Form 1 Question changes made throughout testing cycle

---

Overview

Cover
Doc Control Info
Instructions
OMB Approval Log

---

Sheet 1: Cover

OMB Approval Log
Version:	v0.01

---

Sheet 2: Doc Control Info

Document Control Information
Document Information
Document Identification		1
Document Name		OMB Approval Log
Project Name		2020 Tracking Survey
Client		Jennifer Hunter Childs
Document Author		<Insert author name>
Document Version		0.01
Document Status		Draft
Date Released		<Insert release date <dd-mmm-yyyy>>

---

Sheet 3: Instructions

Decisions Log

Overview

The OMB approval log tracks all tracking survey questions which need OMB approval

Instructions

1.

Record the details about each question being added to the tracking survey questionnarie.

Question Log Field Details

Field

Description

Summary

Question #

Unique identifier for the question.

Question wording

Exact wording of the question.

Status

Status of the approval, drop down list which includes the following: New In Progress Closed On Hold Canceled

Date

The date of the question submission.

Expected Approval Date

Date the approval is due to be provided. This must be equal to or greater than the current date.

Actual Date of question approved

Date the question was approved was made.

Data Collection Dates

Dates for data collection of this question

Notes

Notes

Notes or comments about the question

---

Sheet 4: OMB Approval Log

OMB Approval Log
Summary							Notes
Question #	Question Wording	Status	Date	Expected Date of Approval	Actual Date of Approval	Data Collection Dates	Notes
Undercount 1	The 2020 census will ask how many people live or stay in your home. If you were completing the census form today and there were a child 4 years old or younger staying in your home most of the time, would you include this child on your census form, would you not include this child, or are you not sure? Would include, would not include, don't know	Closed				January 6-13, 2020
Undercount 2	The 2020 census will ask how many people live or stay in your home. If there was a child 4 years old or younger staying in your home most of the time, would you include this child on your census form, would you not include this child, or are you unsure?	Closed				March 30-April 5, 2020
Context Experiment	Do you think the 2020 Census questionniare will or will not ask…	Closed	1/6/2020	1/10/2020		January 13-26, 2020	With 1400 cases, split equally into citizen-question and education-question groups, we should be able to detect a Difference-in-Differences of at least 0.076 (using a logistic regression and testing the significance of the interaction term against alpha = 0.10). With 2800 cases, we should be able to detect a difference-in-differences of 0.055. Thus, we are increasing the data collection period to 2 weeks. This analysis assumes that respondents will be equally likely to correctly answer either question, that the mean proportion of respondents answering “definitely/probably will respond” is the same for both questions, and that these parameters remain constant across the study (and do not deviate from the patterns found in the December 2019 data). This analysis does not incorporate survey weights.
	Which people living in your household are U.S. citizens?
	What about the highest education level of each person living in your household? Do you think the census will ask that?
	What about the social security number of each person living in your household?
	What about the date of birth of each person living in your household?
Debunk Experiment (WEB ONLY)	The Department of Commerce requested the 2020 Census ask which people living in each household are U.S. citizens. The U.S. Supreme Court later decided the 2020 Census could not ask that.	Closed	1/6/2020	1/24/2020		January 27-Feb 23, 2020	Legal approved this question. Initially, Nancy had mentioned we wanted to be able to detect a 5% difference for the Intent question and the Citizenship question. At the time of this analysis (using November 2019 web numbers), these proportions were 75% and 74%, respectively. Using these proportions, n=643 or n=662 would be enough to detect a difference, which we rounded up to 700 for our first proposed sample size. To be on the safe side, we also assumed a proportion of 50% in case we wanted to detect differences for other questions or if the proportions for our questions of interest changed. This resulted in a sample size equal to 872, which we rounded up to 900 for our proposed sample size. The initial power analysis I did last December used the standard 2100 GP cases collected over one week, and we determined 900 additional GP cases needed to be collected for the treatment group over the course of the same week to detect a 5% difference. The subgroup analysis from yesterday assumes we continue to collect an additional 900 GP treatment cases each week (on top of the 2100 GP cases we will already be collecting each week). Based on the web quota targets from yesterday, this would mean the following approximate sample sizes for each of the subgroups over the course of 4 weeks: · Hispanic (15%): n=1260 control, n=540 treatment · Black (12%): n=1008 control, n=423 treatment · Asian (11%): n=924 control, n=396 treatment
	Are you aware the 2020 Census will not ask about citizenship?
CORONAVIRUS series	There are multiple methods that you or someone in your household could use to answer the census questions. If you do not respond, a census taker will come to your door to collect a response. What is your preferred way to take the census? [READ] Online, Mailing back a paper questionnaire, talking to an interviewer over the phone, waiting for an interviewr to come to my home.	In Progress	3/13/2020	3/16/2020		3/16/20 - end of collection	This quesiton will replace MODE in an effort to better understand movement in preferred mode of response given COVID-19 and an increased probability of an outbound calling operation.
	If you do not respond to the census in time and a census taker comes to your door, how likely is it that you or someone in your household will respond to the census taker? [READ]	In Progress	3/13/2020	3/16/2020		3/17/2020 - end of collection	This question aims to understand whether or not the population would decline to answer the door, given COVID-19. The Census Bureau seeks to monitor the trend while it is an issue of concern. Extending through the end of the data collection to continually measure public perception of a personal visit as NRFU dates get closer.
	As you may know, a new strain of a virus called coronavirus was identified by the World Health Organization. How concerned are you about the coronavirus? [READ]	In Progress	3/13/2020	3/16/2020		3/17/2020 - end of collection	This question aims to understand whether or not the population would decline to answer the door, given COVID-19. The Census Bureau seeks to monitor the trend while it is an issue of concern. Extending through the end of collection to use as a potential predicter to indended/reported response and receptiveness of a NRFU visit
	In the past two weeks: [READ]	Closed	3/13/2020	3/16/2020		3/17/2020 - 5/4/2020	This series intends to measure changes in media consumption habits that could impact advertising buy decisions during the 2020 Census. Extending through 5/4 measures that still apply given the stay at home orders, both to measure media intake and to use as a composit measure of social distancing as it relates to intended/reported response and receptiveness of a NRFU visit.
Relates out of home ads, drive-time radio, and TV day parts.	ADD_CE_Q8. Have you worked from home or worked remotely	Closed	3/13/2020	3/16/2020		3/17/2020 - 4/7/2020	Stopped 4/7 because it's no longer a valid social distancing measure given "stay at home" orders
	1. More than usual
	2. About the same
	3. Less than usual
	4. Cannot work from home
	98. VOL: Don’t know
	99. VOL: Refused
Relates to out of home ads.	ADD_CE_Q9: And in the past two weeks , have you gone to public places like restaurants or movie theaters	Closed				3/17/2020 - 5/4/2020	This series intends to measure changes in media consumption habits that could impact advertising buy decisions during the 2020 Census. Extending through 5/4 measures that still apply given the stay at home orders, both to measure media intake and to use as a composit measure of social distancing as it relates to intended/reported response and receptiveness of a NRFU visit.
	1. More than usual
	2. About the same
	3. Less than usual
	98. VOL: Don’t know
	99. VOL: Refused
Relates to out of home ads.	ADD_CE_Q10: Have you used public transit	Closed				3/17/2020 - 4/7/2020	Stopped 4/7 because it's no longer a valid social distancing measure given "stay at home" orders
	1. More than usual
	2. About the same
	3. Less than usual
	4. Do not use public transit
	98. VOL: Don’t know
	99. VOL: Refused
Relates to cable ads	ADD_CE_Q11: Have you watched cable or satellite TV	Closed				3/17/2020 - 5/4/2020	This series intends to measure changes in media consumption habits that could impact advertising buy decisions during the 2020 Census. Extending through 5/4 measures that still apply given the stay at home orders, both to measure media intake and to use as a composit measure of social distancing as it relates to intended/reported response and receptiveness of a NRFU visit.
	1. More than usual
	2. About the same
	3. Less than usual
	4. Do not watch cable or satellite TV
	98. VOL: Don’t know
	99. VOL: Refused
Relates to digital ads	ADD_CE_Q12: Have you watched TV using a streaming service on the internet	Closed				3/17/2020 - 5/4/2020	This series intends to measure changes in media consumption habits that could impact advertising buy decisions during the 2020 Census. Extending through 5/4 measures that still apply given the stay at home orders, both to measure media intake and to use as a composit measure of social distancing as it relates to intended/reported response and receptiveness of a NRFU visit.
	1. More than usual
	2. About the same
	3. Less than usual
	4. Do not watch TV using streaming services
	98. VOL: Don’t know
	99. VOL: Refused
Relates to digital ads	ADD_CE_Q13: Have you used the internet to find information, including news	Closed				3/17/2020 - 5/4/2020	This series intends to measure changes in media consumption habits that could impact advertising buy decisions during the 2020 Census. Extending through 5/4 measures that still apply given the stay at home orders, both to measure media intake and to use as a composit measure of social distancing as it relates to intended/reported response and receptiveness of a NRFU visit.
	1. More than usual
	2. About the same
	3. Less than usual
	98. VOL: Don’t know
	99. VOL: Refused
Relates to partnership events	ADD_CE_Q14: Have you avoided large public events or gatherings	Closed				3/17/2020 - 5/4/2020	This series intends to measure changes in media consumption habits that could impact advertising buy decisions during the 2020 Census. Extending through 5/4 measures that still apply given the stay at home orders, both to measure media intake and to use as a composit measure of social distancing as it relates to intended/reported response and receptiveness of a NRFU visit.
	1. More than usual
	2. About the same
	3. Less than usual
	98. VOL: Don’t know
	99. VOL: Refused
Relates to statistics in schools	ADD_CE_Q15: Have you kept children home from school (either voluntarily or due to school closures)	Closed				3/17/2020 - 4/7/2020	Stopped 4/7 because it's no longer a valid social distancing measure given "stay at home" orders
	1. More than usual
	2. About the same
	3. Less than usual
	4. Do not have school-aged children
	98. VOL: Don’t know
	99. VOL: Refused
Relates to Digital ads	ADD_CE_Q16: Have you used or visited social media sites	Closed				3/17/2020 - 5/4/2020	This series intends to measure changes in media consumption habits that could impact advertising buy decisions during the 2020 Census. Extending through 5/4 measures that still apply given the stay at home orders, both to measure media intake and to use as a composit measure of social distancing as it relates to intended/reported response and receptiveness of a NRFU visit.
	1. More than usual
	2. About the same
	3. Less than usual
	4. Do not use social media
	98. VOL: Don’t know
	99. VOL: Refused
Relates to intention to participate in the census and awareness	ADD_CE_Q17. Since the news of the coronavirus (COVID-19) pandemic started, have you experienced any of the following, or not? [Randomized response options]	Closed				4/13/2020-5/4/2020	This item intends to measure changes on coronavirus related social stigma that could impact the ways in which people intend to participate in the 2020 Census, their awareness about the census, and in general about their willingness to cooperate with government based data collection. In particular, we are interested in measuring willingness to open the door to an enumerator if the person has experienced stigma. Data from this question will be used primarily for the partnership program to aid in communication. There has been a documented increased in coronavirus-related discrimination and social stigma particularly against people of Asian descent, people who have traveled, emergency and healthcare respondents. See CDC's page dedicated to social stigma and coronavirus: https://www.cdc.gov/coronavirus/2019-ncov/daily-life-coping/reducing-stigma.html
	1. Verbal harassment Yes/No
	2. People refusing to interact with you Y/N
	3. People acting as if they are afraid of you Y/N
	4. Physical violence Y/N
	5. Denial of healthcare Y/N
Census Nonresponse	WEB ONLY (ask of those who say they haven’t completed the 2020 Census, after answered, after intent, after mode_pref):	Closed				5/5/2020-5/7/2020	To further understand census non-response to see if there are additional reasons we should be addressing/set up to measure. This was approved for collection 5/3-5/5 but we postponed them to reduce burden by waiting until the other COVID items ended on 5/4.
	You said you have not completed the 2020 Census.  What are the main reasons you have not completed it?
Undercount 3	The 2020 census will ask how many people live or stay in your home. If there was a child 4 years old or younger staying in your home most of the time, would you include this child on your census form, would you not include this child, or are you unsure?	Closed				6/1/2020 - 6/7/2020	Given the extension of the self-response phase, the undercount of young children research team wants to add a data point to help influence advertising and partnership activity for the next few months. In addition, the team would like to add an open ended response if a person says that they are not sure or would not include the child to further inform outreach efforts.
	If no or unsure Why do you say that? (open end)	Closed				6/1/2020 - 6/7/2020
Face Masks	Split-ballot such that half the sample gets either a or b.	In Progress				6/10/2020-7/8/2020
	a. If the census taker came to your door and was wearing a face mask, would you be more likely to respond, less likely to respond, or would it not make a difference?	In Progress				6/10/2020-7/8/2020	This question would be used to inform both advertising (whether ads should be made with enumerators in masks and 2) field staff instruction, which currently is to abide by state and local guidelines. In the web data, 172 non-responders would be split between two questions, so ~85 non-responders per mask question. In the phone data, 282 non-responders would be split between the two questions, so ~140 non-responders per mask question. Based on our power analysis using these n-sizes, we have proposed the above listed length length of time in order to detect approximately a 15% difference at the subgroup level.
	b.If the census taker came to your door and was NOT wearing a face mask, would you be more likely to respond, less likely to respond, or would it not make a difference?	In Progress				6/10/2020-7/8/2020