HomeMy WebLinkAboutWORK ORDER - RFP - 7226 SURVEY SERVICES - MISC PUBLIC OPINION SURVEYSWORK ORDER FORM
PURSUANT TO AN AGREEMENT BETWEEN
THE CITY OF FORT COLLINS
AND
National Research Center
DATED: 5/26/2011
RFP
Work Order Number: 7226%�
Purchase Order Number: 32 51
Project Title: Natural Areas Survey and Visitor Count
Commencement Date: May 26, 2011 C
Completion Date: August 31, 2012
Maximum Fee:4 J-y 60-d . d
Project Description: Survey residents about their perceptions of the Natural Areas Program and count
the number of visitors to each site.
Scope of Services: Attached.
Service Provider agrees to perform the services
identified above and on the attached forms in
accordance with the terms and conditions contained
herein and in the Services Agreement between the
parties. In the event of a conflict between or
ambiguity in the terms of the Services Agreement and
this work order (including the attached forms) the
Services Agreement shall control.
The attached forms consisting of 6vG ( E� )
pages are hereby accepted and incorporated herein
by this reference, and Notice to Proceed is hereby
give.
National Research Center, Inc.:
By:� Q
Date: S l- ao I I
CC: Purchasing
City of Fort Collins: 1
Submitted by: _Zoe Whyman
Project Manager
Reviewed By:
Date: 5/26/11
Date:
Approved by: G�
N� ural Areas Program Mgr.
Date:
Approved by:
Director of Purchasing and Risk Management
(if over $60,000.00)
Date:
City of Fort Collins Natural Areas Program Survey and Visitor Count
Scope of Work for 2011 Natural Areas Survey
1) Implement a visitor count study
Work with NAP to understand the complexities of the natural areas and their points of entry, develop a
visitor count protocol to capture seasonal variation in visitation.
We will work with NAP to devise the observation schedule and data collection protocols. We will work with
NAP to understand all the various areas to be included, what the data entry points are, and any issues that
may arise during observation. We will help to devise a data collection schedule to allow the most efficient
use of observation hours and other tools such trail/ vehicle counter and other previously collected data. We
will want to capture the most data, but to ensure representation from both high and low use areas.
Help with training on visitor count study.
We will work with NAP to train those collecting the data, but our organization will not do the data
collection.
Tabulate the visitor counts, calculate estimates of total visitation, summarize the results and methods in
the final report. Provide a spreadsheet of the count data to NAP.
We will analyze and summarize the results and methods. We will use the available data to create the most
accurate estimates possible of total visitation, and visitation by site and season.
2) Implement a resident survey
Task 1: Development of Survey Questions and Mailing Materials
It is our experience that, to achieve the best results, time must be invested up -front to craft the right survey
instrument and to tie survey design to the overall goals and objectives of the project. We regard the
questionnaire development process as an iterative one, with drafts being reviewed and revised by a project
oversight committee or point person and our team.
NRC will work with the City to develop a comprehensive questionnaire that covers the objectives of the
study. We will start with the 2006 visitor survey instrument and the list of desired modifications/additions
from the RFP to draft the questionnaire. We will work with the City in an iterative process until a final
questionnaire is developed that is approved by everyone. The survey will be 3 pages in length.
Task 2: Sampling
Selecting survey recipients
In survey research, it is desirable for every household (or every adult person living in a household) to have
a known and equal chance of participating in the survey. A sample of all households is selected to be given
an opportunity to participate. Sampling is based on probability theory; the idea that if respondents are
chosen randomly from the larger population, the results from that random sample will be very close to
what we would get by interviewing every member of the population.
The challenge is to find a list of all households from which survey recipients can be randomly chosen. For a
mailed survey, mailing lists exist from the United States Postal Office which include all households with a
mailing address.
All households located within the boundaries the City determines are appropriate for this study would be
eligible for the survey. We will geocode the location of each address to assure it is within the study
boundaries (as well as any desired subareas within the overall boundaries, if desired).
National Research Center, Inc.
e e City of Fort Collins Natural Areas Program Survey and Visitor Count
We will use systematic sampling from the list of eligible households to select the sample of households.
Systematic sampling is a procedure whereby a complete list of all possible addresses is culled, selecting
every Nth one until the appropriate number of households is sampled. Not only will NRC scientifically and
randomly sample households,.but we also will select without bias the household member to participate. In
cases where more than one household member is 18 or older, a procedure called "the birthday method"
will be used to select the survey respondent without bias. The birthday method requests that the
respondent be the eligible adult in the household who most recently had a birthday (irrespective of the
year of birth). This methodology helps ensure the attitudes expressed by our respondent "sample" closely
approximate the attitudes of all adult residents living in the study area. Without this step in survey
research, it is likely that results will be biased towards those who are more sedentary and those without
jobs (who may have different opinions about some services).
Choosing the number of completed surveys ("sample size")
The relationship between sample size and precision of estimates (the 95% confidence interval or margin of
error) is shown in the adjacent table. The margin of error or 95% confidence interval is determined solely
by the number of completed surveys, and not the size of the .
population from which the sample is drawn, as long as the
"population" size is greater than about 4,000.
The table can be interpreted as follows: If 61% of 400
respondents said they would support a mandatory vehicle
emissions program in Fort Collins even though it is not
Federally required, we can be confident that, had we gotten
ratings from all residents, somewhere between 56% and
66% would have supported mandatory vehicle emissions.
We will send surveys to 1,600 households, for a likely final
completed sample size between 400 and 600, which would
provide a 95% confidence interval of plus or minus 5 percent.
� Sample '
,-_Size
Mf argin:oError"��,
(95q Confidence4nterval) \.
100
- ±9.8%
zoo
±6.9%
400
+4 9%
600
±4.0%
800
±3.5%
11200
±2.8%
2,000
±2.2%
Task 3: Survey Administration
Below is outlined our survey administration procedures. We encourage our clients to inform the public
about the survey through local newspapers, newsletters, community meetings and any other appropriate
media. The publicity provides an opportunity to inform them about the purposes of the survey and the uses
- for programming and policy - to which it will be put. The publicity, furthermore, will assist the survey
process by increasing willingness of potential respondents to participate.
Mailed Survey Data Collection
As described above, a sample of addresses from within Fort Collins and/or Larimer County will be selected
to be part of the survey. This database of addresses will be processed for certification and verification. NRC
vendors use CASS"/NCOA software that relies on the USPS National Directory information to verify and
standardize the address elements and assign each a complete, nine -digit zip code where possible. In
addition, the software will sort and barcode the addresses, allowing significant postage discounts. We
propose mailing all materials pre-sorted first class. This ensures a fast delivery time of the materials, but
provides an opportunity for a discount on the full first-class postage rates.
Our mailhouse will oversee the printing and preparation of the prenotification postcards, survey packets
and reminder postcards. As a part of the quality control process, proofs of the survey materials are
reviewed by NRC staff before the final job is printed. We will include a point person at Fort Collins as a
"seed" in the mailing list, so they will receive the mailing materials at the same time as the sampled
recipients.
National Research Center, Inc.
City of Fort Collins Natural Areas Program Survey and Visitor Count
We propose three contacts with each sampled household:
1) A prenotification announcement, informing the household members that they have been selected to
participate in the community survey will be sent to each household. This announcement should be
scheduled to arrive about a week before the survey packet.
2) Approximately one week after mailing the prenotification, each household would be mailed a survey
containing a cover letter (ideally signed by the mayor or other appropriate high-level official or staff
member) enlisting participation. The packet would also contain a postage paid return envelope that
would send all completed surveys directly for data entry.
3) A reminder letter and survey, will be scheduled to arrive 1 week after the first survey. The second cover
letter asks those who have not completed the survey to do so and those who have already done so to
refrain from turning in another survey. We mail the survey twice because anonymity is promised in the
cover letter to enhance the likelihood of honest responses. We take this implicit contract with
respondents as a serious principle of the survey trade, which, if violated, harms the survey research
industry no less than the client or respondent. We can also check for duplicates if we include a survey
code for later mapping of responses. Generally we find only one or two duplicates in our responses.
When the survey administration mode has been chosen, a sample size selected, and the survey materials
finalized, NRC will produce a memo for the City outlining the survey protocols and containing the final
survey materials.
Task 4: Data Tabulation
Survey processing
Mailed surveys will be returned to an NRC affiliate directly via postage -paid business reply envelopes
where staff will assign a unique identification number to each survey. Additionally, each survey will be
reviewed and "cleaned" as necessary. For example, a question may have asked a respondent to pick two
items out of a list of five, but the respondent checked three; NRC staff would choose randomly two of the
three selected items to be coded in the dataset. We have found that very little cleaning is needed on most
surveys due to our expertise in question construction and survey formatting. However, extensive cleaning
tends to be needed in surveys containing complicated question structure or skip patterns.
Once all surveys have been assigned a unique identification number, they will be entered into an electronic
dataset. This dataset will be subject to a data entry protocol of "key and verify," in which survey data are
entered twice into an electronic dataset and then compared. Discrepancies are evaluated against the
original survey form and corrected. Range checks as well as other forms of quality control will also be
performed.
Weighting the Data
The last step in.preparing the data for analysis is to weight the data to reflect the demographic profile of the
Fort Collins/Larimer County population. Weighting is an important method to adjust for potential non -
response bias. In general, residents with certain characteristics (for example: those who are younger or
rent their homes) are less likely to participate in surveying, whatever the data collection mode. Weighting
allows us to look at the demographic profile of residents who returned the survey compared with the US
Census profile. We consider these disparities along with others and accordingly increase or decrease the
weight of each respondent to mimic as closely as possible the population demographics. The weighting
variables to be considered will be all those demographics included on the survey, as well as any geographic
variable used for sampling. NRC has extensive experience with complex weighting schemes required with
stratification.
National Research Center, Inc.
City of Fort Collins Natural Areas Program Survey and Visitor Count
Basics of Data Analysis and Reporting
It is paramount that the result of this study be more than a nice looking report that sits on a shelf, and as
such we will work with the City of Fort Collins to ensure that we ask the right questions to feed into an
actionable analysis plan.
For quantitative analysis, we rely on the Statistical Package for the Social Sciences (SPSS). We believe that
analysis must be replicable and leave a clear path. To this end, we keep every label and command run in
SPSS in a syntax file available for audit and re -running, as necessary. We also have trained clients on SPSS
analysis and, for small recurring analyses, how to use Microsoft® Excel.
We will code open-ended responses using both an emergent approach, where themes are revealed through
the analysis, combined with a deductive approach, where a scheme or codes are predetermined and
applied to the data. Techniques include word counts, content analysis, comparative analysis, componential
analysis and taxonomic analysis, among others and codes are applied based on the unit of analysis. Our
qualitative analysis of verbatim responses will be done using Microsoft® Access, Microsoft® Excel or QSR
qualitative software, depending on the types and complexity of the open-ended questions.
The data and report will undergo a thorough quality assurance review. We will audit the original data files,
our syntax/analysis files, compare automatically generated output to the formatted output in the report
and data check all numbers and text prior to submitting the report.
We will provide basic descriptive statistics such as frequencies and means summarizing the responses to
the survey questions. If it makes sense to create scales or indices that summarize information from a
number of survey items, we can use factor analysis and other psychometric testing to create valid factors.
Additionally, NRC staff often have employed other advanced data analysis techniques to transportation
survey findings to determine what the key drivers of respondent satisfactions are (using linear or logistic
regression to demonstrate the factors that are most important in respondent satisfaction), cluster analysis
(market segmentation) to identify groupings of respondents whose members share characteristics but
whose clusters differ in important way. For example, in a study we did for the Regional Air Quality Council
on ground level ozone pollution, we found there were four clusters or segments of the population in
regards to attitudes towards and behaviors to ameliorate ground level ozone, ranging from the "Negative
Normans" (comprising 12% of the population) who did not feel ground level ozone was a problem and were
unlikely to make any behavior changes to decrease ground level ozone to "The Eager" (comprising 22% of
the population) who reported already making lifestyle changes to reduce ground -level ozone pollution and
who were ready to be influenced to change behavior.
Upon completion of the data analysis, NRC can provide an electronic copy of the dataset to the City of Fort
Collins in a variety of formats. Typically, clients prefer to receive the data as an Excel worksheet, but we can
also provide an ASCU dataset, an Access dataset, or an SPSS dataset. We will provide information about the
layout and value labels of the data so that you will have no trouble accessing the raw data. As a note of
caution: while we can provide the dataset in Excel, including the appropriate weight for each case, applying
the weights in Excel is quite difficult. SPSS or another statistical application maybe required to apply the
statistical weights and derive the identical results to what would be displayed in the report of results.
However, we keep all the data files on our own server, and will be happy to run additional crosstabulations
for you (at no additional cost) as questions come up as you use and disseminate the report.
Task 5: Reporting the Results
Report of Results
NRC understands that most readers want to understand the broader patterns and interpretations of data,
so our professionally designed graphics are engaging and pop with clear information. Our reports are
comprehensive and include technical and detailed numbers and information, but not at the expense of the
National Research Center, Inc.
.,May 20"2011 City of Fort Collins Natural Areas Program Survey and Visitor Count
day-to-day usefulness of the documents. We are careful about our communication. You will not need a
degree in statistics to understand the survey results. All the technical details are in appendices for those
who wish to read them (and so that the survey can be repeated by any vendor you choose), but we make
sure the report body tells the story of the survey results in a stylish, colorful, informative way. For further
ease, we also include an executive summary that gives a quick overview of results while highlighting key
findings. We will prepare the report in Microsoft® Word (as well as convert documents into a PDF format).
Drafts of the report will be provided to the City and the final report will incorporate any comments we
receive from you. We will meet with you to review the draft report, and answer any questions or comments
you have.
We will provide an electronic copy of the report in Microsoft Office Word® format along with a version in
PDF format (readable by Adobe Acrobat). We will also provide a hard copy. of the report. .
Consulting and Continued Support '
We know that an effective report will be reviewed and discussed by many.people in Fort Collins after we
have completed the survey projects, and that new questions may arise. We commit to supporting these
discussions by offering crosstabulations and other additional analyses of results when needed. And, though
we will not abandon you with your data, we always will provide you with the electronic dataset of your
responses for your future reference and use.
National Research Center, Inc.