SEPS ss new part B

B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS
(If your collection does not employ statistical methods, just say that and delete the following five
questions from the format.)
1. Describe (including a numerical estimate) the potential respondent universe and any
sampling or other respondent selection method to be used. Data on the number of entities
(e.g. establishments, State and local governmental units, households, or persons) in the
universe and the corresponding sample are to be provided in tabular form. The tabulation
must also include expected response rates for the collection as a whole. If the collection has
been conducted before, provide the actual response rate achieved.
The absence of federal fishing licenses in the U.S. Caribbean requires the use of trip ticket and
fishermen census databases to build sampling frames.1 The sampling frame was developed by
merging Puerto Rico’s DNER and U.S. Virgin Islands F&W trip ticket databases with the Puerto
Rico’s and U.S. Virgin Islands’ fishermen census databases, respectively. The trip ticket
database provided us a list of active fishermen (i.e., fishermen actively landing fish) while the
censuses supplied us with their most current addresses.
Because of Puerto Rican fishermen animosity towards the newly instituted Commonwealth
regulations, there has been a pointed decrease in the number of the trip ticket forms returned in
recent years. Therefore, we decided to include trip ticket data from 2001-2004 in the design of
the sampling framework. Our intention is minimize any bias caused by the low reporting
compliance with the industry reporting requirements (i.e., capture trip ticket non-respondents).
Unlike in Puerto Rico, trip ticket reporting has remained constant in the U.S. Virgin Islands
during the last years. For consistency, we also used 2001-2004 data in the U.S. Virgin Islands.
The sampling design requires that stratified random sample of 1,000 fishermen. We plan to
survey 700 fishermen in Puerto Rico and 300 fishermen in the U.S. Virgin Islands. Gear type
will be used to stratify the sample. We weighed the number of sub-gears to be sampled by their
contribution to total landings.2 For instance, if bottom line contributed 40% of the total landings
then we weighted the sample appropriately.
Tables 2 and 3 summarize the number of fishermen who participate in the hook and line, net and
dive fisheries in Puerto Rico and U.S. Virgin Islands, respectively. Note that the same fisherman
may use multiple gears during a given year. Agar et al (2005) reported an average response rate
of 0.8 in a recent study (see, Agar, J. J., M. Shivlani, J. R. Waters, M. Valdés-Pizzini, T. Murray,
J. Kirkley and D. Suman, 2005. U.S. Caribbean Fish Trap Fishery Costs and Earnings Study.
NOAA Technical Memorandum NMFS-SEFSC- 534, 127 p.)

1

The only exception is the HMS permit, which is required for those vessels harvesting tunas, swordfish, and sharks
in the Atlantic Ocean, including Gulf of Mexico and Caribbean waters.
2
For instance, hook and line gear is made up of four sub-gears: bottom line, longline, rod and reel, and trolling.

Table 2: Sampling design for gear operating in the Commonwealth of Puerto Rico.
Geographical
Area
Puerto Rico

Gear type

Population
Size

Bottom line
Longline
Trolling
Rod and line
Beach Seine
Trammel
Gillnet
Cast net
Scuba diving
Skin diving

Survey
Sample

2550
277
918
61
201
222
1158
352
1317
321

Expected
Response Rate
325
18
90
26
26
20
139
21
188
23

0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8

Total

Target
Number of
Surveys per
Strata
260
14
72
21
21
16
111
17
150
18
700

Table 3: Sampling design for gear operating in the U.S. Virgin Islands.
Population
Size

Target
Number of
Surveys per
Strata

Expected
Response
Rate

Survey
Sample

Geographical Area

Gear type

St. Thomas/St. John

Gillnet
Cast net
Seine net
Trammel
Line
Longline
Scuba diving
Skin diving

3
16
28
0
128
0
20
23

3
16
26
0
101
0
16
18

0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8

3
13
21
0
82
0
13
14

St. Croix

Gillnet
Cast net
Seine net
Trammel
Line
Longline
Scuba diving
Skin diving

44
8
19
1
169
2
96
34

26
10
21
1
63
0
84
31

0.8
0.8
0.8
0.8
0.8
0.8
0.8
0.8

21
8
17
1
50
0
67
25

Total

300

2

2. Describe the procedures for the collection, including: the statistical methodology for
stratification and sample selection; the estimation procedure; the degree of accuracy
needed for the purpose described in the justification; any unusual problems requiring
specialized sampling procedures; and any use of periodic (less frequent than annual) data
collection cycles to reduce burden.
One time, voluntary surveys will be used to elicit costs and earnings information. The method of
sampling will be stratified random. The survey will encompass different gear and sub-gears (e.g.,
trolling and longline belong to the hook and line gear category) and geographical strata (Puerto
Rico, St. Thomas & St. John, and St. Croix). The stratification is needed because 1) the fleet’s
economic performance is likely to vary by gear used (e.g., scuba diving is more efficient than
skin diving) and geographically due to the spatial and temporal availability of various reef fish
species, 2) the proposed regulations could inadvertently marginalize or exclude some of the user
groups.
To minimize the burden on fishermen, a list containing a random sample of fishermen for each
sub-gear and area will be provided to the selected contractor. The list will contain the following
information: fisherman name, address, and phone number. A list in excess of 1,000 names will
be provided to take into account the possibility that some fishermen will decline to participate in
this voluntary survey. Should a fisherman decline to participate in the survey, the contractor
could then select an additional fisherman from the list until the survey goal for the given stratum
is reached.
The data collected will be used for descriptive and analytical purposes. Descriptive uses include
the estimation of average harvesting costs per trip and total harvesting costs for the fleet. The
procedures for estimating harvesting costs in the sampling universe will be based on the standard
equations available in various statistical texts such as Thompson (1992).3 For a description of
analytical purposes the reader is directed to section A.2.
3. Describe the methods used to maximize response rates and to deal with non-response.
The accuracy and reliability of the information collected must be shown to be adequate for
the intended uses. For collections based on sampling, a special justification must be
provided if they will not yield "reliable" data that can be generalized to the universe
studied.
Several steps will be taken to maximize the response rates. First, the contractor will be selected
for survey experience and familiarity with local fishing communities and practices. Second,
trained interviewers will conduct in-person surveys at times and places that are convenient to
fishermen. This will minimize any potential disruption to fishermen’s fishing practices. Third,
the contractor will work with local authorities to ensure that the wording facilitates
understanding and reflects local idioms. Last, surveys will be available in English and Spanish to
further reduce any burden to non-English speaking fishermen. To deal with non-response we will
use call-backs and two-phase sampling procedures as described in textbooks such as Lohr’s.
(See, Lohr’s, S., 1998. Sampling: design and analysis). The sample size of 1,000 will provide
reliable estimates of the cost structure of the industry.
3

Thompson, Steven K., 1992. Sampling. John Wiley and Sons, Inc., New York, 343 p.

3

4. Describe any tests of procedures or methods to be undertaken. Tests are encouraged as
effective means to refine collections, but if ten or more test respondents are involved OMB
must give prior approval.
We pre-tested the survey with 5 fishermen to make instrument easier to understand and
complete.
5. Provide the name and telephone number of individuals consulted on the statistical
aspects of the design, and the name of the agency unit, contractor(s), grantee(s), or other
person(s) who will actually collect and/or analyze the information for the agency.
Drs. Jim Waters and Juan Agar, social scientists employed by the NMFS, were consulted on the
statistical design.4 NMFS social scientists and CFMC staff will use the data for regulatory
analysis.

4

Drs. Jim Waters and Juan J. Agar, Southeast Fisheries Science Center, National Marine Fisheries Service and be
reached at 252-728-8710 and 305-361-4218, respectively.

4