Seeing Significance: New Chart Showing Confidence Intervals for Nonfarm Employment Changes

Last year I wrote about how we’ve been using more charts and maps to explain our data. In another post I wrote about how BLS deals with uncertainty in our measures and how we explain the strengths and limitations of our data. Today I want to tell you about a new chart that will help you see measures of uncertainty in one of our most closely watched series—nonfarm employment.

We recently fielded a survey to ask data users about our ideas for creating charts that show confidence intervals. A confidence interval is a measure of the uncertainty in our estimates. We asked data users to tell us if confidence intervals would give them useful information. If so, how can we present confidence intervals in a clear, visually appealing way? Based on your responses, we will add a chart showing confidence intervals to the package of interactive graphics we update each month with The Employment Situation report. We will post the new chart when we release the November 2016 national employment numbers on Friday, December 2.

The responses to our survey confirm you want to know more about the limitations of the employment data. The red dots in the chart show the over-the-month employment changes for total nonfarm and the major industries. The blue bars show the 90-percent confidence intervals of the employment changes for each of these groups.

Chart showing nonfarm employment changes for major industries in October 2016 and the confidence intervals for those changes.

What does this chart tell us? The red dot for total nonfarm employment shows a gain of 161,000 jobs in October, as we reported on November 4. That number is an estimate based on our monthly sample survey, rather than a complete count of jobs each month. Different samples of employers might give us different estimates of employment change. We can measure the sampling error, the variation that occurs by chance because we collected the number from a sample of employers instead of all employers. With our measure of sampling error, we can calculate a confidence interval. The blue bar for total nonfarm shows the 90-percent confidence interval ranged from 46,800 to 275,000. We call this a 90-percent confidence interval because, if we were to choose 100 different samples of employers, the October nonfarm employment change would be between 46,800 and 275,000 in 90 of those samples.

We also learn from this chart whether an employment change is statistically significant. A change is statistically significant if the blue bar in the chart does not cross the zero line. For example, the confidence interval for construction includes zero, so we can’t say with confidence that construction employment increased in October. For education and health services, however, the confidence interval does not include zero, so we can say more confidently that employment in the industry rose over the month.

The chart here shows the 1-month employment changes. When we begin publishing it on December 2, we will also let you choose charts that show 3-, 6-, and 12-month changes.

Check out the new charts and let us know what you think in the comments below. We’re always looking for better, clearer ways to explain our data, and I welcome you to share your ideas.

Labor Market Status of U.S. Military Veterans

As we continue to celebrate our veterans this month, here are our most up-to-date statistics about veterans in the civilian labor force.

  • After reaching 9.9 percent in January 2011, the unemployment rate for veterans was 4.3 percent in October 2016.
  • The unemployment rate for Gulf War-era II veterans — who served on active duty at any time since September 2001 — reached 15.2 percent in January 2011 and was 4.7 percent in October 2016.
  • The peak unemployment rate for nonveterans was 10.4 percent in January 2010; their rate was 4.5 percent in October 2016.
  • There were 471,000 unemployed veterans in the United States in the third quarter of 2016; 22 percent of them were ages 18 to 34.
  • More veterans work in government than in any other industry; 21 percent of all veterans and 27 percent of Gulf War-era II veterans work for federal, state, or local government. By comparison, 13 percent of employed nonveterans work in government.
  • After government, the next largest employers of veterans are manufacturing and professional and business services (about 12 percent each).

Now let’s take a look at some data that may help veterans who are looking for work or considering a career change.

Looking to move?

In 2015, the unemployment rate for veterans varied across the country, ranging from 1.9 percent in Iowa to 7.7 percent in the District of Columbia.

Map of unemployment rates for veterans by state in 2015

What industries have the most job openings?

There were 5.5 million job openings in September 2016. Here’s how they break down by industry.

Chart showing job openings by industry in September 2016

What are the fastest-growing jobs?

Thank you, veterans, for your service. Check out our website at www.bls.gov 24/7 or give our information office a call at 202.691.5200. We also have regional information offices available to help you. BLS has the data you need to make wise decisions.

BLS Microdata Now More Easily Accessible to Researchers across the Country

I am pleased to announce that BLS is now part of the Federal Statistical Research Data Center Network.

Researchers at universities, nonprofits, and government agencies can now go to 24 secure research data centers across the United States to analyze microdata from our National Longitudinal Surveys of Youth and our Survey of Occupational Injuries and Illnesses. Before, researchers had to visit our headquarters in Washington, D.C., to use these data.Image of researchers examining data.

Making our underlying data more accessible for researchers from coast to coast is a huge step forward, and I hope it will lead to a surge in research using BLS data. I believe that having more researchers use BLS data not only will showcase new uses of the data but improve our products by encouraging researchers from BLS and other organizations to collaborate. It also supports transparency because external researchers can analyze inputs to our published statistics.

Another key benefit to having BLS data alongside datasets from the U.S. Census Bureau and the National Center for Health Statistics is that researchers can combine data from two or more agencies. Using multiple datasets allows researchers to match data to answer new questions with no more burden on our respondents. Put simply, more data = better research = better decisions that rely on research.

Researchers are enthusiastic about adding BLS data to the research data center network.

“We at the Federal Reserve Bank of Atlanta are excited that more BLS microdata are available to researchers. Policy questions are usually complicated. Matched data from different sources can give researchers a much better understanding of economic relationships. That will help us provide more informed policy advice,” said John Robertson, senior policy adviser at the Federal Reserve Bank of Atlanta.

Over the next year, we will add more BLS data to the research data centers based on user demand.

Researchers can also still visit us at our D.C. headquarters to access our full suite of microdata. To learn more and to apply, see our BLS Restricted Data Access page.

The Value and Influence of Labor Statistics in the 21st Century

What’s in the “DNA” of BLS—what were we born with? Not so long ago, as I prepared to become BLS Commissioner, I read the First 100 Years of the Bureau of Labor Statistics. The first chapter describes how BLS was created (in 1884) during a time of severe economic upheaval and industrial unrest. Policymakers of the time realized that a key barrier to peace and shared prosperity was the lack of trustworthy information about the economy. What has struck me ever since is how we can trace some of the distinguishing features of today’s modern BLS directly back to those first days, to the vision of one of our founders. This post links that past to the BLS of today.

Carroll D. Wright, first BLS Commissioner

Commissioner Carroll D. Wright

In 1893, sometime after becoming the first Commissioner of Labor, Carroll D. Wright set forth a mission for the agency. He was a pioneer in the search for truth and a better understanding of labor statistics by the public. In his Value and Influence of Labor Statistics (later published in the 54th Bulletin of the Bureau of Labor), he described our mission as collecting “information upon the subject of labor in the United States, its relation to capital, the hours of labor, and the earnings of laboring men and women, and the means of promoting their material, social, intellectual, and moral prosperity.”

Today, our mission is much the same as it was then. Commissioner Wright established a modern statistical agency long before the Internet made it possible for anyone to access our data and read our publications on demand. These days we say that our mission is “to collect, analyze, and disseminate essential economic information to support public and private decision-making.” While the wording has evolved with the times, the core meaning remains the same. Furthermore, in support of our mission for the past 132 years, BLS has practiced what Commissioner Wright termed “the fearless publication of the facts without regard to the influence those facts may have upon any party’s position or any partisan’s views.”

Wright developed much of the vision and practices that he instilled here while working for the Massachusetts Bureau of Statistics of Labor from 1873 to 1878. There he launched several studies to provide the people of Massachusetts with accurate labor market data. One of the largest studies was to find out the true unemployment level in Massachusetts. At the time, many people believed there were 200,000–300,000 people unemployed in the state and 3,000,000 unemployed in the entire country. Alarmists spread word through newspapers, speeches in Congress, and political resolutions until these figures were widely believed as fact, despite no previous attempt to measure unemployment. Wright’s staff canvassed the state twice to discover if the rumored number was accurate. The Bureau of Statistics of Labor of Massachusetts determined the true number of unemployed in the state was 28,508 skilled and unskilled laborers in June 1878; by November there were fewer than 23,000 unemployed, while the national number could not have been more than 460,000 unemployed. Wright explained that “The figures published by the report were used all over the country, and completely reversed the popular belief relative to the vast number of the alleged unemployed in the country.”

Today, you can see a parallel between Wright’s efforts to learn and classify the number of unemployed workers in Massachusetts and how BLS has expanded its offering to include six alternative measures of unused or underused labor. We call these measures U-1 through U-6. BLS not only calculates these alternative measures nationally, but also for each of the 50 states, the District of Columbia, and two large metro areas. This ensures that the American public, researchers, and policymakers have a wide range of data to understand the health of the labor market and make important decisions.

Also similar is our enduring focus on specific populations in the workforce. Under Wright’s leadership, state Bureaus of Labor investigated the use of child labor and uncovered the “evils it entailed upon the community.” The Bureaus published the number of young children (those under 10 years old) who worked in factories and workshops. Because of these studies, the numbers declined significantly. Time and again, Wright sought out the facts and ensured the American people had the information they needed to make decisions. Wright said, “It is only through rigid, impartial, and fearless investigations that any community can know itself in many directions.”

Today, we continue to seek new and better measures about particular groups in our economy and society. For example, in recent years BLS expanded the scope of the Current Population Survey to include six new questions to identify people with disabilities. These data provide insight into the labor market challenges of people with disabilities. The data aid individuals, nonprofit organizations, employers, and policymakers in making decisions affecting the lives of Americans with disabilities. Our monthly Employment Situation report now includes information about the employment status and labor force participation of the more than 30 million Americans age 16 and older living with a disability.

Our “DNA,” that is, our mission, our vision, and our understanding of the value of the statistics we produce, is as important today as it was in 1884. We continue our determined work to impartially collect, analyze, and publish essential economic information to support private and public decision-making. Today BLS provides a wide variety of information that benefits all Americans. I am certain that Commissioner Wright would be pleased that our reports, charts, and data are far more accessible than he ever could have imagined. Whether you’re exploring a new occupation, starting a business, looking for the change in consumer and producer prices, identifying average wages by occupation, or learning how Americans spend their time, there’s a stat for that. For all these situations and many more, BLS helps Americans make smart decisions in their lives. The cost of providing this valuable information may come out to less than $2 per person each year, but its positive impact remains priceless.

A Brief Labor Market Update for Labor Day 2016

A diverse group of people in a variety of occupationsIn 1894, President Grover Cleveland signed the law designating “Labor Day” as the first Monday in September. This national holiday pays tribute to American workers. A decade before Labor Day existed—since the creation of the U.S. Bureau of Labor Statistics in 1884—we began reporting on how the labor market is faring. So, what’s up as we reach Labor Day 2016?

  • Our monthly payroll survey shows that employment continues to expand—now nearly 6.2 million jobs above the January 2008 peak.
  • Although job growth continues, it has been slower in 2016 than in the last couple of years. The average monthly job gain in 2016 has been 182,000, compared with 229,000 in 2015 and 251,000 in 2014.
  • At 4.9 percent in August, the unemployment rate has changed little since August 2015. During late 2006 and early 2007, the unemployment rate was at its recent low, 4.4 percent. In October 2009, the rate reached 10.0 percent.
  • The number of long-term unemployed people (those jobless for 27 weeks or more) was 2.0 million in August. That was 26.1 percent of the total unemployed, down from the recent peak of 45.5 percent in April 2010, but still above the 16-percent share seen in late 2006 and early 2007.
  • July unemployment rates were uneven among the states. South Dakota (2.8 percent) and New Hampshire (2.9 percent) had the lowest rates, while Alaska (6.7 percent) and Nevada (6.5 percent) had the highest.
  • Among major worker groups, the unemployment rate for teenagers was 15.7 percent in August, while the rates were 4.5 percent for both adult women and adult men. The August unemployment rate for African Americans was 8.1 percent, compared with 5.6 percent for Hispanics or Latinos, 4.4 percent for Whites, and 4.2 percent for Asians.
  • The labor force participation rate—the share of the population working or seeking work—has been trending down since the early 2000s and even more rapidly since 2008. The rate was 62.8 percent in August 2016, down from rates around 66 percent that prevailed from late 2003 to 2008.
  • Real (adjusted for inflation) average hourly earnings for all employees increased 1.7 percent from July 2015 to July 2016. Real earnings have finally started to grow in 2015 and 2016, after several years of little change.
  • Among workers in private industry, 64 percent had access to paid sick leave in March 2016, and 76 percent had access to paid vacations.
  • Labor productivity in nonfarm businesses decreased at a 0.6-percent annual rate in the second quarter of 2016. Although labor productivity has fallen recently, it has grown by 330 percent since 1947.
  • There were 4,821 workers in the United States who died from an injury suffered at work in 2014. That was the highest annual total since 2008 but still below the numbers of workplace deaths in the 1990s and early 2000s.
  • The rate of nonfatal occupational injuries and illnesses has declined over the past several decades in the private sector. The rate in 2014 was 3.2 cases per 100 full-time workers, down from 9.2 cases per 100 full-time workers in 1976.
  • From 2014 to 2024, 7 of the 10 occupations with the fastest projected growth are related to healthcare, but there will be opportunities in a variety of fields.

The U.S. economy is large, complex, and evolving. So, BLS works hard to provide good information to help Americans make better informed decisions. We’ve been doing this for over 130 years and plan to keep serving America’s information needs for many decades to come!