s due to budgetary constraints [See sad note to this effect].

Here is a nice illustration from Philip Cohen’s Family Inequality blog on using these data to find out how many World War II “war brides” are still alive.

How many WWII war brides are still living?
Philip Cohen | Family Inequality blog
April 14, 2014

If you don’t like his definition of a war bride, make your own and write it up in your own blog.

The Effects of Rurality on College Access and Choice

Via: Chronicle of Higher Education, The Ticker

By: Michael Andrew Koicich, Texas Tech University
Presented at:
AERA 2014 Annual Meeting
April 4, 2014

Abstract:

Students from rural areas have a history of lower educational attainment at both the K-12 and postsecondary levels, yet this population remains understudied. This study seeks to update past studies of rural youth by examining college attendance and choice decisions for students from non-metropolitan counties. Logistic regression is used to study the postsecondary attendance and institutional choice for rural students. Data are taken from two national data sets, the Education Longitudinal Study of 2002 and the 2004 County Typologies published by the U.S. Department of Agriculture, which are then combined to conduct this analysis. The results of this study reveal clear disadvantages in postsecondary attendance, institutional level, selectivity, and other indicators resulting simply from living in a non-metropolitan county. The effect of most student characteristics did not vary substantially between rural and non-rural areas, but the analysis did reveal one interesting finding: The relationship between college attendance and choice and a student’s socioeconomic status was weaker for rural students than for non-rural students. Some systematic variation across rural communities was revealed, as well.

Full paper (PDF)

Working Papers from NBER

Using Social Media to Measure Labor Market Flows
by Dolan Antenucci, Michael Cafarella, Margaret C. Levenstein, Christopher Re, Matthew D. Shapiro #20010
Abstract; PDF

Why Stars Matter
by Ajay Agrawal, John McHale, Alexander Oettl #20012
Abstract; PDF

Preferences, Selection, and Value Added: A Structural Approach
by Saziye Pelin Akyol, Kala Krishna #20013
Abstract; PDF

The Impact of Emerging Climate Risks on Urban Real Estate Price Dynamics
by Devin Bunten, Matthew E. Kahn #20018
Abstract; PDF

The ACA: Some Unpleasant Welfare Arithmetic
by Casey B. Mulligan #20020
Abstract; PDF

How Much Favorable Selection Is Left in Medicare Advantage?
by Joseph P. Newhouse, Mary Price, J. Michael McWilliams, John Hsu, Thomas G. McGuire #20021
Abstract; PDF

Behavioral and Descriptive Forms of Choice Models
by Ariel Pakes #20022
Abstract; PDF

Input Diffusion and the Evolution of Production Networks
by Vasco Carvalho, Nico Voigtlaender #20025
Abstract; PDF

Which Peers Matter? The Relative Impacts of Collaborators, Colleagues, and Competitors
by George J. Borjas, Kirk B. Doran #20026
Abstract; PDF

Census Bureau is considering adding a new Middle East/North Africa ethnic category

Source: Pew Research Center, Fact Tank
By: Jens Manuel Grogstad

From the FactTank story:

The new category would be broader than the Arab ancestry data collected by the Census Bureau since 1980. The Arab-American population is small but growing, and its exact size is disputed. The Census Bureau estimates there are 1.8 million Arab-Americans in the U.S., up 51% since 2000. But the Arab American Institute Foundation estimates there are nearly 3.7 million Arab Americans living in the country. The Arab-American population is also diverse, with people claiming ties to 22 countries and various religious backgrounds.

Read the full story
A story from earlier in the month about other Census form race and ethnicity changes

Working Papers from the NBER

Equalizing Superstars: The Internet and the Democratization of Education
by Daron Acemoglu, David Laibson, John A. List #19851
Abstract; PDF

Institutions, Human Capital and Development
by Daron Acemoglu, Francisco A. Gallego, James A. Robinson #19933
Abstract; PDF

Heterogeneous Paths Through College: Detailed Patterns and Relationships with Graduation and Earnings
by Rodney Andrews, Jing Li, Michael F. Lovenheim #19935
Abstract; PDF

Breastfeeding and Child Disability: A Comparison of Siblings from the United States
by George L. Wehby #19940
Abstract; PDF

Effects of Prescription Drug Insurance on Hospitalization and Mortality: Evidence from Medicare Part D
by Robert Kaestner, Cuiping Long, G. Caleb Alexander #19948
Abstract; PDF

The Sad Truth About Happiness Scales
by Timothy N. Bond, Kevin Lang #19950
Abstract; PDF

Negative Tests and the Efficiency of Medical Care: What Determines Heterogeneity in Imaging Behavior?
by Jason Abaluck, Leila Agha, Christopher Kabrhel, Ali Raja, Arjun Venkatesh #19956
Abstract; PDF

Matching Methods in Practice: Three Examples
by Guido W. Imbens #19959
Abstract; PDF

Empowering Women: The Effect of Schooling on Young Women’s Knowledge and Use of Contraception
by Mabel Andalon, Jenny Williams, Michael Grossman #19961
Abstract; PDF

Aspirations and Inequality
by Garance Genicot, Debraj Ray #19976
Abstract; PDF

Does grief transfer across generations? In-utero deaths and child outcomes
by Sandra E. Black, Paul J. Devereux, Kjell G. Salvanes #19979
Abstract; PDF

Instrumental Variables: An Econometrician’s Perspective
by Guido W. Imbens #19983
Abstract; PDF

Education, Health and Wages
by James J. Heckman, John Eric Humphries, Gregory Veramendi, Sergio Urzua #19971
Abstract; PDF

Who Benefits when the Government Pays More? Pass-Through in the Medicare Advantage Program
by Mark Duggan, Amanda Starc, Boris Vabson #19989
Abstract; PDF

Risk Adjustment of Health Plan Payments to Correct Inefficient Plan Choice from Adverse Selection
by Jacob Glazer, Thomas G. McGuire, Julie Shi #19998
Abstract; PDF

Option Value of Work, Health Status, and Retirement Decisions in Japan: Evidence from the Japanese Study on Aging and Retirement (JSTAR)
by Satoshi Shimizutani, Takashi Oshio, Mayu Fujii #20001
Abstract; PDF

Does Classroom Time Matter? A Randomized Field Experiment of Hybrid and Traditional Lecture Formats in Economics
by Theodore J. Joyce, Sean Crockett, David A. Jaeger, Onur Altindag, Stephen D. O’Connell #20006
Abstract; PDF

Smoking rate stays high among the poor

While smoking began as a practice among the well-off, according to a study published in Population Health Metrics, the habit is now much more prevalent in the working and poor classes.

New York Times article

Full text of the study

Mapping Twitter

Source: Pew Research Internet Project
By: Marc A. Smith, Lee Rainie, Ben Shneiderman, and Itai Himelboim

Mapping Twitter Topic Networks: From Polarized Crowds to Community Clusters

Conversations on Twitter create networks with identifiable contours as people reply to and mention one another in their tweets. These conversational structures differ, depending on the subject and the people driving the conversation. Six structures are regularly observed: divided, unified, fragmented, clustered, and inward and outward hub and spoke structures. These are created as individuals choose whom to reply to or mention in their Twitter messages and the structures tell a story about the nature of the conversation.

Overview
Complete Report (PDF)
Infographic: The six types of Twitter conversations

Big Data Initiative at NIH-OBSSR

From the Connector blog post:

The NIH Big Data to Knowledge (BD2K) initiative is designed to address these issues and facilitate broad use of biomedical big data through new data sharing policies, catalogs of datasets, and training. Behavioral and social scientists should be aware of several recently-issued RFAs. In these RFAs NIH is requesting applications for Centers of Excellence, Data Coordination Centers, training enhancement, and data facilitation. If you are involved in mHealth, this might be a great opportunity for you, or if you are pooling data for the purposes of GxE interaction studies in the behavioral and social sciences this initiative might also fit you well. Critically consider your current research and ways that Big Data may already be part of your portfolio.

Read the full post
NIH Big Data to Knowledge (BD2K) website

Obama Budget Proposes Only Modest Increases for Research Spending

By: Paul Basken
Source: Chronicle of Higher Education

The budget proposes an increase of $30.4 billion, which when inflation is accounted for, is a cut of 1%. NSF, NASA and the Department of Defense fare little better.

Read the full Chronicle article here