Census 2020 looking to go online

Reaching younger Americans is part of the reason plans are underway to move parts of the decennial 2020 census online:

Millennials (born from 1981 to 1996) and Generation Z (born after 1996) account for about 35 percent of the approximate 325 million people in the U.S., according to estimates, and census officials say their traditional means of outreach — mail-in questionnaires, landline phone calls and door-to-door surveys — are failing to connect with this significant segment of the population.

The Census Bureau plans to conduct its first-ever online headcount, which it predicts will generate 60 percent of the total responses for 2020…

However, social scientists suggest that millennials and Generation Z could have a hard time appreciating the importance of the census, having grown up amid a distorted media landscape of instant online gratification, “fake news” and a culture of likes on social networks…

Last month, census communications chief Burton Reist was quoted as saying endorsements from celebrities such as LeBron James are being considered. He described a hypothetical situation in which the NBA superstar urges young people during halftime to pull out their cellphones and “answer the census.”

Moving data collection online would seem to offer a lot in terms of lower costs and easier data tabulation. But, as the article suggests, it brings along its own issues such as cutting through the online clutter and working with celebrities to pitch the online data collection.

On one hand, this might lead to the conclusion that it is still difficult to use web surveys to collect information on a broad scale. Unless a research company has a panel of possible participants in a recruited and relatively representative panel, reaching the broader public on a voluntary basis is hard.

On the other hand, perhaps this should be taken as a good sign: the Census Bureau clearly indicates their data collection has to match what people actually use. Going door to door may not be feasible going forward. If people are online or using devices for hours a day, online surveys might be more attractive.

Almost regardless of how this turns out in the 2020 count, it will be an interesting experiment to watch. What will the online response rate be? How will the Census Bureau have to go about advertising online data entry?

Census income figures misreported based on gender norms

The Census measures numerous important features of American life. Yet, accurate measurement is difficult. A new report suggests reported income can not be the most truthful when women make more money than their husbands:

Researchers found that when wives are the bigger breadwinners, husbands report making an average of 2.9 percent more than what’s in their tax filings. Meanwhile, women who make more than their husbands report earning 1.5 percent less than their actual income…

So why does this phenomenon happen? Researchers say they suspect societal expectations about the roles each person plays in a marriage could be a main factor.

“When married couples . . . violate the norm that husbands outearn their wives, the survey respondents reporting the couples’ earnings appear to minimize the violation by inflating the earnings of the lower-earning husbands and deflating the earnings of the higher-earning wives,” researchers wrote in their findings.

If the misreporting is due to gender norms, might we expect this to go away as more women earn more money? Already, “In about one out of four couples surveyed, wives made more money than their husbands.” Give this a few decades and this misreporting might disappear.

On the other hand, social norms can be last a long time even after society has changed quite a bit from when the social norm arose. If the misreporting continues or even increases, it would be interesting to see how the Census and other surveyors adjust their figures.

Race, ethnicity, ancestry – and which one people identify with

I have followed an interesting urban sociology listserv discussion involving a recent New York Times editorial by Herbert Gans where he notes several mistakes the Census Bureau makes in measuring race and ethnicity. It strikes me that sociologists and others really want to measure three different things and then a fourth piece of information we could gather would help us better understand which three traits are more important. Here is what we might measure:

  1. Race. Largely based on skin color in the United States. A long history of black and white with groups in between.
  2. Ethnicity. Largely based on cultural or national groups. Has become more prominent in recent decades with the Census moving in 2000 to a separate question about Hispanic or Latino ethnicity or discussions about a Middle Eastern or Northern African background.
  3. Ancestry. This could align with the two categories but not necessarily. This typically refers a country or people group in a family lineage.
  4. In addition to the three pieces of information, shouldn’t we ask which category is most important to people? It is true that race in the United States has dominated social relations for centuries. At the same time, race on its own is simplistic. A few examples might suffice. A white Jewish person with ancestry in Russia. A non-white Brazilian with ancestry in Africa. A Chinese person from Singapore. A white person from Tennessee who says their ancestry is American (though it may be in Wales and Germany). Different people will see different traits as more essential to their own understanding as well as how they would like others to see them.

Of course, having four categories like this would complicate the study of trends and groups. But, as more people marry across groups and new groups continue to come to the United States, we need a more nuanced understanding of how these traits come together and matter to people.

“The most detailed map of the United States’ racial diversity”

Check out a new map that shows population by race and ethnicity at a very detailed level: SocScape. Curbed provides a brief description of the project:

Adapting a grid-charting system used for mapping the craters of Mars for NASA, Stepinski and his postdoctoral researcher Anna Dmowska, have created the most detailed map of the United States’ racial diversity—ever. The interactive tool displays enormous volumes of census information through more granular units, each representing 323 square feet. The result is a visual presentation that’s more accurate and useful to analysts interested in exploring geographic shifts in population and racial diversity.

Stepinski is already picking up on trends in the data from the 1990, 2000, and 2010 censes: Generally, white neighborhoods have become more diverse, Asian and Hispanic populations appear to be concentrating in distinct geographic pockets, while largely black neighborhoods have not increased in diversity.

Here is a view of much of the Chicago metropolitan region:

Firefox_Screenshot_2017-05-01T22-13-48.452Z

SocScape, Chicago MSA, 2010 Census by race and ethnicity

From this image, it looks like an improved version of the racial dot maps as it has more geographic specificity. The tool also has some added data layers – here is the same region with the 1990 race and ethnicity data:

Firefox_Screenshot_2017-05-01T22-18-02.117Z

SocScape, Chicago MSA, 1990 Census by race and ethnicity

Quite a bit of change over a twenty year stretch with increasing numbers of non-white residents living in the suburbs.

The ongoing politics of the 2020 Census

The dicennial Census is not just a counting exercise; it is a political matter as this commentary suggests.

According to recent documents from the Census Bureau and the Government Accountability Office, the bureau plans to substantially cut back on door-to-door surveying and, instead, use the internet, the Post Office and other means to determine who is living where.

The bureau thinks the 2020 survey will cost $5.2 billion less than the last one (an estimate the GAO questions), but the accuracy could be called into question. There will also likely be worries about fraud because many of the conclusions will be drawn through “imputations” — educated guesses.

In fact, fraud could affect the House of Representatives elections for years to come if someone isn’t watching.

During a recent hearing before the House Oversight Committee, which maintained control over the Census Bureau after the Obama-Emanuel caper, a key technology officer for the 2020 decennial admitted that a fraud prevention system won’t be fully in place until just a few months before the polling starts.

If the Census Bureau – often led by sociologists and other social scientists who have expertise in collecting and analyzing data – is fraudulent because certain parties don’t like the result, what can be left alone?

Sampling and estimation alone does not have to be a problem. Just because the Census can’t reach everyone – and they have certainly tried at points – doesn’t mean that there is room for fraud. If done well, the estimates are made based on accurate samples – meaning they generally match the proportions of the total population – and responsible people reporting on this data will always note that there is not 100% certainty in the data.

The clustering of wealthy counties in the United States

With recently released data, the Census Bureau describes the patterns in the wealthiest counties in the United States:

A Census Bureau report on the “highlights” of the data released yesterday noted that the nation’s wealthiest counties are disproportionately in the corridor of territory that runs from Virginia and Maryland and then north along the East Coast.

“Seventy-seven counties had a median household income within the highest range ($81,129 to $125,900),” said the “highlights” report. “Forty-two of these high-income counties were located in the Northeast region, Maryland and Virginia.”…

Nationwide, the median household income in 2015 was $55,755, according to the Census Bureau. That means the local median household income in each of the nation’s three richest counties—all of which are Washington suburbs in Northern Virginia—are more than twice the national median household income.

Of the Top 20 richest counties in the nation, nine are suburbs of the city that serves as the seat of a federal government

It then wouldn’t be too hard to look for patterns in other demographic data across these wealthier counties. One marker – noted in this article – is that many of these wealthier counties are suburban. But, I’m guessing these counties are also well educated and largely white.

It would also be interesting to see how those concerned with inequality would deal with county level data. Many American counties don’t have a lot of control compared to municipalities or states. There can be a lot of variation within counties, both really wealthy and really poor pockets. Usually, recommendations about poverty or affordable housing are made at a municipal or regional level. Is there a way to leverage counties to address particular issues?

Local note: it appears that three Chicago area counties – Lake, DuPage, and Kendall – fit into the highest range in the data. See page 3 of the Census highlights.

Push continues to add Middle Eastern race/ethnicity to Census

This project has been in the works for at least a few years and now the White House is joining the effort to add a Middle Eastern race/ethnicity to government forms:

Under current law, people from the Middle East are considered white, the legacy of century-old court rulings in which Syrian Americans argued that they should not be considered Asian — because that designation would deny them citizenship under the 1882 Chinese Exclusion Act. But scholars and community leaders say more and more people with their roots in the Middle East find themselves caught between white, black and Asian classifications that don’t fully reflect their identities…

On Friday, the White House Office of Management and Budget advanced the proposal with a notice in the Federal Register, seeking comments on whether to add Middle Eastern and North African as a separate racial or ethnic category, which groups would be included, and what it should be called.

Under the proposal, the new Middle East and North African designation — or MENA, as it’s called by population scholars — is broader in concept than Arab (an ethnicity) or Muslim (a religion). It would include anyone from a region of the world stretching from Morocco to Iran, and including Syrian and Coptic Christians, Israeli Jews and other religious minorities.

But the Census Bureau, which has been quietly studying the issue for two years, also has gotten caught up in debates about some groups — such as Turkish, Sudanese and Somali Americans — who aren’t included in that category. Those are issues the White House is trying to resolve before adding the box on 2020 census forms.

This is a good reminder of how social science categories can develop: through a lot of effort. There are social changes to account for here: how groups understand themselves changes over time. In this particular context, immigrating to the United States and facing particular challenges and opportunities leads to groups that wouldn’t necessarily group themselves together to now desire such a thing. Does the Middle East as Americans often understand it line up with how the region is understood elsewhere (let alone in that part of the world)? There are politics involved: who gets to define such groups? How is the data used? There are measurement issues: who would count in such a category? What are the boundaries? Is it a racial category or ethnicity? How many racial/ethnic categories are needed to understand the world?

All that said, the category of white – typically based on skin color – is very broad and often not very useful considering the social changes in the United States in the last few decades.