No Kids in the House: A Historical Look at Adults Living Without Children

Written by: Emily Schondelmyer, Social Economic Housing Statistics Division

Over the last 50 years, the percentage of adults living without children has climbed 19 percentage points from 52.5 percent in 1967 to 71.3 percent today based on an analysis of Current Population Survey data.

The Current Population Survey holds a wealth of information on the state of American families. We know that the population is aging, and that fertility is not as high as it was some decades ago, but how does this affect the composition of the group of adults who do not live with children in their home? In this blog, we highlight the historical changes among adults who do not have children in the home using data from CPS.

A higher proportion of all adults live without children in their home now than in 1967.

Among the 64.7 million adults living without children in 1967, 59 percent shared their household with a spouse. Another 11 percent identified as an adult child of the householder, and 14 percent as living in some other arrangement, such as with a boarder or roommate. Another 14 percent lived alone and less than 1 percent lived with an unmarried partner.

By 2016, the living arrangements of adults living without children had shifted. The biggest change was a decline in the proportion living with a spouse — down to 44 percent. This decline was due to increases in the percentage living alone (20 percent) and those living with a partner (8 percent). Another 11 percent were identified as the adult child of the householder, and 16 percent were in some other living arrangement.

The percentage of 25- to 34-year-olds with no children living with them more than doubled since 1967.

When considering changes that occurred between 1967 and 2016, keep in mind that the median age at first marriage increased from 20.6 to 27.4 for women and from 23.1 to 29.5 for men. Age at first birth increased as well. Most babies are born to a married couple, so it is natural that we see shifts in the percentage of adults who live with no children when we look at particular age groups.

We saw the largest change in the proportion of adults living without children among those who were ages 18 to 24 and ages 25 to 34. In 1967, less than half of 18- to 24-year-olds did not have children living with them (46.7 percent) but by 2016, the majority, 68.8 percent, did not have children living with them.

We see even more dramatic changes for 25- to 34-year-olds. In 1967, 23.9 percent of 25- to 34-year-olds did not have children living with them. By 2016, the percentage of 25- to 34-year-olds without children in the home more than doubled to 61.5 percent. See Figure 1 for the percentage of all age groups living without children.








Most adults living with no children are married.

In both 1967 and 2016, a larger proportion of those who did not have children living with them were married (including married, spouse absent), compared to all other marital statuses, including never married. In 1967, 60 percent of those living without children were married and 19 percent of those living without children were never married. In 2016, 46 percent of those without children living with children were married, while 32 percent reported they had never married (See Figure 2). Conversely, if we examine only those who were never married, we find that a higher percentage in 2016 did not have children living with them, compared with 1967, 80 percent and 67 percent, respectively.








More than half of adults living without children have more than a high school degree.

In both 1967 and 2016, the largest share of those who did not live with children had a high school degree only (28 percent in 1967, 31 percent in 2016 [see Figure 3]). This makes sense, given that younger people, who may have not yet earned a college degree, were also likely not to have had children yet. In 1967, about 20 percent of people who did not have children living with them had more than a high school degree. By 2016, that had increased to 58 percent. Overall, the share of those with at least some college education has increased over the years for all adults.

When we examine those with a college degree and whether or not they had children living with them, we see even starker changes. By 2016, 68 percent of those with a college degree did not have children living with them at the time of the interview. This is an increase from the 49 percent of college graduates in 1967.








It is important to keep in mind that these demographic changes over time only reflect changes for those who had no children living with them at the time of the interview. Further, this group is specifically identified by the lack of own children in the house. Own children are identified as biological, step or adoptive, never married children 18 years and under. These groups could have had children previously who no longer live with them, they could have gone on to have children, or they may never have had children.

Given that the total fertility rate has decreased over the years and the nation is aging, we should expect to see more Americans living without children.

For more information on the living arrangements, marital status and household composition of adults and children, please see the America’s Families and Living Arrangements table package.

Posted in Uncategorized | Leave a comment

What’s in a Name

Despite the nation’s growing racial and ethnic diversity, the five most frequent American surnames in 2010 remained the same as in 2000 and were mainly reported by whites and blacks.

Smith, Johnson, Williams, Brown and Jones were the most common last names, according to a Census Bureau analysis of the 2010 Census.

However, the growing Hispanic population has pushed surnames reported by Hispanics up the rankings since 2000. Garcia was the sixth most common name, up from eighth in 2000. Two more Hispanic surnames are in the top 10 — Rodriguez and Martinez.

Census respondents reported about 6.3 million surnames in 2010. Of these, only 11 were reported more than a million times each. The vast majority — 62 percent — were reported only once.

“Some are unique or unique variations of more frequent names,” Joshua Comenetz, assistant chief of the Population Division’s Population Geography staff and manager of the surnames project, said.

The rising frequency of Latino and Asian surnames is a reflection of growth in those populations. However, the high ranking of some Hispanic names nationally is partly a result of less surname diversity in that group.

“There is more surname clustering among Hispanics,” Comenetz said. “Twenty-six surnames cover a quarter of the Hispanic population and 16 percent of Hispanic people reported one of the top 10 Hispanic names. The pattern is similar for Asians and blacks.”

“Names are not distributed among racial and ethnic groups the same way the population is distributed,” Comenetz said. “Also, it takes fewer names to cover a large segment of the Hispanic, Asian or black populations, compared to the white population, which has higher surname diversity.”

Most names are dominated by one Hispanic origin or race group. For example, more than 98 percent of people named Xiong are Asian, more than 87 percent of those named Washington are black, more than 96 percent named Barajas are Hispanic, and almost 98 percent named Yoder are white.

The most rapidly increasing surnames among the top 1,000 in both 2000 and 2010 are mostly Asian: Zhang (up 111 percent), Li (93 percent), Ali (66 percent), Liu (64 percent) and Khan (63 percent) top the list.

Three of the 15 fastest growing were Hispanic: Vazquez (up 63 percent), Bautista (59 percent) and Velazquez (59 percent).

The Census Bureau’s 2010 Census surnames product is the third in a series of data releases based on names recorded in the decennial census. Previous products tabulated name frequency from the 1990 and 2000 Censuses. The complete product, along with a methodology document, is available at <>.

Posted in Uncategorized | Leave a comment

Collaboration Is Key to Producing Timely School District Poverty Estimates

Written by: Carolyn Gann, Social, Economic and Housing Statistics Division

Many factors play a role in a child’s educational success, and it is the mission of the U.S. Department of Education to promote excellence in education and ensure equal access. To help achieve this goal, the Department of Education plans to provide over $15 billion in federal funds to qualifying Title I school districts for the next school year. These funds will improve the academic achievement of children living in poverty. To identify school districts with children in poverty, the Department of Education partners with the U.S. Census Bureau’s Small Area Income and Poverty Estimates (SAIPE) program. The SAIPE program produces the only source of single-year statistics of childhood poverty for the nation’s more than 13,000 public school districts.

To estimate the number of children living in poverty within a school district, the SAIPE program pays special attention to school district boundary changes. Today’s release of the 2015 SAIPE reflects the most recent school district boundary update.

In the latest update, for 2015 and 2016, there are 13,245 U.S. public school districts. This is down from 13,486 school districts from the previous update for 2013 and 2014, a decrease of 241 school districts. Changes are typically the result of school districts shifting, splitting or consolidating boundaries, often driven by state or local policy changes.

Figure 1 shows a hypothetical secondary (often grades 9-12) school district (Sec. D) that covers the same geographical boundaries as three distinct elementary (often grades K-8) school districts (Elem. A, Elem. B, and Elem. C). As they stand in the figure, each school district would be financially responsible to provide services to school-aged children within their district’s grade range. For these four hypothetical school districts, the Department of Education would use SAIPE along with other data sources to determine Title I eligibility and funding allocations.







We often see elementary and secondary school districts (like those in Figure 1) consolidated into one unified school district. In the latest boundary update, this type of consolidation occurred throughout Vermont as well as other states. Additionally, many neighboring unified school districts consolidated to form larger entities. Figure 2 displays a side-by-side comparison of the unified and elementary school district boundaries in Vermont before and after the latest boundary update. Before the latest boundary update, Vermont had 276 school districts. Now Vermont has only 60 school districts.

Figure 2 illustrates the kind of widespread boundary changes that can occur after a school district boundary update. When a school district boundary changes, there is a change in the number of school-aged children living in the district, including children who are living in poverty. Capturing the latest school district boundary changes is crucial for the SAIPE program to produce timely and accurate estimates of childhood poverty.

Given the SAIPE data’s timeliness and adaptability, we can examine how childhood poverty in school districts changes over time. To illustrate these changes, the recent SAIPE data release includes an animated map (also seen below in Figure 3) displaying data from 2007 through 2015. The year 2007 serves as a base year before the most recent recession. In the animated map, the lighter colors represent lower poverty rates, and the darker colors represent higher poverty rates. The animation allows us to see the changes in childhood poverty before, during and after the recession.

Figure 4 focuses on school districts with high poverty (at or above 20.0 percent). This figure displays annual SAIPE data back through 1999 and constructs a fuller picture of pre- and post-recession trends for school districts with high poverty. Since the most recent recession, there has been an increase in the percentage of school districts with high poverty. Since 2013, we are only just beginning to see that percentage decrease.

As the composition of childhood poverty changes over time, it is necessary for the Department of Education to identify all Title I qualifying districts and distribute funding where it is needed most. This work by the School District Review and SAIPE programs are two of many examples of how agencies work together to meet administrative needs for delivering resources, while ensuring the quality and accuracy of the information available to all stakeholders.

Posted in Income, Poverty, Uncategorized | Leave a comment

A Glance at the Age Structure and Labor Force Participation of Rural America

By: Jennifer Cheeseman Day, Donald Hays, and Adam Smith

Rural America is older than urban America. The median age of all people living in rural areas is 43 years, compared with 36 years for urban areas. We know that age is an underlying factor driving many phenomena, such as labor force participation, educational attainment and earnings. With the release of the 5-year American Community Survey this year, we are exploring rural-urban relationships such as these.

When looking at the populations of both rural and urban America, we find a distribution with two peaks, baby boomers in their 50s and 60s forming one, and “millennials” in their late teens and twenties forming the second (Figure 1). While there are more people in the millennial generation in urban areas, baby boomers form the higher peak in rural areas.

Looking at the percentage distributions by age in rural and urban areas, we see notable differences particularly among people in their prime working ages (16 to 64 years). Rural areas have much older working-age populations.

Generally, an older working-age population points to lower overall labor force participation (i.e., the percentage of the overall population in the civilian labor force), as workers begin to drop out of the labor force past age 50. Rural areas have lower labor force participation overall (59.2 percent compared with 64.2 percent for urban areas). Looking at labor force participation by age groups, we find that that men and women living in rural areas have lower labor force participation than their urban counterparts at similar ages (Figure 2). The difference does not seem to stem from the availability of full-time, year-round employment. Among all workers, proportions of rural male and female workers with full-time, year-round employment are comparable with those of their urban counterparts (with some exceptions among women of childbearing age) (Figure 3).


Since earnings tend to rise with age, we might expect rural workers (as an older group) to have higher median earnings than urban workers. However, median earnings for rural male and female workers who worked full time, year-round were about $47,000 and $35,000 compared with $50,000 and $40,000 for urban male and female workers (Figure 4). Compared with their urban counterparts, rural workers have lower median earnings in nearly every age group.

Also, we see at each age group, men have higher median earnings than women at similar ages. On average, median earnings for women were 75.0 percent of median earnings for men in rural areas. In urban areas, the gap between earnings for men and women was smaller, with median earnings for women at 80.3 percent of median earnings for men (Figure 5). However, looking at the earnings ratio between women and men for each age group, we find that for younger ages, the earnings ratio is higher (closer to equal) for urban workers than rural workers. Between the ages of 45 and 64, both urban and rural women workers had median earnings about 74.0 percent of men at similar ages.

To understand earnings differences between urban and rural workers, we can consider educational attainment (Figure 6). Specifically, men and women living in rural areas had lower rates of higher attainment (bachelor’s degree or more) compared with their urban counterparts across all ages. Ostensibly, educational attainment translates into higher earnings.

This exploration is only possible by the extraordinary depth of data provided by the American Community Survey. For more information about how the Census Bureau defines urban and rural geographies, see Defining Rural at the U.S. Census Bureau.

Posted in Uncategorized | Leave a comment

Life Off the Highway: A Snapshot of Rural America

Written by: Alison Fields, Kelly Ann Holder, and Charlynn Burd, Social, Economic and Housing Statistics Division

We are a nation of communities. The 2011-2015 American Community Survey statistics released today helps tell the story of us all — each city, town and neighborhood. This year we are taking a detour down a road driven by only one in five Americans. We are exploring what the latest U.S. Census Bureau statistics can tell us about the 60 million people who live in rural America. Rural areas can take on varying characteristics affected by regional differences across the country. The rural population is not the same everywhere except in its distinction of not being urban. American Community Survey statistics are the only source of annual data for some rural residents, and are vital for local decision making by elected officials, city and county planners, and businesses. Today, the Census Bureau released several products highlighting selected characteristics of the rural population of America.

The term rural means different things to different people. For many, it conjures images of rolling hills and farmlands, with infrequently clustered houses. For our purposes, we explore rural statistics based on the official Census Bureau classification. Urban and rural areas are defined after each decennial census, every 10 years, by applying specific criteria related to population thresholds, density, distance and land use. The same geographic designations are then used in the American Community Survey each year for the next decade.

Urban areas represent densely developed territory, and encompass residential, commercial and other nonresidential urban land uses. The Census Bureau identifies two types of urban areas: urbanized areas with 50,000 or more people and urban clusters with at least 2,500 people and fewer than 50,000 people. Rural areas encompass all population, housing and territory not included within an urban area. Generally, rural areas are sparsely populated with a low housing density, are not built up, and are at a distance from urban areas. Urban areas and urban clusters make up 3 percent of the entire land area in the country but contain 80.7 percent of the population while rural areas spread across 97 percent of the land area and contain 19.3 percent of the population (see Table 1). For more information about how the Census Bureau defines urban and rural geographies, see our new brief Defining Rural at the U.S. Census Bureau.

Over the past century, the nation’s urban landscape has changed, and with it, so have our rural areas. As population change has transformed our urban areas and the criteria used to define it, the rural population has decreased from 54.4 percent of the total in 1910 to 19.3 percent by the 2010 Census (see Figure 1). 







Source: U.S. Census Bureau, 1910 to 1990 Censuses, available at; 2000 Census, Table P002; 2010 Census, Table P2.

Where today’s 60 million rural Americans live might actually surprise you. American Community Survey statistics show that Maine and Vermont have the highest proportions of population living in rural areas (about 61.0 percent) and California has the lowest (4.9 percent). While the Midwest and West may appear to have a larger share of wide-open rural spaces from the view outside of an airplane window, the majority of the nation’s rural population (64.4 percent) actually lives east of the Mississippi River. Only 10.0 percent of the total population of the West lives in rural areas. Nearly half (46.7 percent) of all people living in rural areas are in the South. That is about 28.0 million people, more than the populations of Washington, Massachusetts, Arizona and Indiana combined.

The Census Bureau’s American Community Survey data products for rural areas are available on in American FactFinder for the nation, regions and states. By using the 2010 Census as a guide, we can demonstrate how to look at American Community Survey statistics at the county level based on a concentration of rural population. County-level statistics are suitable for this exercise because counties are a geography that most people can conceptualize, the statistics are easily accessible from American FactFinder, and they are useful for many policy purposes.

By calculating the percentage of the population in each county who lived in rural areas in 2010, we can create one measure of “rurality” that we can then apply to the American Community Survey statistics (see Table 2 and Figure 3). Counties run somewhere along a spectrum of completely rural to completely urban. Of the 3,143 counties and county equivalents in the United States as of 2010, 704 consisted of populations that were completely rural by the 2010 Census designation. At the other end of the urban/rural spectrum, only 29 counties in the country had no rural residents. The majority of counties fall in between.

American Community Survey 1-year statistics are available for about one-quarter of the counties in the nation, whereas American Community Survey 5-year statistics are available for all counties and county equivalents. The 704 completely rural counties, currently home to about 5.3 million people — almost equivalent to the entire population of Colorado — rely solely on American Community Survey 5-year statistics. Several of our products this year focus specifically on these counties. 

*Total counties/county equivalents as of 2015. Bedford city, Va., was consolidated with Bedford County, Va., after the 2010 Census. For a list of county names, see county lookup table. Source: U.S. Census Bureau, 2010 Census, Table P2.

It is also important to point out that the Census Bureau’s definitions of urban and rural differ from the classifications of metropolitan and nonmetropolitan. Often nonmetropolitan, or nonmetro, is used synonymously with rural, and while there is overlap, these geographic entities are not identical. Over half (54.4 percent) of people living in rural areas live within a metro area. Only 59.2 percent of the nonmetro population lives in rural areas.

The infographic, Our Changing Landscape, that will be released later displays national-level statistics that allow a comparison of people in urban and rural areas on a select number of characteristics.  With the data products released today, we add to the narrative of our nation and our communities. To learn more about the American Community Survey including other topics covered, please visit <>.

Posted in Uncategorized | Leave a comment