How We Measure Health Insurance

Bookmark and Share

Written by: Jonathan Rodean

Next week, the Census Bureau is releasing two sources for health insurance statistics in the United States:  the Annual Social and Economic Supplement to the Current Population Survey (CPS ASEC) and the American Community Survey (ACS).  While both surveys have questions that attempt to measure the same phenomena, they go about it in different ways.

Since 1987, the CPS ASEC has collected health insurance statistics every year, making it one of the most widely used sources of statistics on health insurance coverage in the United States.  It provides statistics on health insurance status (insured or not insured), as well as type of coverage, for the whole nation, by demographic groups, and shows us trends over the last couple of decades.  By combining two years of CPS sample, this survey can also provide state statistics on health insurance.

Starting five years ago, the Census Bureau also began asking about health insurance coverage using the ACS.  With its much larger sample size, we can see health insurance statistics for a broader range of geographic levels including all 50 states, all counties, metro areas, and many other small geographies.

Both surveys obtain a person’s health insurance status by asking if a person has insurance through a number of different sources: an employer, directly through an insurance company, Medicare, Medicaid, Veterans Administration and other public sector insurance and the military.  However, the surveys differ in both their time of collection and reference period.

Like the other topics in the CPS ASEC, which is conducted in February, March, and April, respondents answer questions about the previous calendar year.  Specifically, we ask if they were covered at any time in the previous calendar year.  CPS ASEC, thus, measures if a person was insured on any day during the previous year.  They are considered “uninsured” only if, for the entire year, they had no coverage under any type of health insurance.

In contrast, the ACS is a rolling sample of households collected continuously all year long.  We ask if a person is currently covered by any of the listed types of health insurance.  So, ACS measures health insurance of the population based on whether people are insured at the point-in-time that they answered the survey during the year of collection.

There is also a variety of differences in the survey logistics.  While the ACS is most often administered online or through a paper survey that is mailed to respondents, the CPS ASEC is administered by phone.  Because of space limits within the paper survey, the ACS asks fewer and less detailed questions than the CPS ASEC.  The ACS asks about the insurance coverage of each household member specifically, while the CPS ASEC asks if anyone in the household is covered, and if so, who that is.   

health insurance graphicWith these variations and others, the CPS ASEC and ACS produce slightly different statistics on health insurance coverage.  The two surveys show similar trends (as shown in the graph below), though the CPS ASEC shows higher uninsured rates for each year.  This degree of consistency between the two surveys collected under such different conditions gives us confidence that these statistics are useful for those who need to understand the state of health insurance in America.

Many people contact us each year asking how to know which estimate to use for a particular purpose. For national statistics, we recommend the CPS because it provides a consistent historical time series at the national level back more than a decade. The CPS can also be used to look at limited state-level trends. However, because of the larger sample size and smaller sampling errors, we recommend using the ACS for subnational geographies.

Posted in Health Care | Tagged , | Leave a comment

How the Census Bureau Measures Income and Poverty

Bookmark and Share

Written by: David Johnson

Note: This is an updated  version of a blog that ran last year. It provides important information on when to use statistics from the Current Population Survey and when to use them from the American Community Survey for income and poverty.

Income, poverty and health insurance statistics  for 2012 from the Current Population Survey (CPS) will be released Tuesday, Sept. 17, 2013. One-year statistics from the 2012 American Community Survey (ACS) will be released on Thursday, Sept. 19, 2013.

In all likelihood, the national statistics from these two sources will not be identical. Why not? Which is correct? Well, it’s complicated.

The Current Population Survey serves as the nation’s primary source of statistics on labor force characteristics. A supplement to the survey provides the official annual statistics on the nation’s income and poverty levels as well as statistics on age, sex, race, marital status, educational attainment, employee benefits, work schedules, school enrollment, health insurance, noncash benefits and migration.

The American Community Survey, on the other hand, is the only source of small-area statistics available on a wide range of important social and economic characteristics for all communities in the country. In addition to income, poverty and health insurance, other topics include education, language ability, the foreign-born, marital status, migration, homeownership, the cost and value of our homes and many more.

Statistics from these two surveys differ for a number of reasons. First, income questions on the CPS are much more detailed than the summary questions asked on the ACS.

Second, the reference periods for the two surveys are very different. The CPS asks respondents to report on their income in the previous calendar year. The ACS asks about income in the prior 12 months. Since the ACS is a continuous survey administered throughout the year, some respondents to the 2012 ACS (those who fill out the survey in January) are reporting income received between January 2011 and December 2011 while other respondents (those who fill out the survey in December) are reporting income received between December 2011 and November 2012.

Third, for the CPS, trained interviewers administer the survey while people primarily respond to ACS questions over the Internet or by mail. (Trained interviewers follow-up with households who do not respond to the ACS online or by mail.)

These differences often result in different national statistics for such key indicators as poverty, median income and inequality. Despite differences in the “levels” of these indicators, the trends over time tend to be very similar across the two surveys. The following graphs show median household income and poverty rates from the ACS compared with statistics from the CPS for previous years. The red line adjusts the CPS for the differences in reference periods.

 Comparison of CPS ASEC and ACS Median Household Income and Poverty Rates: 2000 to 2011

Many people contact us each year asking which estimate to use for a particular purpose. For national statistics, we recommend the CPS because it provides a consistent historical time series at the national level back, in some cases,  more than half a century. The CPS can also be used to look at limited state-level trends. However, because of the larger sample size and smaller sampling errors, we recommend using the ACS for subnational geographies.

The next two charts show the volatility of the single year CPS ASEC statistics relative to the ACS statistics for two smaller states: Arkansas and Maryland. Since the CPS has a smaller sample size, you see more volatility in these smaller states.


Comparison of CPS ASEC and ACS Poverty Rates: 2000 to 2011: Arkansas

Comparison of CPS ASEC and ACS Household Median Income: 2000 to 2011 Maryland

If you are interested in a longer time series for a small state than is available from the American Community Survey, we recommend using two- or three-year averages from the CPS ASEC.



Posted in Income, Poverty | Tagged , | Leave a comment

Who is a STEM Worker?

Bookmark and Share

Written by: Liana Christin Landivar

Did you know that the most common science, technology, engineering, and math (STEM) occupation is software developer? More than 80 percent of STEM workers today are employed in computer occupations or engineering.

Two new Census Bureau reports look at the educational background and the demographic characteristics of STEM workers.

In 2011, there were 7.2 million people employed in STEM occupations, accounting for 6 percent of the workforce. An additional 7.8 million people worked in occupations related to STEM, such as health care.

When we look at the size and composition of the STEM workforce, we can see how it has changed over time, rising from 4 percent in 1970 to 6 percent in 2011.

Engineering employed the most STEM workers until the 1990s. After a surge of employment linked to the expansion of the Internet and computers in daily use, computer occupations surpassed engineering. Currently, engineers make up about a third of STEM employment, while computer workers comprise half of the STEM workforce.

Employment in STEM occupations from 1970 to 2011

Women’s representation in STEM employment has increased since the 1970s, but they remain underrepresented in engineering and computer occupations. In addition, the most recent decades show less growth in STEM employment among younger women. Most of the growth in women’s share of STEM employment among those under 40 occurred between 1970 and 1990.

Employment in STEM also varies by race and Hispanic origin. About 71 percent of STEM workers are non-Hispanic white, followed by Asian (15 percent), Hispanic (7 percent) and Black (6 percent). (The estimates for Black and Hispanic employment in STEM occupations are not statistically different.)  

Standardizing STEM

Industry, government, and academic leaders have increasingly focused on STEM as a key component of innovation and increased economic prosperity. However, until recently, the definition of STEM lacked standardization, making cross-organizational and cross-national comparisons of the STEM workforce difficult.

In April 2012, the Standard Occupational Classification Policy Committee, a consortium of nine federal agencies charged with standardizing occupational definitions, issued a recommended STEM occupation classification. This classification places workers into three primary occupational domains: STEM, STEM-related, and non-STEM. STEM includes computer workers, engineers, mathematicians and statisticians, life scientists, physical scientists, and social scientists. STEM-related includes architects and health care workers. Non-STEM includes all other occupations. This classification is designed to make comparisons of the STEM workforce more consistent across federal agencies and organizations. The Census Bureau reports issued today are consistent with the new STEM occupation classification.

Posted in Education, Technology | Tagged | Leave a comment

How the Census Bureau Measures the Foreign-Born Population and Immigration

Bookmark and Share

Written by: Elizabeth M. Grieco

Census Bureau statistics can be used to answer many questions about the foreign-born population in the United States. For example: How many foreign-born people are here? Where are they from? Where do they live? Are they naturalized U.S. citizens or noncitizens? Do they speak English? How many have a college degree? Are they working? How much do they make? What percent are in poverty? How many have health insurance coverage?

Data from the American Community Survey can be used to answer these and many other key questions. As the nation’s flagship survey, the American Community Survey is designed to provide statistics on demographic, social, economic and housing characteristics of American communities. This includes the immigrant population.

According to the 2011 American Community Survey, there were 40.4 million foreign-born people living in the United States. More than half (53 percent) were born in Latin America and more than one-fourth (29 percent) in Asia. Most (56 percent) lived in just four states: California, New York, Texas and Florida. Less than half (45 percent) were naturalized citizens.

The majority (84 percent) of the foreign-born residents age 5 and older spoke a language other than English at home, and among those who did, about two in five (40 percent) spoke English “very well.” More than one in four foreign-born residents 25 and older had a bachelor’s degree or higher education.

In addition, two-thirds (67 percent) of immigrants 16 and older were in the labor force. Foreign-born households had a median income of about $46,000, but about one-in-five immigrant families (19 percent) were in poverty. About one-third (34 percent) had no health insurance coverage.

While the American Community Survey can tell us a great deal about immigrants, what about immigration? How many foreign-born people enter the United States each year? How many of today’s immigrants arrived in the last 10 years, during a previous decade, or even during an earlier period, say, before 1980?

The Census Bureau does not count the number of immigrants entering the United States. Rather, that’s the task of the Department of Homeland Security, the principal source of administrative data on immigration. While invaluable, the Homeland Security data include only a limited number of variables, providing little detailed information for analysis.

Data from the American Community Survey, however, can be used to estimate the number of foreign-born arrivals in two ways. First, the survey asks: Where did you live one year ago? Residence one year ago helps determine the number of immigrants who were living outside the United States a year before the survey. For example, in 2011, there were 1.1 million foreign-born people age 1 and over who lived abroad 12 months before the survey.

Second, the American Community Survey also asks: When did you come to live in the United States? Data on year of entry helps estimate the number of today’s foreign-born people who came to this country during specific periods. For example, in 2011, of the 40.4 million foreign-born, 14.4 million came to live in the United States in 2000 or later; 10.8 million arrived between 1990 and 1999; and 15.2 million entered before 1990.

In contrast to administrative data on immigration, the American Community Survey offers a wealth of additional social, economic, demographic and housing information. This extra information gives policy makers and researchers the opportunity to draw important conclusions about the characteristics of immigrants as well as trends over time.

For instance, by combining residence one year ago with country of birth, the American Community Survey data show that immigration from Mexico has recently declined, from 369,000 in 2005 to 161,000 in 2011. By analyzing when the foreign-born came to live in the United States, the American Community Survey also shows that recent arrivals — those who came in 2008 or later — are more likely to be from Asia and less likely to be from Central America than earlier arrivals.

Also, the longer the foreign-born reside in the country — as measured by the years since they came to live in the United States — the more integrated they are into American society. The survey data clearly show that the longer immigrants live in the United States, the more likely they are to speak English “very well,” have good-paying jobs and be naturalized citizens, and the less likely they are to live in poverty and be uninsured.

For additional information on the foreign-born population in the United States, visit the Census Bureau’s website. In addition, please see a recent report on married-couple households with a foreign-born spouse.

Posted in Marriage, Population | Tagged | Leave a comment

Hispanic College Enrollment Surges in 2012

Bookmark and Share

Written by: Julie Siebens

Hispanic college enrollment numbers increased from 2011 to 2012, even while white non-Hispanic and black enrollment numbers declined.  Is this simply the product of a growing U.S. Hispanic population, or are there other forces at work?

The adult Hispanic population in the United States has risen faster than both the white non-Hispanic and black populations in recent years.  The Hispanic population increased by more than 2.2 million from 2011 to 2012, while the black population grew by only 800,000 and the white non-Hispanic population remained stable. Even without a change in the probability of being enrolled in college, we would expect growth in Hispanic enrollment to outpace that of these other groups.

In addition to sheer population growth, however, the Hispanic population also increased its rate of college enrollment.  From 2011 to 2012, the enrollment rate of Hispanics 18 and older increased from 8.8 percent to 9.5 percent, even while the rate for the total (Hispanic and non-Hispanic) over-18 population fell from 8.7 percent to 8.3 percent.

One reason for high enrollment might be the large proportion of young people in the Hispanic population, so it’s important to check how rates are changing by age.  Among young adult Hispanics (age 18 to 24), the rate of college attendance has also been growing.  From October 2007 to October 2012, the percentage of the young adult Hispanic population in college grew from 27 percent to 37 percent.  In the same period, overall college enrollment for people this age went from 39 percent to 41 percent.

What type of colleges did these young adult Hispanic students attend?  Forty-three percent of the Hispanic college students age 18 to 24 were attending a two-year college in 2012, while only 27 percent of similar age non-Hispanic students did so.  Hispanic college students were also more likely than non-Hispanic students to attend part time (22 percent compared with 12 percent).  Eighty-nine percent of them were attending a public college, compared with 81 percent of non-Hispanic students.

Since 1961, the Census Bureau has released annual estimates of enrollment from the October Enrollment Supplement to the Current Population Survey.  This year’s release includes statistics from 2012, historical estimates covering enrollment from 1955 to 2012  and a new report using the 2011 American Community Survey.

Posted in Education | Tagged , | Leave a comment