Families in America presents a wide selection of information from the American Community Survey that helps us describe American living arrangements, relationships, marriages, births, children, and incomes. Each section includes a large selection of information for the United States, the 50 states, and the District of Columbia. This is followed by a more limited selection of data for 381 metropolitan areas, 980 counties with populations of 50,000 or more, and 795 cities with populations of 50,000 or more.
Families in America includes details about both family and nonfamily households and includes topics such as multi-generational households, same-sex partner households, grandchildren living with grandparents, and nonrelatives in family households. In addition, information related to age, sex, race, Hispanic origin, income, poverty, and health insurance for various household types is included.
About the Author
Deirdre A. Gaquin has been a data use consultant to private organizations, government agencies, and universities for over 25 years. Prior to that, she was Director of Data Access Services at Data Use & Access Laboratories, a pioneer in private sector distribution of federal statistical data. A former President of the Association of Public Data Users, Ms. Gaquin has served on numerous boards, panels, and task forces concerned with federal statistical data and has worked on four decennial censuses. She holds a Master of Urban Planning (MUP) degree from Hunter College. Ms. Gaquin is also an editor of Bernan Press's Congressional District Atlas; The Who, What, and Where of America: Understanding the American Community Survey; Places, Towns and Townships; and The Almanac of American Education.
Mary Meghan Ryan is a senior research editor with Bernan Press.
Table of Contents
Part A.Living Arrangements
Table A-1. States
Table A-2. Counties
Table A-3. Metropolitan Area
Table A-4. Cities
Table B-1. States
Table B-2. Counties
Table B-3. Metropolitan Area
Table B-4. Cities
Part C. Marriages and Births
Table C-1. States
Table C-2. Counties
Table C-3. Metropolitan Area
Table C-4. Cities
Part D. Children
Table D-1. States
Table D-2. Counties
Table D-3. Metropolitan Area
Table D-4. Cities
Part E. Income, Poverty, and Health Insurance
Table E-1. States
Table E-2. Counties
Table E-3. Metropolitan Area
Table E-4. Cities
Appendix A. Source Notes and Explanations
Appendix B. Core Based Statistical Areas and Components (as defined February 2013)
Appendix C. Cities by County
Most Americans live in families, defined by the Census Bureau as persons related by birth, marriage, or adoption. Family households can include persons who are not family members, and nonfamily households often include persons with close personal relationships. This book includes a selection of data from the American Community Survey that helps us describe American living arrangements, relationships, marriages, births, children, and incomes. Part A includes information on types of households. Part B is about relationships, defining the variety of people who can make up a household. Part C includes details on marital status, marital history, and recent births. Part D summarizes the living arrangements of children and the employment patterns of their parents. Part E includes information about income for the various types of households. Each section includes a large selection of information for the United States, the 50 states, and the District of Columbia. This is followed by a more limited selection of data for 381 metropolitan areas, 980 counties with populations of 50,000 or more, and 795 cities with populations of 50,000 or more.
Although most Americans have always lived in family households, the proportions have changed dramatically over the past half-century. In 1950, nearly 90 percent of households were family households and the proportion remained over 80 percent in 1960 and 1970. The very high birth rates of the baby boom in the 1950s and 1960s were followed by historically low birth rates in the 1970s. The proportion of family households fell to 73.2 percent in 1980 and by 2010 it had dropped to 66.4 percent. As the baby boom population formed new households, the proportion of one-person households increased from 9.3 percent in 1950 to 22.7 percent in 1980. The millennial generation continued this trend and one-person households rose to 26.7 percent in 2010. Family households with children were more than half of families in 1950 through 1980, but dropped to 44.8 percent of families by 2010, as birth rates remained low and hit even lower levels than in the 1970s.
Decennial census household data for the United States, 1950-2010
Average household size
Percent one-person housholds
Percent in group quarters
Family households as a percent of households
Percent of families with own children under 18 years
The 2010 census was different from any census in recent memory. All American households answered a simple questionnaire with ten questions. No longer did some people get the “long form” with dozens of detailed questions about employment, education, income, previous residence, housing characteristics, and more. The data gleaned from these important questions have long been used by federal, state, and local governments to evaluate their populations and program needs; by large and small businesses and nonprofit organizations for a variety of planning and location purposes; and by academic researchers to study trends in social and economic conditions. The “long form” has been replaced by the American Community Survey (ACS). Under development for more than a decade, the ACS is an ongoing survey of the American people that is ushering in a new era in social and economic data analysis. The census “long form” provided detailed estimates of social and economic characteristics every ten years. The ACS collects this same information on a rolling basis. It takes 5 years of ACS responses to accumulate a sample almost as large as the census “long form” collected at a single point in time. But data users now have the ability to study these characteristics and trends throughout the decade.
Because the ACS is a sample survey, large numbers of sample cases are needed before reliable estimates can be made for small populations. Each year’s sample is large enough to produce estimates for the nation, all the states, most metropolitan areas, and many counties and cities.
The richness of the ACS data can be accessed in varying degrees. Much more subject matter detail is available for large geographic areas partly because reliable estimates for large areas can be produced with smaller samples, and partly because more data must be suppressed for the smaller areas to protect the confidentiality of the respondents. The tables in this book include the ACS 3-year estimates for the period 2011 through 2013. Although 1-year estimates are available for most of the areas in this book, the 3-year estimates were chosen because the book includes very detailed information and the reliability of the estimates is better with the 3-year data.
This book is designed to include a sampling of key information about families, but also to guide users through the process of using the Census Bureau’s website to expand on the information included here. The state tables in this book include nearly 500 data items. The metropolitan area, county, and city tables include about 100 data items. The data in the tables are a small selection that show what is available for the counties and cities in the book. Every column includes an ACS Table Number that enables users to find the original data on the Census Bureau’s website. Much more information is available for analysis of cities and counties or for analysis of specific racial or ethnic groups if those groups have large populations in a particular city or county. Furthermore, the 5-year data include most of the same information for all cities and counties in the nation, no matter how small.
One of the most notable differences between the census “long form” and the ACS is the time frame of the estimates. We are accustomed to the census data that give us specific information every ten years, a snapshot of the country on April 1. The ACS multiyear estimates are different. The data in this book are from the ACS 3-year, 2011–2013 estimates. They are not averages, nor do they represent 2012, the midpoint of the 3-year estimates. They are period estimates with data spread evenly throughout the 3-year survey time period.
Finally, it is always critical to remember that all estimates are subject to sampling error. On the Census Bureau’s website, every ACS number is accompanied by its margin of error. In the interests of space and simplicity, this book does not include the margins of error, but all users are encouraged to consult the Census Bureau’s website and to understand some basics: small differences are very likely to represent no difference at all; do not draw conclusions from small numbers; use these numbers as a starting point to explore the wealth of information from the ACS.