An official website of the United States government
Here’s how you know
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
Secure .gov websites use HTTPS
A lock (
) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.
This report analyzes data gathered from an experiment that was carried out in the 2020 Census. A new training module was designed for bilingual enumerators to aid in administering the census questionnaire to Spanish-speaking households in the nonresponse follow-up operation. An objective is to study the association between the training and response rate. A multinomial regression with continuation-ratio logit link is used to model the sequence of contact attempts to affected households. Several log-odds ratios are considered to provide inference about this relationship. We do not find significant evidence to conclude that the training is associated negatively with response rate; i.e., training appears not to harm response rate. We suspect that a stronger conclusion can be made about the training and improvement to response rate in applicable contact situations; however, a more controlled experiment with associated data would be needed to make this assessment. Additionally, a power analysis which was carried out in advance of the experiment is now revisited. Here, power of the selected log-odds ratios is evaluated under the experiment as carried out in practice using the realized sample size.
Share
Some content on this site is available in several different electronic formats. Some of the files may require a plug-in or additional software to view.
Top