Author Archive for lisan

Page 4 of 25

Reproducibility Initiative: It’s not just for cancer

Reproducibility Initiative logo

The following are links to related efforts in Open Science. The first is about funding for a “Reproducibility Initiative” to validate 50 landmark cancer studies. Frankly, this can/should apply to population research as well. Included are links from The Economist and Nature about the importance of replication.

In general, there is a move towards “Open Science” across all disciplines. In fact, a different initiative, “The Reproducibility Project” is an effort to identify the predictors of reproducibility among published studies in psychology – a field that contributes far too much to the “Retraction Watch” website.

Reproducibility Initiative
Science Exchange News
October 16, 2013

Initiative gets $1.3 million to verify findings of 50 high-profile cancer papers
Richard Van Noorden | Nature News Blog
October 16, 2013

Unreliable research: Trouble at the lab
The Economist
October 19, 2013
Scientists like to think of science as self-correcting. To an alarming degree, it is not.

The governments of the OECD, a club of mostly rich countries, spent $59 billion on biomedical research in 2012, nearly double the figure in 2000. One of the justifications for this is that basic-science results provided by governments form the basis for private drug-development work. If companies cannot rely on academic research, that reasoning breaks down. When an official at America’s National Institutes of Health (NIH) reckons, despairingly, that researchers would find it hard to reproduce at least three-quarters of all published biomedical findings, the public part of the process seems to have failed.

If a job is worth doing, it is worth doing twice
Jonathan Russell | Nature
April 3, 2013

Reproducibility Project
Large-scale open collaboration to estimate the reproducibility of a sample of studies in psychology

Retraction Watch
Tracking retractions as a windo into the scientific process

Center for Open Science
A non-profit organization, which provides infrastructure tools for open science.

Dear Congress: Why are you doing this?

It is likely that the government shutdown and debt ceiling crisis will be resolved this week, but there has still been harm to the data and research infrastructure. Bookmark this post and use it as notes for your next letter to your representatives.

The Government Shutdown was Temporary, Its Damage to Science Permanent
Andrew Rosenberg | Scientific American
October 18, 2013

Federally funded science allows us to do things as a country that we could never do alone. But the threat of shutdown, combined with inconsistent funding from Congress, leaves America’s scientific enterprise in the lurch.

Shutdown: It ain’t over when it’s over
Jeff Neal | Federal News Radio
October 15, 2013
Author notes that the shutdown is not a toggle switch, where we can easily switch the government back to “on.” There rare many repercussions of the shutdown, detailed in the post.

Sunday Shutdown Reader: Harold Varmus on Self-Destruction in the Sciences
James Fallows | The Atlantic
October 13, 2013

Closed Question
Editorial | Nature
October 9, 2013
The US shutdown is damaging science, and Congress must be called to account.
There are more specific stories, linked to the end of this editorial. In case, they don’t remain linked, here they are:
NASA missions struggle to cope with shutdown
08 October 2013
US Antarctic research season is in jeopardy
04 October 2013
NIH shutdown effects multiply
02 October 2013
US government shuts down
01 October 2013

Cancelled NIH study sections: a subtle, yet disastrous, effect of the government shutdown
Rafael Irizarry | StatsBlogs
October 10, 2013
(This article was originally published at Simply Statistics, and syndicated at StatsBlogs.)

The New York Times has a series of editorials, all tagged with “Government Shutdown.” I’ll link to one of them on funding the Census Bureau.

To Stop the Craziness in Washington, Fund the Census
Teresa Tritch | New York Times
October 4, 2013

And, finally, most readers of this blog probably received an Action Alert from Population Association of America (PAA). When it shows up on the PAA website, I’ll link to it here.

Time for a Legal Prohibition on Data Re-identification?

This is a very thorough blog post on respondent re-identification issues. The author takes to task the re-identification rainmakers, who have made careers out of exposing re-identification risks – often overstating the risks. He calls for a well-designed legal prohibition on data re-identification.

In fact, may of the restricted data contracts PSC users operate under have an “inadvertent discovery” clause. Here’s the language from LAFANS, which prohibits broadcasting the “find” to others.

Ethical Concerns, Conduct and Public Policy for Re-Identification and De-identification Practice: Part 3 (Re-Identification Symposium)
Daniel Barth-Jones | Columbia University
October 2, 2013

Janet Yellen, Economic Demographer

Janet Yellen was nominated as the first female head of the Federal Reserve yesterday [note, that Rand Paul has put a hold on the nomination] . Here is a paper she and her husband George Ackerloff wrote almost 20-years ago on the increase in unmarried childbearing:

An Analysis of Out-Of-Wedlock Births in the United States
George Ackerloff and Janet Yellen | Brookings Review
Fall 1996
This Policy Brief was prepared for the Fall 1996 issue of the Brookings Review and adapted from “An Analysis of Out-of-Wedlock Childbearing in the United States,” which appeared in the May 1996 issue of the Quarterly Journal of Economics.

Also of interest is a news story about Ackerloff in the mid-1980s on “efficiency wages” based on their experience hiring babysitters:

Why Unemployment Sometimes Lingers On Stirs Renewed Interest
Alan Murray | Wall Street Journal
December 26, 1985
Note that a young Larry Summers (age 30) is mentioned in this piece. Sticky wages are also mentioned in this summary of her appointment in the New York Times:

Yellen’s Path From Liberal Theorist to Fed Voice for Jobs
Binyamin Appelbaum | New York Times
October 9, 2013
As thorough as this piece is, it fails to mention that Charlie Brown was a teaching assistant for her at Harvard.

A Sad Day for Demographers

The government shutdown has affected many government websites and data operations. And, in fact, it is useful to look at this sortable table to see how many furloughed workers there are by agency (Commerce is 87%). Below are links to some compilations – some might be useful as images to use in presentations:

[Slideshow of shuttered government websites] via @phylogenomics
[Fact-checking websites shutdown] via @PolitiFact

Here’s a re-cap by the WSJ on how to keep track of the economy during the government shutdown. Losing access to the federal statistical system is quite crippling.

How to Track the Economy During A Government Shutdown
Josh Mitchell and Jeffrey Sparshott | Wall Street Journal
October 1, 2013

Finally, there is a work-around to reaching shuttered government websites via the internet archive. However, the access is slow and some of the data-access tools don’t work. For instance CDC Wonder gets stuck in an “I agree” loop.

internet archive feature

The internet archive has even hard-coded archival links to many of the shuttered websites, but I find this access slow – perhaps because everyone is using the same portal. Still, this is a useful link for posterity:

Blacked Out Government Websites Available Through Wayback Machine
Posted on October 2, 2013 by brewster

Counting Prisoners

The New York Times had another editorial on this issue:
Prison-Based Gerrymandering
Editorial Board | New York Times
September 26, 2013

A search on its site shows that this has been a common editorial/story topic
[Counting Prisoners Editorials/Stories]

The PSC Infoblog has had a previous post on this topic as well, which included the Census Bureau’s response to the issue. The Census Bureau released group quarters data in time for redistricting.

Another excellent source on this topic is the National Academy of Sciences book, which is available in the PSC library:
Once, Only Once and in the Right Place: Residence Rules in the Decennial Census
Daniel Cork and Paul Voss, Editors | The National Academies
2006

Two updates from The Census Project

Sorry, Come Back Later (Make an Educated Guess in the Meantime)
Terri Ann Lowenthal | The Census Project Blog
October 7, 2013
This piece reflects on the government shutdown – reflecting that Congress has done what the House couldn’t do – shut down the federal statistical infrastructure.

Thanks to the government shutdown, the Census Bureau’s work has come to a grinding halt. No harassing phone calls to unwitting, over-burdened citizens. No pesky, door-knocking surveyors invading the privacy of hard-working Americans who just want to live a quiet, government-free life (as soon as someone fills that pothole down the street). Even the duty-bound who want to cooperate (however grudgingly) from the comfort of their own computers are out of luck; online survey response is closed for business.

Losing Sleep (While Counting Sheep)
Terri Ann Lowenthal | The Census Project Blog
October 1, 2013
This is a timely piece on the first day of a government shutdown. The Census Bureau needs funding from Congress so that it can do the necessary research for a smarter 2020 Census. The review of the fiscal situation and Congressional gridlock is not pretty.

budget uncertainty is causing significant concerns for the 2020 census program as we enter that period during which it is crucial to conduct tests so that we can begin applying new technologies and methods … We have already delayed planned research and testing activities to later years … We cannot further delay critical research that will help us make critical design decisions for those systems. [John Thompson]

. . . the Census Bureau needs money to figure all of this out in time. The bureau can execute a fundamentally redesigned 2020 census for the 2010 census price tag (plus inflation), Director Thompson says. Invest now, save later – that’s the bottom line.

And, a nice closing line: “Did I mention that the next census starts in less than six years? The Census Bureau can do a lot of things, but it cannot stop the clock. I bet Director Thompson is having a few sleepless nights, too.”

I might note that according to this link, 87% of Commerce employees are furloughed. Obviously, Census2020 planning is not happening today. I wonder if it shuts down the data collection operations for the ACS, CPS, etc.?

census.gov is #shutdown but you can read about Census 2010 research

The Census Bureau website is down with the government shutdown:

Census shutdown message

But, you can read all about some research based on the 2010 Census. Here is a sampling:

Misclassifying New York’s Hidden Units as Vacant in 2010: Lessons Gleaned for the 2020 Census
Joe Salvo and Peter Lobo | Population Research and Policy Review
August 6, 2013
This is a great article if you are interested in the details of the history of the Census Bureau’s master address file; how it gets created, corrected, updated, etc.

This piece traces the puzzling number of vacancies in two areas of New York City during the 2010 Census, which resulted in a lower census count than New York City had expected. It is a nice piece of detective work. As a reminder, Peter Lobo was a PSC trainee who I always quote as saying “I worked with Ren Farley at Michigan, and the time I spent there were some of the best years of my life.”

Quality and the 2010 Census
Hogan, Howard et.al. | Population Research and Policy Review
April 5, 2013
This is a nice summary of ways to evaluate census quality – 2 of the 5 authors are PSC trainees (Howard Hogan and Victoria Velkoff).

There is a companion press conference on the Census Bureau website, which will be linked to when the #shutdown is over.

The rest of the articles in this special issue devoted to the 2010 Census are here:

Population Research and Policy Review
Volume 32, Issue 5, October 2013
Special issue on New Findings from the 2010 Census
Guest Editor: William P. O’Hare

The Census as a Luxury

The following are a collection of news articles, editorials, and reports on the likely possibility that the UK will be scrapping its census – to be replaced by a survey and a sweep of commercial data sources and administrative records.

In for the count: Arguments for scrapping UK census do not add up
Editorial | Fiscal Times
September 2, 2013

Quotes:

The census is Whitehall’s window on British society. If you are not counted, you do not count.

Yet the government believes that the census is a luxury that Britain can no longer afford. When it was last conducted in 2011 it required a 35,000-strong army of researchers and cost £480m. This is cheap by international standards – the US census costs more than three times as much per head – and a drop in the £7tn-odd ocean of public spending that will, over the course of a decade, be influenced by the results. Census data can save ministers from costly mistakes. Abolishing it may prove penny-wise but pound-foolish.

And, the Canadian problem:

At stake is not only the accuracy of the census itself, but also that of countless sample-based surveys that it is used to calibrate.

And, a nice, pithy conclusion:

The carpenter’s rule is to measure twice and cut once. The government should reflect on this advice before ditching an indispensable yardstick of social change.

Ending the national census would make us blind to our society
Danny Dorling | The Guardian
September 2, 2013

This is a shorter version of a paper published in Radical Statistics. This paper has useful references.

The 2011 Census: What surprises are emerging . . . cancellation is stupid
Danny Dorling | Radical Statistics

The next two articles are from the Fiscal Times. This is not an open access journal. You may be able to read the articles by answering a survey question (on cloud computing, smart phone usage, etc.).

Researchers in UK count cost of plan to scrap census
Kate Allen | Fiscal Times
September 1, 2013
This piece emphasizes that neither a survey or currently available data (administrative/proprietary commercial) would provide the geographical detail or the content scope that a census would.

Loss of census seen as threat to UK historical insight
Kate Allen | Fiscal Times
September 1, 2013
This piece emphasizes what a loss of a census would mean for historical research, including genealogy.

Natural Experiments: In Education

A former PSC student, Mike Lovenheim, has several recent papers that have made the news because they shed light on educational policy. I’m only going to highlight two of these, but link to several of his working papers for those who want to see the creative ways he is examining the economics of education and issues in local taxation. And, a new paper has just been added to this entry. It uses lottery-data to evaluate the NY small school movement.

Small High Schools and Student Achievement: Lottery-Based Evidence from New York City
Atila Abdulkadiroğlu, Weiwei Hu, Parag A. Pathak | NBER
October 2013
Abstract | Paper

Early Retirement Incentives and Student Achievement
Maria D. Fitzpatrick & Michael F. Lovenheim | NBER Working Paper, No. 19281
August 2013
Abstract | Paper

[Popular Press]
Student Test Scores Rose When Teachers Retired Early
Khadeeja Safdar | Wall Street Journal
August 19, 2013

Does Federal Financial Aid Affect College Enrollment? Evidence from Drug Offenders and the Higher Education Act of 1998
Michael F. Lovenheim and Emily G. Owens | NBER Working Paper, No. 18749
February 2013
Abstract | Paper

[Popular Press]
Financial Aid for College Students With Drug Convictions
Freakonomics
February 7, 2013