Delivery alert

There may be an issue with the delivery of your newspaper. This alert will expire at NaN. Click here for more info.

Recover password

Census map shows NM likely undercounted

Copyright © 2017 Albuquerque Journal

New Mexicans and Alaskans are the least cooperative Americans when it comes to filling out and returning federal census surveys, according to a recent analysis.

Researchers at City University of New York took the survey return rates for the 2010 census along with recent population estimates and built them into a searchable map that shows most of New Mexico is considered a hard-to-count area.

What that means is the state population and demographics are likely undercounted.

And undercounting results in a loss of federal funding and accompanying private sector attention, the analysis says.

In 2015, the federal government used 2010 census data to distribute more than $6 billion in aid, mostly for Medicaid, to New Mexico. That’s about $3,000 per person.

“So, what would it mean to New Mexico if we had a 1 percent undercount? Well, at 1 percent, that’s 20,000 people; that’s $60 million. Carried over for 10 years (until the next census) that’s $600 million,” said Robert Rhatigan, associate director for Geospatial and Population Studies at the University of New Mexico.

Statewide, Catron County had the lowest documented survey return rate, with only 45.1 percent of residents returning surveys. The other least cooperative areas, excluding tribal areas, included tracts and neighborhoods in San Juan, Rio Arriba, San Miguel, Cibola, Socorro, Lincoln, Sierra, Grant and Otero counties.

In the Albuquerque metro area, Corrales residents returned the fewest surveys, with only a 55.3 percent return rate. The majority of the metro area had a return rate high enough to not register as hard to count.

Alaska, in comparison, had only one tract with the lowest category of returns, with a 28.6 percent return rate. The majority of the state is considered hard to count the way New Mexico’s tribal areas are, according to the map and analysis, which was conducted by the Center for Urban Research at the Graduate Center of CUNY.

In addition to funding, census counts are used to determine the number of federal representatives each state can send to Congress.

Rhatigan says New Mexico isn’t in jeopardy of losing or missing out on a representative because of undercounting.

“But an undercount will affect state legislative districts. And so many businesses decide where to invest based on census data,” Rhatigan said. “As they say, ‘bad data makes bad decisions.'”

Hard for NM

While the 2020 census will use internet-based surveys, which the agency hopes will increase the national return rate, New Mexico is still likely to have low returns considering poor internet access.

The Census Bureau considers New Mexico one of the least internet connected states, with about 68 percent of homes with access to high-speed internet. The national state average is about 78 percent.

In addition, New Mexico has a high population of groups that have historically not engaged in the census process.

“Historically, the census has undercounted young children, people of color, rural residents, and low-income households at higher rates than other population groups,” the CUNY report says. “Also … ‘linguistically isolated’ households; frequent movers; foreign born residents; households below the poverty line; large (i.e. overcrowded) households; low educational attainment households; and single-parent headed households … and people who distrust government authorities and/or have been or could be targets of law enforcement or heightened surveillance … ”

For people who don’t send in or submit their survey, the bureau dispatches an in-person “enumerator” to interview residents at the target home. If that fails, the enumerator interviews neighbors.

And that process is very expensive, especially for a state with high population of nonparticipating groups, or for a state with extremely rugged or sparsely populated territory, like Alaska.

“We happen to have a lot of those people in New Mexico,” Rhatigan said.

And that has affected past census attempts.

In the 2000 census, there was a certified 2 percent undercount, which “by many accounts a disaster,” Rhatigan said. The 2010 census was not as bad, he said, but for 2020, Rhatigan is “wanting to hit the panic button.”

Difficult count

Rhatigan echos the concerns of the Government Accountability Office, which listed the 2020 Census on its 2017 high risk report for projects that will likely fail and/or be wasteful.

That report says the census is underfunded and underprepared for the 2020 endeavor, but it says the bureau is attempting some cost-cutting measures by shrinking administration, hiring contractors and attempting to use technology.

Technicians at a firm in Indiana have been hired to scour government satellite images to identify new housing units that need to be added to the survey mailing list. This job used to be done by an army of pedestrians hired to tour every inhabitable block of the U.S. to report new houses or abandoned ones.

Rhatigan said this process has cut down the walking area to 25 percent of what it was in 2010. And the bureau is relying on census contacts in academia or local governments to help verify mailing address lists, to cut down on the bureau’s list responsibility.

It’s working on email surveys to replace the paper surveys, hoping that increases the response rate and reduces how many in-person interviews must be done. According to the GAO, the bureau still needs to do a dress rehearsal of its IT system for the online surveys.

And for those cases in which an enumerator goes into the field and still fails to interview a household or its neighbors, they will now be able to search government databases. So if someone has registered for Medicaid or Social Security or other government program, those documents can be used by an enumerator to fill in data about how many people live in the home and their demographics.

In 2010, this process cost $12.3 billion, about double the cost of the 2000 census, which was about double the cost of the 1990 census.

But this year, the bureau does not get double the budget. Its budget now is about $12.5 billion, unless provided with more money.

“The budget is getting pulled out from under them,” Rhatigan said. “So it falls on people like me to be the advocate for the state, to say ‘did they really capture all the growth?'”