Demographic analysis estimated that the 2010 Census had a 4.6 percent net undercount of children under the age of 5 (Hogan et al. 2013). Young children had a higher net undercount than any other age group. The Census Bureau is looking at data from a variety of sources to better understand this coverage problem and reduce the undercount of young children in the 2020 Census. This report analyzes response data from the Census Bureau’s 2020 Census Barriers, Attitudes, and Motivators Study (CBAMS) survey. The 2020 CBAMS provides an opportunity for the Census Bureau to determine if the attitudes, barriers, and motivators for households with young children differ in any important ways from those of households without young children. In this report, “young children” are defined as children age 5 and under.

There are several important limitations associated with this analysis. Response and nonresponse errors may distort some of the conclusions. The CBAMS survey had a response rate of about 39 percent, and it is possible that the barriers, attitudes, and motivators of nonresponding households differ from those of responding households. In addition, errors will exist if households with young children did not correctly report in the CBAMS survey as having young children. Finally, the report does not try to isolate the effect that young children in the household might have on attitudes, barriers, and motivators. Differences in the demographic, economic, and social characteristics of households with and without young children likely contribute to the differences summarized in this report. For additional discussion of limitations, refer to section 4.4 on page 7.

In general, the responses for people who have a young child in the household are very similar to those without a young child in the household. The report includes detailed information on both similarities and differences. We also highlight issues that we believe are likely to be important in informing an education and outreach campaign for households with young children.

The results based on all households with young children compared with all households without young children revealed the following:

• Only 60 percent of respondents with young children said they were extremely or very likely to respond to the 2020 Census compared with 68 percent of respondents without a young child in the household. The communications campaign needs to address the likelihood of lower self-response for respondents with young children to achieve success in reaching young children in these households.
• Most respondents (with and without young children) indicated that funding for public services was the most important reason to participate in the census. The specific public services identified by respondents with young children differed from those identified by respondents without a young child. Respondents with young children indicated that the most important services and programs were:
  • Day care for children.
  • Schools and the education system.
  • Job training programs.
  Given the connection between census counts and federal funding, making this connection could be a powerful way to motivate respondents with young children to complete the census. For details, see https://www2.census.gov/programs-surveys/decennial/2020/program-management/working-papers/Uses-of-Census-Bureau-Data-in-Federal-Funds-Distribution.pdf.
• Respondents with young children indicate they have very high internet access. (Only 2 percent say they never use the internet.) The very high use of smartphones to access the internet and a strong preference for online reporting also set households with young children apart from those without young children. These trends are likely because of the younger ages of the respondents in households with young children. Having an internet response option in 2020 that will allow self-response by smartphone could help in gaining response from these households. Messaging should emphasize the ease of accessing and completing the census questionnaire online. Advertising through the internet is likely to successfully reach many of the respondents in households with young children.
  
  It is worth noting, however, that the CBAMS survey found that internet use is lower for non-Hispanic Blacks, Hispanics, and low-income respondents with young children. This is important because Census Bureau research using demographic analysis methods estimates a higher net undercount for Black and Hispanic young children compared with non-Hispanic White young children (O’Hare 2014). Additional research using administrative records finds higher net undercoverage for racial and ethnic minority young children and young children living in low-income families (Fernandez et al. 2018).
• Households with young children are less familiar with the census and have lower levels of knowledge about the census and its uses. While all households share some misconceptions about the census, households with young children are, on average, younger and have less experience with a previous census. They may need clear education about what is true and what is not true about the census, its process, and its uses.

A more detailed look at the universe of households with young children highlighted the variation in responses by social, economic, and demographic characteristics. Survey responses varied greatly with respect to race and Hispanic origin, education, income, and language. For example,

• Looking at households with young children, respondents with the highest incomes and the highest levels of education differed markedly from those of respondents with lower incomes and less education. The highest levels of concern (about confidentiality, data sharing, and potential negative repercussions) and the lowest levels of trust in the federal government were found in households with young children with the lowest incomes and the least education. These lower income and lower educational attainment groups were also less likely to report an intention to respond to the census.
• The survey responses from non-Hispanic White respondents with young children in their household differed in many ways from those of respondents in households with young children of other race and Hispanic origin groups. Differences are seen in some barriers and motivators to response. For example, the enforcement of civil rights laws was far more important to non-Hispanic Black respondents compared with non-Hispanic White respondents. Hispanic respondents were more likely than non-Hispanic White respondents to rate each of the CBAMS-identified uses of census data as important.
• Within households with young children, respondents who spoke a language other than English at home and respondents who were not proficient at speaking English responded differently to many questions in the CBAMS survey than those only speaking English at home and speaking it very well. They were less likely to intend to respond to the census and more likely to identify barriers to their responses. They were more likely to prefer paper over online questionnaires. Non-English language materials (i.e. questionnaires and messaging) and language assistance are required for these populations.
  

Given these differences, the authors of this report urge that social, economic, and demographic characteristics be taken into account in any planning for messaging to households with young children.