RESPONSES TO OMB QUESTIONS ON
“NATIONAL PARK SERVICE COMPREHENSIVE SURVEY
OF THE AMERICAN PUBLIC, 2007-2008”
October 9, 2007
1. Please provide a copy of the methodology/technical report from the America the Beautiful survey, since it is cited several times.
The greatest concentration of methodological information from the America the Beautiful project is contained in the technical report to the DOI. We have attached the relevant methodological section from this report. The full report is published on the DOI website at http://www.doi.gov/initiatives/rec_fees/NRP%20Survey%20Final.pdf.
2. What comparisons to NPS Visitor Services Project surveys can/will be done – e.g., in the regional reports?
The VSP surveys and the NPS Comprehensive Survey are not intended to be comparable. Methodologically, the two surveys differ in several ways that make comparisons between them problematic. Significant differences include:
Data collection method—the VSP uses a mailback questionnaire, while the Comprehensive Survey is a telephone interview.
Respondent universe—the VSP samples populations of visitors at a non-representative selection of individual parks. These respondents do not represent any population except the visitors at the parks where they are contacted. The Comprehensive Survey samples the population of U.S. households with telephones. This includes both visitors and non-visitors.
Units of analysis—in VSP surveys, the units of analysis are either the individual visitor or the visitor’s group (depending on the question). In the Comprehensive survey, the unit of analysis for most questions is the individual respondent.
Sampling frame—in VSP surveys, the sampling frame is composed of people16 years of age or older visiting a park during a single 7-10 day period. In the Comprehensive Survey, the sampling frame is composed of people 18 years of age or older reached by RDD and who visited (or did not visit) a park within the previous two years.
The major planned comparisons using Comprehensive Survey data are between NPS administrative regions and between visitors and non-visitors. Where data permit, a limited number of comparisons will be made with the 2000 NPS Comprehensive Survey, which was conducted using a similar methodology.
3. What is the visitation trend at other Federal (e.g., BLM) lands and state parks? Does NPS believe it understands the relationship between visitation patterns among these sites and NPS sites and/or does this need to be probed more in this survey?
In general, visitation at other public lands in the U.S. and Canada has been trending downward. In Task Report #2 for the America the Beautiful project (published on the U.S. Department of Interior website), the following trends in park visitation for eight selected state park systems are described.
State 2000 2004
California 98,519,949 82,031,611
Florida 16,672,326 19,117,944
Massachusetts 12,775,063 10,050,913
Oregon 38,563,461 45,144,475
Texas 18,750,576 9,715,728
Utah 6,737,083 5,867,074
Virginia 5,717,026 6,125,745
Wisconsin 15,470,176 14,967,419
Overall, the state park visitation rate for the United States decreased from 1016 per 100,000 population to 943 per 100,000 from FY 2000 to FY 2004.
Parks Canada also reported a decline in visitation numbers from 22,516,817 in FY 2001 to 21,483,422 in FY 2005.
Preliminary data from the U.S. Forest Service’s National Visitor Use Monitoring Program indicates a 35% decrease in visits to the National Forest System between 2000 and 2006.
The NPS is interested in reasons underlying these trends in visits to public lands. It does not appear to be the case that people stop visiting one category of public land in favor of visiting public lands under a different jurisdiction. The technical advisory committee of the NPS Comprehensive Survey of the American Public included the Senior Vice President for Research of the Travel Industry Association of America and the NPS National Tourism Director. Both felt that recent declines in visitation to the National Park System are due in part to increasing competition from commercial segments of the leisure and travel industries. This belief was reinforced in consultations with senior researchers at D.K. Shifflit, one of the leading travel-market research firms in the U.S.
4. Just to clarify, is it the case the RDD survey will use up to 10 callbacks over an eight-week period? If not, please clarify.
The RDD survey will use at least ten callbacks over the eight-week period of calling. The Wyoming Survey and Analysis Center follows the best practices recommendations of the American Association for Public Opinion Research (AAPOR). For unanswered telephone calls, WYSAC will call back at least ten times before eliminating a phone number from the call list.
For calls which are answered, but for which a soft refusal is recorded (e.g., “I can’t talk now, and I can’t schedule a callback now either”), WYSAC has trained refusal converters to call back and attempt a telephone interview completion. Those calls for which an irate refusal is recorded are not called again.
5. According to OMB guidelines for federal surveys, NPS needs to plan a specific nonresponse bias analysis study beyond simply comparing general demographic variables with census data for weighting adjustment. There are several options that would meet this requirement, such as conducting a separate set of follow-ups of a subsample of nonresponders (Keeter et al., 2006), attempting to collect a very small number of questions of hard refusals at time of refusal, or performing zip code or other geographically-based analysis (Johnson et al., 2006). The 2006 special edition of the Public Opinion Quarterly which focused specifically on nonresponse bias is a great resource for explaining these various methods and their costs and limitations.
This is an excellent policy for all survey researchers to follow. Our plan for studying nonresponse bias included a geographic analysis of area codes of non-responders and a separate analysis of converted and unconverted soft refusals, examining the demographic characteristics of each group. However, after speaking with our Survey Center call manager, we have the following suggestion.
Refusals are quite varied. There are the irate refusals, who receive no call-back. There are soft refusals, who receive a call-back, at which time an attempt is made to conduct the entire interview. If a soft refusal says “no,” a second time, these are termed “hard refusals,” and we make no other attempt to call back. However, for the soft refusal that says “no” for the second time, we will attempt to encourage them to answer a very limited number of questions that would take less than one minute. The interviewer will use the following script:
Interviewer: “I understand that you don’t want to take this survey, but just for our records could I ask you two questions that will take less than one minute of your time? No one will call back in either case, but your answers to these two questions will help us understand who is not answering our survey questions.”
If “yes,” proceed to questions.
If “no,” terminate call-back
Interviewer: “The National Park System consists of all the units managed by the National Park Service, including national parks, historic and cultural sites, and national monuments. How many times in the past two years have you visited any units of the National Park System?
Interviewer: Please consider all your experiences to date with National Park System units, including national parks, historic or cultural sites, or monuments. Using a 10-point scale on which “1” equals “very dissatisfied” and "10" equals “very satisfied,” how satisfied are you with the National Park System?
Interviewer: “Thank you so much.”
By using this approach, answers from the hard refusals will give us the ability to make comparisons. This plan has been added to the Supporting Statement Part B #1.
6. How will you address criticism that a single focus group will definitionally be regionally homogeneous and possibly ethnically (e.g., mostly Caribbean) homogeneous too?
The focus groups for this survey supplement previous work conducted by the NPS and other federal agencies on under-served populations (e.g., African Americans, Hispanic Americans). Thus, the focus groups are not the only sources of relevant information used to develop the questionnaire. The intent is to further refine questions dealing with issues covered in previous work by further probing patterns and reasons for use and non-use of the National Park System by these populations. Examples of previous reports include:
Solop, F., Hagen, K, & Ostergren, D. 2003. Ethnic and Racial Diversity of National Park System Visitors and Non-Visitors. Tech. rep. prepared from the 2000 NPS Comprehensive Survey of the American Public. Washington, D.C., National Park Service.
(http://www.nature.nps.gov/socialscience/pdf/Ethnic_and_Racial_Diversity_Report_12_2003.pdf)
Floyd, M. 1999. Race, ethnicity and use of the National Park System. Social Science Research Review 1(2). Washington, D.C.: National Park Service.
(http://www1.nature.nps.gov/socialscience/pdf/SSRR_2.pdf)
Gramann, J. 1996. Ethnicity, Race, and Outdoor Recreation: A Review of Trends, Policy, and Research. Misc. Paper, R-96-1, Vicksburg, MS: US Army Engineer Waterways Experiment Station.
(http://corpslakes.usace.army.mil/employees/ethnic/pdfs/mpr96-1.pdf)
Gramann, J., Torres, C., & Floyd, M. 2002. Improving Outdoor Recreation at Corps Lakes for Under-served Populations: Results of Six Focus Groups. Tech. rep. prepared for Environmental Laboratory, Engineering Research and Development Center. Vicksburg, MS: US Army Engineer Waterways Experiment Station.
(not available online)
In addition, our recent experience indicates that even small focus groups conducted in a single location can be heterogeneous. In the focus group of nine African Americans which we have already completed in Denver, CO, four of the nine members had recently moved to Colorado, three from the eastern U.S., and one from California.
7. What mode of survey administration will occur in the focus groups?
In the focus groups, we typically give the participants a hard copy of the survey instrument, and ask them to fill out the questionnaire as if it were a paper and pencil self-interview. After 15 minutes, we move to self-introductions of all the members, and then begin discussing the questionnaire.
8. The $70 incentive for both cognitive interviews and focus groups is too high. We typically agree to $35 for a cognitive interview and $50 for a focus group. Higher amounts would need to be justified.
Upon OMB’s recommendation, cognitive interview participants will receive $35 compensation and focus group participants will receive $50 compensation. This change is reflected in the updated Supporting Statement Part A #9. In addition, the estimated burden for the cognitive interview has been changed to 105 minutes (60 minutes for the cognitive interview and 45 of travel time). We suspect that we overestimated the initial burden. This burden change will be updated in ROCIS and in the Supporting Statement Part A #12.
9. Why does NPS us the phrase “opinion poll”? We would prefer “survey.”
“Opinion poll” has been changed to “survey” in the revised interview script (attached).
10. What prior use of testing was done in 2000 on the 2-year recall period for a “recent visit”? Or what could be added for 2007 cognitive work? This seems like a long time, and I wonder about the accuracy of recall.
We have no first-hand knowledge of prior testing done in 2000 on the 2-year recall period. There is no description of testing in the 2000 survey’s technical report.
Recall bias is always a consideration in surveys of this type. Research that addresses recall bias in recreation participation has produced mixed results; however, several studies report that longer recall periods tend to lead to a “telescoping effect” in which the number of days participating is over-estimated. Some telescoping may have occurred in the 2000 survey. When first asked how many times they had visited a unit of the National Park System in the past two years, 52% of respondents reported at least one visit. However, when subsequently asked to name the unit they had visited, only 32% could do so accurately.
While some of this discrepancy may be caused by initially over-estimating actual visits, undoubtedly a large amount is due to the respondents confusing national park units with national forests, state parks, and similar areas. The many designations employed by the NPS (e.g., park, monument, memorial, historic site, battlefield, seashore, recreation area, etc.) contribute to this confusion. Asking respondents to name the national park site they visited is an important mechanism for dealing with this issue and is also one check on recall bias.
In 2000, a “recent visitor” was defined as one who had made at least one visit within the previous two years to a national park site that he/she could accurately name. This conservative measure will be used again in 2007, and there is great interest among stakeholders in the trend in this measure. For this reason, we would like to preserve the 2000 wording. However, the actual number of visits in the previous two years is not as important as whether or not at least one visit was made.
11. Why use a five year window for questions 25-28?
There is no five-year window for questions 25-28. The introductory questions #5, #5b, and #5c provide time lines for data analysis. With the information from these questions we will be able to analyze the answers by those individuals who visited an NPS site within the last two years, from 3–5 years ago, ever, and never.
12. At various points, the burden is expressed as “about 12 minutes” and “10 to 15 minutes.” If the burden is an average of 15, then that should be explained to respondents.
This discrepancy is addressed in the final version.
13. Question 17 seems like a very long list for a telephone survey, and we wonder about the unique contribution of each sub-question. In addition, we would prefer the use of first person.
With the addition of statements new to this year’s questionnaire, the list has become longer. However, we have received more queries from Congress and stakeholders concerning trends in these barriers to visitation than any other part of the 2000 survey. Therefore, as much as possible, it is important to repeat the wording in the new survey. However, in order to deal with the length issue, we will randomize the list and ask each respondent to respond to ten of the 16 statements.
13a. Specifically, in item k, we do not believe it is wise to essentially ask someone to speak for his entire racial group. Please consider asking simply about someone’s comfort level and cross-tabulating the information with the racial and ethnic data you will be collecting on the demographic section.
The technical advisory committee recommended other changes to this item, so comparability with 2000 is not an issue. The suggested change can be made:
“National Park System units are unpleasant places for me to be.”
13b. Related, we would prefer to see item p be less leading. For example, the question could say something like “I don’t want to be around park rangers during a vacation because they remind me of police.”
Since this is a new item, this change can be made.
13c. 17o seems very narrow, especially since there is no other question about alternative activities – is there a literature behind this question? Isn’t there some literature that people just want to create their own private oasis at home – so maybe they are spending more time on their yards, etc., rather than in front of the TV, as this question implies?
The question is: “I prefer to spend my free time doing electronic activities, like watching videos, enjoying computer games, or surfing the Internet.”
This item is drawn from recent literature on “nature-deficit disorder,” the hypothesis that involvement in electronic media causes children and adults to have less direct experience with nature. Popular books (Louv, 2006) and scientific articles based on secondary data (Pergams & Zaradic, 2006) have advanced this notion, and it has become a major topic of discussion and policy debate in public resource management agencies at all levels of government. However, primary data on age trends among national park visitors partially contradict the argument (Gramann, et al., 2007), at least in its application to children. This question provides a first direct examination of the notion that use of electronic media is associated with a reduced preference for visiting national parks among persons 18 years of age and older.
Louv. R. 2006. Last Child in the Woods: Saving Our Children from Nature-deficit Disorder, Chapel Hill, NC: Algonquin
Pergams, O. & Zaradic, P. 2006. Is love of nature in the US becoming love of electronic media? 16-year downtrend in national park visits explained by watching movies, playing video games, internet use, and oil prices. Journal of Environmental Management 80(4): 387-393
Gramann, J., Hollenhorst, S., Littlejohn, M., & Le, L. 2007. Last child in the parks? Age trends in US national park visitation. Paper presented at the biennial meeting of the George Wright Society, St. Paul, MN. April, 19.
14. Question 25 – what is the source of this scale? Is there a demonstrated difference between “very” and “extremely”?
This type of question is relatively new to the outdoor recreation literature and is still being refined. This particular scale was developed by the investigators for an onsite interview at Rocky Mountain National Park. It was successfully used to analyze what types of outdoor sensory experiences anticipated by the visitors. A similar question was employed by researchers from the U.S. Geological Survey to assess sound levels in the park. The cognitive interviewing for this question will help determine whether or not respondents perceive a difference between “very” and “extremely.”
15. Questions 26 and 28 ask directly about what NPS policy should be. We would prefer that they ask about the person’s own opinion/enjoyment as question 25 does.
We have modified these questions to ask about people’s own opinions. Question 26 now asks individuals to give their opinions about the appropriateness of noise at parks and question 28 asks whether the individuals personally agree or disagree with a list of statement. Thus, these questions no longer inquire about NPS policies. These updated questions appear in the copy of the survey instrument included below.
16. D3 – What is the origin of the response category “living with life partner”? It is the only response category that explicitly indicates who is “lived with.” Since you are planning to compare with census data, please consider using a Census Bureau form of this question.
Clearly, what we are trying to distinguish is married (legally) versus two adults living together as partners (whether they are the same sex or not). The categories in the Census Bureau’s marital-status question do not address this social reality, although other national surveys do. For example, the Multimedia Audience Research Systems (MARS) survey, a national survey of adults 18 years of age and older, asks respondents if they “are currently living with an unmarried/domestic partner.” And the National Center for Health Statistics includes the category “living with partner” in its marital-status question.
Alternatively, we could ask the marital-status question using the response categories employed by the Census Bureau (now married, widowed, divorced, separated, never married) and follow this with a separate question about living with a life partner. This would allow us to determine how “life partner” responses cross-tabulate with legal marital status. Other research (e.g., MARS 2004) showed that people living with domestic partners reported their legal marital status in several ways, i.e., as “divorced” (23.6%), “single” (21.7%), “separated” (12.6%), “widowed” (5.9%), or “married” (1.1%).
17. D5 should read “Latino” not “Latina” in the question body.
We thought we had caught this in an earlier draft, but obviously not.
ATTACHMENT A
Selection of Methods Sections from the Final Report for the New Interagency Pass
2. Methods
2.1. Mode of data collection
The survey was conducted from late February through late April, 2006, using WYSAC’s facilities for Computer-Aided Telephone Interviewing (CATI). Most telephone numbers were automatically dialed by the CATI software, but with a live interviewer on the line at all times so that respondents were unaware of the automatic dialing. Numbers identified (by the sample vendor, or in WYSAC’s initial calling) as likely cell phones were dialed manually, as required by federal law.
The interview script was programmed into the software, so that as respondents were asked and answered the questions, the computer screens automatically moved through the questions following the appropriate skip logic (as indicated in Appendix A). Therefore, respondents were not burdened with written data entry, or with being asked questions irrelevant to their particular circumstances. The CATI software was also programmed to randomize the presentation of various possible prices for the new pass (see below).
The telephone mode of data collection was particularly appropriate because of the tight time constraints faced by the sponsoring federal agencies to obtain information for pricing the pass. Also, the restrictive definition of the target population stipulated by the agencies meant that many ineligible households could only be screened out based on answers to preliminary questions during the survey itself. These two considerations made other modes of data collection (mail-out survey, web survey, face to face interviewing) impracticable. Discussions were held with representatives of the federal agencies and with our external consultants about the possibility of using an Internet-based survey of recreationists, with a compensated panel of respondents. However, an Internet panel would be highly self-selected and not representative of the general population, precluding a sound statistical estimate of key parameters of interest in this study.
2.2. Sampling
The potential universe of contacts consisted initially of all landline telephone numbers in the United States with an area code, three-digit prefix, and working 100-bank (next two digits) that included at least one listed residential telephone. The latest available estimates indicate that this conventional sampling frame for Random Digit Dialing (RDD) covers about 90 percent of all U.S. households.
The sponsoring agencies specified that the survey should include current pass holders as well as non-pass holders. However, the benchmarking part of the larger project revealed a low rate of pass sales nationally, suggesting that the RDD sample would yield relatively few pass-holding households. That sample was therefore augmented by a sample drawn randomly by WYSAC from a list of U.S. telephone numbers (landline or cell) recorded by the National Parks Foundation (NPF) as having purchased a National Parks Pass between April of 2004 and March of 2005 (i.e., one to two years prior to the survey). Since most pass sales occur at a park site rather than through NPF, the NPF population list covers only about 6% of the National Parks Passes that were sold in that 12 month period. However, a substantial majority of the NPF households were still pass holders one to two years later, when the survey was conducted, whereas very few in the RDD sample were current pass holders. Use of the NPF sample to augment the RDD sample therefore ensured a sufficient number of pass holders to permit examination of such households.
In this report, as in our other analyses of the survey findings, the RDD and NPF results are presented separately. By design, the RDD results are more representative of the general U.S. population.
Existing annual passes to federal recreational lands provide access for all occupants of one private vehicle. Similarly, it is anticipated that the NRP will be valid for up to four adults traveling together (plus children under 16). Therefore, the relevant unit of analysis for this study was the household, and no within-household sampling or respondent selection was necessary. The particular individual who was interviewed was asked to provide information on behalf of the responding household, rather than as an individual-level respondent. In the case of a one-person household, there was no distinction between informant and respondent.
The RDD sample initially consisted of 20,000 telephone numbers purchased from a national vendor (Marketing Systems Group). It was proportionally stratified by the nine major geographic divisions defined by the U.S. Bureau of the Census (as was the NPF sample). The initial RDD sample was pre-screened by the vendor to eliminate, insofar as possible, disconnects, businesses, and other known ineligibles in the RDD numbers. More than half of the initial RDD numbers were eliminated through this pre-screening.
The remaining sample was randomly subdivided into waves before calling began, and each wave was thoroughly worked in the survey calling (15 or more callbacks per un-reached number) before deciding whether the next wave was needed. This step reduced the potential over-representation of households easiest to reach by phone. In some cases, there were more than 40 callback attempts. At least one refusal conversion attempt was made in the case of terminated calls or other soft refusals, but not for irate refusals. Households where WYSAC’s interviewers encountered a language barrier were called back by a bilingual interviewer and, if possible, interviewed in Spanish.
Response rates can be defined in various ways. OMB prefers what the American Association for Public Opinion Research (AAPOR) calls RR3. This formula assumes that the proportion of eligible households where eligibility could not be determined was the same as the proportion eligible among households that were definitively identified as either eligible or ineligible. This is generally a conservative assumption in RDD, since with 15 or more callbacks spread over several weeks, most of the repeated no-answer numbers and other non-contacts were probably not working phones.
Here, we had two distinct standards of eligibility to consider. First, the phone number must have reached a private U.S. household, not a business or government office and not a disconnected number. Those numbers were eligible for the “screener” or filter questions, as discussed in the next section. Second, households that were screened in, based on those filter questions, were eligible for the “full” survey. As expected, at both stages dialing of random-digit phone numbers yielded lower proportions eligible and higher proportions of unknown eligibility than did sampling from the NPF population list.
Based on these definitions, in the RDD screener, RR3 = 26.6% while in the NPF screener,
RR3 = 47.1%. For the RDD full survey, RR3 = 65.1%, and for the NPF full survey, RR3 = 54.6%.
Although the RDD screener response was lower than had been hoped, overall the response rates are consistent with current standards for telephone surveys. Comparing the RDD respondents to U.S. Census data revealed modest differences in non-response by geographic area and somewhat lower response among low-income and minority households. As discussed in section 2.4, below, weighting was used to adjust, insofar as possible, for any non-response bias that might be associated with such demographic differences.
Another relevant figure is the completion rate, which may be defined as the proportion of known eligibles who finished the full survey (at least through the key question about willingness to pay for the NRP). By this measure, the RDD completion rate was 96.0% and the NPF completion rate was 98.5%. Item non-response was generally very low, except for a very few questions (e.g., income). The analyses for this report exclude missing data, except as noted, but Appendix A includes full details on item non-response for each question in the survey.
2.3. Filter questions
Screening within the survey reduced both sample pools substantially, especially the RDD. (See Appendix A for the full survey script, including introductory and filter items.) The survey design called first for information regarding an individual’s availability to answer questions and, so as to determine initial eligibility for the screener, whether the call had reached a private household. The respondent was then given information about the voluntary nature of the study. Upon the advice of OMB strict confidentiality was not promised. However, the survey was, for all practical purposes, anonymous.
The survey design also required information about age and disability. A household with anyone age 61 or over in 2006, or with a disabled household member, would (by 2007) qualify for a $10 Golden Age Passport or a free Golden Access Passport. Such households were therefore screened out as unlikely to be part of the potential market for the new pass. As a double check, additional questions late in the interview asked about ownership of these two lifetime passes. Households reporting that they currently had either pass or that they expected to obtain one of them in the next 12 months were treated as ineligible for the full survey, regardless of their responses on the earlier filter questions.
Finally, the sponsoring federal agencies specified that the target population for the survey consisted of households that had visited federal recreational lands in the past two years. Another filter was therefore used to screen out long-term non-visitors as unlikely to be part of the market for the new pass. The two-year time frame was longer than in subsequent questions, which focused on a 12-month period. Also, this target population was not limited to those who had visited lands that charge an entry fee. Thus, the target population includes potential pass users, some of whom had paid a fee to visit federal lands in the past few months, and some of whom may have visited as long ago as two years earlier and paid no fee.
Households identified as ineligible by any of these screening criteria were not asked the core survey questions about the NRP. However, they were asked to report their household’s demographic characteristics, to allow for appropriate weighting of the data (see Appendix H). Based on these criteria, screening reduced the RDD sample from 1799 to 556 households. Similarly, the NPF sample dropped from 1974 to 1524 households due to screening.
Simple random samples (SRS) of 556 and 1524 would yield statistical margins of error of roughly 4.5 and 2.5 percentage points, respectively, with 95% confidence. Neither sample is strictly SRS, however. Both were proportionally stratified geographically, which would tend to decrease the margin of error, and both were also weighted as described in the next section, which would tend to increase it. The margin of error also varies with the parameter being estimated. For example, estimating a percentage when the true population value is 10% yields a smaller margin of error than when the population value is 50%. Thus the margins of error just mentioned are offered only as general guidelines. This report is mainly descriptive and, as such, presents no formal significance tests or confidence intervals. The econometric report from this project includes tests for statistical significance that treat the RDD and NPF samples as SRS.
2.4. Weighting
Before applying the within-survey screens to eliminate households that were not part of the target population, both samples of households with completed screener interviews were weighted (separately) to improve the representativeness of the data. No additional weighting could be done after eliminating screened-out households. Because of the restrictive definition of the target population, there is no known population distribution toward which to further adjust the screened-in sample distributions.
The probability of selection in an RDD sample is proportional to the number of answerable landline telephone numbers that ring into a household. Therefore, the number of landlines was ascertained in the survey, and its reciprocal was applied to weight the complete, unscreened RDD sample. This first step in the weighting was not necessary for the NPF sample, because there the sampling frame consisted of purchasers of the National Parks Pass with phone numbers known to NPF, not randomly generated landline phone numbers. Duplicate phone numbers within either sample had been eliminated before calling began, and the one phone number that appeared in both samples was eliminated from the RDD pool to ensure the independence of the two samples.
Both the unscreened RDD and NPF samples were weighted to adjust the geographic distribution of responding households to the known geographic distributions of the corresponding populations. Household data from the U.S. Census Bureau provided the external reference for the RDD sample, and the NPF population list served that purpose for the NPF sample. No further weighting of the NPF sample was undertaken, because of a lack of external reference information about other characteristics of this population.
The unscreened RDD responses were next weighted to approximate the Census estimates of the distribution of the U.S. population by Hispanic origin, race, and household income. Then, using DOI figures on annual sales of the Golden Eagle and National Parks Passes, a further weight was applied to reduce an apparent over-representation of pass-holding households in the RDD sample.
As usual, successive weighting altered the geographic distribution somewhat. Therefore, the final step was to adjust the geographic weights to compensate. Rather than iterating this “raking” process further, the final weights retained a slight under-representation of minorities and low-income households. Weight trimming (truncation) was unnecessary, since the range of the final weights fell within conventional guidelines.
Comparison of the weighted and unweighted results in Appendix A shows that the weights generally have very little impact on the findings, with the partial exception of demographic variables used explicitly in the weighting. This report, like our other analyses of the survey, relies on the weighted data unless otherwise indicated. The weights used are appropriate for analyses at the household level; somewhat different results would be obtained by adjusting toward the population distribution of individual adults, rather than households.
2.5. Measuring willingness to pay
To gather information for pricing the new federal recreational pass, households were offered a random set of pass prices. The initial bid ranged from $25 to $165 in $20 increments. A second bid question was also asked, higher than the initial bid value if the respondent said yes to the first price, or lower than the initial value if the respondent said no. This approach allowed a wide range of possible bids to be used, reduced the potential of starting-point bias, and at the same time reduced respondent burden since each household was generally asked only two bid questions.
The values in the bid vector were determined based on results obtained by pre-testing the questionnaire in the focus group part of the full project. Considering only the first bid, one of the following eight different bid prices was posed initially:
$25 $45 $65 $85 $105 $125 $145 $165.
By design, the bid vector included $65, which is the current price for the Golden Eagle Passport. Because the Golden Eagle pass is very similar to the NRP, and because information was also obtained in the survey as to whether the household purchased the Golden Eagle pass, we were able to contrast hypothetical purchasing decisions for the NRP at $65 with actual purchasing decisions for the Golden Eagle pass.
For respondents who said yes initially, the second bid value was randomly selected from two higher possibilities: $20 more than the initial bid, or $5 less than double the initial bid. For those who said no initially, the second bid value was randomly selected from two lower possibilities: $20 less than the initial bid, or half of an amount equal to the initial bid plus $5. The $5 adjustments ensured that respondents would not perceive a trick in getting a second bid exactly double, or exactly half, the initial value. The $5 adjustments also aligned the bid possibilities on the second bid with those on the first bid, while providing more detail at both the high end and the low end.
Considering both the first and the second bids, there were 19 different bid prices posed, as follows:
$5 $15 $25 $35 $45 $55 $65 $75 $85 $95 $105 $125 $145 $165 $185 $205 $245 $285 $325.
Respondents who said no to both the initial bid price and the second, lower bid amount were then asked whether they would be interested in the pass for free. In effect, this added zero as the twentieth point on the bid vector. Doing so identified potential “protest” bidders who may have an objection in principle to the pass, to the current management practices on federal lands, or the like. Those who would reject the pass even if it were free were asked why, to determine the nature of their objection. Protest bidders defined themselves out of the potential market for the pass, and needed to be identified so that their refusal to take the pass at any bid price, even zero, did not distort the estimated demand curve.
3. Selected comparisons of the two samples
We will not attempt here a thorough discussion, much less a complete analysis, of the survey data. Appendix A provides descriptive results on all variables in the questionnaire, and a separate econometric report analyzes the willingness-to-pay data in detail. The results we highlight below are limited to selected comparisons of the screened RDD and NPF samples.
With total National Parks Pass (NPP) sales of approximately half a million passes sold per year, the NPF population list of just 30,629 households covers only about 6% of the households that purchased an NPP between one and two years prior to the survey. The NPF population, and therefore the NPF sample drawn from that population, mainly represents households that purchased the pass on-line during that time. The NPF sampling frame omits the much larger number of households that purchased the NPP in-person at a park, except for those that chose to provide their phone numbers by mailing a reply card to NPF, requesting a park “owner’s manual,” or sending the NPF a donation. It also omits the roughly 50,000 annual purchasers of the Golden Eagle Pass.
Thus, results from the NPF sample are of interest in that they reflect a small but noteworthy group of supporters of the national parks, while the RDD sample comes as close as our methodological and budget constraints allow to being representative of the full spectrum of the roughly 110 million U.S. households. Recall that both samples were then screened to eliminate households not part of the target population as defined by the sponsoring federal lands agencies, based on age, disability, and non-visitation to federal lands.
ATTACHMENT B
Revised Comprehensive Survey Script
NOTATIONS:
Notations in blue after a question indicate whether it was asked in the 2000 Comprehensive Survey or whether it is a new question.
DK/NS = Don’t Know/Not Sure
Print in bold face is read to the respondent.
Highlights in yellow indicate a revision in response to review comments from OMB.
1. CELLINTRO:
(ask only if sample = cell phone supplement, following manual dial;
if sample = RDD, skip this item and go directly from autodial to 1A, Intro)
Hello, I’m doing a survey for the National Park Service, and I am not selling anything. Have I reached you on a cell phone? (new)
(Yes) Skip to Cell
(No) Skip to SorryLL
(Hangs up without speaking) End interview; autocode as cell soft refusal
(Needs Spanish interview) End interview; autocode for cell bilingual callback
(Other - assign disposition code) End interview; go to disposition coding screen
1A. INTRO:
Hello, I’m calling on behalf of the National Park Service, and I am not selling anything.
Your phone number was randomly chosen for a nation-wide survey. The purpose of the survey is to help the National Park Service improve its services to you and people like you. My name is __________ [first name] and I only need about 15 minutes to ask you some important questions about our national parks, historic sites, and recreation areas. Would you be able to help me out with this? (similar to 2000 intro)
(SUPPLEMENTAL INTRO, if respondent hesitates: The U.S. Office of Management and Budget has approved this research under the Paperwork Reduction Act. All of your answers are completely voluntary. Responses to this study will be used only for statistical purposes. The reports prepared will summarize findings across the sample and will not associate responses with a specific individual. We will not provide information that identifies you to anyone outside the study team, except as required by law. The University of Wyoming is conducting this study for the National Park Service, which will use the results to better serve the public.)
(If needed: It’s really important that we hear from all kinds of households, whether you think you know much about national parks or not. The National Park Service wants input from the American people about managing parks. No action may be taken against you for refusing to supply the information requested. No personal data will be recorded that will identify you. Your phone number will be separated from your answers, so the final data will be anonymous. Participation in this survey is expected to average about 15 minutes per household. U.S. Code 16-1a-7 authorizes collection of this information. The OMB approval number is __________, with an expiration date of _________. You may direct comments on any aspect of this survey toll-free to the University of Wyoming at 1-866-966-2715.)
(Yes or maybe) Skip to Private
(No – PERSUADE) Skip to Callback
(Hangs up without speaking) End interview; autocode as RDD soft refusal
(Needs Spanish interview) End interview; autocode for RDD bilingual callback
(Reached on a CELL PHONE) (proceed to Cell)
1B. CELL:
I’m not allowed to interview you if you’re driving, and I don’t want to use your minutes. May I keep going, or should I call back some other time? (new)
(If not now: What would be a good time to call back? PROBE for time zone.)
(OK to keep going) Skip to LandLine
(Not now, call at another time) Go to scheduler screen; autocode as cell callback
(Safe to talk, but says No – PERSUADE) (proceed to Callback)
(IRATE refusal) End interview; autocode as 221, final cell refusal
1C. CALLBACK:
You might only qualify for a few of the questions. Can I ask those, and you can tell me to stop whenever you want? (new)
(If needed: Could I call back at a more convenient time, or speak to someone else now? It’s really important that we hear from all kinds of households, whether you think you know much about national parks or not. The U.S. government needs input from the American people about managing federal lands. The University of Wyoming is conducting the survey for the U.S. National Park Service.)
(OK to keep going) (proceed to Private; except if Cell=3, skip to LandLine)
(Not now, call at another time) Go to scheduler screen; autocode as callback
(Someone else comes to the phone) Skip back to #1A, Intro
(Other - assign disposition code) End interview; go to disposition coding screen
2. PRIVATE:
Have I reached you at a private household in one of the 50 United States or the District of Columbia? (new)
(If needed: Private household means a residence where one or more individuals or families live. It could be a house, an apartment, or a mobile home, but not a business or government office where nobody lives. A retirement community, boarding house, or other group quarters is not considered a private household unless this phone number rings directly into the living quarters of a particular individual or family, rather than ringing into a central switchboard or a phone shared with other residents.)
(Yes) Skip to Adult
(Not private or not U.S.) Skip to SorryPH
(Volunteered: Reached in the U.S., but on a cell phone) Skip back to Cell
(Don’t know/Not sure) Skip to SorryPH
(No answer/Refused) Skip to SorryPH
2A. LANDLINE:
In the household where you live, is there at least one residential landline phone that can be answered by a person, or does your household only have cell phones? (new)
(If needed: A landline phone that is only used for business, or only for a computer or fax machine, is NOT considered a residential phone.)
(Household has one or more residential phones) (see conditional skips below)
(Household only has cell phones) (see conditional skips below)
(Don’t know/Not sure) (see conditional skips below)
(No answer/Refused) (see conditional skips below)
If sample = Cell and LandLine = 1 or 3 or 4, skip to SorryLL. If sample = Cell and Landline = 2, proceed to Cell 2B. But if sample = RDD, skip to Adult regardless of response on LandLine.
2B. CELL2B:
Your phone number was randomly chosen for a nation-wide survey. The purpose of the survey is to help the National Park Service improve its services to you and people like you. My name is __________ [first name] and I only need about 15 minutes to ask you some important questions about our national parks, historic sites, and recreation areas. (similar to 2000 intro)
(SUPPLEMENTAL INTRO, if respondent hesitates: The U.S. Office of Management and Budget has approved this research under the Paperwork Reduction Act. All of your answers are completely voluntary. Responses to this study will be used only for statistical purposes. The reports prepared will summarize findings across the sample and will not associate responses with a specific individual. We will not provide information that identifies you to anyone outside the study team, except as required by law. The University of Wyoming is conducting this study for the National Park Service, which will use the results to better serve the public.)
(If needed: It’s really important that we hear from all kinds of households, whether you think you know much about national parks or not. The National Park Service wants input from the American people about managing parks. No action may be taken against you for refusing to supply the information requested. No personal data will be recorded that will identify you. Your phone number will be separated from your answers, so the final data will be anonymous. Participation in this survey is expected to average 15 minutes per household. U.S. Code 16-1a-7 authorizes collection of this information. The OMB approval number is __________, with an expiration date of _________. You may direct comments on any aspect of this survey toll-free to the University of Wyoming at 1-866-966-2715.)
All who reach this point, skip to Adult
2B. SORRYLL:
I’m sorry, but for this part of the survey we’re supposed to interview households that only have cell phones. Thanks anyway!
End interview, autocode as Not a cell-only household.
2C. SORRYPH:
I’m sorry, but for this part of the survey we’re only supposed to interview private households in the U.S. Thanks anyway!
End interview, autocode as Not a private U.S. household.
2D. ADULT:
And I need to ask, are you at least 18 years old? (new)
(Yes) Skip to NUMAD
(No) (proceed to SorryAd)
(Don’t know/Not sure) Skip to SorryPH
(No answer/Refused) Skip to SorryPH
2E. SORRYAD:
I’m sorry, but I’m not allowed to interview anyone under 18. Is someone there who is 18 or older that I can speak to now, or could I call back some other time? (new)
(No adult lives here) End interview; autocode as No adult in household
(Not now, call back another time) Go to scheduler screen; autocode as callback
(Someone else comes to the phone) Skip back to #1A, Intro
(Other - assign disposition code) End interview; go to disposition coding screen
3. NUMAD:
Including yourself, how many people age 18 or older currently live in your household? (new)
(Open-ended, record response)
(1-26 adults in the household)
(code 27 for 27 or more)
(28=DK/NS)
(29=Refused)
If NumAd=1, set marker text to “only adult”. If sample = cell and NumAd is not 1, set marker text to “adult cellphone user.” If sample = cell, skip to #4F. If sample = RDD and NumAd=1, skip to #5.; if sample = RDD and NumAd = 2 through 27 or DK or Refused, proceed to #4.
4. To be sure our survey represents everyone equally, my computer is going to randomly select one person in your household for me to interview.
If NumAd = 27 or DK or Refused, proceed to #4A. Otherwise, randomize: one-third to 4A, one-third to 4B, and one-third to 4C. Store the random value as a variable.
4A. It says I need to talk to the person currently living in your household who is 18 years of age or older and who has had the MOST RECENT birthday. Would that be you or someone else? (part of 2000 intro)
(set marker text = “adult who has had the most recent birthday”)
(Self) Skip to #5
(Someone else) Skip to #4D
4B. It says I need to talk to the person currently living in your household who is 18 years of age or older and who will have the NEXT birthday. Would that be you or someone else? (new)
(set marker text = “adult who will have the next birthday”)
(Self) Skip to #5
(Someone else) Skip to #4D
4C. (set marker text = “oldest adult,” or “second oldest adult,” etc., randomly chosen with equal probabilities from the number of adults reported on NumAd)
It says I need to talk to the person currently living in your household who is 18 years of age or older and who is the [insert marker text]. Would that be you or someone else? (new)
(Self) Skip to #5
(Someone else) (proceed to #4D)
4D. May I please speak to the member of your household who is the [insert marker text]? (part of 2000 intro)
(Yes, someone comes to the phone) (proceed to 4E)
(Not now, call back another time) Go to scheduler screen; autocode as RDD callback
(Other - assign disposition code) End interview; go to disposition coding screen
4E.
You have been randomly chosen for a nation-wide survey. The purpose of the survey is to help the National Park Service improve its services to you and people like you. My name is __________ [first name] and I only need about 15 minutes to ask you some important questions about our national parks, historic sites, and recreation areas. Would you be able to help me out with this? (similar to 2000 intro)
(SUPPLEMENTAL INTRO, if respondent hesitates: The U.S. Office of Management and Budget has approved this research under the Paperwork Reduction Act. All of your answers are completely voluntary. Responses to this study will be used only for statistical purposes. The reports prepared will summarize findings across the sample and will not associate responses with a specific individual. We will not provide information that identifies you to anyone outside the study team, except as required by law. The University of Wyoming is conducting this study for the National Park Service, which will use the results to better serve the public.)
(If needed: It’s really important that we hear from all kinds of households, whether you think you know much about national parks or not. The National Park Service wants input from the American people about managing parks. No action may be taken against you for refusing to supply the information requested. No personal data will be recorded that will identify you. Your phone number will be separated from your answers, so the final data will be anonymous. Participation in this survey is expected to average about 15 minutes per household. U.S. Code 16-1a-7 authorizes collection of this information. The OMB approval number is __________, with an expiration date of _________. You may direct comments on any aspect of this survey toll-free to the University of Wyoming at 1-866-966-2715.)
(Yes or maybe) Skip to #5
(Not now, call back another time) Go to scheduler screen; autocode as RDD callback
(Needs Spanish interview) End interview; autocode for RDD bilingual callback
(Other - assign disposition code) End interview; go to disposition coding screen
4F. USDC:
(ask only if sample = cell; use area code to define residence location if sample = RDD)
Do you currently live in one of the 50 United States, or in the District of Columbia? (new)
(Yes) (proceed to State)
(No) Skip to SorryPH
(Don’t know/Not sure) Skip to SorryPH
(No answer/Refused) Skip to SorryPH
4G. STATE:
In what state do you currently live, or do you live in D.C.? (new)
(list 50 states plus D.C. on the screen, in alphabetical order, followed by DK and NA)
5. The National Park System consists of all the units managed by the National Park Service, including national parks, historic and cultural sites, and national monuments. How many times in the past two years have you visited any units of the National Park System?
(2000, #2)
__________ (Insert number of visits, 0-997)
If zero, skip to #5a2
5a. Which National Park System unit did you last visit? (2000, #2a)
______________ (Code response using list)
Display list on the screen, most-visited units on one screen in alphabetical order; less-visited units on other screens, in alphabetical order.
If response doesn’t match list, enter response in a text field for Other/specify and proceed to #5a1; if response matches a unit on the list, skip to 5d.
5a1. I’m not finding a unit by that name on my list. The National Park System includes parks like Yellowstone National Park, but also battlefields, monuments, and smaller urban sites. For example, _____________________ and ______________ are both National Park System units.
Can you name any sites such as these that you’ve visited in the past two years? (new)
______________ (Code response using list)
Fill in blanks with NPS units from the respondent’s state
If response still doesn’t match list, skip to #5b; if response matches a unit on the list, skip to 5d.
5a2. A lot of people don’t realize that the National Park System includes not only parks like Yellowstone National Park, but also battlefields, monuments, and smaller urban sites. For example, _____________________ and ______________ are both National Park System units.
Can you name any sites such as these that you’ve visited in the past two years? (new)
______________ (Code response using list)
Fill in blanks with NPS units from the respondent’s state
If response still doesn’t match list, proceed to #5b; if response matches a unit on the list, skip to 5d.
5b.How about three to five years ago? Did you visit a unit of the National Park System during that period? (new)
(Yes) Skip to #5d
(No) (proceed to 5c)
(DK/NS) (proceed to 5c)
(NA/Refused) (proceed to 5c)
5c. Have you ever visited any unit of the National Park System? (new)
(Yes) (proceed to #5d)
(No) Skip to #6
(DK/NS) Skip to #6
(NA/Refused) Skip to #6
5d. How old were you the first time you visited a unit of the National Park System? (new)
(Record actual age)
(DK/NS)
(NA/Refused)
6. Please consider all your experiences to date with National Park System units, including national parks, historic or cultural sites, or monuments. Using a 10-point scale on which “1” equals “very dissatisfied” and "10" equals “very satisfied,” how satisfied are you with the National Park System? (2000, #4)
(Record number)
(DK/NS)
(NA/Refused)
7. Do you strongly agree, somewhat agree, somewhat disagree, or strongly disagree with the following statement: “I plan to visit a unit of the National Park System within the next 12 months.” (2000, revision of #5)
(Strongly agree)
(Somewhat agree)
(Somewhat disagree)
(Strongly disagree)
(DK/NS)
(NA/Ref)
8. We’re interested in what kinds of vacation trips you like to take when you spend at least one night away from home.
In the past two years, have you taken any overnight vacation trip away from home? (new)
(Yes)
(No)
(DK/NS)
(NA/Refused)
If No, DK, or NA, skip to #9
I’m going to list some different kinds of vacation trips that people might take. For each trip, please tell me how much you like it. Use a scale from one to four, where 1 means you “don’t like it at all,” 2 means you “like it very little,” 3 means you “like it pretty much,” and 4 means you “like it a lot.” The first one is …. (new)
randomize order of presentation; for each one, include the following note:
(If needed: Would you say you “don’t like this at all,” “like it very little,” “like it pretty much,” or “like it a lot”?) (8 = DK/NS; 9=NA/Ref for each item)
A trip to visit friends or relatives
A trip to an out-of-town sporting event
A out-of-town shopping trip
A trip to a theme park, such as Disney or Six Flags
A trip to experience art, music, or other cultural activities
A trip to a national park site to experience nature
A trip to a national park site to see historical exhibits
A trip to a state park or national forest
A trip to a casino or other gaming place
A trip to another country
A trip to a spa or resort
A trip on a cruise ship
If no answer on 5a and on 5a1 and on 5a2, skip to #17]
Now I'd like to ask you a series of questions about your last visit to a National Park System unit, which you said was ___________ [Insert name from #5a or 5a1 or 5a2].
Thinking about your last visit to [insert park name], what were your two or three main reasons for visiting there? (2000, #6, most common open-ended responses listed, in order of frequency)
(Do NOT read choices; code up to 3 responses)
(Sightseeing)
(Vacationing with guests, family, company, relatives)
(Viewing exhibits, park information, educational sites)
(Day hiking)
(Go just because it’s there, proximity)
(Camping)
(Visiting a cultural or historic site)
(Playing sports, recreation, exercise)
(Fishing)
(Viewing wildlife)
(Picnicking)
(I thought it would be fun) (new)
(Shopping in nearby stores) (new)
(Riding ATVs or snowmobiles in nearby areas) (new)
(Any other reason that cannot be coded above)
(DK/NS)
(NA/Refused)
We’d like to know how much the following experiences may have added to your enjoyment during your last visit to [insert park name]. Please rate each one on a scale from one to four, where 1 means it “added no enjoyment at all,” two means it “added very little enjoyment, ” 3 means it “added some enjoyment,” and 4 means it “added a lot of enjoyment.” (new)
randomize order of presentation; for each one, include the following note:
(If needed: Would you say it “added no enjoyment at all,” “added very little enjoyment,” “added some enjoyment,” or “added a lot of enjoyment”?) (8 = DK/NS; 9=NA/Ref for each item)
Learning more about history and culture
Learning more about nature
Getting exercise
Getting away from the noise back home
Getting away from the bright lights back home
Seeing distant or unobstructed views
Enjoying the sounds of nature
Relaxing physically
On your last visit to [insert park name], which one of the following best describes the type of group you were with? Were you …
(Read each choice, and mark only ONE) (new)
Alone Skip to #14
With your family
With friends
With family and friends
With an organized group or tour (see instruction in #13)
(DK/NS)
(NA/Refused)
Including yourself, how many people were in your travel party when you visited [insert park name]?
If traveled with an organized group or tour on #12, also display:
I need the size of your personal group, not the total number of people in your tour.
(Open-ended, record response) (new)
(Record number, 1-96)
(97=97 or more)
(98=DK/NS)
(99=NA/Refused)
On your last visit to [insert park name], did you or any member of your personal group participate in any of the following in the park? (2000, #7, but added “any member of your group” and “in the park” to introduction)
randomize order of presentation; for each one, include the following note:
(If needed: On your last visit, did you or any member of your personal group do this?) (1=Yes; 2=No; 8 = DK/NS; 9=NA/Ref for each item)
Go walking or day-hiking for at least 30 continuous minutes? (added “for at least 30 minutes,” the Surgeon General’s recommendation for healthful activity)
Go jogging or running for at least 30 continuous minutes? (new)
View or photograph animal life, including birds? (new)
View or photograph plant life, including flowers or trees? (new)
Engage in snow sports, such as skiing, snowmobiling, or sledding? (new)
Engage in water activities, such as swimming or boating? (new)
On your last visit to [insert park name], did you or any member of your personal group use any of the following programs or services?
randomize order of presentation; for each one, include the following note:
(If needed: On your last visit, did you or any member of your personal group do this?) (1=Yes; 2=No; 8 = DK/NS; 9=NA/Ref for each item)
Attend a ranger-led activity, such as a tour or talk (2000, combined #7d & #7e)
Talk informally with a ranger (new)
View outside exhibits (new)
View indoor exhibits (new)
Attend a cultural demonstration or performance (2000, #7f)
Read the park brochure or newspaper (new)
Go to the visitor center (new)
Watch movies or videos about the park (new)
Participate in the Junior Ranger program (new)
If Yes to at least two of these, proceed to 15a; otherwise, skip to #16
15a. Which ONE of those programs or services added the most enjoyment to your visit to [insert park name]? (new)
(Do NOT read choices; code ONE response)
(Attend a ranger-led activity, such as a tour or talk)
(Talk informally with a ranger)
(View outside exhibits)
(View indoor exhibits)
(Attend a cultural demonstration or performance)
(Read the park brochure or newspaper)
(Go to the visitor center)
(Watch movies or videos about the park)
(Participate in the Junior Ranger program)
(Other)
(DK/NS)
(NA/Ref)
16. On your last visit to [insert park name], how did you travel from your home to the park? Please tell me all forms of transportation you used to reach the park on that visit.
(Read choices, one at a time; mark ALL that the respondent mentions) (new)
Car, truck, or SUV
Recreational vehicle or motor home
Airplane
Tour bus or tour van
City bus or subway
Train or long-distance passenger bus
Cruise ship or other water transportation
(Any other travel mode that cannot be coded above)
(DK/NS)
(NA/Refused)
If Yes to car/truck/SUV or to RV/MH, or both, proceed to 16b; otherwise, skip to 17.
16a. For that visit, did you use …
(Read both choices, one at a time; mark ALL that apply) (new)
A rented car, truck, or SUV
A rented RV or motor home
(No rented vehicle)
(DK/NS)
(NA/Refused)
17. Now we’re interested in why people don’t visit National Park System units more often. I’m going to read a series of statements. I’d like you to think of your own experiences, and tell me if you strongly agree, somewhat agree, somewhat disagree, or strongly disagree with each statement:
randomize list, each respondent presented with 10 of the 16 items in random order
for each one, include the following note:(If needed: Would you say you “strongly agree,” “somewhat agree,” “somewhat disagree,” or “strongly disagree”?) (8 = DK/NS; 9=NA/Ref for each item)
a. Entrance fees are too high (2000, #14)
b. Other costs of visiting National Park System units, such as hotel and food costs, are too high (2000, #16)
c. National Park System units are not safe places to visit (2000, #17)
d. It takes too long to get to any National Park System units from my home (2000, #18)
e. National Park System units are too crowded (2000, #19)
f. It is difficult to find a parking space within National Park System units (2000, #20)
g. National Park System units are not accessible to persons with physical disabilities (2000, #21, changed from “to the physically disabled”)
h. I just don’t know that much about National Park System units (2000, #22)
i. Reservations at National Park System units have to be made too far in advance (2000, #23)
j. National Park Service employees give poor service to visitors (2000, #24)
k. National Park System units are unpleasant places for me to be (2000, #25, but substantially modified; “for persons of my race and ethnicity or gender” dropped; “uncomfortable” changed to “unpleasant.”)
l. There isn’t enough information available about what to do once inside a National Park System unit (2000, #26)
m. My kids are not interested in visiting national parks (new)
n. I have already been to all the parks I want to see (new)
o. I prefer to spend my free time doing electronic activities, like watching videos, enjoying computer games, or surfing the Internet (new)
p. I don’t want to be around park rangers during a vacation because they remind me of police (new—from African American focus group)
In your opinion, what is the ONE most important thing the National Park Service can do to encourage you to visit units within the National Park System?
(2000, #16, most common open-ended responses listed)
(Do NOT read choices; code the ONE most important thing)
(Nothing, no suggestions)
(Keep up the good work)
(Advertise/publicize more)
(Lower park and transportation costs)
(More parking)
(Accessible, closer in proximity)
(More lodging in parks)
(Keep parks clean)
(More benches, restrooms, better maintenance)
(More variety in programs and events)
(Reduce commercialization)
(Reduce traffic congestion and crowds)
(Improve security, safety, protection)
(More accessible to people with disabilities)
(More Internet/cellphone/wireless connectivity) (new)
(Protect natural/cultural resources, wildlife, water, vegetation) (new)
(Include more cultural sites/stories of interest to my race/ethnic group)(new)
(Provide more space for organized activities and games such as basketball, soccer, etc.) (new)
(Increase the number of historical sites and monuments which represent people of my race/ethnicity) (new)
(Increase motorized riding in the parks, including snowmobiles and ATVs) (new)
(Allow hunting in the parks) (new)
If no answer to 5a and to 5a1 and to 5a2, skip to D1.
During your last visit to [insert park name], did you stay overnight there, either in the park itself, in a neighboring community, or both? (2000, #9, but mode of community stay specified in a separate question in 2007)
(No, did not stay overnight) Skip to #22, Sources
(Stayed in park only)
(Stayed in neighboring community only) Skip to #21, Community
(Both in park and neighboring community)
(DK/NS) Skip to #22, Sources
(NA/Ref) Skip to #22, Sources
20. Where in the park did you stay?
(Read choices one at a time, and mark ALL that apply) (new)
A lodge or hotel within the park
A campground for tents or RVs
Overnight camping by backpack, horseback, boat, or aircraft
(Any other park lodging that cannot be coded above)
(DK/NS)
(NA/Ref)
If #19=3 (park only), skip to #22, Sources
Where in the community did you stay?
(Read choices one at a time, and mark ALL that apply)
A hotel, motel, inn, or bed & breakfast (2000, #9, added B&B)
A neighboring campground for tents or RVs (2000, added “Tent or RV”)
With family or friends (2000, #9)
(Any other community lodging that cannot be coded above)
(DK/NS)
22. Thinking about your last visit to [insert park name], please tell me if you used any of the following information sources to plan your visit, either before you arrived or during your trip.
randomize order of presentation; for each one, include the following note:
(If needed: For your last visit, did you use information from …) (1=Yes; 2=No; 8 = DK/NS; 9=NA/Ref for each item)
Friends or relatives (2000, #10g)
Your own previous visit (new)
Travel agent (2000, #10j)
State tourism office (new)
Tour operator (2000, #10k, dropped “Local”)
Guidebook, magazine, or newspaper (2000, combined #10h & #10i)
National Park Service Internet website (www.nps.gov) (2000, #10c)
Other Internet website (2000, #10d)
Park radio (new)
23. Before this survey, were you aware of any of the following ways that people can help national parks? (new)
23a. Volunteering time to do needed jobs in parks.
(Yes)
(No)
(DK/NS)
(NA/Refused)
23b. Making a monetary donation.
(Yes)
(No)
(DK/NS)
(NA/Refused)
23c. Donating things (such as equipment, search and rescue services, and historical artifacts to parks).
(Yes)
(No)
(DK/NS)
(NA/Refused)
23d. Joining a park friends association.
(Yes)
(No)
(DK/NS)
(NA/Refused)
If “Yes” to #23a, #23b, #23c, or- #23d, proceed to #24; otherwise, skip to text before #25)
24. Have you actually helped a park system unit in any of these ways? (new)
(Do NOT read choices; check ALL that respondent reports)
(Volunteered time)
(Donated money)
(Donated things)
(Joined a friends association)
(DK/NS)
(NA/Refused)
Now, please think about a large national park like Yellowstone, Grand Canyon, or Great Smoky Mountains. These parks have a variety of natural sounds, like birds singing, water flowing in rivers, and sounds of wildlife, including elk, foxes, or frogs. These parks also may have other sounds, such as vehicle and aircraft engines and the sounds of construction and maintenance equipment.
25. We’d like to know how important it is to hear the sounds of nature for enjoying an experience in the wild or undeveloped areas of a large national park. For you personally, is it … (new)
(Read choices; mark ONE response.)
Not important at all
Somewhat important
Very important
Extremely important
(DK/NS)
(NA/Refused)
26. How much do you agree or disagree with the following statement? I should be able to go a national park and not hear mechanized sounds like engine noise and cell phones when I am out hiking in wild or undeveloped areas. (new)
(Read choices; mark ONE response.)
Strongly agree
Somewhat agree
Somewhat disagree
Strongly disagree
(DK/NS)
(NA/Refused)
Now think about the historical parks, like Gettysburg, Valley Forge, or the cliff dwellings at Mesa Verde. Parks like these honor historic events or preserve early lifestyles and cultures. The exhibits there may have cultural and historical sounds, such as musket fire, folk songs, or Native American music.
27. We’d like to know how important it is to hear cultural and historical sounds like that for enjoying an experience in one of those parks. For you personally, is it … (new)
(Read choices; mark ONE response.)
Not important at all
Somewhat important
Very important
Extremely important
(DK/NS)
(NA/Refused)
The large national parks like Yellowstone, Grand Canyon, and Great Smoky Mountains are known for their natural resources. For example, they have plants, animals and fish, remote areas and wilderness, fresh air, lakes and rivers, and starry night skies. I’m going to read you some statements about these parks and ask you how much you personally agree or disagree with each statement. (new)
randomize order of presentation; for each one, include the following note:
(If needed: Would you say you “strongly agree,” “somewhat agree,” “somewhat disagree,” or “strongly disagree” that the National Park Service should do this in our large national parks?) (8 = DK/NS; 9=NA/Ref for each item)
Non-native plants that do not occur naturally in these parks should be removed
Non-native animals or fish that do not occur naturally in these parks should be removed
Wild animals or fish that used to occur naturally in these parks should be restored
Aircraft flights over wild and undeveloped areas of these parks should be allowed
These national parks are places where there should be no air pollution from communities and industries
These national parks are places I should be able to go and see the night sky without interference of artificial lights from nearby communities.
These national parks are places where there should be no water pollution from communities, industries, and agriculture
Basic visitor facilities in these parks, such as roads, trails, restrooms, and water fountains, should be provided
Developed visitor facilities in these parks, such as lodges, restaurants, and stores, should be provided
The number of private vehicles in these parks during the busiest periods should be limited
All types of transportation should be allowed in these parks, including jet-skiing and snowmobiling
(DK/NS option for each item)
D1. To finish, I need to ask you some questions so we can be sure our sample is representative.
If LandLine was not asked, skip to D1C (because this is a landline interview with location known from area code). If Landline = 1 or 2 or 3 or 4, proceed to D1A to determine place of residence (because this is a cell interview with location unknown).
D1A. ZIP:
What is the 5-digit zipcode where you currently live? (new)
(open-ended; 88888=DK; 99999=NA)
If NumAd=1 and sample = cell, skip to D1D. If sample = RDD, skip to D1C; otherwise, proceed to NumCell.
D1B. NUMCELL:
How many of the adults, age 18 and older, who currently live in your household have a cell phone on which they can receive calls? (new)
(If needed: Please count any adult in your household who is currently the main user of a cell phone, even if someone else is paying for it.)
(Open-ended, record response)
(1-26 of the adults in the household)
(code 27 for 27 or more)
(28=DK/NS)
(29=NA/Refused)
(30=“All”)
If sample = cell, skip to D1D.
D1C. This question is about the residential phones in your household, not counting cell phones, business lines, or numbers that are only used for a computer or a fax. Including the phone number that we’re talking on right now, how many different residential phone numbers ring into this household and can be answered by a person? (new)
(0-6 residential phone numbers)
(code 7 for 7 or more)
(8 for DK/NS)
(9 for NA/Refused)
D1D. Are you, or is anyone in your household, a National Park Service employee? (2000 #1)
(Yes)
(No)
(DK/NS)
(NA/Refused)
D2. What is the highest grade of school or year of college that you have completed?
(If needed, read choices) (2000, #D1, combined “Masters” and “Doctoral/Law”)
(Up to 8th grade)
(9th to 11th grade)
(High school graduate or GED certificate)
(Some college, no degree)
(Degree from technical school or community college)
(University degree—BA/BS)
(Some graduate school, no advanced degree)
(Graduate degree—MA/MS/JD/MD/PhD, etc.)
(DK/NS)
(NA/Refused)
D3. Are you single, married, living with a life partner, divorced, separated, or widowed?
(If needed, read choices) (2000, #D2)
(Single)
(Married)
(Living with a life partner)
(Divorced)
(Separated)
(Widowed)
(DK/NS)
(NA/Refused)
D4. How often do you use the Internet? Do you use it many times per day, two or three times a day, only once or twice a week, a few times each year, or not at all? (2000, #D3a, deleting “No use yet, but would like to”)
(Many times per day)
(Two or three times a day)
(Once or twice a week)
(A few times each year)
(Not at all)
(DK/NS)
(NA/Refused)
D5. Are you Hispanic or Latino? (2000, #D4)
(If respondent is female, say Latina.)
(Yes)
(No)
(DK/NS)
(NA/Refused)
D6. I’m going to read a list of racial categories. Please select one or more to describe your race. Are you . . .
(Read choices one at a time, and mark ALL that apply) (2000, #D5, “Other race” deleted from response list and reworded to conform to current format on OMB website)
American Indian or Alaska Native
Asian
Black or African American
Native Hawaiian or other Pacific islander
White
(DK/NS)
(NA/Refused)
D7. In what year were you born? (2000, #D6)
(Record response)
(DK/NS)
(NA/Refused)
D8. Does anyone in your household have a disability or impairment that could cause them to encounter access or service problems during a visit to a unit of the National Park System? (new)
(Yes)
(No) #D9, Children
(DK/NS) #D9, Children
(Refused) #D9, Children
D8a. What kind of disability or impairment? (If needed, read choices) (new)
(Sensory—blindness, severe vision impairment) (American Community Survey)
(Sensory—deafness, severe hearing impairment) (ACS)
(Physical—substantial limits on walking, climbing, reaching, lifting, carrying due to chronic medical disorder or other physical condition) (ACS, adding “due to chronic . . .” from Dept. of HHS)
(Mental—sustained [6 months or more] difficulty learning, remembering, or concentrating) (ACS)
(Any other condition that cannot be coded above)
(DK/NS)
(NA/Refused)
D9. How many children under the age of 18 currently live in your household? (2000, #D8)
(0-26 children in the household)
(code 27 for 27 or more)
(DK/NS)
(NA/Refused)
D10. Which one of the following income groups best describes your total household income in 2006 before taxes? Please stop me when I read the correct category. (2000, #D10, response categories changed to match Census reporting)
Less than 10,000 dollars
10,000 to 25,000 dollars
25- to 50,000 dollars
50- to 75,000 dollars
75- to 100,000 dollars
100,000 to 150,000 dollars
Over 150,000 dollars
(DK/NS)
(NA/Refused)
D11. (Gender) (2000, #D10)
(Respondent’s gender; code without asking, unless unclear)
(If needed: And I’m required to ask, are you male or female?)
(Male)
(Female)
That concludes the survey. Thank you very much for participating! Do you have any questions for me?
Thanks again
D12. (Language of interview) (new)
(Code without asking)
(English only)
(Mainly English, some Spanish)
(Both English and Spanish, about equally)
(Mainly Spanish, some English)
(Spanish only)
| File Type | application/msword |
| File Title | RESPONSES TO OMB QUESTIONS ON |
| Author | Gaia |
| Last Modified By | mmcbride |
| File Modified | 2007-10-17 |
| File Created | 2007-10-17 |