Communitysurvey SS 081508 part B

SUPPORTING STATEMENT
West Coast Community Economic Data Collection
NOAA Fisheries - Northwest Fisheries Science Center
OMB CONTROL NO. 0648-xxxx

B. COLLECTIONS OF INFORMATION EMPLOYING STATISTICAL METHODS
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.
Data will be collected from a random sample of the owners and operators of businesses,
households, and visitors to 8 small fishing engaged communities on the West Coast. Table 1
provides population and commercial fish landings for all West Coast ports with commercial fish
landings in 2006. The data in Table 1 indicates that there were 41 small (population less than
10,000) fishing engaged communities on the West Coast in 2006.
Table 1 --- Population and Commercial Fish Landings for all West Coast
Fishing Engaged Communities

Region
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA

N CA
N CA
N CA
N CA

Port Name
ALBION
ALAMEDA
POINT ARENA
BERKELEY
BOLINAS
FORT BRAGG
CRESCENT CITY
EUREKA
FIELDS LANDING
OAKLAND
OTHER HUMBOLDT
COUNTY PORTS
OTHER MENDOCINO
COUNTY PORTS
OTHER S. F. BAY
AND SAN MATEO
COUNTY PORTS
OTHER SONOMA
AND MARIN
COUNTY OUTER
COAST PORTS
PRINCETON / HALF
MOON BAY
RICHMOND
POINT REYES

Population (2006)
5,000
70,699
473
101,555
1,246
6,785
4,006
25,435
5,000
397,067

Total Value of
Commercial Fish
Landings in 2006
$34,861.80
$28,134.92
$432,434.19
$55,716.24
$172,427.05
$5,326,336.88
$22,755,525.73
$11,662,259.10
$53,195.10
$19,773.84

NA

$84,126.80

NA

$5,835.08

NA

$228,457.74

NA

$61,607.61

12,308
102,120
5,000

$4,779,232.54
$11,955.25
$93,941.04

1

N CA
N CA
N CA
N CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA

S CA
S CA

S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
S CA
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR

SAN FRANCISCO
SAUSALITO
TOMALES BAY
TRINIDAD
AVILA
BODEGA BAY
SANTA CRUZ
DANA POINT
PORT HUENEME
LONG BEACH
MONTEREY
MOSS LANDING
MORRO BAY
NEWPORT BEACH
OTHER SANTA
BARBARA AND
VENTURA COUNTY
PORTS
OTHER OR
UNKNOWN
CALIFORNIA PORTS
OTHER SANTA
CRUZ AND
MONTEREY
COUNTY PORTS
OCEANSIDE
Other LA and Orange
Cnty Ports
OTHER SAN DIEGO
COUNTY PORTS
OTHER SAN LUIS
OBISPO COUNTY
PORTS
OXNARD
SANTA BARBARA
SAN DIEGO
SAN PEDRO
TERMINAL ISLAND
VENTURA
WILLMINGTON
ASTORIA
BANDON
BROOKINGS
CANNON BEACH
Charleston (Coos Bay)
PSUEDO PORT CODE
FOR COLUMBIA
RIVER
DEPOE BAY
FLORENCE
GOLD BEACH
GEARHART SEASIDE
NEWPORT
NEHALEM BAY
NETARTS BAY

2

744,041
7,207
5,000
314
5,000
1,423
54,778
35,945
21,814
472,494
28,803
300
9,998
70,032

$6,962,700.82
$31,026.08
$4,780.40
$3,074,629.96
$1,022,452.63
$5,453,483.26
$609,372.11
$1,547,747.92
$4,266,545.86
$562,317.00
$869,063.04
$4,876,692.76
$1,911,555.30
$724,598.06

NA

$27,089.26

NA

$65,693.46

NA
165,803

$35,264.49
$1,584,437.16

NA

$940,495.66

NA

$2,964,186.94

NA
184,463
85,681
1,256,951
100,000
100,000
106,000
50,000
9,917
2,901
6,344
1,720
15,999

$7,752.75
$2,927,576.59
$6,499,934.72
$2,565,696.46
$18,217,183.39
$10,880,334.54
$5,255,403.88
$148,047.76
$32,971,394.46
$11,047.00
$8,067,632.89
$19,025.25
$20,187,661.01

NA
1,361
8,122
1,907

$2,633,705.11
$146,646.25
$149,356.08
$316,666.54

1,106
9,896
208
744

$99,885.15
$33,014,185.19
$5,303.25
$3,304.80

OR
OR
OR
OR
OR
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA

WA
WA
WA
WA
WA
WA
WA
WA
WA
WA

PORT ORFORD
PACIFIC CITY
TILLAMOOK/GARIB
ALDI
WINCHESTER BAY
WALDPORT
ANACORTES
BELLINGHAM BAY
BLAINE
COPALIS BEACH
EVERETT
FRIDAY HARBOR
GRAYS HARBOR
LA CONNER
LA PUSH
ILWACO/CHINOOK
NEAH BAY
OTHER COLUMBIA
RIVER PORTS
OLYMPIA
OTHER NORTH
PUGET SOUND
PORTS
OTHER SOUTH
PUGET SOUND
PORTS
OTHER OR
UNKNOWN
WASHINGTON
PORTS
OTHER
WASHINGTION
COASTAL PORTS
PORT ANGELES
SEATTLE
SEQUIM
SHELTON
TACOMA
PORT TOWNSEND
WILLAPA BAY
WESTPORT

1,164
1,027

$3,155,756.49
$73,090.71

4,424
488
2,051
16,633
75,150
4,508
489
98,514
2,103
70,900
791
500
997
794

$4,120,818.90
$1,298,485.38
$65,409.85
$7,022,950.28
$25,249,191.93
$6,009,712.51
$2,129,393.03
$1,968,435.94
$624,210.65
$515,669.96
$2,687,221.71
$2,975,957.21
$19,787,492.06
$6,610,814.68

NA
44,645

$2,761,172.79
$10,679,761.57

NA

$2,061,058.97

NA

$10,675,507.53

NA

$339,380.01

NA
18,984
582,454
5,688
9,236
196,532
9,134
50,000
2,499

$6,942,789.90
$419,800.34
$9,391,682.60
$1,355,369.58
$24,139,614.45
$3,731,873.14
$3,078,973.90
$19,245,946.68
$27,710,594.39

The 8 communities surveyed in this project were selected from the population of 41 small fishing
engaged communities through the use of a stratified weighted random sampling method. Two
communities were selected from each of four strata (Washington, Oregon, Northern California,
and Southern California).
Each community’s probability of selection into the study was weighted by the percentage of the
total value of landings that are accounted for by the ports with populations fewer than 10,000
inhabitants. The probability of each port being selected with in each region was:
Pnr=ln/Lr
(1)
where P is the probability of selection, l is the total landings in each port n within the given region
r, and L is the total regional landings within region r. The total coast wide probability that any
given port was selected for inclusion was:
3

Pn=(ln/Lr)/k
(2)
where k is the number of regions (in this case 4).
Table 2 presents the total coast wide probability that any community will be selected for inclusion
in the study. The communities will be randomly selected for inclusion in the study based on these
probabilities.
Table 2 --- Probability of Selection for Small West Coast Fishing Engaged Communities
Region
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
N CA
S CA
S CA
S CA
S CA
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA

Port Name
ALBION
POINT ARENA
BOLINAS
FORT BRAGG
CRESCENT CITY
FIELDS LANDING
POINT REYES
SAUSALITO
TOMALES BAY
TRINIDAD
AVILA
BODEGA BAY
MOSS LANDING
MORRO BAY
ASTORIA
BANDON
BROOKINGS
CANNON BEACH
DEPOE BAY
FLORENCE
GOLD BEACH
GEARHART - SEASIDE
NEWPORT
NEHALEM BAY
NETARTS BAY
PORT ORFORD
PACIFIC CITY
TILLAMOOK/GARIBALDI
WINCHESTER BAY
WALDPORT
BLAINE
COPALIS BEACH
FRIDAY HARBOR
LA CONNER
LA PUSH
ILWACO/CHINOOK
NEAH BAY
SEQUIM
SHELTON
PORT TOWNSEND
WESTPORT

4

Overall (Coastwide)
Probability of Selection
0.000273
0.003381
0.001348
0.041639
0.177893
0.000416
0.000734
0.000243
0.000037
0.024036
0.019271
0.102786
0.091915
0.036029
0.098695
0.000033
0.024149
0.000057
0.000439
0.000447
0.000948
0.000299
0.098824
0.000016
0.000010
0.009446
0.000219
0.012335
0.003887
0.000196
0.015472
0.005482
0.001607
0.006918
0.007661
0.050941
0.017019
0.003489
0.062145
0.007927
0.071339

Data collection will involve in-person interviews and/or mail questionnaires sent to selected
members of each of the different survey groups. In many cases, individuals may receive the
questionnaire in advance to allow them to prepare their responses but may be interviewed via
telephone or in person to ensure the clarity of their responses. To the extent practicable, the data
collected will be that which the respondents maintain for their own business purposes. Therefore,
the collection burden will consist principally of transcribing data from their internal records to the
survey instrument and participating in personal interviews. In addition, current data reporting
requirements will be evaluated to determine if they can be modified to provide improved economic
data at a lower cost to the Agency and with reduced burden on potential respondents.
The eight communities selected with this methodology were Westport, Blaine, Newport,
Brookings, Crescent City, Fort Bragg, Bodega Bay, and Moss Landing. Table 3 provides
population, number of households, number of businesses, total employment, payroll, and
recreational visitors for each of these eight communities.
Table 3 --- Eight Communities Selected for West Coast Community Economic Survey
Zip Code - City

Population

Households

Businesses

Payroll

106

Employ
ment
1,357

$39,162,000

Recreational
Visitation
35,000

98595 Westport, WA
98230 - Blaine,
WA
97365 Newport, OR
97415 Brookings, OR
95531 - Crescent
City, CA
95437 – Fort
Bragg, CA
94923 – Bodega
Bay, CA
95039 –Moss
Landing, CA

2,856

1,347

4,508

1,818

377

3,313

$143,117,000

32,000

9,896

4,398

634

5,609

$134,103,000

64,220

6,344

2,758

480

4,293

$103,766,000

16,000

4.006

1,669

416

3,689

$89,233,000

20,000

6,785

2,887

535

4,203

$102,290,000

24,500

1,423

674

49

537

$12,687,000

70,000

300

125

47

672

$47,925,000

8,400

Data Source: Population figures are 2006 estimates prepared by each state, based upon 2000 Census values.
Household figures were obtained by taking the persons per household from the 2000 Census and applying the figure to
the 2006 population estimate to obtain an estimate of the number of households. Data on number of businesses,
employment, and payroll was obtained from the Census Bureau’s 2005 Zip Code Business Patterns. Visitation data is
estimated from data taken from Wen-Huei Chang and R. Scott Jackson, Economic Impacts of Recreation Activities at
Oregon Coastal and River Ports, ERD/EL TR-03-12, U.S. Army Corps of Engineers, August 2003.

The total sample universes for businesses and households are the total numbers of each in each of
eight small fishing engaged communities. Total number of households have been determined from
U.S. Census records and addresses were obtained from public records searches. The total number
of businesses by ZIP code and by 2-digit North American Industry Classification System (NAICS)
was obtained from the U.S. Economic Census and from County Business Patterns. The sample
universe of recreational visitors is estimated from a study of visitors to Oregon ports done by the
U.S. Army Corps of Engineers (Wen-Huei Chang and R. Scott Jackson, Economic Impacts of
5

Recreation Activities at Oregon Coastal and River Ports, ERD/EL TR-03-12, U.S. Army Corps of
Engineers, August 2003).
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.
Households and business
Households and businesses within each of the selected communities will be randomly selected for
inclusion in the study. Names, address, and telephone numbers for businesses and households will
be obtained from local government records and from public record searches. The formula for
calculating the sample size for a simple random sample without replacement is as follows:
2

⎛ zσ ⎞ ⎛ z ⎞
n=⎜
⎟ =⎜ ⎟
⎝ E ⎠ ⎝m⎠

2

where,

z is the z value (e.g., 1.645 for 90% confidence level, 1.96 for 95% confidence
level, and 2.575 for 99% confidence level);
σ is the standard deviation of the population;
E is the acceptable bound on the error or the “margin of error”

m is the margin of error expressed as a proportion of the standard deviation (e.g.,
.05 = + or – 5%, .07 = + or – 7%, and .1 = + or – 10%);
For the purposes of this study, we are using a 95% confidence level and a allowable error of +/10%.
The Finite Population Correction (FPC) factor is routinely used in calculating sample sizes for
simple random samples. In fact, many sample size formulas for simple random samples include the
FPC as part of the formula. It has very little effect on the sample size when the sample is small
relative to the population but it is important to apply the FPC when the sample is large (10% or
more) relative to the population. The sample size equation solving for n ' (new sample size) when
taking the FPC into account is:

n' =

n
1+

n
N
6

where,
n is the sample size based on the calculations above, and
N is population size.
The n’ estimate of sample size will then be multiplied by the estimated response rate to obtain the
actual number of surveys that will need to be mailed out.
Table 4 provides the number of households, the household sample size n’ calculated using the FPC
factor, the expected response rate, and the corresponding number of expected respondents to the
household survey in each community. Table 5 provides the number of business establishments,
the business establishment sample size n’ calculated using the FPC factor, the expected business
response rate, and the corresponding number of expected respondents to the business survey in
each community.
Table 4 --- Household Survey Sample Size, Response Rate, and Respondents
Zip Code - City
98595 - Westport, WA
98230 - Blaine, WA
97365 - Newport, OR
97415 - Brookings, OR
95531 - Crescent City, CA
95437 – Fort Bragg, CA
94923 - Bodega Bay, CA
95039 – Moss Landing, CA
TOTAL

Number of
Households

Household
Sample
Size
299
317
353
337
312
339
245
84
2,297

1,347
1,818
4,398
2,758
1,669
2,887
674
125
15,676

7

Household
Response
Rate
.6
.6
.6
.6
.6
.6
.6
.6
.6

Number of
Responses
179
190
212
202
187
203
147
57
1,378

Table 5 --- Business Survey Sample Size, Response Rate, and Respondents
Zip Code - City
98595 - Westport, WA
98230 - Blaine, WA
97365 - Newport, OR
97415 - Brookings, OR
95531 - Crescent City, CA
95437 – Fort Bragg, CA
94923 - Bodega Bay, CA
95039 – Moss Landing, CA
TOTAL

Number of
Business
Establishments
106
377
634
480
416
535
49
47
2.644

Business
Sample
Size
83
190
239
213
200
224
43
42
.1235

Business
Response
Rate
.7
.7
.7
.7
.7
.7
.7
.7
1,201

Number of
Responses
58
133
167
149
140
157
30
29
864

Visitors

Estimates of the total number of recreational visitors will be determined by collecting data on total
visitor occupancy in local hotels and then surveying respondents at numerous locations and times
throughout the city to determine the ratio of visitors staying in hotels and those not staying in
hotels. The total number of visitors (N) can then be determined by the following calculation:

TS
HS
Where HT is the total number of visitors staying in hotels, TS is the total number of visitors
surveyed, and HS is the number of visitors surveyed that stayed in hotels. This method uses two
pieces of information --- the number of visitors staying in hotels and the percentage of visitors
staying in hotels --- to estimate the total number of visitors. The total number of visitors staying in
hotels will be determined from locally available hotel occupancy rates and by surveying hotel
guests (to determine the number of visitors per occupied hotel room). The percentage of visitors
staying in hotels will be determined from the visitor survey. It is important that the sample for the
visitor survey be representative of the visitor population in terms of the percentage of visitors
staying in hotels. As a result, the visitor survey will be fielded in each community at multiple
locations and at multiple times of the day and days of the week.
N = HT *

The initial questionnaire for visitors contains only four short questions which are estimated to lake
less than a minute to answer in total. If the respondent is willing the surveyor would ask the
individual the four questions. The respondent would then be asked if they would be willing to
answer an additional longer 15 minute questionnaire in exchange for a token gift (NOAA Fisheries
tee shirt). If the respondent is not willing they will be asked if they would take the questionnaire
home and complete it at their leisure, then return it in a prepaid envelope that is provided. If they
are not willing to do this, we thank them for their time and wish them a pleasant day. Impartiality
in selection for interviewing is stressed in interviewer training.
Table 6 provides the estimated number of visitors, the visitor sample size, the expected visitor
response rate to the initial short questionnaire, the number of short survey respondents, and the
number of longer questionnaire respondents for each community. Using the same sample size
calculation from above, the total number of visitor interviews needed is as follows (potential
8

universe size is estimated from Wen-Huei Chang and R. Scott Jackson, Economic Impacts of
Recreation Activities at Oregon Coastal and River Ports, ERD/EL TR-03-12, U.S. Army Corps of
Engineers, August 2003). The response rate for the longer survey (not shown in the table) is
assumed to be the same 60% as the response rate for the initial short questionnaire. That is, the
estimates in Table 6 assume that 60% of the visitors contacted will complete the short
questionnaire, and that of those visitors completing the short questionnaire, 60% will complete the
longer follow-up questionnaire.
Table 6 --- Visitor Survey Sample Size, Response Rates, and Respondents
Zip Code - City

Annual
Recreational
Visitors

Visitor
Sample
Size

Visitor
Response
Rate

Initial Visitor
Questionnaire
Responses

98595 - Westport, WA
98230 - Blaine, WA
97365 - Newport, OR
97415 - Brookings, OR
95531 - Crescent City, CA
95437 – Fort Bragg, CA
94923 - Bodega Bay, CA
95039 – Moss Landing, CA
TOTAL

35,000
32,000
64,220
16,000
20,000
24,500
70,000
8,400
270,120

380
380
382
375
377
378
382
367
3,021

.6
.6
.6
.6
.6
.6
.6
.6
.6

228
228
229
225
226
227
229
220
1,813

Longer
Visitor
Questionnaire
Responses
137
137
137
135
136
136
138
132
1,088

Expected Response Rates:
Based on previous studies of households and businesses, a response rate of about 60% for
households and 70% for businesses is expected. These response rates are consistent with those
reported in Dillman (1974), Dillman (2007), and Fox et al. (1988). For visitors, it is expected that
60% of the people contacted will be willing to answer the short four question survey. It is then
expected that 60% of the people who answer the initial questionnaire will respond to the longer
survey. These are similar to response rates that the USDA Forest Service (2002) received with
their National Visitor Use Monitoring (NVUM) study. Additionally, the aforementioned WenHuei Chang and R. Scott Jackson study also received a 60% response rate for visitors to Oregon
ports.
Additionally, adherence to the Dillman method, the use of social exchange, and garnered support
from local officials and business leaders will ensure high response rates.
3. Describe the methods used to maximize response rates and to deal with nonresponse. 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.

Cooperation from industry representatives has been garnered as well as support of government
officials, commercial leaders, and the local population. A “social exchange” framework was
utilized to emphasize the potential benefits of responding (greater understanding of the local
economy and how to foster desired levels of economic growth) and to reduce the potential time
cost to the boat owners. Social exchange is mentioned by Dillman (2007) as a crucial component
of any social research survey and is intended to highlight the benefits of responding to the survey
9

while stating how the survey has been designed to reduce the time and effort costs to the
respondents.
A modified Dillman Tailored Design Method (Dillman 2007) will be employed to for the
household survey and the business survey. Personalizing correspondence, a respondent friendly
questionnaire, multiple contacts with survey participants through multiple modes, and a stamped
return envelope will be utilized to increase response rates. The business survey and the household
survey will utilize the following protocols:
1. Mailing of an information letter three to five days prior to the mailing of the survey. This
letter describes the kind of information that the survey will ask, describes how the
information will be used, and highlights the benefits of the survey to the respondent.
Correspondence will be personalized wherever possible. The household survey
correspondence will be addressed to the head of household. The business survey
correspondence will be directed (where appropriate) to the business owner. In cases where
the business owner is deemed unlikely to be at the local mailing address (such as a large
national chain store), the letter will be sent to the store manager rather than a specific
individual.
2. Three to five days after the information letter is mailed, the actual survey instrument will
be mailed with a detailed cover letter explaining the purpose of the study, the survey
population, and the expected benefits.
3. Two weeks after the survey is mailed, a thank you/reminder post card is mailed
4. Two weeks after the post card is mailed, a replacement survey and cover letter will be
mailed to nonrespondents
5. Two weeks after the replacement surveys are mailed, calls will be made to nonrespondents.
Nonrespondents to the household survey will be asked if 1) they have received the survey,
2) whether the survey was sent to the correct person in the household, and 3) if they need
help in completing the survey. Up to a maximum of five attempts (made at different times
of the day on different days of the week) will be made to contact household survey nonrespondents. Messages will be left only on odd numbered attempts. Nonrespondents to the
business survey will be asked if 1) they have received the survey, 2) whether the survey has
been sent to the correct contact person, and 3) if they need any help in completing the
survey. If the survey was not initially sent to the correct contact person, information on the
correct contact person will be collected and survey materials will be mailed directly to that
person. While only five attempts will be made to contact household survey nonrespondents when no answer is obtained, more than five calls may be made to business
survey recipients in cases where improved contact information is obtained.
To reduce the possibility of unit non-response bias, a chi square test for structural differences will
be employed to ensure that non-respondents from the survey of businesses are not systematically
different from the population as a whole in known attributes such as business size (as measured by
number of employees) and business type (as measured by NAICS code). A similar analysis will be
performed on households to ensure that respondents are not systematically different from nonrespondents in known attributes such as household size and income stratification.
Sample post-stratification methods will then be used to generate weighting classes if structural
differences are found.
10

For the visitor survey, a token gift will be offered to respondents willing to fill out the 15 minute
survey. The token gift will be a tee shirt designed for this project, the total value not exceeding $5.
Data collection will begin approximately two months after OMB approval is received. If approval
is received by September 1, 2008, data collection will begin in November 2008. Data collection
will be completed in all eight communities by the end of summer 2009.
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.

None
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.

Carl Lian, Ph.D.
Economist
NOAA Fisheries
206-302-2414
Philip Watson, Ph.D.
Economist
University of Idaho
208-885-6934
Don English, Ph.D.
Economist
US Forest Service
202-205-9595
Eric White, Ph.D.
Economist
US Forest Service
541-750-7422.

11