SUPPORTING STATEMENT
SOCIOECONOMICS OF CORAL REEF CONSERVATION
OMB CONTROL NO. 0648-xxxx
U.S. Department of Commerce
National Oceanic and Atmospheric Administration
Coastal Services Center
Coral Reef Conservation Program
The potential respondent universe for this study is adults, eighteen years or older, who live near, and may use, coral reefs affected by activities related to the NOAA’s Coral Reef Conservation Program. The total population (all individuals) of the potentially impacted area is 11,244,759. Respondents will be classified into seven geographical jurisdictions and 22 reporting units, as defined in Table 5. In American Samoa, face-to-face interviews will be conducted and in the remaining jurisdictions, a combination of internet and phone based interview will occur. Each of the geographical jurisdictions is expected to be surveyed once every three to four years. Respondents will be randomly selected from the target audiences. Based on previous NOAA surveys of the target populations, we anticipate that the response rate will be 50 percent for mail surveys, 40% for internet surveys and 80 percent for in-person surveys.1 Dillman et al. (2009) considers a response rate above 50% a high response rate for mail surveys.2
Table 5: Study Jurisdictions and Reporting Units
Jurisdiction |
Reporting units |
Population |
|
Puerto Rico |
|
|
3,725,7893
|
Florida
|
|
|
5,784,0434 |
U.S. Virgin Islands |
|
|
106,4055 |
Guam |
|
159,3586 |
|
American Samoa
|
|
|
55,0707 |
Main Hawaiian Islands |
|
Maui County |
1,360,2118 |
Commonwealth of the Northern Mariana Islands (CNMI) |
|
Saian Municipality |
53,8839 |
Total |
11,244,759 |
||
For each of the jurisdictional populations, we intend to select a random sample of individuals over the age of eighteen, stratified geographically as described in Table 6. The random sample will be obtained from the selected survey firm using standard sample selection tools. The sample frame will be developed from telephone directories, mailing lists obtained and maintained by the survey firms and other sources as needed, depending onr the coverage of these sources. These strata have been designed to account for the differing sizes of the populations in the areas close
to coral reefs. We have used the standard approach to estimating sample size for a stratified population:
[t2 N p(1-p)] / [t2 p(1-p) + a2 (N-1)]
Where N is the size of the total number of cases, n is the sample size, a is the expected error, t is the value taken from the t distribution corresponding to a certain confidence interval, and p is the probability of an event. The final sampling size will be based on available resources.
Table 6: Sampling Requirements by Geographical Jurisdictions
Jurisdiction |
Total Sample |
Sample Size by Strata |
||
1. American Samoa |
550 |
400 |
Tutuila |
Island |
50 |
Ta’u |
County |
||
50 |
Olosega |
County |
||
50 |
Ofu |
County |
||
2. CNMI |
500 |
400 |
Saipan |
Municipality |
50 |
Tinian |
Municipality |
||
50 |
Rota |
Municipality |
||
3. Guam |
1,000 |
100 |
Agat |
Municipality |
100 |
Piti |
Municipality |
||
100 |
Asana |
Municipality |
||
100 |
Talafolo |
Municipality |
||
100 |
Merizo |
Municipality |
||
250 |
Tamuning |
Municipality |
||
250 |
Mangilao |
Municipality |
||
4. Hawaii |
1,500 |
350 |
Hawaii |
County |
750 |
Honolulu |
County |
||
100 |
Honolulu |
County |
||
300 |
Maui |
County |
||
5. Florida |
4,000 |
250 |
Monroe |
County |
1,250 |
Miami-Dade |
County |
||
500 |
Martin |
County |
||
1,000 |
Broward |
County |
||
1,000 |
Palm Beach |
County |
||
6. Puerto Rico |
2,500 |
1,000 |
South & West |
PR |
1,000 |
North & East |
PR |
||
250 |
Western Islands |
PR |
||
250 |
Eastern Islands |
PR |
||
7. US Virgin Islands |
750 |
350 |
St. Croix |
Island |
350 |
St. Thomas |
Island |
||
50 |
St. John |
Island |
||
Total |
10,800 |
|||
In addition to asking the questions regarding the impact of the Coral Reef Conservation Programs, the surveys will collect information on socioeconomics and demographics. This additional information will be used to sort and categorize the survey results in order to control for as many variables as possible. This approach will ensure a large enough respondent pool (particularly in more populated jurisdictions) to make comparisons between strata where required.
In
each of the jurisdictions, we intend to hire local surveying
contractors with databases of contact information in order to allow
for the greatest possible randomization of survey participants. NOAA
will also work with these contractors to select the most
cost-effective survey methodology which will resonate with the
population measured. Survey participants will be selected in
American Samoa for face-to-face interviews due to the very low
incidence of either cellular phones or land lines; in other
locations, local opinion poll contractors will select participants at
random using a combination of internet and telephone polling. The
methodology to be employed by jurisdiction can be found in Table 7.
Table 7: Survey Methodology by Geographical Jurisdictions
Jurisdiction |
Geographic scope |
Estimated Response Rate (based on previous NOAA surveys) |
1. American Samoa |
Face-to-face (and mail as back-up) |
50-80% |
2. CNMI |
Telephone |
50-80% |
3. Guam |
Telephone |
50-80% |
4. Hawaii |
Telephone or Internet |
50-80% |
5. Florida |
Telephone or Internet |
50-80% |
6. Puerto Rico |
Telephone or Internet |
50-80% |
7. US Virgin Islands |
Telephone or Face-to-face |
50-80% |
We do not intend to compare survey results between jurisdictions (though comparisons between the larger regional strata are possible), so there is no concern about comparability issues between methodologies.
As can be seen from Table 7, we have selected a number of different methods to collect data from different jurisdictions. Table 8 highlights the percent of population classified as internet users for the seven jurisdictions. In general we will attempt to collect data using a mixture of internet and telephone methods. The one exception is American Samoa where an in-person household survey backed-up by mail surveys will be conducted due to the extremely low level of internet usage in this jurisdiction (i.e., approximately 6 percent). In addition, the average internet use in CNMI, Guam, Puerto Rico and the US Virgin Islands is 39 percent as compared to 79 percent for Hawaii and Florida. As a result we will support the internet survey in these jurisdictions with lower internet usage with a telephone survey, or in the case of USVI with face-to-face interviews, to capture non-internet users.
Table 8: Internet Usage in Survey Jurisdictions
Jurisdiction |
Population |
Percent of Population Classified as Internet Users |
1. American Samoa |
55,070 |
6% |
2. CNMI |
53,883 |
30% |
3. Guam |
159,358 |
56% |
4. Hawaii |
1,360,211 |
79% |
5. Florida |
5,784,043 |
80% |
6. Puerto Rico |
3,725,789 |
40% |
7. US Virgin Islands |
106,405 |
28% |
Source: Data from Hawaii and Florida US Census 2010. Other data “Internet World Statistics”, American Samoa data March 2011, CNMI data from August 2010, Guam data from June 2010, Puerto Rico data from, June 2011, and US Virgin Islands from December 2002 (see http://www.internetworldstats.com/).
We expect that there will be some language issues. Table 9 shows there are several major languages spoken beyond English by the populations of each jurisdiction.
Table 9: Languages Spoken in Survey Jurisdictions
Jurisdiction |
Major Languages Spoken |
1. American Samoa |
English, Samoan |
2. CNMI |
English, Chamorro, Carolinian, Tagalog, Chinese, Korean, Japanese |
3. Guam |
English,Chamorro, Tagalog, Chinese, Korean, Japanese |
4. Hawaii |
English, Hawaiian pigin |
5. Florida |
English, Spanish |
6. Puerto Rico |
English, Spanish |
7. US Virgin Islands |
English, Negerhollands, Virgin Islands Creole |
This language issue will be ameliorated by the use of polling specialists who speak the local language. These contractors will also be used to ensure that the questions posed in the survey are translated into the proper cultural contexts. Responses will be tracked to see if there are statistically significant differences in the survey results between those who speak English at home and those who do not. In addition, mail and internet surveys will be translated into local languages.
We also expect that there is some risk of sample selection bias towards those of higher incomes, particularly for the telephone and internet surveys. In areas where access to phone and internet services are not widely available, this bias may be more than minimal. To the greatest extent possible, we hope that this can be corrected through the use of telephone surveys. If responses appear to favor high-income groups we will use various weighting procedures in the post-survey analysis to adjust for bias. Specifically, we will overweight the underrepresented groups if expected responses are not obtained. We will identify ‘control totals’ for the population that the survey is aiming to reach and calculate weights to adjust the sample totals to the control totals. For example, suppose the distribution of income groups in the population is as follows:
Income |
Percent of Population |
$0-$25K |
28.00 |
$25K-$50K |
27.00 |
$50-$75K |
18.00 |
$75K-$100K |
11.00 |
$100K plus |
16.00 |
Total |
100.00 |
However, the response distribution is:
Income |
Percent of Population |
$0-$25K |
5.00 |
$25K-$50K |
27.00 |
$50-$75K |
18.00 |
$75K-$100K |
11.00 |
$100K plus |
39.00 |
Total |
100.00 |
In response to this disparity, we would weigh the sample for the “$0-$25K” and the “$100K plus” groups to bring it in to line with the proportion of that group in the population as a whole. That is, we would apply a weight of 5.6 to the results for the “$0-25K” group (i.e., 28 percent divided by 5 percent) and a weight of 0.41 to “$100K plus” group to deal with their overrepresentation.
This survey will be conducted every three to four years to minimize the cost burden.
While the surveys conducted in person are expected to yield standard rates of response (80% based on previous NOAA surveys of targeted population),10 there is some concern about the potential for non-response in the telephone/internet surveys. While response rates for many surveys have been declining in the United States for years, previous studies have indicated that the low response rates commonly associated with internet polling can be somewhat improved with the use of pre-poll telephone calls. To accomplish this, polling representatives ask respondents whether they are willing to participate in the online study and then direct them via a secure link or email. In addition we will conduct extensive online advertising to encourage response. Research has shown that under these conditions internet and telephone surveys can reach similar response rates as those found in mail surveys.11
A variety of techniques have been incorporated into this study to maximize response rates. The surveys are user-friendly, with clear, easy to comprehend questions. Each questionnaire is short and can be completed in a short period of time (see Part A). The survey topic and related questions were developed to be interesting to respondents. Each survey makes use of listing options to allow the respondent to answer questions by checking appropriate boxes, which may aid in recall and analysis.
In person surveys will be conducted at respondent’s homes and participants will be given the opportunity to receive and /or return the survey by mail if they are unable to complete the surveys at the time of interview. These individuals who complete the survey by mail will receive a pre-addressed stamped return envelope.
The implementation of the mail surveys is based on the Dillman Tailored Design Method.12 This approach includes multiple steps and points of contact. The initial mailing will include the questionnaire, a pre-addressed stamped envelope and a detailed cover letter. The cover letter will explain the project, why a response is important, a statement indicating that all personal information will be kept confidential, and instructions for completing and returning the completed survey (via mail/fax/email). Addresses on envelopes will be handwritten, and colored envelopes will be used to make them stand out. Surveys will be tracked using individual identification numbers. A follow-up thank you postcard will be sent seven to nine days after the questionnaire. The postcard will express appreciation for participating and will indicate that if the completed questionnaire has not yet been mailed, it is hoped that it will be returned soon. Three weeks after the initial mailing, a second mailing will be sent to all who have not returned the survey. This follow-up will consist of a different cover letter, another copy of the questionnaire, and another pre-addressed stamped envelope.
For internet surveys we will use a number of techniques13 to increase response including:
Subject lines on contact emails will clearly indicate the purpose of the survey and will explicitly avoid SPAM language in the subject line or body of the message (I.e. title all caps)
Information on how the respondents name was obtained, the survey intention, the use of the data, guarantees of anonymity
Personalized messages
Use of a .gov reply email address
Indication of how long the survey takes to complete and the cutoff date.
Use of only clean and updated email lists
Scheduled regular reminders and follow-ups.
Cross-cultural research faces additional methodological challenges that, if not properly addressed, may considerably increase the risk of inferential errors during the administration of surveys.14 Specifically, concepts may entail culture-specific attributes and meanings which need to be explicitly taken into account to ensure sound interpretation of cross-cultural data.15 As discussed above (see Question 2), we will address this cross cultural issue by using polling specialists who speak the local language to conduct in-person and phone surveys. These polling specialists’ knowledge of local culture and idioms are anticipated to have a positive impact on survey response rates.
In terms of increasing response rates for telephone surveys, we will use a number of techniques. First, we will work with survey firms to ensure that we have an accurate up-to-date list of phone numbers from which to draw potential respondents. Second, we will use a combination of proven approaches to increase surveys response, including conducting interviewer training so interviewers are sensitive to cultural issues and know how to administer the survey, setting clear establishment of researcher credentials in the introduction, and increasing call attempts and targeting call times. These methods have proven to be an effective approach in increasing response rates for telephone surveys. 16 In addition, we will ensure that we have multilingual telephone staff available for specific calls where language may be an issue.
Contracted polling groups will be asked to demonstrate their survey administration techniques on nine participants prior to execution of the full survey. This will include participants interviewed by those speaking their languages. This sample test will allow for the refinement and correction of any methodological issues that are identified.
Christy
Loper, Ph.D.
Coral Reef Conservation Program
US National
Oceanic and Atmospheric Administration
on detail
to:
Office of the Assistant Secretary for Conservation and
Management
1401 Constitution Avenue, Suite 6224
Washington,
DC
20230
202-482-5143 (office)
240-429-7044 (cell)
[email protected]
Individuals consulted on the statistical aspects of the design:
Victoria Adams, Ph.D. Economist Booz Allen Hamilton 8283 Greensboro Drive McLean, VA, 22102 Telephone: (703)-377-4942 |
The individuals and firms that will collect and analyze the data have not yet been identified.
Cantor, D. and Cunningham, P. (2002) “Methods for Obtaining High Response Rates in Telephone Surveys” in “Studies of Welfare Populations: Data Collection and Research Issues Panel on Data and Methods for Measuring the Effects of Changes in Social Welfare Programs”, Eds.Ver Ploeg, M, Moffitt, R.A. and Citro, C.F. , Committee on National Statistics Division of Behavioral and Social Sciences and Education National Research Council.
Dillman, D., J. Smyth and L. Christian. (2009) Internet, Mail and Mixed‐Mode Surveys: The Tailored Design Method. New York: John Wiley & Sons.
Dillman, D. A. (2000). Mail and Internet surveys: The total design method (2nd ed.). New York:
Wiley. Division of Instructional Innovation and Assessment, The University of Texas at Austin. “Guidelines for Maximizing Response Rates.” Instructional Assessment Resources. 2007. http://www.utexas.edu/academic/diia/assessment/iar/teaching/gather/method/surveyResponse.php
Loomis, David K., “Beach Users Perceptions Concerning Zuma Beach Restoration”, University of Massachusetts Amherst, 2009.
NOAA, “2011 National Marine Recreational Fishing Expenditure Survey” 2011.
Peng, T. K., Peterson, M. F., & Shyi, Y.-P. (1991). Quantitative Methods in Cross-National Management Research: Trends and Equivalence Issues. Journal of Organizational Behavior, 12(2), 87-107.
“Public Perception and Attitudes about the Hawaiian Monk Seal, Survey Results Report”, Sustainable Resources Group International, Inc., prepared for NOAA Fisheries Service Pacific Islands Regional Office, April 2011.
Singh, J. (1995). Measurement Issues in Cross-Cultural Research. Journal of International Business Studies, 26(3), 597-619.
“Washington-Oregon-California Purse Seine Survey”, NOAA, 2007, OMB Control #: 0648-0369, Gulf States Marine Fisheries Commission (GSMFC).
1 See “Public Perception and Attitudes about the Hawaiian Monk Seal, Survey Results Report”, Sustainable Resources Group International, Inc., prepared for NOAA Fisheries Service Pacific Islands Regional Office, April 2011. “Beach Users Perceptions Concerning Zuma Beach Restoration”, David K. Loomis, University of Massachusetts Amherst, 2009.
“Washington-Oregon-California Purse Seine Survey”, NOAA, 2007, OMB Control #: 0648-0369, Gulf States Marine Fisheries Commission (GSMFC). NOAA,“2011 National Marine Recreational Fishing Expenditure Survey” 2011.
2 Dillman, D., J. Smyth and L. Christian. (2009) Internet, Mail and Mixed‐Mode Surveys: The Tailored Design Method. New York: John Wiley & Sons.
3 http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml [Census Total Population figure, 2010]
4 http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml [Census Total Population figures for 5 counties, 2010]
5 http://2010.census.gov/news/xls/cb11cn180_vi.xls [Three islands only]
7 http://2010.census.gov/news/xls/cb11cn177_as.xls [Eastern & Western Districts; Ofu, Olosega, Tau Counties, 2010]
8 http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml [Census Total Population figures for Hawai’i, Honolulu, Kauai, and Maui Counties, 2010]
9 http://2010.census.gov/news/xls/cb11cn178_cnmi.xls [Three municipalities]
10 See “Public Perception and Attitudes about the Hawaiian Monk Seal, Survey Results Report”, Sustainable Resources Group International, Inc., prepared for NOAA Fisheries Service Pacific Islands Regional Office, April 2011, “Washington-Oregon-California Purse Seine Survey”, NOAA, 2007, OMB Control #: 0648-0369, Gulf States Marine Fisheries Commission (GSMFC) and NOAA, “2011 National Marine Recreational Fishing Expenditure Survey” 2011.
11 See, 2000, “Complementary Methodologies: Internet versus Mail Surveys”, DSS Research, Inc.
12 Dillman, D., J. Smyth and L. Christian. (2009) Internet, Mail and Mixed Mode Surveys: The Tailored Design Method. New York: John Wiley & Sons.
13 See Dillman, D. A. (2000). Mail and Internet surveys: The total design method (2nd ed.). New York:
Wiley. Division of Instructional Innovation and Assessment, The University of Texas at Austin. “Guidelines for Maximizing Response Rates.” Instructional Assessment Resources. 2007. http://www.utexas.edu/academic/diia/assessment/iar/teaching/gather/method/surveyResponse.php
14 Singh, J. (1995). Measurement Issues in Cross-Cultural Research. Journal of International Business Studies, 26(3), 597-619.
15 Peng, T. K., Peterson, M. F., & Shyi, Y.-P. (1991). Quantitative Methods in Cross-National Management Research: Trends and Equivalence Issues. Journal of Organizational Behavior, 12(2), 87-107.
16 This approaches have been shown to have a positive impact on response rates, see Cantor, D. and Cunningham, P. (2002) “Methods for Obtaining High Response Rates in Telephone Surveys” in “Studies of Welfare Populations:
Data Collection and Research Issues Panel on Data and Methods for Measuring the Effects of Changes in Social Welfare Programs”, Eds.Ver Ploeg, M, Moffitt, R.A. and Citro, C.F. , Committee on National Statistics Division of Behavioral and Social Sciences and Education National Research Council.
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | sarah.brabson |
| File Modified | 0000-00-00 |
| File Created | 2021-01-31 |