Part B - SUPPORTING STATEMENT 12-27-10

SUPPORTING STATEMENT

U.S. Department of Commerce

U.S. Census Bureau

National Survey of Fishing, Hunting, and Wildlife-Associated Recreation (FHWAR)

Cell Phone and Debit Card Test

OMB Control Number 0607-<XXXX>

Part B – Collections of Information Employing Statistical Methods

Question 1. Universe and Respondent Selection

This study’s universe is the household addresses for which telephone numbers are unavailable. The frame is the 34,064 2011 FHWAR cases (minus the 5,154 CAPI cases selected for the full survey) that have no available telephone numbers selected from the July 2010 Master Address File (MAF) which is maintained by the U.S. Census Bureau.

Three panels of 500 sampled cases will be selected for this test. These cases will be distributed proportionally to the states based on the phone number look-up rates within each state from the 2011 FHWAR survey. We will look at response rate estimates at the national level. These sample sizes within the three different treatment groups will allow us to detect changes in response rates between the groups as low as 5.2% at the 90% confidence level.

We wish to measure the response rates for these three panels which are based on different means of reaching respondents. These rates will be compared to the 2011 FHWAR response rate to determine if any of the methods will increase the overall response rate.

Question 2. Procedures for Collecting Information

Data collection for the test panels will be conducted at the Census Bureau’s decentralized telephone centers using the same instrument used for the 2011 FHWAR production. The first wave of the 2011 FHWAR Survey will be conducted April to June 2011 and consists of a screener interview and two detailed interviews. The advance letter will instruct household respondents to begin calling the telephone centers on April 1, 2011. Interviewers will also begin calling the cell phone sample in April. The household respondent will be asked to provide a household roster, demographics, and details on the household’s hunting, fishing and wildlife-associated activities. All persons in the household who hunted and/or fished in 2011 by the time of the screening interview will be sampled for the Wave 1 – Sportsperson Detailed Interview. All persons in the household who have participated in wildlife-associated activities in 2011 by the time of the screening interview will be sampled for the Wave 1 – Wildlife Watcher Detailed Interview. (Note: those household members who indicate they will participate in hunting, fishing, and wildlife-associated recreation later in 2011 will not be interviewed in Waves 2 and/or 3 since the test will only be conducted in Wave 1.) The reference period for sampling into either detailed interview is January 1, 2011, through the date of the screener interview.

Estimates from the three panels will not be used in the final 2011 FHWAR estimates. The purpose for collecting these test data is to analyze response rates to determine if one of the data collection methods can be implemented to help improve the response rates for future surveys without impact on the cost comparison and data quality.

We will look at the national weighted response rate of each panel and the sportperson participation rates through the first wave for those addresses where no phone number is available. The estimation procedure will reflect the inverse of the probability of the address being selected into the panel. Population controls for the number of addresses where no phone number is available are unknown. Thus, ratio controls to known populations will not be used in this study.

There are no unusual problems requiring specialized sampling.

This is a one-time data collection operation that is coincident with the 2011 FHWAR data collection.

Question 3. Methods to Maximize Response

The Census Bureau’s telephone interviewing staff will perform standard procedures to keep the non-interview rate at the lowest possible level.

Wording on each sample case’s advance letter will encourage household respondents to call the telephone centers to conduct an interview.

The first panel will receive a prenotice post card letting the household know that they will receive a cell phone in the next few days as a means of communication between their household and the Census Bureau to conduct a demographic survey. A few days later, the household will receive a cell phone with a specialized advance letter explaining the survey and the purpose of the cell phone. Interviewers will have a cell phone number to reach the Cell Phone Sample. Interviewers will make multiple call attempts at various times of the day over the interviewing period in an effort to reach the Cell Phone Sample.

The second panel will receive an advance letter and a $25.00 debit card. The advance letter will request that a household member call the telephone center to complete an interview and accept the debit card as a “thank you” for participating.

The third panel will receive only an advance letter that requests a household member call the telephone center to complete an interview.

Interviewers will not have a telephone number for the Debit Card Sample and the Advance Letter Only Sample. The only way Census Bureau interviewers can obtain an interview with these samples will be if a household respondent calls the telephone centers and is willing to conduct an interview.

We have two additional panels (Panel 4 and 5 in the table on the next page) of all the households from our production sample for quality comparison. The first additional panel will be a subsample of the cases from our Computer-Assisted Personal Interviewing (CAPI) sample where we could not locate a phone number. This panel will receive the advance letter and be part of the personal visit sample. The second additional panel will be the cases from our Computer-Assisted Telephone Interviewing (CAT) sample where we located a phone number. This panel will receive the advance letter and be part of the telephone interview sample. The table below outlines the panel sample sizes.

2011 FHWAR Research Panels
Panel	Sample	Sample Size	Phone Numbers Available?
1. Advance Letter and Cell Phone	Test	500	Cell phone number only
2. Advance Letter and Incentive	Test	500	No
3. Advance Letter Only	Test	500	No
4. Advance Letter with Personal Visit Follow-up	Production	5,154	No
5. Advance Letter with Telephone Follow-up	Production	47,891	Yes

The five panels will help us determine whether there are viable alternatives to collecting data through a personal visit when we have an address but no telephone number. First, we would like to decide if any of these alternatives are feasible. If the operation of collecting the data works, (i.e., test respondents call the telephone centers and/or contact is made via the cell phone), we want to determine which alternative works best in terms of response rates, data quality, and cost. We will look at the response rates of the five panels. We will consider Panel 4 (Advance Letter with Personal Visit – the current method) as the base line for our study. We will compare the response rate from this panel to the response rates of Panels 1, 2, and 3. These comparisons will help us determine if any of the alternative procedures produce response rates equal to or greater than the current method for cases without phone numbers. Within the three alternative methods, we will consider Panel 3 as a base line and compare its response rate to the response rates of Panels 1 and 2. This test will determine which of the three alternatives produces the highest response rate. In another analysis, we will compare the response rate of the cases with no available phone number to the response rate of the cases with a phone number (Panel 4 versus Panel 5). This will help to determine whether conducting a CATI interview differs from conducting a CAPI interview in terms of response rates.

The data from these three test panels will be analyzed separately for research purposes, and therefore will be excluded from the final 2011 FHWAR survey data.

As part of this test, we will study the number of respondents who called the telephone centers during the Wave 1 interview period, the reasons why they called, and the number of interviews obtained from these calls. Collecting these data may help us find alternative methodologies that could increase response rates in future FHWAR surveys.

The Census Bureau will not attempt personal visit interviews for the test cases because we are investigating different modes to increase the response rate by telephone interviewing without personal interviews.

Question 4. Tests of Procedures or Methods

The Census Bureau is not introducing new procedures or methods with this collection. All questions in the 2011 FHWAR have been fielded in a previous version of the survey, with the exception of one. Four other questions were changed slightly or the answer categories were modified slightly. In order to comply with the Census Bureau’s pretesting policy, the Census Bureau’s Statistical Research Division was asked to provide an expert review of these questions.

We are coordinating cognitive testing for the postcard and letters with the Census Bureau’s Center for Statistical Methods and Research (CSRM). The purpose of the testing is to evaluate the clarity of these materials and to evaluate respondent reaction regarding the idea of receiving a cell phone or debit card in a national survey.

Question 5. Contacts for Statistical Aspects and Data Collection

You may consult the following Census Bureau individuals for information regarding sample

design and data collection:

Sample Design: David Hornick 301-763-4183

Thomas Moore 301-763-5997

Data Collection: Denise Pepe 301-763-3785

Deborah Kinnaman 301-763-3817

Created by DSD/SSB Page 4 2/1/2021