Responses to OMB questions on the ICR

U.S. Census Bureau Response to the

Office of Management and Budget (OMB) Questions on the

2010 Census Coverage Followup Telephone (CFU) Operation

July 22, 2009

1. Is a report available detailing the results of the 2008 CFU operation for the Dress Rehearsal? Please provide a copy of the report or the expected date when the results will be available.

See the attached reported entitled, “2008 Census Dress Rehearsal Coverage Followup Assessment Report,” 2008 Census Dress Rehearsal Memoranda Series, No. 65, July 30, 2009, pp. 1-90.

2. Please provide more information on how Census expects the approximately 8 million households that will be included in the CFU telephone operation will come from each of the following:

1. Census returns with more than six people:

1.2 million households

2. Census returns with count discrepancies:

2.2 million households

3. Census returns with the overcoverage and/or undercoverage question being
   checked 'yes:'

4.5 million households

4. Census returns matched administrative records showing an undercount for the household:

30,000 households

5. Census returns for which computer matching has determined possible duplicates.

50,000 households

3. If the caseload exceeds the projected 8 million households, how will Census prioritize among the above different sources in selecting cases for CFU?

We will manage the workload in “buckets” based on the case type. Lower priority “buckets” will not be called during the start of operations. Higher priority "buckets" will be called. During that time, we will assess how good the workload estimates are and how well the phone centers are doing. If the phone centers meet their productivity goals, they will be able to handle the lower priority "buckets" workload. If the phone centers cannot handle additional workload (because the caseload exceeds the projected estimate or the phone centers are less productive than expected), then those “buckets” that are lower on the priority list will not be called. The following is the priority order:

• Count Discrepancies

• Large Households (households with more than six people)

• Overcount Coverage Question – College category

• Overcount Coverage Question – Nursing Home category

• Returns matching to Administrative Records showing an undercount

• Overcount Coverage Question – Jail category

• Overcount Coverage Question – Multiple categories selected for a housing unit

• Overcount Coverage Question – Military category

• Undercount Coverage Question – Someone Lives Here Temporarily category

• Undercount Coverage Question – Relatives category

This ordering is based on expected roster changes per call based on evidence from studies earlier in the decade.

4. On page 4, it was noted that the "2010 Residence Rule" will be available pending approval. What is the status of this? Please provide the rule if it has been approved or provide more information on when this will be finalized.

The residence rule for the 2010 Census is below:

Residence Rule

The residence rule is used for determining where people should be counted (which means tabulated) in the 2010 Census.

Residence Rule: Count people at their usual residence, which is the place where they live and sleep most of the time. People in certain types of group quarters (GQ) on Census Day should be counted at the GQ. These GQ types are listed in the box below. People who do not have a usual residence or cannot determine a usual residence, and who are not in one of the GQs types listed below, should be counted where they are on Census Day.

5. On page 4, it was noted that Census interviewers will call respondents using telephone numbers provided by respondents. What is the protocol, if there is no telephone number provided or if the number is no longer working or incorrect?

A telephone lookup operation is scheduled to obtain phone numbers for cases missing this information and gathers alternate phone numbers (up to three) for all cases. Thus, if the number provided by the respondent is no longer working and an alternate phone number was identified during telephone lookup, then that number will be used to attempt to contact the respondent.

If we still do not have a phone number or the phone number is not working or incorrect, we will be unable to complete the interview. The case will be closed and no additional attempts will be made to reach the respondent.

6. B.2 states that CFU cases "will stay in the DRIS telephony channel until they are resolved," but please clarify different ways a case will be resolved. Please provide more information about the minimum number of calls and the timing of those calls that will be made. Is there a maximum number of calls that will be made? Will special interviewers and special protocols be used to convert refusals?

A case can be resolved if a respondent is reached and either completes the CFU interview or refuses to participate (two attempts maximum). There are also cases where the respondent is never reached, even after multiple attempts. Examples of this occur when the telephone number is no longer working; it is a fax machine, has a busy signal, or rings but it is never answered. For the examples given, these cases are attempted a maximum of 25 times. Note that in 2008, the average number of attempts on the cases called was six; we expect 2010 to be similar. Every case with a telephone number will be attempted at least once.

Calls will be made Monday through Saturday, between 9 a.m. and 9 p.m. local time and on Sunday between 11 a.m. and 9 p.m. local time. An automated dialing system will be used to prioritize what calls are made on each day and the time when the call is attempted. Calls are made randomly so that a household may be called on different days of the week and at different times of the day. The system also allows for respondents to make appointments when it is convenient for them. No called will be placed if there is no interviewer available.

The CFU will employ refusal conversion specialists who are interviewers will either have previous interviewing skills or be identified during training as having good soft skills. No additional training is planned beyond what they receive in interviewer training (which includes training on soft skills, Frequently Asked Questions and answering respondents concerns), although it is currently being discussed.

7. Please provide more information about why CFU is planned to continue into mid-August. Is CFU being done on NRFU or VDC cases?

The Non Response Followup cases are eligible to be selected for CFU. Therefore, CFU continues into mid-August to provide these cases with sufficient time to be called. Additionally, cases identified using administrative records are included in the CFU program in July, allowing interview to continue for several weeks after this delivery provides an opportunity to reach respondents and complete interviews.

8. Please explain what is meant by "Residence Coding of CFU collected data as part of data integration."

While CFU collects information to help determine a person's residence, the determination of the final residence status (or the residence coding) of the persons in the households that completed a CFU interview is done after CFU. Using the information gathered during CFU for that particular case, residence coding determines whether a person should or should not be considered a census day resident of the household.

9. What evaluations are planned to examine the effectiveness of the CFU telephone operation?

An assessment of the operation will be conducted in 2010, similar to the attached 2008 report, which will identify how often we added or deleted someone as a result of the interview.

Additionally, an experiment will be conducted to determine how to improve the CFU interview. Questions are those presented in Mod Q as previously submitted in the Instrument Requirements. These questions are asked when a respondent indicates that someone may be missing or counted in error (using the coverage questions) but then fails to tell us anything about that situation in CFU. We will use the Mod Q questions to help understand why we did not end up changing a roster when we had expected to and understand the living situations of those people.

July 31, 2009 4