Methodology Responses

Key issues for the Coast Guard to address on the NRBS:

OMB Comment:

2. For the boating survey, Coast Guard and ICF Macro should make every attempt to obtain administrative records from the States on registered boat owners. It was noted that by Coast Guard in B.1.1.1.1 that “Most states make their registration databases available to the public.” Clearly, this information should be obtained if it is publicly available. If not, Coast Guard or ICF Macro should attempt to enter into agreements with States to obtain the information for use in surveying boat owners and providing important policy information at the State-level. Such statistical use of administrative records is widely carried out by Federal agencies, and is compatible with the purpose for which the information was originally obtained. Furthermore, OMB recently issued Memorandum M-11-02 (http://www.whitehouse.gov/sites/default/files/omb/memoranda/2011/m11-02.pdf), which encourages greater use of administrative records for statistical purposes. Previous surveys have obtained these registration lists from the vast majority of states. With this information ICF Macro can design and draw samples, mail surveys to sampled persons, conduct nonresponse follow-up, and use information from the records to examine potential nonresponse bias, either directly or indirectly by matching to other external datasets.

OMB considers the revised plan to ask states to make first contact with potential respondents and request that they “opt-in” to do a survey before they can be contacted by ICF Macro to be a major change in methodology that will severely negatively impact the quality of the information that can be obtained. There is no quality assurance for States to draw appropriate samples or appropriately calculate response rates, and achieved response rates for this “opt-in” approach will be considerably lower than would be achieved through direct survey mailings by ICF Macro. Furthermore, no information will be available about nonrespondents to enable nonresponse bias analyses. Thus, these revised plans are not approvable as currently outlined and justified.

The Coast Guard and ICF Macro should revisit this approach and propose to either return to the original methodology or propose an alternative method that is sufficiently rigorous that it will obtain information of appropriate quality and meet OMB standards.

Coast Guard’s Answer:

The Coast Guard has not fundamentally changed its original methodology. Its contractor ICF Macro will acquire the data that states will make available to the public for the purpose of designing the sample, drawing samples, and mailing survey questionnaires to recreational boaters selected in the sample. Follow-up telephone calls to non-respondents as well as the nonresponse analysis will be carried out by ICF Macro. However, prior to starting the mail survey, the Coast Guard will obtain proper documentation issued by the states, authorizing the use of registration data to contact recreational boaters selected to participate in the National Recreational Boating Survey.