Multilingual Research for the 2020 Census

Helping reduce costs and encourage greater response rates through expertise in survey methodology, multilingual instrument development, and mobile data collection

As the United States gears up for the 2020 Census, RTI is leading research into innovative ways to enumerate the U.S. population and encourage greater response rates. By focusing on language, technology, and difficult-to-reach populations, we aim to help the U.S. Census Bureau reduce costs while maintaining coverage and data quality.

The Challenge of Collecting Data on More Than 300 Million People

Enshrined in the nation’s Constitution, the United States Census is one of the core functions of our government. It’s how we determine voting district boundaries from Congress down to the local level, and it plays a key role in government functions as well as academic research.

It’s also an enormous undertaking. In fact, counting and collecting data on more than 300 million people represents the largest operation conducted in the U.S. The actual count takes place once per decade, but the planning, research, and support behind the decennial Census never ends.

The Census Bureau has set a goal to conduct the 2020 Census at a lower inflation-adjusted cost per household than the 2010 Census without sacrificing coverage or data quality. Reaching this goal means finding innovative ways to enumerate every person in America and persuade them to respond to the Census. It requires expertise in surveys, multilingual data collection, and sensitivity to cultural differences.

Researching Languages, Technology, and Difficult-to-Reach Populations

In preparing for the 2020 Census, the Census Bureau chose RTI to conduct research concerning three areas: language, technology, and populations that are difficult to reach or enumerate.

We have pretested Census questionnaires in seven of the most frequently spoken languages in the U.S.: English, Spanish, Chinese, Korean, Vietnamese, Russian, and Arabic. We have explored the “bring your own device” concept that will allow people to respond to the Census using mobile devices. And we have evaluated new and revised Census content with Americans who have a rural route address, same-sex cohabiting couples, renters and owners in complex living situations, and Arab Americans born in the U.S. and abroad.

We are uniquely suited to each of these avenues of research. Our multilingual staff of survey researchers evaluated translated Census questionnaires and facilitated improvement using the results of cognitive interviews and focus groups. Our experts in survey methodology and mobile data collection conducted usability tests of the online questionnaires, again considering differences in language. We blended cultural sensitivity and usability expertise to adapt the Census materials to different populations.

Informing National and Local Policy and Advancing Survey Research

The ultimate goal of our work is better, more accurate coverage of the U.S. population, a result that has far-reaching implications for politics, government, and the economy. As the Census Bureau prepares for the 2020 count, our findings on linguistically and culturally appropriate content, enhanced usability, and messages to encourage participation will become part of the operational design.

Our work also allowed us to contribute to the field of survey research. Since 2014, our researchers, often collaborating with their Census Bureau counterparts, have authored more than 10 peer-reviewed publications on improved methods of multilingual research, usability research, and Census coverage. These are just the latest in more than a decade of publications that have grown out of our research for the Census Bureau.

For 2020 and beyond, the Census Bureau is working to conduct an accurate yet cost-efficient measurement of our population. This means taking advantage of new technologies and the improvements they offer. Our tests of innovative practices will help ensure that the Census remains the preeminent source of information about the U.S. population in an increasingly data-driven world.