E7-23401. Collection of Information Under Review by Office of Management and Budget: OMB Control Number: 1625-0089

AGENCY:

Coast Guard, DHS.

ACTION:

Thirty-day notice requesting comments.

SUMMARY:

In compliance with the Paperwork Reduction Act of 1995, this request for comments announces that the U.S. Coast Guard is forwarding one Information Collection Request (ICR), abstracted below, to the Office of Information and Regulatory Affairs (OIRA) of the Office of Management and Budget (OMB) requesting re-instatement, with change, of a previously-approved collection of information: 1625-0089, National Recreation Boating Survey. Our ICR describes the information we seek to collect from the public. Review and comments by OIRA ensure we only impose paperwork burdens commensurate with our performance of duties.

DATES:

Please submit comments on or before January 3, 2008.

ADDRESSES:

To make sure your comments and related material do not enter the Coast Guard docket [USCG-2007-29070] or are received by OIRA more than once, please submit them by only one of the following means:

(1) Electronic submission. (a) To Coast Guard docket at http://www.regulations.gov.

(b) To OIRA by e-mail to: [e-mail].

(2) Mail or Hand delivery. (a) To Docket Management Facility (M-30), U.S. Department of Transportation, West Building Ground Floor, Room W12-140, 1200 New Jersey Avenue, SE., Washington, DC 20590-0001. Hand deliver between the hours of 9 a.m. and 5 p.m., Monday through Friday, except Federal holidays. The telephone number is 202-366-9329.

(b) To OIRA, 725 17th Street, NW., Washington, DC 20503, to the attention of the Desk Officer for the Coast Guard.

(3) Fax. (a) To Docket Management Facility at 202-493-2251.

(b) To OIRA at 202-395-6566. To ensure your comments are received in time, mark the fax to the attention of Mr. Nathan Lesser, Desk officer for the Coast Guard.

The Docket Management Facility maintains the public docket for this notice. Comments and material received from the public, as well as documents mentioned in this notice as being available in the docket, will become part of this docket and will be available for inspection or copying at room W12-140 on the West Building Ground Floor, 1200 New Jersey Avenue, SE., Washington, DC, between 9 a.m. and 5 p.m., Monday through Friday, except Federal holidays. You may also find this docket on the Internet at http://www.regulations.gov.

A copy of the complete ICR is available through this docket on the Internet at http://www.regulations.gov. Additionally, copies are available from Commandant (CG-611), U.S. Coast Guard Headquarters, (Attn: Mr. Arthur Requina), 2100 2nd Street, SW., Washington, DC 20593-0001. The telephone number is [phone].

FOR FURTHER INFORMATION CONTACT:

Mr. Arthur Requina, Office of Information Management, telephone [phone] or fax [phone], for questions on these documents. Contact Ms. Renee V. Wright, Program Manager, Docket Operations, [phone], for questions on the docket.

SUPPLEMENTARY INFORMATION:

The Coast Guard invites comments on the proposed collection of information to determine if it is necessary in the proper performance of Departmental functions. In particular, the Coast Guard would appreciate comments addressing: (1) The practical utility of the collection; (2) the accuracy of the estimated burden of the collection;

(3) ways to enhance the quality, utility, and clarity of information subject to the collection; and (4) ways to minimize the burden of collection on respondents, including the use of automated collection techniques or other forms of information technology.

Comments to the FDMS or OIRA must contain the OMB Control Number of the ICR addressed. Comments must contain the docket number of this request, [USCG 2007-28578]. For your comments to OIRA to be considered, it is best if they are received on or before the January 3, 2008.

Public participation and request for comments: We encourage you to respond to this request by submitting comments and related materials. We will post all comments received, without change, to http://www.regulations.gov. They will include any personal information you provide. We have an agreement with DOT to use their Docket Management Facility. Please see the paragraph on DOT's “Privacy Act Policy” below.

Submitting comments: If you submit a comment, please include the docket number [USCG-2007-28578], indicate the specific section of the document to which each comment applies, providing a reason for each comment. We recommend you include your name, mailing address, and an e-mail address or other contact information in the body of your document to ensure you can be identified as the submitter. This also allows us to contact you in the event further information is needed or if there are questions. For example, if we cannot read your submission. You may submit your comments and material by electronic means, mail, fax, or delivery to the Docket Management Facility at the address under ADDRESSES; but please submit them by only one means. If you submit them by mail or delivery, submit them in an unbound format, no larger than 81/2 by 11 inches, suitable for copying and electronic filing. If you submit them by mail and would like to know that they reached the Facility, please enclose a stamped, self-addressed postcard or envelope. We will consider all comments and material received during the comment period. We may change the documents supporting this collection of information or even the underlying requirements in view of them.

Viewing comments and documents: Go to http://www.regulations.gov to view documents mentioned in this notice as being available in the docket. Click on “Search for Dockets,” and enter the docket number (USCG-2007-28578) in the Docket ID box, and click enter. You may also visit the Docket Management Facility in room W12-140 on the West Building Ground Floor, 1200 New Jersey Avenue, SE., Washington, DC, between 9 a.m. and 5 p.m., Monday through Friday, except Federal holidays.

Privacy Act: Anyone can search the electronic form of all comments received in dockets by the name of the individual submitting the comment (or signing the comment, if submitted on behalf of an association, business, labor union, etc.). You may review the Privacy Act Statement of DOT in the Federal Register published on April 11, 2000 (65 FR 19477), or you may visit http://DocketsInfo.dot.gov.

Previous Request for Comments

This request provides a 30-day comment period required by OIRA. The Coast Guard has published the 60-day notice (72 FR 38839, July 16, 2007) required by 44 U.S.C. 3506(c)(2). That notice elicited 12 comments.

The Coast Guard issued an OMB Information Collection supporting statement for its National Recreational Boating Survey for public comment on July 16, 2007. The proposed information collection activities are based on recommendations from a Scientific Advisory Committee (SAC) as well as a Collaboratory of Partners (COP), two groups that a grant recipient and the Coast Guard put in place to assist with the development of the National Recreational Boating Survey. The SAC was a group of methodologists whose role was to design the survey. The COP, on the other hand, was a collaboration involving groups such as various government agencies, boater associations, and the National Association of State Boating Law Administrators. The primary responsibility of the COP was to help Coast Guard define the content of its survey questionnaires.

We reviewed each of the comments received with diligence, and made some changes to our survey and its supporting statement where it was deemed appropriate. The present document provides a summary of public comments, our responses thereto, and changes made to the survey and its supporting statement.

1. General Supportive Comments

Several comments in support of the National Recreational Boating Survey indicated it has been substantially revised to reflect the need for more targeted data in response to the elements included in the National Recreational Boating Safety (RBS) Program's Strategic Plan, which calls for collection of participation/exposure data to develop reliable national/state-level measures of risk incidental to recreational boating. In fact, valid comparisons of injury or fatality rates across states or other geographic entities, which have always been of interest, require the use of participation/exposure data as a common base for calculating rates' denominators. This survey will make exposure data available to the boating community, in addition to collecting various other boating participation data broken down by boat type and length.

2. Weighting of Survey Data

One commenter, while supporting the proposed survey process and the idea of conducting it more frequently, indicated the suggested fixed number of 400 per state would not yield valid national estimates. The commenter's rationale is that the number of boats varies considerably per state, and some sort of data weighting is warranted. Another commenter pointed out the lack of discussion about weighting matters. We do not intend to obtain a fixed predetermined number of 400 respondents per state. Our intention is to obtain approximately 30,000 respondents from the mail survey of registered boat owners, and 20,000 respondents from the Random Digit Dialing (RDD) data collection targeting households that do not own a registered recreational vessel. Each of these surveys is based on stratified samples, with proportional allocation as described in the supporting statement for this survey. We agree with the commenter that the survey must be weighted to account for differential selection probabilities. We added an entire section in the supporting statement that provides a detailed description of the weighting process.

3. General Survey Design

One commenter expressed a concern that we did not adopt a rotating panel design for our National Recreational Boating Survey. The commenter stated the Coast Guard should justify its proposed continued use of an “antiquated” cross-sectional survey approach, which he feels will prevent the agency from obtaining useful and actionable data on net changes in how individuals alter their boating-related behaviors. Further, he opined that it will only allow for the estimation of gross flows (or changes).

We disagree with the commenter that cross-sectional surveys provide estimates of “gross” changes and not estimates of “net”. The cross-sectional surveys we are planning will provide estimates of “net” changes needed to observe trends, and not “gross” estimates. A “net” change represents, for example, the difference in overall boating participation levels between two years (years 1 and 2); while a “gross” change quantifies specific movements of year 1 boaters (e.g. those who stopped this activity in year 2). Consequently, obtaining “gross” change estimates requires tracking of individual level adjustments over time, which has traditionally been achieved with panel surveys. States may conduct local panel studies to further look into the “net” changes revealed by Coast Guard's National Recreational Boating Survey. The use of a rotating panel design is primarily justified if a key objective of the survey program is to provide reliable information on “gross” as well as “net” changes. That is not the case with the National Recreational Boating Survey. Nevertheless, we added a section in the supporting statement that discusses the issue of change estimation to provide a better justification of the proposed design.

4. Dual-Frame Issues

One commenter raised a concern about the use of the dual-frame approach, and how sample data collected by telephone will be compared to or combined with the mail survey data. The commenter would like us to explain the handling of the overlap between the two approaches and justify the use of two sampling frames.

In states that will provide boat registration data, we will implement a dual-frame survey with two separate components:

• The first component is a mail survey of households with a member who owns a registered recreational vessel.
• The second component is an RDD survey of boating households with no registered recreational vessel owner.

The mail survey using registration data is an effective way to collect the desired boating data with the possibility of targeting users of a particular type of watercraft. However, users of unregistered vessels constitute a significant portion of the boating population. Although some unregistered vessel users and owners are in households that also own registered vessels and are therefore included in the mail survey target population, a sizeable number are believed to reside without owning any registered recreational vessel. Since the mail survey does not cover households that do not own a registered vessel, an RDD household survey must be conducted to target them. The RDD sample will be screened, and a sufficiently large sample of boating households with no registered boat will be interviewed. It is a well known fact that the dual-frame approach can be highly efficient for surveying rare populations. For example, obtaining statistics on personal watercrafts could be difficult if one has to rely solely on a random national sample of households. Using the state boat registration data, one can target specific boats more effectively. As far as combining data from the mail and RDD surveys is concerned, we will weight the units of analysis from each component independently and obtain national/state level estimates by calculating the sums.

In states that will not provide boat registration data, the National Recreational Boating Survey will be based exclusively on an RDD sample; households, boats, and boaters will be weighted accordingly. National-level estimates will be obtained by summing all corresponding state-level estimates.

5. Mail Survey's Response Rates

A commenter indicated the projected response rate of 35 percent for the mail survey is unduly low and cannot be expected to yield valid estimates. He also stressed that some states will not provide any boat registration data to the Coast Guard, leading to a poor and incomplete sampling frame. Other concerns were also raised, ranging from not referencing Dr. Dillman's works on survey response rate improvement to failing to discuss standardization. For the 2002 National Recreational Boating Survey, the response rate of the mail survey was 49 percent, while that conducted telephonically was more than 61 percent. We anticipate higher response rates in 2007 due to a increased data collection budget, and a more systematic approach for converting non respondents. Our estimate of 35 percent represents the response rate with respect to the number of initial contacts, which include eligible as well as ineligible households. Survey response rates as defined by the American Association for Public Opinion Research (AAPOR) are calculated with respect to the number of eligible sample units. For the purpose of quantifying the response burden, we used a response rate with respect to the initial contacts (many of which are ineligible), and deliberately decided to adopt a conservative approach by minimizing our projections. When calculated with respect to the eligible sample size, the response rate will be higher. Based on past experience, we believe the proposed approach for reducing non-response will be effective. Concerning the standardization of studies, we believe some flexibility must be given to the data collection contractor implementation of specific protocols to improve survey response rates, and, not provide very detailed specifications to achieve this goal.

6. Survey Questionnaires

• A commenter suggested the tabularized format of some questions may lead different survey vendors to translate questions into different Computer-Assisted Telephone Interview (CATI) questions. When developing the survey questionnaires, our goal was not to write detailed specifications for a CATI programmer. Our primary objective was to provide questionnaires that are sufficiently clear for any CATI programmer to understand the exact nature of data items to be collected. Moreover, different CATI programmers may organize questions in different ways without it being problematic so long as the data item needed is properly collected.
• A commenter suggested the timeframe for collecting the data should be from October of the initial year to September of the following year, which will supposedly reduce the recall bias. We are not aware of any study which would support the commenter's statement.
• The commenter indicated the survey questionnaires are flawed based on the following issues:
• The absence of “Don't know” or “Refusal” options in the yes/no questions;
• The number of household members listed on the questionnaire; and
• The special order in which household members are listed.

We appreciate these comments and will work with selected contractors to address these concerns. The proposed survey questionnaire is not to be seen as a detailed specification memorandum to be sent by mail to a CATI programmer, but, should rather be considered as a document that will be explained and discussed with the data collection contractor.

Concerning question 5 of the screener questionnaire for states not sharing registration data, an answer (yes or no) is mandatory since that information is used to determine eligibility for the detailed survey. Therefore the “Don't know” option is unacceptable. The interviewer may need to talk to a more knowledgeable person if necessary. For those survey questions we can modify prior to selecting the contractor, we did so. Here are the changes:

• Concerning the collection of data on ethnicity, we have modified the questionnaires to comply with OMB standards.
• In the screener and detailed questionnaires, an adult is now defined as someone aged 16 or older. This modification was made following a comment by the same commenter.
• The number of home-use telephone numbers in the household is now collected.
• A commenter raised concerns about the pre-testing of the questionnaires. The National Recreational Boating Survey was last conducted in 2002, and many questions in the 2007 questionnaire were taken and thoroughly tested. The other questions in the 2007 questionnaire not used in the 2002 version were also used on several occasions by various boating researchers to collect subject data. The collection contractor is expected to conduct a limited pre-test to identify possible unforeseen problems.

7. Data Analysis

A commenter indicated that very little was said in the supporting statement about how the data collection contractor will analyze the data. In response to this comment, we expanded the data analysis section to show how national, state, and regional estimates will be calculated. However, the contractor will essentially provide the Coast Guard with basic contingency tables showing weighted counts describing various aspects of the boating population and their activities during 2007. We may conduct further analyzes internally after receiving the micro-data file.

Information Collection Request

Title: National Recreational Boating Survey.

OMB Control Number: 1625-0089.

Type of Request: Reinstatement, with change, of a previously approved collection for which approval has expired.

Affected Public: Recreational boating participants and owners of recreational vessels.

Abstract: The Federal Boat Safety Act of 1971 determined the framework of the Coast Guard RBS program. This program as set forth in 46 U.S.C., Chapter 131, requires the Coast Guard to “encourage greater state participation and uniformity in boating safety efforts, and particularly to permit the states to assume the greater share of boating safety education, assistance, and enforcement activities.” See 46 U.S.C. 13101. The Coast Guard Office of Boating Safety achieves these goals by providing timely and relevant information on activities that occur in each respective jurisdiction. The boating information provided by the Coast Guard enables each state agency to tailor and implement safety initiatives addressing specific needs of boaters in local jurisdictions. The primary objective of this collection is to provide the Coast Guard with the required information in a format suitable to effectively manage the program.

Burden Estimate: This is a biennial requirement. In the year the survey is conducted, the burden is estimated to be 67,619 hours.

Dated: November 26, 2007.

D.T. Glenn,

Rear Admiral, U.S. Coast Guard, Assistant Commandant for Command, Control, Communications, Computers and Information Technology.