Is a Veteran With a Service-Connected Disability the Same as a Disabled Veteran?

Bookmark and Share

By Kelly Holder, Social, Economic and Housing Statistics Division

The Census Bureau’s American Community Survey (ACS) provides a variety of information critical to understanding the needs of America’s veterans. This information is relied upon by the Department of Veterans Affairs (VA) to provide services to the 19 million veterans living in the U.S. and Puerto Rico. One topic of particular importance to VA and other veterans’ advocates is disability status.

There are several definitions that can be used to address issues facing America’s disabled veteran population. The ACS has included two distinct question series related to disability status since 2008 to provide key statistics for use primarily by VA.

The first question series measures functional disabilities and difficulties with activities of daily living among the general population. The second series measures VA service-connected disability status and ratings. A service-connected disability is a disability, disease or injury incurred or aggravated during active military service. The degree of disability is graduated from 0 percent to 100 percent in increments of 10 percent.

The combination of disability data with the detailed socioeconomic characteristics available from the ACS improves how VA measures the demand for health care services. ACS allows VA to cross-classify service-connected disability status with income to determine which veterans might qualify for priority health care enrollment. ACS also provides market-level statistics for VA to determine the location of health care facilities.

In 2014, about 28.8 percent of all veterans had any type of disability, as defined by the ACS, and 19.6 percent of all veterans (about 3.7 million) had a VA service-connected disability rating. Looking at the two concepts in conjunction, 8.6 percent of all veterans had both a service-connected disability and an ACS-defined disability. Eleven percent had a service-connected disability and no ACS-defined disability, while 20.2 percent had no service-connected disability but did have an ACS-defined disability (see figure below).


So, is a veteran with a service-connected disability the same as a “disabled” veteran?

Not necessarily. Veterans with a service-connected disability rating were more likely at all ages to have an ACS-defined disability than veterans without a service-connected disability rating. However, even those veterans with the highest ratings did not all report having an ACS-defined disability. About 35 percent of veterans with a service-connected disability rating between 0 and 40 percent, and 54.6 percent of veterans with ratings of 50 percent or higher had an ACS-defined disability. In this sense, having a service-connected disability rating does not equate to having a disability for these veterans.

The two measures of disability on the ACS are correlated but are not interchangeable. Therefore, it is important for data users to know how they want to define “disabled” when categorizing veterans. Questions about disability status generally measure prevalence of disability in the population. The series of six questions does not capture all types of disability or account for the severity of an individual’s impairment. It is also not possible to determine when a disability occurred, which is relevant to the discussion of service-connected disability status in the ACS. The concept of service-connected disability status generally measures participation in a benefits program administered by the VA.

Using data from the American Community Survey, the Census Bureau’s latest working paper on veterans, “The Disability of Veterans”, highlights the differences in the measurement of these disability concepts and further discusses the prevalence and types of disabilities among the veteran population.

For more information on veterans, visit the Census Bureau’s veterans statistics home page.

Posted in Uncategorized | 1 Comment

Census Bureau Statistics Allow for Deeper Dive Into Rising Costs of Child Care

Bookmark and Share

Written By: Lynda L. Laughlin, Fertility and Family Statistics Branch

In the last quarter century, out-of-pocket costs for child care have nearly doubled. During the same period, the types of arrangements that families use have also changed. Results from the Census Bureau’s Survey of Income and Program Participation (SIPP) show us how usage patterns in child care vary over time based on a family’s income.

Adjusting for inflation, families with employed mothers spent $84 per week on average in 1985, which increased by 170 percent to $143 per week in 2011. Since 1997, the use of both organized day care centers and father-provided care increased while the proportion of children in nonrelative care in the provider’s home has decreased.

Recent research suggests that the rise in out-of-pocket costs might be driven by higher income families purchasing more expensive forms of child care typically found in child care centers. Historically, higher income families are also more likely to pay for child care services.

Beyond differences in who pays for child care and how much is paid, SIPP provides detailed data on primary child care arrangements from 1993 to 2011 under three categories: families with incomes below poverty, those with incomes between 100 and 199 percent of the federal poverty threshold, and families with incomes at least 200 percent of the federal poverty threshold. For ease of discussion, we will refer to these groups as poor, low-to-moderate, and moderate-to-high income families, respectively.

Below, in Figure 1, we explore trends in three commonly used primary child care arrangements by family income: grandparent care, center-based care and nonrelative child care. Historical data on other types of primary arrangements for children under 5 can be found on The primary child care arrangement is defined as the arrangement that children spent the most time in on a regular basis.

Figure 1

Figure 1 presents data on selected primary child care arrangements of preschoolers with employed mothers between 1993 and 2011, regardless of family income. Since 1997, the use of center-based care has increased while the proportion of children in nonrelative care decreased. In 1993, 17 percent of preschoolers were primarily cared for by a grandparent, which gradually rose to 21 percent by 2011.

Figure 2 shows the variation in the use of grandparent care between 1993 and 2011 by income status.

Figure 2

It is important to note that poverty thresholds vary by family size and composition. For example, in 2011, a four-person family with two adults and two children was considered poor if their annual household income was less than or equal to $22,350. A family of similar size and composition was considered moderate to high income if their annual household income in 2011 was greater than or equal to $44,700.

Between 1993 and 2011, the rates of grandparent care ranged between 19 to 24 percent for poor families. Peak uses of grandparent care as a primary child care arrangement for this income group occurred in 1999 (24 percent) and 2010 (23 percent). Moreover, rates of grandparent care for low-to-moderate income families fluctuated between 1993 and 1999 and then remained around 20 percent between 2002 and 2011. In 2011, 20 percent of preschoolers living with poor families primarily used grandparent care, which is not statistically different from the rate for low-to-moderate (20 percent) or moderate-to-high income families (21 percent).

Center-based child care (Figure 3) includes care in a day care center, nursery, preschool or federal Head Start program. With the exception of 1995, moderate to high-income families were more likely to use center-based child care than poor or low-to-moderate income families. While this gap has fluctuated over time, the gap was largest in 2010. Readers should use caution when making comparisons between years because of changes in the SIPP design, such as changes in the number of response options, modes of data collection and seasonality issues related to work schedules and changes in the number of response options for the questions.

Figure 3

Lastly, nonrelative child care (Figure 4) includes family day care homes or other nonrelative caregivers in the child’s or provider’s home (baby sitters, nannies, etc.). The proportion of children in nonrelative care has decreased for all income groups. Moderate-to-high income families are more likely to use nonrelative care than poor or low-to-moderate income families.

Figure 4

Examining variations in child care usage by family income is just one way of understanding the relationship between child care arrangement type and out-of-pocket child care costs for families. Trends suggest that moderate-to-high income families are more likely than poor or low-to-moderate income families to utilize formal arrangements such as child care centers or family day care homes, arrangements that often require monetary payments.

For more information on child care arrangements, visit For further information on the source and accuracy of the estimates, please visit the SIPP Web page.

Posted in Uncategorized | Leave a comment

Visit Us at the 2016 Allied Social Science Association and American Economic Association Meeting in San Francisco

Bookmark and Share

By Randy Becker, Center for Economic Studies

Census Bureau economists will present results from their research at the annual meeting of the Allied Social Science Association (ASSA) and American Economic Association (AEA) in San Francisco Jan. 3-5, 2016. This meeting brings together more than 11,000 economists and scholars in related fields from around the world and showcases ongoing research in economics. Census Bureau economists will also serve as discussants of related papers in their fields of expertise, act as panelists and recruit doctoral candidates interested in careers at the Census Bureau.

Economists at the Census Bureau play a key role in creating and improving statistical products that are essential to policymakers, researchers and the public. These products come from a variety of sources, such as survey microdata on businesses and households, linked employer-employee data and confidential microdata from federal and state administrative and statistical agencies. Our economists apply these data to the study of income and labor dynamics, industrial organization, household structure, health and disability, international trade and other topics.

This year, the ASSA/AEA meeting includes 21 papers with Census Bureau co-authors showcasing recent findings on the following diverse range of topics.

Labor and earnings dynamics: Labor and earnings dynamics continues to be an area of fruitful research at the Census Bureau. Much of this research uses linked employee-employer data from the Longitudinal Employer-Household Dynamics (LEHD) program. Papers include examinations of firm performance and the volatility of worker earnings (Juhn, McCue, Monti and Pierce), downward wage rigidity (Kurmann, McEntarfer and Spletzer), the effects of public housing demolitions on long-term earnings of children (Pollakowski, Andersson, Haltiwanger, Kutzbach and Palloni) and the impact of workplace characteristics on worker earnings (Barth, Davis and Freeman). Other papers examine changes in marriage and earnings patterns using data from the Survey of Income and Program Participation (SIPP) linked to administrative data on earnings (Juhn and McCue) and the duration of self-employment for those becoming self-employed during the Great Recession (Luque and Jones).

Mobility: Three papers focus on geographic mobility. They include an investigation of the recent decline in residential mobility and job switching (Hyatt, McEntarfer, Ueda and Zhang), an examination of the roles of employment, earnings and living costs in migration (Janicki, Kutzbach, Nowak and Sandler) and a look at whether low income housing tax credits provide a path to better neighborhoods (Brummet and Bartalotti). Two other papers focus on intergenerational mobility; one examines occupation mobility from 1850 to 2000 (Ferrie, Massey and Rothbaum) and another documents the tendency of fathers to share employers with their sons and daughters (Stinson and Wignall).

Business economics: Research using microdata on businesses is also a central focus at the Census Bureau. Two papers focus on productivity in the manufacturing sector, including a discussion of a collaborative effort with the Bureau of Labor Statistics to produce statistics on within-industry variation on productivity (Foster, Grim, Pabilonia, Stewart, Wolf and Zoghi) and an examination of how best to estimate plant-level productivity (Foster, Grim, Haltiwanger and Wolf). Other business-oriented research concerns the role of financing in firm growth (Earle and Brown) and the decline in business dynamism and entrepreneurship in recent decades (Decker, Haltiwanger, Jarmin and Miranda).

New and better data: Developing new data products and improving existing ones is often a by-product of the research we do. At other times, it is the explicit focus of our economists’ research. For example, one paper compares reported and imputed earnings from the Survey of Income and Program Participation (SIPP) to the earnings from the Social Security Administration’s Detailed Earnings Record (Chenevert, Klee and Wilkin). Another explores an alternative imputation methodology to impute missing income values in the Current Population Survey (CPS) Annual Social and Economic Supplement (ASEC) (Hokayem, Raghunathan and Rothbaum). The results of a link between the Science and Technology for America’s Reinvestment – Measuring the Effects of Research on Innovation, Competitiveness and Science (STARMETRICS) data and the 2010 Census, covering female graduate students and professors in STEM programs, will also be presented (Buffington, Harris, Jones and Weinberg).

Modernizing federal economic statistics: Ron Jarmin, assistant director for research and methodology, and William Bostic, associate director for economic programs, will discuss the possibility of modernizing federal economic statistics by leveraging information on economic activity held by the private sector. Such efforts can improve the quality and timeliness of official statistics, make possible new data products and reduce respondent burden.

More: In addition to these and other papers by Census Bureau co-authors, there will be presentations of research papers based on Census Bureau microdata, written by researchers using the Federal Statistical Research Data Center (FSRDC) network.

For further details on the papers to be presented at the ASSA/AEA meeting, including a preliminary program with abstracts, please see

For more information on working papers by Census Bureau researchers and FSRDC researchers, please see

For presentations by Census Bureau researchers at previous ASSA meetings, and at other major professional meetings, please see


Posted in Uncategorized | Leave a comment

In What Ways are D.C.-Area Neighborhoods With Rail Transit Different From Other Neighborhoods?

Bookmark and Share

Written by: Brian McKenzie, Social, Economic and Housing Statistics Division

A look around the Washington, D.C., metro area reveals a series of neighborhoods with varying demographic profiles. Researchers in the social sciences have produced a rich body of literature illuminating residential patterns for several key demographic characteristics, such as income, race and age across metro areas. A small but growing body of research explores the relationship between neighborhood demographic profiles and the myriad resources and amenities across communities.

A new working paper explores the extent to which the population profile of workers living near rail transit differs from those of other workers within the Washington region. The paper, “Transit Access and Population Change: the Demographic Profiles of Rail-Accessible Neighborhoods in the Washington, D.C. Area,” explores demographic changes in rail-accessible neighborhoods using two American Community Survey (ACS) three-year datasets for comparison, 2006-2008 and 2011-2013. The ACS is an important source of information for research focused on neighborhoods and other community-based analysis.

The analysis includes the six counties or county-equivalents in the region that had at least one Metrorail stop during the study period: Washington, D.C., the city of Alexandria, Va., Arlington and Fairfax counties in Virginia, and Montgomery and Prince George’s counties in Maryland. Workers with rail access are those who live in a census block located within a half-mile from a Metrorail stop.

Findings reveal that young adults, recent movers, white workers, highly educated workers and workers with high earnings all disproportionately live near rail stops in Washington and the five surrounding counties with at least one Metrorail stop.

Figure 1 shows that within Washington, white workers are disproportionately represented in neighborhoods near rail stops. For the 2011-2013 period, 56 percent of workers living near rail stops were white, whereas 38.3 percent of workers who did not live near rail stops were white.

Figure 1.


Black workers are underrepresented near rail stops, relative to neighborhoods without rail access. Between 2006-2008 and 2011-2013, the proportion of black workers within rail-accessible neighborhoods declined from 32.9 percent to 24.1 percent, whereas the proportion of all other groups either increased or did not experience a statistically significant change. Unlike their counterparts in Washington, rail-accessible neighborhoods in the five-county surrounding area have a lower proportion of Hispanic workers than do neighborhoods without rail access.

Another finding revealed that younger workers disproportionately reside near rail stops, and their prevalence near rail has increased in recent years. Within Washington and the surrounding areas, about four out of 10 workers living in a rail-accessible neighborhood were between ages 25 and 34 for the 2011-2013 period (see Figure 2). Moreover, between 2006-2008 and 2011-2013, the proportion of workers in this age group increased at similar rates for Washington and the surrounding counties at about 8 percent. Neighborhoods without rail access have a more even distribution of workers across age groups, both in Washington and the surrounding area.

Figure 2.


The presence of a rail stop may play an important role in defining neighborhood characteristics well beyond the central city. Furthermore, the rail-accessible population in Washington is similar to the rail-accessible population of the counties that surround it along several socio-economic indicators, such as age and educational attainment. In both cases, young and highly educated workers disproportionately reside near rail.

It is important to note that the Washington metro area experienced considerable population growth and economic development during the study period. The influx of younger workers into the region may exacerbate competition associated with living near a rail stop, sharpening differences in population profiles between rail-accessible neighborhoods and other neighborhoods. Demographic shifts, changes in the spatial distribution of wealth and changes in the built environment have blurred familiar notions of a rigid “urban” and “suburban” divide in the Washington region and others.

In increasingly complex and diverse metropolitan landscapes, information about how rail stops shape neighborhood identities and boundaries contributes to our understanding of emerging socio-spatial patterns. An improved understanding of such patterns requires examining communities through multiple lenses, including transportation infrastructure.

This paper is available at

Posted in Uncategorized | 1 Comment

Improving Our Nation’s Data on Race and Ethnicity

Bookmark and Share


Written by: Nicholas Jones, Director of Race and Ethnic Research and Outreach

Since the 1970s, the Census Bureau has conducted decennial content tests to research and improve the design and function of different questions, including questions on race and ethnicity. As our population grows more diverse, we want to ensure that the data we collect accurately reflects how the people living in our country identify themselves.

Recently, the Census Bureau undertook a critical middecade study to test alternative versions of the race and ethnicity questions. The goal is to improve question design and data quality for the 2020 Census. Building on the foundation and successful strategies of the 2010 Census Alternative Questionnaire Experiment Research on Race and Hispanic Origin, the 2015 National Content Test is helping refine our previous efforts to address race and ethnicity reporting issues and important racial/ethnic community concerns.

During the 2015 National Content Test, we are focusing on three crucial goals for race and ethnicity:

  • Increasing the accuracy and reliability of reporting in major race/ethnic categories.
  • Lowering item nonresponse rates.
  • Collecting detailed data for myriad groups.

This research will enable the 2020 Census to provide critical racial and ethnicity statistics about our nation’s population. To learn more about the ongoing research, click here.

Outreach and Engagement Over the Past Year                                   

Coinciding with this extensive research, the Census Bureau continues ongoing engagement and discussions about race and ethnicity with the U.S. Office of Management and Budget, federal statistical agencies and many stakeholder groups. Over the past year, our research team reached out to many stakeholders to make them aware of the 2015 National Content Test research plans, discussed their questions and obtained their feedback.

This ongoing outreach includes collaboration and dialogue with:

  • Office of Management and Budget and other federal agencies through interagency meetings.
  • Department of Justice, Office of Civil Rights about redistricting.
  • Equal Employment Opportunity Commission on comparing data trends.
  • White House Initiative on Asian Americans and Pacific Islanders on data disaggregation.
  • State Department on racial identity in the Americas.

Additionally, the agency engaged in dialogues with many other stakeholders, including:

  • The Census Bureau’s National Advisory Committee on Racial, Ethnic and Other Populations.
  • Conferences with Afro-Latinos.
  • Summits with the National Congress of American Indians.
  • Briefings with congressional caucuses for Blacks, Hispanics, Asians and Pacific Islanders.
  • Legislative briefings with Caribbean-Americans.
  • Conferences with Native Hawaiians and Pacific Islanders.
  • Forums with Middle Eastern and North African scholars.

We heard from thousands of individuals, organizations and community groups, and their feedback helped us with our research plans.

Next Steps

The results of the 2015 National Content Test, public feedback and continued dialogues with the Office of Management and Budget and external stakeholders and experts will guide the proposed race and ethnicity question design for the 2020 Census. The research seeks to improve data on race and ethnicity and inform recommendations to the Federal Interagency Working Group for Research on Race and Ethnicity, regarding the 1997 OMB Standards and Guidance on Race and Ethnicity. By early 2017, the 2020 Census topics must be submitted to Congress, with final question wording due the following year in April 2018.

Posted in Uncategorized | 1 Comment

Census Bureau Opens 20th Research Data Center in Wisconsin

Bookmark and Share

Written by: Barbara Downs, Ph.D., Center for Economic Studies

As we celebrate World Statistics Day and the importance of access to sound data, it’s a good time to recognize the recent opening of the Federal Statistical Research Data Center at the University of Wisconsin-Madison. Research Data Centers like this give qualified researchers on approved projects access to restricted data that are vital to increasing our understanding of the U.S. population and economy.

In addition to its own onsite researchers, the Census Bureau draws upon the expertise of the research community throughout the United States via the Federal Statistical Research Data Center (FSRDC) network. Research from this network has led to new uses of existing Census Bureau data, improved content in existing Census Bureau surveys and the development of new surveys designed to capture the dynamic U.S. economy. All of these activities are in support of the Census Bureau’s mission to serve as the leading source of quality data about the nation’s people and economy. The results of this research have been published in many top peer-reviewed journals, including the American Economic Review, Demography, the Journal of Political Economy and the Quarterly Journal of Economics.

In existence since 1994, the network represents a partnership between participating federal agencies and non-profit research institutions. Research data centers are secure facilities that provide access to restricted-use microdata collected by the Census Bureau and other agencies such as the National Center for Health Statistics and the Agency for Healthcare Research and Quality.  Researchers with approved projects can conduct research that benefits the Census Bureau by improving measures of the economy and people of the United States.

Before gaining access to the FSRDC network, researchers must submit proposals to the Census Bureau. The review process ensures that proposed research is feasible, has scientific merit and benefits Census Bureau programs. In addition, the research data center network’s operating procedures, strict security and strong legal safeguards ensure the confidentiality of data, as required by law. Researchers, for instance, must pass a full background investigation and are sworn for life to protect the confidentiality of the data they access, with violations subject to significant financial and legal penalties.

The FSRDC network currently hosts about 700 researchers working on more than 200 different projects.  The Wisconsin Federal Statistical Research Data Center joins 19 other Census Bureau research data center locations in the United States. These facilities are located at 14 colleges and universities — Baruch College (New York City), Cornell University (Ithaca, N.Y.), Duke University (Durham, N.C.), University of Michigan (Ann Arbor), University of Minnesota (Minneapolis), Penn State University (University Park, Pa.), University of Southern California (Los Angeles), Stanford University (Palo Alto, Calif.), Texas A&M (College Station), UC-Berkeley (Calif.), UC-Irvine (Calif.), UCLA (Los Angeles), University of Washington (Seattle) and Yale University (New Haven, Conn.) — and in the cities of Atlanta; Boston; Chicago; Raleigh, N.C.; and Suitland, Md.

Centers expected to open soon will be in Kansas City, Mo., as well as the University of Maryland (College Park), the University of Missouri (Columbia) and the University of Nebraska (Lincoln).

You can learn more about the FSRDC program and our newest location, the Wisconsin Federal Statistical Research Data Center, at and

Posted in Uncategorized | 1 Comment

Where Did All the Construction Workers Go?

Bookmark and Share

New Census Data on Job-to-Job Flows

By Hubert Janicki and Erika McEntarfer

 Unfilled job openings in the construction industry have risen steadily since 2009 (Job Openings and Labor Turnover Survey, Bureau of Labor Statistics). The rise follows a sharp fall due to the housing bust and the subsequent Great Recession (Job Openings and Labor Turnover Survey, Bureau of Labor Statistics). Anecdotally, many builders and contractors have reported difficulty finding new workers.  While not necessarily indicating a shortage in the market for construction workers, it is possible that tightening labor markets in construction are a drag on employment recovery in this sector.  This post sheds light on the labor market in construction using new Census Bureau Job-to-Job Flows data.

Specifically, we investigate the dynamics of worker flows in the construction industry in an attempt to examine some possible causes of this recent tightening in the labor market. We find that over 60 percent of construction workers displaced by the housing bust are employed in other industries or have left the labor market by 2013. We also find evidence of a persistent drop in hiring of younger workers into construction jobs over the last decade that is likely contributing to the current shortage of skilled workers in construction.

We first look at cross-industry job flows to see if construction workers moved into other industries after the housing boom. Figure 1 shows the sources of net employment growth in the construction industry from the second quarter of 2000 to the fourth quarter of 2013. Net employment growth in construction is employment gained or lost from workers moving in and out of construction (black line in Figure 1) due to workers switching industries (blue dashed line in Figure 1) and workers moving in and out of employment (red dotted line in Figure 1).

Employed workers moving from other industries into construction fueled a large part of construction employment growth during the housing boom. However, when the boom ended in 2006, there was no corresponding employment decline from workers moving into other industries; almost the entire employment decline is workers separating to nonemployment.

Which industries fueled employment growth in construction during the housing boom?

Figure 2 shows net flows into construction due to industry switching (blue dashed line in Figure 1) by industry. Workers moving from manufacturing, mining and especially leisure/hospitality jobs fueled much of the employment growth in construction between 2003 and 2005. Flows from leisure/hospitality jobs are especially large — about half of employment growth during the construction boom came from workers moving from leisure/hospitality into construction. These flows are likely driven by pay differentials between the two industries: average worker earnings in leisure/hospitality jobs are lower than in the construction sector. When the industry contracted sharply after the housing crash, that pipeline reversed (i.e. net flows between construction and leisure/hospitality became negative), but only for a few quarters.

Figure 1: Net Employment Change in Construction: Industry Switchers and Workers Moving In and Out of Employment: 2000Q2 – 2013Q4

Net Employment Change in Construction: Industry Switchers and Workers Moving in and out of employment: 2000Q2-2013Q4

Source: U.S. Census Bureau, beta Job-to-Job Flows (J2J) data, seasonally adjusted, 2014 Q4 release, set of 32 states with complete data from 2000 Q2 to 2013 Q4. “Employed Workers Switching Industries” shows the net job reallocation between construction and other industries from workers changing jobs with little to no employment gap (less than one quarter) between job spells. “Workers Moving In and Out of Employment” shows net employment change in construction from workers moving in and out of longer nonemployment spells (one quarter or more).

One hypothesis for the lack of more sustained worker flows from construction to leisure/hospitality jobs during the Great Recession is the differences in average workers earnings across these two industries. In particular, workers might conduct longer searches or wait out the housing crash rather than go “back down the job ladder” into lower paying work.

Another hypothesis is the relative lack of job security in leisure/hospitality jobs compared with jobs in other sectors. If service sector workers were more likely to lose their jobs in the housing crash and the Great Recession, then construction workers might have less of an incentive to look for jobs in that sector. Consequently, flows of former construction workers back to leisure/hospitality are not as large as one might expect given the long period of inflows from that sector into construction in the preceding years.

Figure 2: Net Employment Change in Construction from Workers Changing Industries: 2000Q2 – 2013Q4

Net Employment Change in Construction from Workers Changing Industries: 2000Q2 – 2013Q4

Source: U.S. Census Bureau, 2015Q2 beta Job-to-Job Flows (J2J) data, set of 32 states with complete data from 2000 Q2 to 2013 Q4. Data are seasonally adjusted. This graph shows net employment reallocation in construction from workers switching industries with little to no employment gap (less than one quarter) between job spells.

What happened to construction workers who entered long nonemployment spells during 2006 to 2009?

The Job-to-Job Flows public-use data do not release outcomes for workers who experienced employment gaps of longer than one-quarter, but we can look to the underlying longitudinal microdata for an answer. For construction workers who entered nonemployment spells of more than three months between 2006 and 2009:

  • About 40 percent were either recalled to their previous employer or eventually found another construction job.
  • Approximately one-third began work in another industry, typically after an employment gap of over a year. The most common destination jobs for these workers are in the trade/transportation and business services sectors. A look at destination jobs by detailed industry suggests many became general laborers, some landscapers, and some truck drivers. Despite the boom in mining during this period, mining accounts for less than 5 percent of new jobs for former construction workers exiting nonemployment.
  • Approximately one-quarter of displaced construction workers have no observed subsequent employment by the end of 2013, five to seven years after displacement. These individuals presumably have left the labor market, although they could be working informally or be self-employed.

Figure 3: Hires of Younger Workers as a Percent of All Hires in Construction 2000Q1 – 2013Q4

Figure 3: Hires of Younger Workers as a Percent of All Hires in Construction 2000Q1 – 2013Q4

Source: Author’s calculations from U.S. Census Bureau’s National Quarterly Workforce Indicators (QWI), 2015Q1 vintage. Data are seasonally adjusted and shares are smoothed using a centered moving average.

Another reason construction firms may have trouble finding skilled workers today is that hiring of young workers in the industry declined after the housing boom. Overall, hires of workers less than 45 years old fell from 73% of construction hires in 2000 to 63% in 2011.

To examine this reason in more detail, we turn to the Quarterly Workforce Indicators data. Figure 3 shows the composition of hires into construction jobs by worker age among younger workers. A striking feature of this figure is a marked decline in the share of total hiring from the 19-24 age group. In particular, the percent of hires accounted for by the 19-25 age group declined from approximately 18 percent at its peak before 2006 to 13 percent in 2012-2013. In comparison, the composition of hires of workers in the 25-34 and 34-44 age groups shows much more modest declines over this time period.

The age 19-24 demographic group of workers is particularly relevant since the participation of that age group in industries other than construction has remained constant at approximately 15 percent throughout this period (Figure 4). In other words, the decline in hires among younger workers in the construction industry is not because of a delayed entrance of 19- to 24-year-olds into the labor market, but rather the entrance of these workers into other industries. One reason for this decline could be unwillingness by construction companies to train relatively young workers relative to other industries in the economy. Another reason for this decline could be unwillingness to hire or work for short-duration contract jobs. Note in Figure 4 that young workers (19-24 age group) are being replaced with workers in the age 45-55 category. This increase in percent of 45- to 55-year-olds employed in the construction sector exceeds the employment share of this age group in all industries.

Figure 4: Decomposing Workforce by Age: Construction versus All Industries, 2000Q1 – 2013Q4

Figure 4: Decomposing Workforce by Age: Construction versus All Industries, 2000Q1 – 2013Q4

Source: Author’s calculations from U.S. Census Bureau’s National Quarterly Workforce Indicators (QWI), 2015Q1 vintage. Shares are smoothed using a centered moving average.

In short, our analysis of the data suggests that 60 percent of displaced construction workers have left the labor market or moved into other industries. Although some former construction workers transitioned quickly to other sectors, for most, a move into another industry occurred after a long spell of nonemployment. Also likely contributing to a shortage of experienced workers is a shift in hiring preferences — during the downturn construction firms hired fewer young workers, fewer young workers gained experience in the industry, and the share of older workers grew faster than in other industries.

You can use these data sources to carry out a more detailed analysis. Job-to-Job Flows Beta release data and Quarterly Workforce Indicators data are available for download through the Longitudinal Employer-Household Dynamics program at

Posted in Uncategorized | 4 Comments

How Well Do You Speak English? Assessing the Validity of the American Community Survey English-Ability Question

Bookmark and Share

Written by: Erik Vickstrom, Social, Economic, and Housing Statistics Division

Using data from the American Community Survey, the Census Bureau estimated that 60.4 million, or 20.7 percent of the population, spoke a language other than English at home in 2013.

Of these people, 58.3 percent reported speaking English “very well.” The Census Bureau generates statistics on English ability by asking respondents who report speaking a language other than English at home to indicate how well they speak English: “very well,” “well,” “not well” or “not at all.” Federal programs use data from this English-speaking-ability question for three purposes:

  • To decide which localities are required to produce voting materials in minority languages.
  • To determine the number of school-age English-language learners in each state.
  • To determine which governmental programs should provide assistance to those who have limited ability to communicate in English.

A new working paper investigates whether this question and its self-assessment of English ability actually measure an individual’s real English ability.

Researchers at the Census Bureau have been interested in measuring English ability since the question first appeared on the 1980 Census questionnaire. In 1982, results from the Census Bureau’s English Language Proficiency Study showed a strong correlation between the English-speaking-ability question and English-proficiency test scores. In addition, results from the 1986 National Content Test found a positive correlation between spoken English ability and reading and writing ability in English. More recently, a National Research Council report called for additional research on the accuracy of this survey item.

My research seeks to provide insights into the validity of the Census Bureau’s American Community Survey English-speaking-ability question by comparing self-assessments of English ability with objective tests of English literacy. Using data from the National Assessment of Adult Literacy, conducted in 2003 by the National Center for Education Statistics, I find that self-reported English ability is consistent with objective measures of literacy.

Table 1: Descriptions of prose proficiency levels, National Assessment of Adult Literacy

Figure 1: Average prose literacy scores by self-reported English ability, National Assessment of Adult Literacy


Respondents to the National Assessment of Adult Literacy completed a literacy assessment that drew from 152 open-ended questions that simulated real-life situations. While these tests allowed the assessment of prose, document and quantitative literacy, my research focused only on prose literacy scores. The prose assessment, which sought to measure the language skills necessary for effective access to public service programs, seemed most comparable to the adult English proficiency tests on the English Language Proficiency Study.

The prose literacy assessment allowed the calculation of a prose score of between 0 and 500 for respondents. The literacy scores correspond to four performance levels: “below basic,” “basic,” “intermediate” and “proficient.” Table 1 shows these performance levels with the corresponding ranges of literacy scores, key abilities and sample tasks for each level. For example, respondents scoring in the “proficient” prose-score range are able to compare viewpoints in two editorials, while those scoring in the “intermediate” prose-score range are able to consult reference materials to determine which foods contain a particular vitamin.

Figure 1 shows the relationship between self-assessed English ability and prose scores among National Assessment of Adult Literacy respondents as measured by the literacy assessment. Respondents who reported speaking only English score 284 on the prose literacy assessment, indicating that they have, on average, intermediate proficiency. Those speakers of languages other than English who report speaking English “very well” have an average prose-literacy score of 269. This score falls in the intermediate performance level. In contrast, speakers of languages other than English who report speaking English “well” have an average score of 233, which puts them in the basic performance level. Those who reported speaking English “not well” or “not at all” scored 151 and 111, respectively, and both groups fall in the below-basic performance level.

Speakers of non-English languages who report speaking English “very well,” like English-only speakers, have average prose-literacy scores that fall into the intermediate performance level. Prose literacy at this level indicates abilities sufficient to read and understand moderately dense, less commonplace prose texts. Thus, the speakers with the best self-reported English ability can perform the same key tasks as the average English-only speaker. Speakers of languages other than English reporting an English ability of less than “very well,” in contrast, have lower language skills on average.

These results suggest that the English-ability question, despite being a self-assessment, does a good job of measuring English ability.

Posted in Uncategorized | 2 Comments

Characteristics of Likely-Transgender Individuals in Administrative Records and the 2010 Census

Bookmark and Share

Written by: Benjamin Cerf Harris, Ph.D., Economist, Center for Administrative Records Research and Applications

Transgender issues and images are increasingly present in popular media, literature, journalism, and the United States legal system. Few data sets, however, include information on the transgender community.

In a new working paper, I evaluate first-name and sex-coding changes in administrative files from the Social Security Administration (SSA) to model which individuals are likely to be transgender. I then match these likely-transgender individuals to their responses in the 2010 Census to create a preliminary dataset for studying the characteristics of likely-transgender individuals.

It is crucial to note that the number of people I identify as “likely-transgender” is not an estimate of the number of transgender individuals in the population, or even in the Social Security data. Gender identity is too complex to be captured only by name changes and sex-coding changes in a particular file. Further, some transgender individuals may not have a Social Security number, may not change their names with the SSA, or may have gender-neutral names. While this approach does not provide exact estimates of who is and is not transgender, it does allow researchers to learn about relative changes in trends over time and cross-sectional differences in characteristics for a subset of likely-transgender individuals.

Beginning in 1936 (the first year Social Security numbers were issued), I find claims for male-to-female and female-to-male name changes, many of which include sex-coding changes in the same directions. Figure 1 shows these “transgender-consistent” claims make up just under 0.02 percent of all claims and remained roughly proportional as the number of people with Social Security numbers grew.

Next, I match likely-transgender individuals to their records in the 2010 Census to explore residential patterns and responses to the question on sex. Figure 2 shows that likely-transgender individuals in the Social Security files are most concentrated in western and northeastern states that had legal protections against discrimination based on gender identity or expression. They are least concentrated in states without such protections.RM2

I also find that likely-transgender individuals answer questions about sex differently than nontransgender individuals. Figure 3 shows likely-transgender respondents report both sexes, or leave the question blank, more often than nontransgender respondents. This is interesting because this question appears everywhere, from surveys to credit card applications to the forms we all fill out at the doctor’s office.


Together, these results help illuminate U.S. transgender history from the 1930s onward, and they help address questions about the characteristics and residential patterns of likely-transgender individuals. This work also demonstrates the potential for integrating data from existing sources, including administrative records, to produce new information products that provide deeper insights into the U.S. population. Despite several limitations to this approach, it is an important step toward learning more about transgender individuals’ experiences in the U.S.

The paper is available at

Note: Please also see the Census Bureau statement on transgender data collection:

Census Bureau Statement on Transgender Data Collection

June 3, 2015 — Content changes in the decennial census and surveys are managed through an interagency process by the Office of Management and Budget (OMB). Decisions on new content are reached through careful consideration and public input and linked to a federal, legislative or programmatic need for the data. At this time the Census Bureau does not have plans to test questions about gender identity or sexual orientation for the 2020 Census or American Community Survey.

The Census Bureau, however, is a member of the Office of Management and Budget’s Federal Interagency Group on Improving Measurement of Lesbian, Gay, Bisexual Transgender (LGBT) populations in Federal Surveys, and the Census Bureau continues to work with the Office of Management and Budget and other federal agencies to examine the changing requirements and data recommended for program implementation. Additionally, a measure of sexual orientation was recently included in the National Health Interview Survey (NHIS) conducted by the Census Bureau and sponsored by the National Center for Health Statistics (NCHS).

The Census Bureau is committed to reflecting the information needs of our changing society. The Census Bureau is constantly examining the effectiveness of census and survey questions to collect accurate data on families and people and works with the National Advisory Committee on Racial, Ethnic and Other Populations and the Census Scientific Advisory Committee on recommendations for changing population needs.

Posted in Uncategorized | Leave a comment

Where Do We Go When We Retire? A Broader Look at Retirement Destinations

Bookmark and Share

Ben Bolender, Population Division, U.S. Census Bureau

Where do you plan to live when you or a loved one reaches the golden years? Do you plan to stay where you are now? Do you plan to move to some place warm and sunny? Our new research from the 2014 population estimates looks at where people age 65 and over are choosing to move.

People moving from place to place are reflected in the measures of migration created every year for the Census Bureau’s population estimates. We use primarily Medicare enrollment and group quarters data to calculate the movement of the population age 65 and over. People move for a variety of reasons. The tendency to migrate is highest for young adults. However, in the ages following retirement (commonly around 65), we see another spike in migration. With the baby boomers being the largest cohort yet to enter these ages, the question of where people move when they retire is becoming increasingly important.

Sumter County, Fla., has been one of the fastest-growing counties in recent years, at least partially because of its planned retirement destination status. Based on our research into older age migration patterns, we see that Sumter attracted just over 11,000 older people from April 1, 2010 (Census Day) to July 1, 2014, from other parts of the U.S. Another county with a strong draw was Maricopa, Ariz., which gained almost 18,000 older people during this same period through migration from other parts of the U.S.

While it is true that these areas do attract the most older-age migrants, they do not reflect the wide variety of places that attract retirement-age people. Examining numeric change tells one story, but it is often useful to look at other measures of change. The picture changes when we group counties by population size and look at percent change in the older population from migration. For example, Sumter County ranks high in this measure, showing an increase of 27.7 percent in the 65-and-older population through just domestic migration from 2010 to 2014. The comparable number for Maricopa County was only 3.9 percent.

For counties with a total population of 50,000 or more in 2014, Sumter leads the list for fastest-growing older population through domestic migration. This was followed by Broomfield, Colo., with a 17.6 percent increase, and Rockwall, Texas (part of the Dallas metro area), with a 16.9 percent increase. Here we see that both midsized mountain towns and suburban areas saw their older populations rise rapidly through migration.

When we look at counties with populations between 20,000 and 50,000, Jasper, S.C., leads the way at 28.1 percent, followed by Kendall, Texas, at 14.9 percent. Many older people choose to move to suburbs and areas adjacent to larger cities. Both of these counties are on the outskirts of larger metropolitan statistical areas.

For counties with fewer than 20,000 people in 2014, looking at percent growth starts to pick up much smaller numeric increases. Stark, Ill., leads the list at 17.3 percent growth, followed by Lake, S.D., at 13.9 percent and Sequatchie, Tenn., at 12.6 percent. These three areas represent different flavors of rural America in their plains, forests, valleys, and fields. Even with the variety of places where you could choose to retire, with the aging of the baby boomers it is likely that when you arrive you will find others who made the same choice.

Posted in Uncategorized | 6 Comments