HomeMy WebLinkAboutStaff Report 05 12/03/2001DEG G, 2 1001
if'
GodbeHosearch & Analysis
Conducted for:
The City of Petaluma
November 29, 2001
60 Slane Pine Road 95 South Marker Street, Suite 390 445 South Figueroa Street, 2699 649 Grand Avenue, Suite G
Hal[ Moon Hay CA 94019-1739 San fuse CA 95113-2350 los Angeles CA 90071-1631 Carlsbad CA 92008-2365
Phone 650/712-3137 Phone 408/289 Phone 213/624-8863 Phone 760/730-2941
Fax 650/712-3131 Fax 408/288-9212 Pax 213/624-8864 Fax 7601720-4706
'Table of cowenls
Table of Contents
Listof Tables .................................................................................................................................................................
i
Listof Figures...............................................................................................................................................................
iv
Introduction..................................................................................................................................................................
l
ExecutiveSummary .....................................................................................................................................................2
Methodology...............................................................................................................................................................13
Issues of Importance - Special Tax Sample...............................................................................................................
20
First Ballot Test - Special Tax Sample.......................................................................................................................
23
Parcel Tax Threshold - Special Tax Sample..............................................................................................................
26
City Sales Tae Threshold - Special Tax Sample........................................................................................................
27
Spending Projects - Special Tax Sample...................................................................................................................
28
Arguments Supporting the Measure - Special Tax Sample.......................................................................................
31
Arguments Opposing the Measure - Special Tax Sample..........................................................................................
34
Second Ballot Test - Special Tax Sample...................................................................................................................
36
Park and Recreation Facilities - Special Tax Sample...............................................................................................
40
Additional Demographic and Behavioral Measures - Special Tax Sample..............................................................
42
Issues of Importance - Assessment.............................................................................................................................
47
First Ballot Test - Assessment Sample........................................................................................................................
50
Assessment Threshold -Assessment Sample..............................................................................................................
53
Spending Projects - Assessment Sample....................................................................................................................
54
Arguments Supporting the Measure - Assessment Sample........................................................................................
57
Arguments Opposing the Measure - Assessment Sample...........................................................................................
60
Second Ballot Test - Assessment Sample....................................................................................................................
62
Park and Recreation Facilities - Assessment Sample................................................................................................
66
Additional Demographic and Behavioral Measures - Assessment Sample...............................................................
68
Chy Of Petaluma-Slreels and Roads Study Godbe Research E Analysis
Pep
List of Ttbhs
List of Tables
Table1.
Methodology................................................................................................................................13
Table2.
Subgroup Labels..........................................................................................................................15
Table 3.
Guide to Statistical Significance - Assessment Sample...............................................................16
Table 4.
First Ballot Test Special Tax Sample by Voting Propensity.......................................................18
Table 5.
'Means' Questions and Corresponding Scales ............................................... .................
............ 19
Table 6.
Issues of Importance by Age.......................................................................................................
21
Table 7.
Issues of Importance by Amount of Information on Local Streets ............................................
22
Table 8.
First Ballot Test by Perceived Condition of Petaluma's Streets..................................................24
Table 9.
First Ballot Test by Propensity to Vote.........................................................................................24
Table 10.
First Ballot Test by Household Party Type...................................................................................
25
Table 11.
First Ballot Test by Homeownership Status and Age..................................................................
25
Table 12.
Proposed Spending Projects by First Ballot Test.........................................................................
29
Table 13.
Proposed Spending Projects by Household Party Type...............................................................
30
Table 14.
Arguments Supporting the Measure by Swing Voters and Gender ............................................
32
Table 15.
Arguments Supporting the Measure by Age................................................................................
33
Table 16.
Arguments Opposing the Measure by Income...........................................................................
35
Table 17.
Arguments Opposing the Measure by Household Party Type .....................................................
35
Table 18.
Second Ballot Test by First Ballot Test.......................................................................................
37
Table 19.
Second Ballot Test by Income.....................................................................................................
37
Table 20.
Second Ballot Test by Propensity to Vote ........................................... .......... ........................
I....... 38
Table 21.
Second Ballot Test by Household Party Type..............................................................................
38
Table 22.
Second Ballot Test by Gender and Condition of Petaluma's Streets .........................................
39
Table 23.
Issues of Importance by Age.......................................................................................................
48
Table 24.
Issues of Importance by Household Party Type..........................................................................
49
Table 25.
First Ballot Test by Propensity to Vote........................................................................................
51
Table26.
First Ballot Test by Age................................................................................................................
51
Table 27.
First Ballot Test by Household Party Type...................................................................................
52
Table 28.
First Ballot Test by Gender and Perceived Condition of Streets ..................................................
52
Table 29.
Proposed Spending Projects by First Ballot Test.........................................................................
55
Table 30.
Proposed Spending Projects by Household Party Type..............................................................
56
Table 31.
Arguments Supporting the Measure by Gender and Swing Voters .............................................
58
Table 32.
Arguments Supporting the Measure by Age................................................................................
59
Table 33,
Arguments Opposing the Measure by Amount of Information on Streets ................................
61
Table 34.
Arguments Opposing the Measure by Party................................................................................
61
Table 35.
Second Ballot Test by First Ballot Test........................................................................................
63
Table 36.
Second Ballot Test by Income.....................................................................................................
63
Cily o(Pelalumn - Saeels and Roads Study Godbe Rereardi & Analysis
Pagc ii
List of Thin
Table 37. Second Ballot Test by Propensity to Vote and Gender ....
Table 38. Second Ballot Test by Household Party Type ..................
Table 39. Second Ballot Test by Age ...............................................
Ciq, of Petaluma -Streets and Roads Stody
Cadbe Research At Analysis
Page ill
List of Figures
List of
Figures
Figure1.
Issues of Importance....................................................................................................................
20
Figure 2.
Condition of Petaluma's Streets, ............................................................... - ................................
22
Figme3.
First Ballot Test............................................................................................................................23
Figure 4.
Parcel Tax Threshold..................................................................................................................26
Figure 5.
City Sales Tax Threshold.............................................................................................................
27
Figure 6.
Proposed Spending Projects.......................................................................................................
28
Figure 7.
Arguments Supporting the Measure............................................................................................
31
Figure B.
Arguments Opposing the Measure...............................................................................................
34
Figure9.
Second Ballot Test........................................................................................................................
36
Figure 10.
Support for the Measure with Park and Recreation Facilities
Without Additional Revenue........................................................................................................
40
Figure 11.
Support for the Measure with Park and Recreation Facilities
with $4 Million in Additional Revenue.......................................................................................
41
Figure 12.
Length at Current Residence.......................................................................................................
42
Figure 13.
Amount of Information on Local Streets....................................................................................
42
Figure 14.
Own Commercial Property or Apartments in the City................................................................
43
Figure 15.
Newspapers that are Read............................................................................................................43
Figure 16.
Iousehold Income.......................................................................................................................43
Figure17.
Gender..........................................................................................................................................44
Figure18.
Zip Cade of Residence..................................................................................................................45
Figure19.
Age................................................................................................................................................
45
Figure20.
Party.............................................................................................................................................
45
Figure21.
Household Party Type..................................................................................................................46
Figure 22.
Issues of Importance....................................................................................................................
47
Figure 23.
Condition of Petaluma's Street...................................................................................................
49
Figure24.
First Ballot Test............................................................................................................................
50
Figure25.
Assessment Threshold..................................................................................................................53
Figure 26.
Proposed Spending Projects........................................................................................................54
Figure 27.
Arguments Supporting the Measure............................................................................................
57
Figure 28.
Arguments Opposing the Measure ...............................................................................................
60
Figure29.
Second Ballot Test........................................................................................................................
62
Figure 30.
Support for the Measure with Park and Recreation Facilities
Without Additional Revenue........................................................................................................
66
Figure 31.
Support for the Measure with Park and Recreation Facilities
with $4 Million in Additional Revenue.......................................................................................
67
Figure 32.
Length at Current Residence.......................................................................................................
68
City of Petaluma - Slreels and Roads Study Codbe Researdi S Analysis
Page it
List of Figures
Figure 33.
Perceived Condition of Petaluma's Streets and Roads .............
Figure 34.
Own Commercial Property or Apartments in the City ..............
Figure 35.
Newspapers that are Read .........................................................
Figure 36.
Household Income .................. ............................................ .....
Figure37.
Gender........................................................................................
Figure 38.
Zip Code of Residence................................................................
Figure39.
Age..............................................................................................
Figure40.
Party...........................................................................................
Figure 41.
Household Party Type... ...... -- ................................. ...............
City ofPelalunta- Stroels and Roads Study Godbe Research &Analysis
Page v
III[] Into
Introduction
Godbe Research & Analysis (GRA) and Shilt's Consulting (SCI) are pleased to present the
results of a voter opinion research project conducted for the City of Petaluma. This report is
organized into the following sections:
Executive summary The Erecudve Summary includes a summary of the 1(e), Findings from the survey and a
Conclusions & RCCDnllnelldalions section, which details our recommended course of
action based an the survey results.
Methodology The Methodology section explains the methodology used to conduct this type of survey
research. This section also explains how to use the detailed crosstabulation tables in Appen-
dices C and D.
Summary of Results hl the body of the report, we present a question -by -question analysis of the survey for each
sample group (Special Tax Sample and Assessment Sample)1. For the readers' convenience,
the reports were crafted so that the reader can turn directly to a section and understand the
results without having to read prior sections. This results in considerable repetition of lan-
guage throughout the report.
Appendices We have included the following four Appendices:
■ Appendi.vil, which presents the questionnaire with topline results for the Special Tax
Sample.
® AppendivB, which presents the questionnaire with topline results for the Assessment
Sample.
e AppendIrC, which presents the complete crosstabulalions for the Special Tax Sample.
e AppendkD, which presents the complete crosstabulations for the Assessment Sample.
'Por a discussion of the Iwo sample types, turn to pagel3
City of Petaluma - Streets and Roads Study Cudbe Research d Anal-sis
Page]
b ealI Ve summery
Executive Summary
Key Findings - Special Tax Sample
Based on an analysis of the survey data, GRA offers the following key findings to the City of
Petaluma for the special tax scenario, which consists only of voters who are likely to partic-
ipate in the November 2002 election:
Issues olImportance - Special Tax The first substantive question of the survey presented respondents with a series of issues fac-
Sample ing residents of the City of Petaluma and asked them to rate the importance of each issue.
Respondenis ranked 'Maintaining the water quality' as the most important issue evaluated,
followed by `Improving the quality of public education', 'Repairing and maintaining local
streets and roads', 'Reducing traffic congestion', and 'Protecting the environment'. It is
worth noting that all issues received an average importance ranking of at least 'somewhat
important', with five of the nine tested exceeding the level of'very important'.
Respondents were next asked to rale the condition of Petaluma's streets. Approximately one
percent of those surveyed in the Special Tax Sample felt that Petaluma's streets were in
'excellent' condition, 19 percent felt they were in 'good' condition, and 24 percent perceived
the streets' condition as 'fair'. The remaining 55 percent of respondents perceived the condi-
tion of Petaluma's streets as 'poor' or 'very poor'.
First Ballot Test - special Tax sample Early in the survey, respondents were presented with the ballot language for the measure and
asked to indicate whether they would support or oppose the measure if the election were held
today. Known as the first ballot test, this question assesses support for the measure without
First priming the respondent with information beyond what is presented in the ballot lan-
guage itself. The first ballot test represents a reliable measure of support for the proposed tax
increase among an 'uninformed' electorate.
Specifically, respondents were asked if they supported a measure that would increase local
taxes to raise 20 million dollars to pay for the repair and maintenance of local streets and
roads in the City of Petaluma. Overall, 51 percent indicated they would 'definitely' or'proba-
bly' support the proposed tax increase at this point in the survey. Approximately 39 percent of
respondents indicated that they would vote 'no' on the measure, whereas nine percent were
undecided or unwilling to state their vote choice. In addition, overall initial support ('defi-
nitely' and 'probably' yes) for the measure was greatest among those who perceive the
streets' condition as 'fair', those with a'medium' propensity to vote (those who voted three,
four, or five times in the past eight elections), individuals who live in single Democratic
households, and respondents between the ages of 50 and 04.
City of Petaluma - Streets and Roads Study Godbe Research & Analysis
Page 2
Cxecmlve Summary
Parcel Tax Threshold - special Tax Respondents were next presented with various parcel tax thresholds. Naturally, the higher
Sample the tax threshold, the less willing some voters are to support the tax. Fifty-one percent of vot-
ers would be supportive of the measure at the highest tax threshold ($80). Survey results
revealed that if voters heard that property taxes were to be increased by $40 per year, 69 per-
cent of the electorate would be willing to support the measure.
City Sales 7axTbreshold- Special Tax Voters in the Special Tax Sample were next presented with various city sales tax thresholds.
Sample Overall, 68 percent of voters indicated they would be supportive of the measure at the highest
lax threshold (one-quarter percent increase). As the tax rates decreased, the percentage of
voters who would support the lax measure increased. If voters heard that the city sales tax
would be increased by ane -tenth percent, 70 percent of 0he electorate would be willing to
support the measure.
Spending Projects- Special Tax Respondents were presented with specific information about the types of projects dial may be
Sample funded by the measure to gauge the impact of this information upon voters' likelihood of
supporting the tax increase. The most strongly supported project was 'Repair and maintain
neighborhood streets', followed closely by 'Resurface major streets', and 'Fix potholes', 'Help
reduce traffic congestion', and 'Repair and maintain major streets in Petaluma, such as Pet-
aluma Boulevard, McDowell Boulevard, Bast Washington Street and Gly Boulevard South'.
Argumoms,rurrounding the Ballot measures do not succeed or fail in a political vacuum. Proponents of a measure will
Measure - Special Tax sample present arguments to try to persuade voters to support the measure, just as opponents will
present arguments to achieve the opposite effect. To simulate a campaign environment and
to gauge the influence of various arguments, the survey presented voters with arguments
both in favor of, and in opposition to, the measure.
The most compelling positive argument tested was 'We need to act now. The longer we wait,
the more expensive it will be to make repairs and the worse the streets will gel', followed by 'A
Citizen's Oversight Committee will be formed to ensure that the funds are spent properly',
'The streets of Petaluma have been neglected for over 30 years, and they are now in desperate
need of repair', and 'Metropolitan Transportation Commission (MTC) rated Petaluma's
streets as the worst in the nine county Bay Area'.
Despite presenting respondents with arguments that were designed to reduce their likelihood
of supporting the measure, two of the arguments tested actually resulted in minimal net
increases in support for the measure and two of [lie arguments resulted in virtually no
change in support. Only the arguments that 'If the measure passes it would commit the City
to as many as 20 years of higher taxes', 'The City will mismanage the money', and 'The
County is considering a sales tax that would fund local street and road repairs, so we
shouldn't double -tax ourselves by supporting the City's measure' actually reduced support
for the measure, more than a negligible amount, among voters as a whole.
Cite of Petaluma - Savels and Roads Study Godbe Rescard S Analysis
Page 3
Executive snnmialY
Second Ballot rest - Special Tac After providing voters with die wording of the proposed measure, the possible tax amounts, a
sample list of projects that may be funded by the measure, as well as arguments both in favor and
against the measure, respondents were once again asked whether they would vote yes or no
on the measure that would increase taxes to raise 20 million dollars to pay for the repair and
maintenance of local streets and roads throughout the City. Approximately 02 percent of
respondents indicated that they would support the measure at this point in the survey, with
31 percent stating that they would not support the measure and seven percent undecided or
unwilling to state their opinion. This represents an 11 percent increase from the first ballot
test. In addition, overall support at the second ballot test ('definitely' and 'probably' yes) for
the measure was greatest among those with an annual household income of $00,000 to
$89,999, those with a 'medium' propensity to vote (those who voted three, four, or five times
in the past eight elections), individuals who live in dual Democratic households, females,
and respondents who perceive the streets' condition as 'fair'.
Park and Recreation FadOties- The final substantive section of the survey asked respondents to indicate their level of support
Special Tax sample for the allocation of funds for the maintenance and operation of park and recreation facili-
ties in addition to the repair and maintenance of local streets and roads within the City of
Petaluma. Overall, 48 percent of likely November 2002 voters surveyed supported a measure
to fund the maintenance and operation of park and recreation facilities in the City in addi-
tion to funding the repair and maintenance of Petaluma streets and roads, 40 percent
opposed, and six percent were undecided or unwilling to state their opinion.
Next, respondents were presented with the information that the funding for the additional
park and recreation projects would require an additional tax increase of fourmillion dollars,
bringing the total tax amount to 24 million dollars. With this information, 41 percent of
respondents indicated they would support the measure and 50 percent would oppose the
measure. The remaining nine percent of voters were undecided or unwilling to state their
opinion.
Key Findings - Assessment Sample
Based on an analysis of the survey data, GRA offers the following key findings to the City of
Petaluma for the benefit assessment scenario, which consists only of voters who own prop-
erly in the City:
Issues of Importance - Assessment The first substantive question of the survey presented respondents with a series of issues fac-
Sample ing residents of the City of Petaluma and asked them to rate the importance of each issue.
Respondents ranked 'Maintaining the water quality' as the most important issue evaluated,
followed by'Improving the quality of public education', 'Repairing and maintaining local
streets and roads', 'Reducing traffic congestion', and 'Protecting the environment'. It is
worth noting that all issues received an average importance ranking of at least 'somewhat
important', with five of the nine issues tested exceeding the level of 'very important'.
City of Petaluma - Streets and Roads Study Godbe Researdi & Analysis
Page 4
6xenalveSummazy
Respondents were next asked to rate the condition of Petaluma's streets. Less than one per-
cent of those surveyed in the Assessment Sample felt that Petaluma's streets were in 'excel-
lent' condition, 18 percent felt they were in 'good' condition, and 24 percent perceived the
streets' condition as 'fair'. The remaining 58 percent of respondents perceived the condition
of Petaluma's streets as 'poor' or 'very poor'.
First Ballot Test -Assessment Sample Early in the survey, respondents were presented with the ballot language for the measure and
asked to indicate whether they would support or oppose the measure if the election were held
today. Known as the Gist ballot test, this question assesses support for the measure without
first priming the respondent with information beyond what is presented in the ballot lan-
guage itself. As such, it represents a reliable measure of support for the proposed lax increase
among an'uninformed' electorate.
Specifically, respondents were asked if they supported a measure that would increase local
taxes to raise 20 million dollars to pay for the repair and maintenance of local streets and
roads in the City of Petaluma. Overall, 50 percent indicated they would 'definitely' or'proba-
bly' support the proposed lax increase at this point in the survey. Approximately 41 percent of
respondents indicated that they would vote 'no' on the measure, whereas nine percent were
undecided or umvilling to state their vole choice. In addition, overall initial support ('defi-
nitely' and 'probably' yes) for the measure was greatest among those respondents between
the ages of 50 and 64, individuals who live in single Democratic households, females, and
those who perceive the streets' condition as 'fair'.
Assessment Threshold - Assessment Respondents were next presented with various assessment thresholds. Naturally, the higher
Sample the threshold, the less willing some voters are to support it. Overall, 51 percent of voters
would be supportive of the measure at the highest threshold ($80). We see that if voters
heard that property taxes were to be increased by $40 per year, 68 percent of the electorate
would be willing to support the measure.
Spending Projects -Assessment Respondents were presented with specific information about the types of projects that may be
Sample funded by the measure to gauge the impact of this information upon voters' likelihood of
supporting the tax increase. The most strongly supported project was 'Resurface major
streets', followed by 'Repair and maintain neighborhood streets', 'Fix potholes', 'Help reduce
traffic congestion', and 'Repair and maintain major streets in Petaluma, such as Petaluma
Boulevard, McDowell Boulevard, East Washington Street and Ely Boulevard South'.
Arguments Surrounding the Ballot measures do not succeed or fail in a political vacuum. Proponents of a measure will
Afeasm e- Assessment sample present arguments to try to persuade voters to support the measure, just as opponents will
present arguments to achieve the opposite effect. To simulate a campaign environment and
to gauge the influence of various arguments, the survey presented respondents with argu-
ments both in favor of, and in opposition to, the measure.
My o(Petaluma - Streets and Roads Study Godhe Reseatrli & Analysis
Page 5
Executive summary
The most compelling support argument tested was 'We need to act now. The longer we wait,
the more expensive it will be to make repairs and the worse the streets will get', followed by'A
Citizen's Oversight Committee will be formed to ensure that the funds are spent properly',
'The streets of Petaluma have been neglected for over 30 years, and they are now in desperate
need of repair', and 'Metropolitan Transportation Commission (MTC) rated Petaluma's
streets as the worst in the nine county Bay Area'.
Despite presenting respondents with arguments that were designed to reduce their likelihood
of supporting the measure, one of the arguments tested actually resulted in a minimal net
increase in support for the measure and one of the items resulted in virtually no change in
support. The arguments that 'If the measure passes it would commit the City to as many as
20 years of higher taxes', 'The City will mismanage the money', and'The County is consid-
ering a sales tax that would fund local street and road repairs, so we shouldn't double -tax
ourselves by supporting the City's measure', 'It's not the voters fault that the roads were not
maintained for the past 30 years, so its not fair to make them pay for it through this mea-
sure', and 'We should be spending money on more pressing issues such as improving educa-
tion' reduced support for the measure, more than a negligible amount, among owners as a
whole.
Second Ballot Ter - Assessment After providing respondents with the wording of the proposed measure, the possible assess -
Sample merit amounts, a list of projects that may be funded by the measure, as well as arguments
both in favor and against the measure, respondents were once again asked whether they
would vote yes or no on the measure that would increase taxes to raise 20 million dollars to
pay for the repair and maintenance of local streets and roads throughout the City. Approxi-
mately 61 percent of property owners indicated that they would support the measure at this
point in the survey, with 33 percent stating that they would not support the measure and
seven percent undecided or unwilling to slate their opinion. This represents an 11 percent
increase from the first ballot test. In addition, overall support at the second ballot test ('defi-
nitely' plus'probably' yes) for die measure was greatest among those with an annual house-
hold income of $90,000 to $109,999, those with a'medium' propensity to vote (those who
voted three, four, or five times in the past eight elections), females, individuals who live in
dual Democratic households, and respondents who are between 50 and 64 years of age.
Park and Recreation Facilities - The final substantive section of the survey asked respondents to indicate their level of support
Assessment Sample for the allocation of funds for the maintenance and operation of park and recreation facili-
ties in addition to the funding of the repair and maintenance of local streets and roads
within the City of Petaluma. Overall, 44 percent of property owners surveyed supported a
measure to fund the maintenance and operation of park and recreation facilities in the City
in addition to funding the repair and maintenance of Petaluma streets and roads, 50 percent
opposed, and six percent were undecided or unwilling to state their opinion.
City of Petaluma - Streets and Roads Study Codbe Ruea¢k k Analms
Page 6
Executive Summary
Next, respondents were presented with the information that the funding for the additional
park and recreation projects would require an additional tax increase of four million dollars,
bringing the total tax amount to 24 million dollars. With this information, 40 percent of
respondents indicated they would support the measure and 52 percent would not support the
measure. The remaining nine percent of voters were undecided or unwilling to state their
opinion.
City ofPela(ume - Sireels and Roads Study G'odbe Rescardi & Analysis
Page 7
eC¢6uflvY Summary
Conclusions and Recommendations
Based on an analysis of the survey data, GRA and SCI offer the following conclusions and
recommendations to the City of Petaluma.
Should the CRY ofPetaimna move The favorability of voters' opinions regarding a revenue measure to fund the repair and
forward whh plans to fund the maintenance of local streets and roads in lire City of Petaluma isjust one of several factors
repair and maintenance of local that the City will undoubtedly consider in deciding whether, and when, to submit a revenue
sireeis and roads by presenting a measure to the voters. However, to the extent that voters' opinions weigh into the City's deci-
revenue measure to the voters? sion on these matters, the results of the survey suggest that the City should take the next steps
toward submitting a revenue measure to the voters.
Although the initial ballot tests were below the required threshold for a special tax this is to
be expected. The ballot language was designed to generate a conservative measure of sup-
port, and respondents were presented the ballot language without additional information
about the proposed measure. Subsequent questions revealed that support for the measure
increased considerably when the various tax thresholds were tested and respondents were
provided with information about the types of projects that may be funded by the measure, as
well as arguments both in support and in opposition to the proposal. For example, when
likely November 2002 voters were informed that the measure would involve a sales lax
increase of a quarter percent, support for the measurejumped to 68 percent. Although the
second ballot test for the special tax measure remained below the required two-thirds thresh-
old (621/0), the percentage of property owners in favor of the measure exceeded the weighted
majority required for success in a benefit assessment, with 61 percent support.
The findings of the survey indicate that the specific content and funding mechanism of the
measure is critical to its success. The current study was designed to evaluate a broad range of
projects, costs, funding mechanisms, and arguments to hone the parameters of the measure.
With this general information in hand, the City should work to develop a more narrow,
refined list of options. With a specific package in mind, the City may also wish to conduct
follow-up research to verify that it has chosen the most appropriate combination of elements
to present to voters.
Based on recent experience conducting studies for similar revenue measures across the state,
the information provided from this study indicates that if packaged correctly and coupled
with a well-financed and effective public information campaign a revenue measure has a
good chance of succeeding.
city arpetaiwna - Streets and Roads Study Godbe Reseetrh g Analysis
Page 8
Cxeculleu Smm111ary
INltich revenue measure options Once again drawing upon the results of the survey as well as GRA's and SCI's experience
should the Chy ofPelaluma conducting studies for similar measures throughout California, we recommend that the City
consider? consider either a sales tax or a benefit assessment as a viable option for a revenue measure
that would fund the repair and maintenance of local streets and roads within the City ofPet-
aluma.
This conclusion is based on several observations from the findings, including:
■ many of the specific proposed spending projects tested in the survey were quite effective at
increasing stated support for [lie measure;
® support levels were considerably high for many of the thresholds tested in both the sales
tax and benefit assessment scenario;
■ support for the measure was fairly broad-based -- that is, there were several key voter sub-
groups that supported the measure, not only specific groups (Le,, Democrats) and
® [here was a substantial increase in support between the first and second ballot tests for
bath the sales tax and benefit assessment scenarios (11% increase in both), which sug-
gests that when provided with specific information (as in a campaign), both voters and
property owners, as a group, become more likely to support the measure,
What are the specific issues A sales tax measure faces significant challenges for the City, but also has the potential to
associated with a sales lax measur0 reap significant rewards. This type of measure has two important advantages over the other
revenue measure options considered in this survey. First, a sales tax can raise significantly
more money than the other options and second, the revenue that is generated from the mea-
sure is paid for by tire beneficiaries of the improved streets and roads and nor just the voters
or property owners in the City of Petaluma. The obstacles associated with a sales tax measure
include the required two-thirds majority support from voters as well as enabling legislation
by the stale that must be passed before any sales tax increase can be implemented by the City.
Another obstacle to the sales tax measure is the possibility that Sonoma County could bring
a transportation -related sales lax measure to the ballot at or near the same time as Peta-
luma's measure. This event would certainly detract support from die City's sales tax proposal
diminishing [he possibility of success for the measure.
I4111at are me specific issues A benefit assessment could be considered if the obstacles associated with the sales lax pro -
associated with a benefit assessment? posal are considered loo great to overcome. One important distinction with this type of reve-
nue measure is a lower threshold for success (majority) in comparison with a sales tax or
any other special lax measure (two-thirds). Because assessment proceedings are not
restricted to occur at normal election times, the City will also be able to more effectively com-
municate with property owners about the measure and its purpose. One of the challenges for
an assessment measure is that commercial property owners, apartment owners, and invest-
ment owners vote on the measure and, in the absence of information, these groups are often
substantially less likely to vote in favor of assessment measures. Without an effective out-
reach campaign, moreover, businesses and apartment owners may actively oppose a mea -
City of Pelaiwna -Streets and Roads Study Cadbe Researdi & Analysis
Page 9
Executive SHMIUry
sure. In Petaluma, it is expected that approximately one-third of the weighted returned
ballots would come from owners of commercial property and/or apartments.
If the City chooses a sales tax, what Naturally, the willingness of voters to support a specific revenue measure is contingent upon
would be the appropriate threshold? the lax rate associated with the measure. Il is critical that the City set [lie lax increase
amount at a level that will be supported by at least two-thirds of voters.
Survey results revealed that 70 percent of likely November 2002 voters would support a sales
tax increase of one-tenth of a percent and support only diminished by two percent when the
sales tax amount was increased to one-quarter of a percent. The survey only tested sales tax
increases between one-tenth of a percent up to a quarter of a percent and all received support
over the two-thirds requirement but not by more than the margin of error associated with
the survey.
In deciding on the appropriate tax rate, GRA recommends that the City keep several consid-
erations in mind:
■ the margin of error for the assessment survey is ±4.50 percent;
■ any state or county level measures that may impact the local measure and
■ current widespread concern about the uncertainty of the economy.
Based on the above considerations and the results of the survey, GRA recommends that the
City set the sales tax increase at a level not to exceed a quarter of one percent. This recom-
mendation reflects the increments that were tested in the survey as well as the belief that it is
often better to be cautious given the possibility of future events and conditions that may
erode overall support for the measure.
If the City chooses an assessment, Should the City look to a benefit assessment as the funding mechanism of choice, it is crili-
what would be the appropriate cal to select an amount that the weighted majority of property owners will support. In tins
threshold? survey, assuming that all residential property owners' votes are of equal weight, approxi-
mately 51 percent of property owners indicated they would support the measure at the tax
rate of 880 per year. Al a rate of 860 per year, the measure receives approximately 57 percent
support.
In addition to the bullet points mentioned in the above section, there are a couple important
considerations the City must keep in mind if a benefit assessment scenario is pursued. It is
imperative that the measure is perceived as affordable by apartment and commercial
property owners and investment owners, as they make up more than one-third percent of
the expected weighted returned ballots in the City. Without their support, a benefit assess-
ment measure has a minimal chance of succeeding. Another important consideration is that
the City must look closely at the expected revenue at the various assessment amounts to
City offelalu na - Streets and Roads Study Godhe Research S Analysis
Page 10
Executive Sununm}'
determine whether or not the amounts supported by the weighted majority of property own-
ers would result in a sufficient amount of funding to complete the key projects.
N9iat are the largest obstacles to the As mentioned previously, any proposed revenue measure (Sales Tax or Benefit Assessment)
success ofa Revenue Aleasure? has a good chance of passing only if the measure is a) packaged appropriately, and b) has a
well-financed and compelling information campaign that conveys to voters the information
they need to make educated decisions about the measure. If either of these conditions is not
met, the measure has a low chance of success in the current environment. By 'packaging',
we principally mean that the measure provides funding for the projects that property owners
indicated are priorities and that the tax rate is kept affordable.
Finally, it is worth emphasizing that this survey identifies voters' support for the proposed
revenue measure in the current economic and political environments. If [lie conditions in
these environments deteriorate, or other significant events occur, it can have an appreciable
impact on support for the measure.
What project areas should be One of the principal tasks of the survey was to identify the projects that voters perceive as pri-
considered prioritiesto be funded by orities for funding. Based on the results from both the special tax and assessment samples,
the measure? the top project priorities are as follows:
® Repair and maintain neighborhood streets
o Resurface major streets
® Pix potholes
® Help reduce traffic congestion
The results of the survey reveal that the projects that are most directly connected to repairing
and maintaining local streets and roads are the most likely to positively influence voters.
Considering the lti projects (hat were evaluated -'Repair and maintain neighborhood
streets', 'Resurface major streets', and 'Fix potholes' were the three most influential projects
in gathering support whereas 'Repair and replace storm drains', 'Provide bike accessibility'
and'Provide access along the Petaluma river' were the three least influential projects in gen-
erating support for the measure. The results from Questions 11 and 12 of the survey (see
page 40 & 41) in which voters were asked if they wanted to include funding for parks and
recreation facilities in the streets and roads revenue measure further reinforces the impor-
tance of focussing on projects that are directly related to repairing and maintaining local
streets and roads and not incorporating peripheral projects into the streets and roads revenue
measure.
How might information affect Voters' opinions about revenue measures are often not well informed, even in this case where
support for the proposed measure? the measure that is being considered has been described and discussed in the local media.
Voters are usually hesitant to support a revenue measure if they are not aware of what
Clly of Petaluma - S'tnets and Roads Sluey Godbe Research & Analysis
Page 11
Etc n live SIII lnmry
exactly the measure will be funding. It is clear from the survey results that without addi-
tional information about the tax rates associated with the measure, the projects that may be
funded by the measure, or compelling reasons to support the measure, a significant percent-
age of voters would be reluctant to vote in favor of funding the repair and maintenance of
local streets and roads in the City of Petaluma. Upon being presented with this information,
however, many voters who initially opposed the measure turned into supporters of the mea-
sure. A well -funded, effective public information campaign is thus critical to providing voters
with the information that many require before they are willing to support a revenue mea-
sure. Based on the first and second ballot test results, it is also critical to the success of the
proposed measure.
Accordingly, GRA recommends that the City lake the next steps toward submitting a revenue
measure to the voters, but only if the measure is packaged correctly and there are organized
groups committed to conducting an effective public information campaign.
Cit' of hoduma - S7eels and Roads Sludy Codbe Researdi & Analpis
Page 12
hielhodology
Methodolo
Research Objectives At the outset of this project, the City of Petaluma and GRA identified several research objec-
tives for this study. Broadly defined, the City of Petaluma was interested in using survey
research to:
a Determine support for a revenue measure to fund the repair and maintenance of local
streets and roads within the City by increasing local taxes to raise 20 million dollars;
■ Identify the lax threshold level at which voters are willing to support the revenue mea-
sure;
s Examine support levels given three different funding mechanisms: parcel tax, sales lax,
and benefit assessment;
■ Determine the issues that are most important to voters and
■ Identify the features of the projects, as well as arguments both pro and con, that resonate
with voters.
Afelhodologh Tablet briefly outlines the methodology employed in this project. The Special Tax Sample
(used in assessing the feasibility of a Parcel Tax and Sales Tax) was comprised of likely
November 2002 voter renters and likely November 2002 voter owners. The Assessment Sam-
ple (used in assessing the feasibility of a Benefit Assessment) included likely November 2002
voter owners as well as unlikely November 2002 voter owners. A total of 600 voters in the City
of Petaluma completed an interview in English, representing a total universe of approxi-
mately 25,620 voters in the City. however, only 588 interviews were included in the analyses
because 12 respondents identified as owners indicated they were now renters. These 12 rent-
ers were not likely voters and thus would not be expected to participate in an upcoming elec-
tion. Interviews were conducted on October 1st through October 7th, 2001, and each
interview, typically lasted 15 minutes.
Table 1. Methodology
Technique Telephone interviewing in English
Interview Length 15 minutes
Universe Voters in the City of Petaluma
Field Dates October 1 through 7, 2001
Sample Size 126 High propensity voter renters;
374 High propensity voter owners;
88 Low propensity voter owners
Sample and Orelghting Choosing the appropriate sampling design for a study is a careful process that involves
detailed consideration of the research objectives. In the present study, one of the main objec-
tives was to determine the feasibility of raising taxes to fund the repair and maintenance of
Chy o(Petaluma - Slmels and Roads Stud, Codbe Research & Analysis
Page 13
Methodology
local streets and roads. In consideration of a special tax, the City was interested in the
November 2002 election as the largest likely voter universe in the near future. It was impor-
tant to assess the level of support for the measure among owners and renters who are likely
to participate in the November 2002 election as well as among voter owners who are not as
likely to participate. Another option to fund the repair and maintenance of local streets and
roads is a a benefit assessment, in which ballots are mailed to all property owners in the Dis-
trict, who then cast their vote and return the ballot by mail. Although the mailing of these
ballots is not tied with a particular scheduled election (i.e., November 2002), GRA has found
that there is a strong relationship between the propensity to turn out at the polls on election
day and to return a mail ballot in a benefit assessment. Based on historical data and the vot-
ing history of each registered voter in the City, GRA identified a total of 25,620 voters to form
the sampling universe for the study.
The portion of the 25,620 City of Petaluma voters in the universe representing high propen-
sity renters (8,356) was stratified on the basis of gender, age, party affiliation, and geogra-
phy. The portion of the universe representing only property owners (17,264) was stratified on
the basis of gender and household party affiliation. In total, six hundred clusters were
formed and voters were randomly selected into a cluster based on their profile. During data
collection, individuals were sampled randomly from each cluster. At the completion of data
collection, the data were weighted on the basis of gender, household party affiliation, and for
the Assessment Sample, expected ballot return rate.
After this weighting, there is a total of 502 respondents within the Special Tax Sample and
462 respondents in the Assessment Sample.
City of Petaluma - Streets and Roads Study Godbe Reward b Analysis
Page 14
Methodology
Subgroup Labels The following subgroup labels are used in (he report and crosstabulation tables:
Table 2. Suborouo Labels
Ace
Respondents were grouped in the following age categories: 18-29,
30-39, 40-49, 50-64, and 65+ (65 and older).
Condition of Peta-
Voters were classified according to whether they felt the condition of
luma's Streets
Petaluma's streets was 'Excellent to Good', 'Fair', 'Poor', or 'Very
Poor'.
First Ballot Test
Voters were grouped based on their position on the measure in the
first ballot Lest: 'Def. yes', 'Prob. yes', 'Prob. no', 'Def. no', and'DK/
NA'.
Gentler
'Male' and 'Female' respondents were identified with separate
abets.
Household Income
Voters were classified into the following household income catego-
ries: Under 540,000, $40,000-559,999, 560,000-589,999, $90,000-
$109,999, $110,000-$149,999, and Over $150,000.
Household Party Type
Individuals were grouped into the following household party types:
'Democrat 1' --one Democrat in the household, 'Democrat 2+' --two
or more Democrats, 'Republican i' --one Republican, 'Republican
2+• --two or more Republicans, 'Other' --Other partisans, 'Mixed' --
household members who differ in their partisanship.
Information on Streets
Respondents were grouped according to the amount of Information:
'A Lot of Information', "Some Information 'A Little Information', or
'No Information' pertaining to the condition of Petaluma's streets and
roads prior to the survey.
Length at Current
Respondents were grouped according to the number of years they
Residence
have lived at their current residence: 'Less than 4 years', '4-8 years',
'8-15 years', and '15 or more years'.
Own Commercial
The subgroup labels of 'Yes' and 'No' were used to classify whether
Property or Apart-
or not respondents owned commercial properly or apartments.
ments
Party
Individuals were grouped according to their political party affiliation:
'Democrat' 'Republican,''Other or 'Decline to State' ('DTS').
Propensity to Vole
Individuals were grouped Into the following according to their pro-
pensity to vole as: 'low'to Indicate they have voted zero, one, ortwo
times in the past eight elections, 'medium' to Indicate they have
voted three, four, or five times In the past eight elections, and 'high'
to Indicate they have voted six, seven, or eight times In the past
eight elections.
Respondent Type
'Renter','Owner - Likely', and'Owner - Unlikely', were each identi-
fied with separate labels.
Second Ballot Test
Voters were grouped based on their position on the measure in the
second ballot test: 'Def. yes', 'Prob. yes','Prob. no', 'Def. no', and
DKINA'.
Swing Voters
Respondents who changed their vote from probably no, definitely
no, or don't know/no answer to definitely yes or probably yes were
labeled 'negative to positive' swing voters. Alternately, Individuals
who changed their vole from definitely or probably yes to probably
no, definitely no, or don't know/no answer were labeled 'positive to
negative' swing voters.
Zip Code
Respondents residing in the zip codes of '94952' and '94954' were
identified accordingly.
Of, of Petaluma - Surely and Roads Study Godbe Research & Analysis
Page 15
Medmdology
Randomization of Questions To avoid the problem of systematic position bias -- where the order in which a series of ques-
tions is asked systematically influences the answers to some of the questions -- several of the
questions in this survey were randomized such that respondents were not consistently asked
the questions in the same order. The series of items in Questions 2, 7, 8, and 9, as well as
question sets 5 and 6, and sets 8 and 9, were randomized in the questionnaire.
Understanding the Wargln of8rror' Because a survey typically interviews a limited number of people who are pari of a larger
population group, by mere chance alone there will almost always be some difference
between a sample and the population from which it was drawn. For example, researchers
might collect information from 400 adults in a town of 15,000 people. Because not all peo-
ple in the population were surveyed, there are bound to be differences between the results
obtained from interviewing the sample respondents and the results that would be obtained if
all people in the population were interviewed. These differences are known as 'sampling
error' and they are expected to occur regardless of how scientifically the sample has been
selected. The advantage of using a scientifically drawn probability sample, however, is that
the maximum amount of sampling error can be estimated with a specified degree of confi-
dence. Sampling error is determined by four factors: the size of the population, the chosen
sample size, a confidence level, and the dispersion of responses to a survey. Of the four fac-
tors, sample size is the most influential variable.
Table 3. Guide to Statistical Siqnificance - Assessment Sample
n
1,90%/:10%
Distribution of Responses
80%120% 70%00% 60%140%
50%/ 50%
1,000
1.80%
2.41%
2.76%
2.95%
3.01%
. 800...._
700
2.03%
2.71%
3.10%
3.32%
3.38%
2.18% 2.90% 3.33% 3.55%
3.63%
600
2.36%
3,14%
3.60%
3.85%
3.93%
500
2.59%
3.46%
3.96%
4.23%
4.32%
462
2;70%1,
3.60%1r,
4;12%
4.41 %
4.50% -:
400
2.91%
3.87%
444%
4.75%
4.84%
300
3.37%
4.49%
5.14%
5.50%
5.61%
-200.
4.13%
5.51%
6.31%
6.75%
6.89%
..100..
'.50
5.86%
7.82%
8.96%
9.57%
9.77%
8.30% 11.07% 12.68% 13.56%
13.84%
Table3 shows the possible sampling variation drat applies to a percentage result reported
from a probability type sample (clustering considered). If a sample of 462 voters (i.e., the
Assessment Sample) is drawn from the universe of voters in the City of Petaluma, one can be
95 percent confident that the margin of error due to sampling will not vary, plus or minus,
City of Petaluma - Streets and Roads Study Godbe Research & Analysis
Page 16
hledlUdolugy
by more than the indicated number of percentage points from the result that would have
been obtained if the interviews had been conducted with all persons in the universe repre-
sented in the sample.
As the table indicates, the maximum margin of error for all topline responses is between 2.70
and 4.50 percent. This means that for a given question widr dichotomous response options
answered by all 462 respondents, one can be 95 percent confident that the difference between
(he percentage breakdowns of the sample population and those of the total population is no
greater than 4.50 percent. The percent margin of error applies to both sides of the answer, so
that for a question in which 50 percent of respondents said yes, one can be 95 percent confi-
dent that the actual percent of the population that would say yes is between 45.50 percent
and 54.50 percent.
The actual margin of error for a given question in this survey depends on the distribution of
the responses to the question. The 4.50 percent refers to dichotomous questions, such as
yes/no questions, where opinions are evenly split in the sample with 50 percent of respon-
dents saying yes and 50 percent saying no. If that same question were to receive a response in
which 10 percent of respondents say yes and 90 percent say no, then the margin of error
would be no greater than 2.70 percent. As the number of respondents in a particular sub-
group (e.g., gender or party type) is smaller than the number of total respondents, the mar-
gin of error associated with estimating a given subgroup's response will be higher.
It is also important to note that the margins of error expected due to sampling apply to both
studies. Therefore, when comparing the results of the two studies, the reader should keep in
mind that there is a +/-2.70% to +/-450% margin of error around both estimates.
How to Read a Crosslabuladon The questions discussed and analyzed in this report comprise a subset of the various
table crosstabulation tables available for each question. Only those subgroups that are of particu-
lar interest or that illustrate a particular insight are included in the discussion on the follow-
ing pages. Should readers wish to conduct a closer analysis of subgroups for a given
question, the complete breakdowns appear in Appendices C and D. These crosstabulation
tables provide detailed information on the responses to each question by many of the demo-
graphic groups that were assessed in the survey. A typical crosstabulation table is shown in
Table4.
A short description of the item appears at the top of the table. The sample size (in this exam-
ple, n=502) is presented in the first column of data under 'Overall'. The results to each pos-
sible answer choice of all respondents are also presented in the first column of data under
'Overall'. The aggregate number of respondents in each answer category is presented as a
whole number, and the percentage of the entire sample that this number represents is just
below the whole number. For example, among overall respondents, 124 people indicated
that they would 'definitely' vote yes on the measure, and 124 represents 24.8 percent of the
City u(Peialuma - Streels and Roads Study Godbe Research & Analysis
Padre 17
Wf Ihodology
total sample size of 502. Next to the 'Overall' column are other columns representing opin-
ions of voters grouped by their propensity to vote. The data from these columns are read in
exactly (lie same fashion as the data in the 'Overall' column, although each group makes up
a smaller percentage of the entire sample.
Table 4. First Ballot Test Special Tax Sample by Voting Propensity
Understanding a 'Mean' In addition to analysis of response percentages, many results will be discussed with respect to
a descriptive 'mean'. 'Means' can be though( of as 'averages'. To derive a mean that repre-
sents perceived importance of local issues (Q.2), for example, a number value is first
assigned to each response category (e.g., 'extremely important' = +3, 'very important' _
+2, 'somewhat important' = +1, and 'not at all important' = 0). The answer of each
respondent is then assigned the corresponding number (from 0 to +3 in this example).
Finally, all respondents' answers are averaged to produce a final number that reflects aver-
age perceived importance of local issues. The resulting mean makes interpretation of the
data considerably easier.
Cly of PVodU11)a - SReels and Roads Study Codbe Research & Analysis
Page 18
hiehhodolugp
Rol, to Read a 'Afeans' Table In tables and charts for Questions 2, 7, 8, and 9 of the survey the reader will find mean scores
that represent answers given by the respondent. The mean scare represents the average
response of each group. The following table shows the scales for each corresponding ques-
tion. Responses of 'don't know' and 'no answer' are not included in calculating the means
for any question.
Table 5. 'Means' Questions and Corresponding Scales
Question
IMeasure
Scale
Values
2
Importance of local issues
0 to +3
0 = Not important at ail
+1 = Somewhat important
+2 = Very important
+3 = Extremely important
7, 8, 9
Likelihood of supporting the
-2 to +2
-2 = Much less likely
measure
-1 = Somewhat less likely
0 = No effect
+1 = Somewhat more likely
'
+2 = Much more likely
ANote an the Tables To present the data in the most accurate fashion, we display the results to the first decimal
point in the tables and figures. For the purposes of discussion, however, conventional round-
ing rules are applied, with numbers that include 0.5 or higher rounded to the next highest
whole number and numbers that include 0.4 or lower rounded to the next lowest whole
number. Because of this rounding, the reader may notice that percentages in the discussion
may not sum to 100 percent. Moreover, the decimal numbers shown in pie charts may vary
somewhat from the decimal numbers shown in the tables due to software requirements that
pie charts sum to exactly 100 percent. These disparities are confined to the first decimal
place.
City offetaluma - Samets and Roads Study Godbe Research & Analysis
Page 19
Issues of Inlpunance Special Tax sample
Issues of Importance - Special Tax
Sample
02. I'm going to read a list ofissues The first substantive question of the survey presented respondents with a series of issues fac-
facing yaur community today and ing residents of the City of Petaluma and asked them to rate the importance of each issue.
for each one, please tell me how This set of questions not only provides insight into how important an issue is on a scale of
important you feel the Issue is to importance, it also provides a relative ranking among the issues. Participants' responses
you, using a scale of"ex9remell�
important', 'veryimportant', were coded using the following scale: 'not at all important' = 0, 'somewhat important' _
somewhat important', or'noi at all +1, 'very important' = +2, and 'extremely important' = +3. The aggregate responses to
important'. each item are presented below in the form of a mean, which is simply a summary statistic
obtained by laking the overall average of the response codes for the entire special tax sample.
A mean of +2, for example, indicates that, overall, respondents felt that the issue was 'very
important'.
To avoid a systematic position bias, the order in which the items were read to respondents
was randomized for each respondent.
As shown in Pigurel, respondents ranked 'Maintaining the water quality' (2.33) as the most
important issue tested, followed by 'Improving the quality of public education' (2.31),
'Repairing and maintaining local streets and roads' (2.10), 'Reducing traffic congestion'
(2.09), and 'Protecting the environment' (2.08). It is worth noting that all issues received an
average importance ranking of at least'somewhat important', with five of the nine issues
exceeding the level of'very important'.
Figure 1. Issues of Importance
@b Llalnluinln9 wnler 9mllly
02. Impmvinn Ihn ppellly of public otlucallan
@I nppelrin9 end malNUlnlnO local streale pntl motls
02e PaCudng Iminc con0eslbn
02. Pmlocling Iho anvlranmenl
QV Prnservin9 open space
02d Malnlolnlnp pnrhs end mcmallnn /ocllllica
., umume Bron h
Q21, Pravanlln9local tar Increases
City of Petaluma - Streets and Roads Study Godbe Research & Analysis
Page 20
Issues of Ingmrim¢a - Special rax Sample
Tablet shows the importance ratings assigned to each issue by respondent age and Table7
shows the ratings by the respondents' prior information about Etre street and road conditions.
Older voters generally gave a higher importance rating to 'Repairing and maintaining local
streets and roads' than their younger counterparts. Those voters who stated they had received
'A lot of information' on the streets and roads of Petaluma were more likely to give a higher
rating of importance to Repairing and maintaining local streets and roads' than their
counterparts who slated they had received less information.
Table G. Issues of Importance by Age
.,._.....,... . Coverall I Age
.�.... .. i
1840 -'.,
i
�
C-
yoars year. Years
y,
yearryears
�
Z 00 7 99 2170 1 95
2.O1 �' 2.04
s 1Bose .-11
Q2h Maintaining wamr
-.quality_ 1
233 �ir 229 240 220
�
2.43 231
Q2c Improving the
quality of puhllc h
2.31 26B p 2.29 a 224
2.27 2.34
education
Q21 Repairing and r
local
}
210 197 203 �,
�-
,
maintaining streets
i; 203
2.22 2.25
[_ and roads �+
I - ��C
I Q2e Ron uesllvpmfflc
2.09 F 209 it 2.04 E Z04
2.16 t 2.07
('-
We Protecting theJill-
environment
Z013 2.17 2.02 2.05 11
p_l
2.10 207
�
i q2f Preserving ape" '_
�y_�-
- y
190 1.98 c 193 1,89 II 8 1.90
space
Q2d Maintalning parks
1.86 1.71 11 2 06 1 83
179 111.80
and recreation faelglies_ (i
I Q29 Limping growthh
1.73 1.451 80 2
1.73 1.09
l Q2h Preventing local lax I
�I;
1 ofi 1.62 II 1.39 1.58 S
1.52 7.69
I -Increases ,
L
L�
L
Cite o(Petaluma - Strecls and Roads Sludy Godba Research & Analpis
Pao 21
Issues of Importance - Specol Tax Sample
Table 7. Issues of Importance by Amount of Information on Local Streets
�V��mM~� t Oveiall I Information on Streets <
t [
4 A lot of:
Some
s
A little
No 1?.
F.
' Info rmatfonl informatla nyInformation
Information
Base
200
31999E
201
i
199
1 2.00
f Q21, Maintaining water
L
�!
quality j
2.33
2.33
2.35
t
2.35
2.26 l
Q2c Improving the
�t
1i t
V quality of public i
2.31
2.26
; 2.22
39
2.45
education
f
Q2i Repalrin9 and
maintaining local streets 11
210
l .29
0.10
1.94
2.00
t and roads
t
i
'r We Reducing traffic p
f - F
congestion
209
t �' 214
t 2.05
206
' 2.13 j
I
t Q2a Protecting the
2 051.15
1 95
P o 13i
213
t
2.14
environment ;; I'
1-62f Preserving open
1 9U
1.73
! 2.03
1.93
1.92
space ,I
`
l
'
Q2d Maintaining parks r
and recreation facilities f
1,66
1 1.65
�-
1.09
1.66
1.62
Q2gtlmltinggnmth �j,;
173
F 1.69
160
1.66
1.76
Q21h Preventing local tax [1
1.56
4' 1.66
1.51
1.54
Ifi�{
1.52 f
Increases '�
L __
44M 4
4
Q3. Generally speaking, lvould you Respondents were next asked to rate the condition of Petaluma's streets. Approximately one
say that the condition orPetaluma's percent of those surveyed in the Special Tax Sample felt that Petaluma's streets were in
streets is excellent, good, fair. poor, 'excellent' condition, 19 percent felt theywere in 'good' condition, and 24 percent perceived
or very poor? the streets' condition as 'fair'. The remaining 55 percent of respondents perceived the condi-
tion a[ Petaluma's streets as 'poor' (24n/o)or'verypoor' (31°%).
2. Condition of Petaluma's Streets
City of Petaluena - Streets and Roads Study Godbe Research & Analysis
Page 22
rlrst Ballot Tut- Special Tax Sample
First Ballot Test - Special Tax Sample
Q4. Next year, Petaluma votersmey One of the primary research objectives of this survey was to determine voters' support for a
be asked to vote on local ballot measure that would raise taxes to pay for the repair and maintenance of local streets and
measures. Lel me read you one of roads throughout the City.
the proposals: In order to pay for the
repair and maintenance oflocal
streets and roads, shall the city of Question 4, also referred to as the first ballot lest, lakes an early assessment of voters' sup -
Petaluma increase local taxes to port for the proposed measure. The purpose of the first ballot test is to take an early assess -
raise iwemytrillion dollars? If ment of respondents' support for increasing taxes to raise 20 million dollars. The reasons for
asked to vote today, wound you vote the strategic placement of the first ballot lest early in the survey are twofold. First, at this
yes or no on this measure? point in the survey, the respondent has not been primed with information about the ballot
measure or project needs beyond what is presented in the ballot language. This situation is
analogous to a voter going to the polls with limited knowledge about the measure in the
absence of an information campaign. Secondly, the first ballot test serves as a baseline from
which to determine the impact of information items upon support for the measure. Subse-
quent questions will assess support for the measure after presenting respondents with addi-
tional information.
3. First Ballot Test
Figure3 illustrates that among respondents overall, 51 percent would 'definitely' or'proba-
bly' support the proposed tax increase at this point in the survey. Approximately 39 percent of
respondents indicated that they would vote 'no' on the measure, whereas nine percent were
undecided or unwilling to state their vote choice.
As one might expect, support for the measure during the first ballot test was not uniform
across the various subgroups of voters identified in the survey. Tables 8 through 11 show how
the distribution of support for the measure varies across a host of voter characteristics. The
tables show that overall initial support ('definitely' plus 'probably'yes) for the measure is
greatest among those who perceive the streets' condition as 'fair', those with a 'medium'
Propensity to vote, individuals who live in single Democratic households, renters, and
respondents between [lie ages of 50 and 64.
ci(y of Petaluma -Streets and Roads Study Codbe Researdi & Analysis
Page 23
Nis I Ballot Test -Special Tax Sample
Table S. First Ballot Test by Perceived Condition of Petaluma's Streets
I
11 Overall :I Condition of Petaluma's
Streets
'
{{
sExcellent'
Poor 9 Very
1 502 '_ 119 116
124 21 27
248h t 17.9% 22.9%
_3
242
71
29.5%
11 to Good Fnir
133 { 40 35
-
Poor
26.5% 336/ 3030A%
�� 21.4% r
I
i Base
`w}➢etlnitely
4 502
102 1 122
121 a
156
vest
X124
16 24
^' n34 41
51
24,8%
15.5% p. 19,5%
28.0% )
32.55.
Probably yes
!i 133
I 20 t. 45
31 j
37
26.6,s
t95 o i H.95.
11 25.4% -
235m
#,
Probably no
I 86
34 14
M 22 t
16 S
I
17.2%
337
178%
10.4%
_`117%
t Definitely no
G 117
18 21
24
41
221%
1Bt/ 22.1%
200% y
285%
`
h 47
14 12
�1 1 °
71 I
OK I NA
E
�..W.. ry�
p 9.4/
133N 9.B%
87% r
1L
Table 9. First Ballot Test by Propensity to Vote
f
City o(Peteluma - Stievis and Roads Study Godbe Research S Analysis
Page 24
Over Propensilyto
Vote
Low fMedium
� High
1 { Base
Definitely Yes
1 502 '_ 119 116
124 21 27
248h t 17.9% 22.9%
_3
242
71
29.5%
Probably es
Y Y
133 { 40 35
-
a 52
26.5% 336/ 3030A%
�� 21.4% r
I
Probably no (
1724 k 182 174%
1
—_
Definitely no
ar 111 �, 20 ( 1B
y
65 s
i 22 _1". 1= 16,7 s 15,6%
26.9% i
( —p'
(` DK I NA
i�....�.�,.._
47 e 18 1 16
9.4 0 13.6% 13.7%
14 i
5A%
City o(Peteluma - Stievis and Roads Study Godbe Research S Analysis
Page 24
Pial Bal Iat Test- Special Tax Sample
Table 10. First Ballot Test by Household Party Type
Table 11. First Ballot Test by Homeownership Status and
s Overall
Overall n
Household
Party Typm^�e ®���I
R �-
Democrat Damaarat 4 Republlc-Rapulan
(2+( I an UI
1 an (2+) Other Mlzsa
( Base
R (.
a 50 100 B4
4
1 75 59 � 142
[1
[ Definitely yes
124 22 25
7
17 �? 14 39
r
24 6 6 22.4'/ I; 2&B% I
16.6%
,E 23.2 23 7,6 6 7.2%
'
k Probably Yes
133 �5 19
26 15% �' 6.0° .9°'_�.
2L1%
j 19 Gt 1 33
�I 251 L 27 1 � 23 0% P
Probably no
86 ( 16 16�t
172 100"
6
12 p 10 Y23
95
�i
� 24990
18.55-
19.0%j
156 F 16.7% 76.4It%
' Prnbahl no Bfi
Probablymc,
q
-0 11 7 II 15 1
7
1 15 35
De
f flnilely no
22.11"�[ 1417'% V 21.194 V
17.01
1 28Z% E 20.2-'h Ii 24 6% I
t DK'I NA
47 10 6 ii
If 9.4% �' 9.9". a 7.7% j�
6
20.0%
ice. 6 e 4 12
7.4% 7.3 6.696
Table 11. First Ballot Test by Homeownership Status and
s Overall
Homeownership,i
Age
(
C
Status
+ 18-29 36,-39 49-09
59 64
65+
Base 562 �f
.�
Renu Own .�
120 391
-_I
years }' —years p years
( 42 166 3 124
years
12a
years
�� 67
Definitely yes z 124
9 Z49
1 36 94
F
l 249/ it 249/
74 27 29
t
767 2'0.7 23.5%
35
274%
! 22 E
26.4
it �i
ya
133
Probably yes
265/
36
31.6%
95
�i
� 24990
15 26 39
i 355 9 26.3 g 304%
35
� :7.4_%
I 13
185%
' Prnbahl no Bfi
Probablymc,
q
�j 206%
6�1
163%
�''
13 ll 15
�� 2096 10.7% 122%
25
194%
16
22.0%
Definitely no Ill
2219
10
I 1
� 12 7
9 0
i1 251 /
7 21 3.
( i
7 26
16 4 '' 26.� 195
1 9
14.590
23
28.1°5 f
IIa,,I
DK 1 NA y 47
.� 19 94,
13
1050
((
ry 34
9.0% It
�i 1 ( 16 10
22 15.6% 7.9%
X7_
k 14
�R 113%
6 —t
69%
Coy uffe(alwna-Sneelss and Roads Study Gudbe Research d /Inin5is
Page 25
Parcel Tax Threslmld - Special Tax Sample
Parcel Tax Threshold - Special Tax
Sample
Q5. Ifyou knew that the proposed One of the central objectives of the study was to estimate the maximum tax rate at which a
measure would be an increased necessary percentage of voters would be willing to support the measure. To achieve this
assessrnentin the annual properly objective, GRA has developed a'Dutch Auction' technique, which has been successful in pre -
ax of 5 would vou roe yes dicting the appropriate tax rate for passing a tax measure. In a'Dutch Auction', file respon-
or no on (his measure?
dents are first presented with the highest proposed tax amount (in this instance, for a parcel
lax) then the next highest lax amount, until the lowest tax amount is read. For each tax
amount, respondents are asked whether they would vote yes or no on the measure. Those
who voted 'definitely yes' for a given lax amount were automatically coded as 'definitely yes'
for all smaller amounts.
Each bar in Figure4 represents one tax threshold and each segment of the bar represents a
voting position on the measure at a particular tax threshold. As shown in the figure, 51 per-
cent of voters would be supportive of the measure at the highest tax threshold ($80). As the
tax rates decreased, the percentage of voters who would support the tax measure increased.
We see that if voters heard that properly taxes were to be increased by $40 peryear, 69 percent
of the electorate would be willing to support the measure.
$80
$70
$60
$50
$40
4. Parcel Tax Threshold
0% 20% 40% 60% 80%
o Definitely yes
la Probably yes
City of Petaluma - Streets and Roads Study Godbe Research & AnalSiis
Page 26
My Sales Tax Tlu''sbohl - Special Tax Sample
City Sales Tax Threshold - Special Tax
Sample
06. Ifyou knew Mal d e proposed Using the 'Dutch Auction' technique as described on the previous page, respondents from the
measure would be an increase in Special Tax Sample were presented with three lax thresholds for a potential citywide Sales
die local sales lax from 7.25 percent Tax. Each bar in Pigure5 represents one tax threshold and each segment of the bar repre-
m_, would yon voteyas Deno seats a votingposition on the measure ata articular lax threshold. As shown in the figure,
on this measure? 68 percent of voters indicated they would bes supportive of the measure at the highest tax
threshold (ane -quarter percent increase). As the tax rales decreased, the percentage of voters
who would support the tax measure increased. If voters heard that the City sales tax would be
increased by one-tenth percent, 70 percent of the electorate would be willing to support the
measure.
Figure 5. City Sales Tax Threshold
0.25
0.15
0.10
0% 20% 40% 60% 80%
El Definitely yes
® Probably yes
City olPeoduma -Streets and Roads Study Godbe Research S, Anab�sls
Page 27
Spewing Projects - Special Tax Sample
Spending Projects - Special Tax Sample
Q7.. The money generated by the The ballot language used in Question 4 slated that the proposed measure would be used to
measure can be used to fund pay for the repair and maintenance of streets and roads within the City of Petaluma. The
various types ofprojecu in the City of purpose of Question 7 was to provide more specific information about the types of projects
Petaluma. would
keels d some o
the munep would be used to that may be funded by the measure, as well as gauge the impact of this information upon
ld to I
would you be marc tiles likely to voters' likelihood of supporting the tax increase. Specifically, respondents were read a brief
vole yes on the measure, or does this description of a project that may be funded by the measure and were subsequently asked
Information have no effect on your whether they would be more or less likely to support the measure given the information. The
vote' order in which each spending project was presented to the respondent was randomized to
avoid a position bias.
To ease interpretation of the results, responses were recoiled and averaged. Individual
responses of'much more likely' were recoiled as +2, responses of 'somewhat more likely'
were recoiled as +1, responses of 'no effect' were assigned a value of 0, responses of 'some-
what less likely' were recoiled as -1, and responses of 'much less likely' were recoiled as -2.
Figure 6. Proposed Spending Projects
orh Repair alrJ malmata pntatW rbYuu SeMpts
9rg uu,.nn,e-,,,qo• nrnele
.Y. Fb t.14mb
We Help Fa.rzg lis epnlrerbn
Qin IyN «Ilpp. ntmnan ✓¢n4�mn IavN....
OJI R.P it a,e) n✓Y.0 :JwclLa
.Th Prov19: WaIXWys iWrrals
111n lN, r nen-pin nnvm ire
O)i povitlu Lih. oavaoMad,
an Mn1de e¢css aterollle Pew ums RNr,
For analyses of this nature, GRA looks for mean score of 1.00 or above which indicates that
voters, as a group, would be at least 'somewhat more likely' to support the measure given
that it would fund the project. Three of the projects tested achieved this threshold: 'Repair
and maintain neighborhood streets' (1.05), 'Resurface major streets' (1.04), and 'Fix pot-
holes' (1.04). Two of the remaining projects, although they did not meet this threshold, were
quite close. 'Help reduce traffic congestion' (0.99) and 'Repair and maintain major streets
in Petaluma, such as Petaluma Boulevard, McDowell Boulevard, East Washington Street
City of Petaluma - Streets and Roads Study Godbe Research & Analysis
Page 28
Spending Pi ojecis - special Tax Sample
and Ely Boulevard South' (0.98) each exceeded a mean of 0.95. The remaining five projects
had means ranging from 0.69 to 0.36.
For the interested reader, Tables 12 and 13 show how the impact of the information items
upon support for the measure varied by voters' responses at the first ballot test and their
household party type.
Table 12. Proposed Spending Projects by First Ballot Test
Overall i'.
, Q4 First ballot
test
Y
Y
Deyes
�I Pryes +Probably
Y
1,"'
Definitely
�l
f(( -
yes
' Y es no
n� o a
DK /'NA
l_
:Base
079 113
�� 1.04 C 0.61
I` 0.09
1.17 1
Q7bRepair and
maintain,neighborhood
1.05 (j 1.51
1 1.42 % 0.82
tl 0.19
V 1.28
( streets -
EQ7g Resurface major
h
r'
streets
1.041 63
-1
lj
1.3G
§ 0.77 �
0.11
1.23
(((
ptp Q7c Fix potholes1
04 pi43
1.34 0.72
0.33
1.41
g7d Help reduce tra
tffic
f_ congestion
4
0.99 1.19
1.22 _ 1.03
1 0.30
1.32
q7a Repair streets
�;
L
�P
such as Petaluma
0.98 i 1.51
1.30 N 0.76
� 0.01
1.36
Blvd_,
1s
4 07f Repair and replace
jj g -p
0.69 0.95
0.87 0.61
0.13
0.95
t sitlewalks
Q7h Provide walkways
II
062 N 078
l
i 0,83 - 0.4--
!i 7
r
i- 0,03
-
i
i.23
and balls
Me Repair and replace'-
Y
pp 0.61 I 0.83
h
`I
0.81 0.47
V
0.08
J
1.02 1
storm drains
071 Provide bike
0 52 N 0 78
75
0. 0,27 (�
-0.10
1.11
accessibility
�i
07i Provide access
along the Petaluma
0.36 0.66
.1
1 0.51 0.20
° -0.20
0.81
'River ..�
City o(Pelalurna -Streets and Roads Study Godbe Research & Anidsis
Page 29
Spending Pmjecls - Special Tax Sample
Table 13. Proposed Spending Projects by Household Party Type
City afPetaluma - Surds and Roads Study Godbe Research & Anallpis
Page 30
Arguments Snpputling the Mensnre special Tax sample
i
Arguments Supporting the Measure -
Special Tax Sample
Q8. In file coming montlhs, you may Ballot measures do not succeed or fail in a political vacuum. Proponents of a measure will
hear supporters of the street and present arguments to try to persuade voters to support a measure, just as opponents will
doadsmeasure abbe been present arguments to achieve the opposite effect. The objective of Question 8 was thus to
Supportersng, talk above the proposal. present respondents with arguments in favor of die measure to identify the impact of the
Supporters say that . After
hearing this, couldyou be more or information items upon their support for (lie proposed measure. Arguments in opposition to
iessbifeiym voteyeson the measure, the measure were also presented and will be, discussed in the next section. It is important to
or does this information hale no note, however, that the order in which respondents received Questions 8 and 9 was random-
eQeci on your vote? ized so that respondents were riot consistently presented with favorable arguments first (or
last).
Again, voters' responses were receded using the -2 to +2 scale, with -2 representing 'much
less likely' and +2 representing 'much more likely' to support the proposed measure. An
overall mean score was then derived for each argument by averaging all recoiled responses.
Thus, for example, a mean score of +1 indicates that respondents, as a group, were 'some-
what more likely' to support the measure upon hearing the argument.
Figure 7. Arguments Supporting the Measure
00tl Lonper we ..If mom ezppnzia. Iu 11x
001 Committee will be IurmoJ lu muum i e.p.nf propmly
fees P.lalum..".a'. naw peon refit -ed lar 50 year
Map RITC mled Pelauma'z streets az went In Csunly
each Aefi-sm.nt commercial property ewlmre will pay az well
Qal As a Wee Ina, l.mlzlz will pay. pmllon of lno m.asur.
091 bad r.ad. co.15]50 p.r year m car damonn
We Basilicas cammoney suppons lie measure
08. City may be I.—d Io IoM. 5's Imm slh.r pmnra,nz
The most compelling positive argument tested was'We need to act now. The longer we wait,
the more expensive it will be to make repairs and the worse the streets will get' (0.92), fol-
lowed by 'A Citizen's Oversight Committee will be formed to ensure that the funds are spent
properly' (0.87), 'The streets of Petaluma have been neglected for over 30 years, and they are
now in desperate need of repair' (0.89), and 'Metropolitan Transportation Commission
(MTC) rated Petaluma's streets as the worst in the nine county Bay Area' (0.81).
City orPetaluma - Streets and Roads Study Godbe Research S Matins
Page, .41
Argunrenis Suppol'Ihigibe Measure Special Tax Sample
Table 14 shows the mean scores assigned to each argument by respondents who were swing
voters" as well as by respondents' gender. Of the swing voters, those changing their vole from
'negative to positive' rated the positive arguments more favorably than did the 'positive to
negative' swing voters. In addition, women were more likely to give a higher rating, overall,
across the arguments than their male counterparts. Tablel5 shows the mean scores
assigned to each argument by respondents' age. Overall, those between the ages of 50 and 64
years, rated the positive arguments more favorably than did respondents in the other age
groups.
Table 14. Arguments Supporting the Measure by Swing Voters and Gender
Imm�� Overfill I'( Swing Votes r
p Gender
h
No[ UW. P, In,. le
le Pas Ilius Negative
Malo 'I Female i
{�
ttt
i Base
0.71
y 090
04806�4
0.77
Longer we wolf
08dw
S f
= orvx F
more ;expensive fa Fl
0.92
it 1.23
0.55 $
p.
p,8� 0.95
OBf Committee Mina.
formed to ensure $'e s
9.87
Y 1.11
0.65 {
0.72 1.01
7
_ spen�erly_Y,�_l
-
OBSPetal.ma.treatia
hove been neglected far
0.84
�� 104
049 !
9,82 0.57
ty;
I 30 years
_
r qgb MTC rated
rp
E Petaluma s streets a5_
0.81
I 097
0.72 €
0.77 0.66
s worst In County
t 7
rr
Obh Assessment
commernlalpmperly
076
�I 098 1
f
045 j
0.73 0,7B
oC
1 owners will pay es wail
II 4
I
OBg As a sales but,
tourists will pay a `
065
h 090
02- 4
0.59 0.71
portion of the meas.;, j
081 Bad reads cast 5350.
0 53-r
0.66
0.55 E
___
p.3B _r 0.66
per year In car tlargon.
I`
fi
OBc Business community',
supports the measure y
.
051
( 065
0.33 [ 0.42 @ 0.59
One City may be forced
to take S's from other
0.49
_
it
12, 0.53y
(
0 33
0.46 0.52
programs
n
r
[
"Respondents who changed their vote from probably no, definitely no, or don't know/n9 answer t9 definitely ycs or probably
yes were laheled'negative to positive'swing voters. Alternately, individuals who changed their vote from definitely or
pmbablyyes to probably no, definitely no, or don't know/no answereere labeled 'positive to negative'swing voters.
City afPelaluma - Streels and Roads Study Godbe Research S Analysis
Page 32
Argnlmmis Sit pput Iing the Measure - Specl0lTax Sample
Table 15. Arguments Supporting the Measure by Age
1
:Dverallyll� : Age
�y
1. 18-29 r 30.39 $_ 40.49 q 50-64_ h 65+
j years r years ryear. (I years I years
Base
071 Me 4 073 11071 � 079 � 086..-.�
qed Longer we wait
i more expensive to fix
t p�
0,92 0 600 85 0 9n x 102 ; 0.93 '
-
Q6f Committee will be
i
formed to ensure S's
t 0.67 e 0181 OM t 0.84 0.91 099
I spent properly
Y
COO Petaluma streets
have been neglected for
it 0,84 j 0,67 Ih 0.77 I: 0.84 [ 0.94 J 0.98
_ 50 years
if(
I 08b MTC rated
Petaluma s streets as
�I I
0 81 0.78 0.83 075 0.90 0.81
worst in County
Clint Assessment
commercial property
0.76 k 052 11 096 1 0.69 0.91 0.63
owners will pay as well
I
qOg As a sales tax,
tourists will a
I J G yI
0.65 0.41 0.73 0.65
pay
l
074k 0�=�
portion of the measure
�� r
`! 001 Bad roods cost $350
0.53 0.38 0.52 0.52 0.03 �� 0.44
1 per - year in car damage
11
QG. Business community
supports the measure
i p
�I 0 of 0. 9 i' 0.49 i 0.49 p 0.56 0.60
Me City may be forced
p �.
to take V. from other'.
I 0 49 0.67 0 59 0 64 0.46 0.11
�Wprograms
City, o(Pefaluma - Streets and Roads Smdy Godbe Research d Anal}sir
Page 33
Q9. In the coming months, you may
hear opponents of the street and
roads measure we've been
discussing lath about the measure.
Opponents say that After
hearing Ibis, would you be more or
less RATIj, to vole yes on the measure,
or does this information have no
effect on your vote?
Argi nlenls Opposing the hleasure - special Tax Sanyle
Arguments Opposing the Measure -
Special Tax Sample
Whereas Question 8 presented respondents with arguments that were designed to elicit sup-
port for the measure, Question 9 presented respondents with arguments that were designed
to decrease support for the measure. The responses were recoiled and averaged in the same
manner as Question 8.
Figure 8. Arguments Opposing the Measure
Oat. M %noule ni sry:rcnno an e:wc uho uft,,ucn
Wb 1mac.im alrtvN• loo 1) nn
WN N nM l:.Ylo n,M in, IM. lolibh
O'x 1laavum.111 nal 11•.%rnuUh Mnas
ORI Coartl' h 4a9 LY-bi o %ales la. na ticubirlU
V_: x'.111 mboor.. %moavy
O`d,9a .,, a% LU Team T hbhel Uves
Despite presenting respondents with arguments that were designed to reduce their likelihood
of supporting the measure, two of the arguments tested actually resulted in minimal net
increases in support for the measure and two of the arguments resulted in virtually no
change in support. Only the arguments that 'If the measure passes it would commit the City
to as many as 20 years of higher taxes' (426), 'The City will mismanage the money' (-
0.20), and 'The County is considering a sales lax that would fund local street and road
repairs, so we shouldn't double -tax ourselves by supporting the City's measure' (415) actu-
ally reduced support for the measure, more than a negligible amount, among respondents
as a whole.
For the interested reader, Tables 16 and 17 show the mean scores assigned to each argument
by respondents' income and household party type.
City or Petaluma - Streets and Roads Study Godbe Research & Analysis
Page 34
Arguments Opposing the Mensine - Special'Pax Sample
Table 16. Arguments Opposing the Measure by Income
Table 17. Arguments Opposing the Measure by Household Party T pe
Overall;] Income ;.
!,JL
HeuseIt Id.
Republicahl
)
�111
p
11 $401, Iv h $601( to $90k to �'$110k to
.a..................a
I Olhvr 1 Mixed
I l $40k it $591, �I $091, $1091, _31491, $150k+�-
ease
0 070 07�TO�B p O.D3 0 02 X0.04 -0.14 1
q9a We shoultl be
� �� ����E I
4
spending on Issues like
i 0.06 D 06 0.24 0.16 0.00 -O.D5 -0.04
education
`t q9b Taxes are already
e p
-.loo high '.',
0.04 f: 0.10 0.08 Om 0 09 �� -0.14 0.04
q9g Its '.nal fair to make
-the vast. pay for this
- '- --- I----� -�?
0.01 L -0.04 0.03 ( DOS it 0.05 -D.oD 0.12 1
09. Mansur. will not
raise enough funds'
u 001 a 007 -0.10 OOe �l -DOa € 0.09 0.03 i
_
aid County Is
considering sales tax, nod
double -tax
_. :_ii r 1.
[ �jI
0.15 031 - -0.07 I- -O.it I�A -0.01 -0.64
r
q9c City Will mismanage
money
-0.20 -0 OB 0.32 i 0 3fi3fi , 0 15 0.10 -OA3
¢ 3
q9f As many as 20 years
of higher taxes
�-H 1l 1,,
0. 60 20 -0.3'0 0.07 ry tOOt6:u5:-:O.:24L
t 1..01
Table 17. Arguments Opposing the Measure by Household Party T pe
City o[Petaluma - Surds and Roads Study Codbe Research & Analysis
Page 35
h. Overall i`_`
Democrat t0emac5tl
r a1i t 12!)'.
HeuseIt Id.
Republicahl
)
�111
Party Typo
Repuhlicanr-�.�
let
.a..................a
I Olhvr 1 Mixed
tP Base �R
007 t, -007 -01
-0.21
0.19
IL -000 0.07_
09a We should ba
f
spending an Issues like
00G 0,15 0.07
0.15
I 0.15
I 027 j 0.17
education
E 09b Takes are already
Ive high
t
004 002 -0.01
!
P 012
.002
019 009
v_�
09g Its not fair to make
p__�y.
the Voters pay for this I�
001 002 -0.07
(? Y
It 0.07
-020
014 0.13
i 09e Measure will not �'
E misveboughfunds041
_
t
013 ° -009
i
-0.15
P
gId:County is
�considadng iex,
��
sales no.M
doehlnlax
p 15 .0 33 j -0-17
I 0L4
-0.21
-0.18 -0.01
{' .9.69 1t
-0.22
-015 = -0.16 1
I 09c City will mismanage
j money N
-0.20 li -0.17 ^� -0.03
ears:
�,
026 -0 4 i .0.22
.0.32
q -0.51
-0.19
(p9f of kigher taxes
II
-013
flflfl
City o[Petaluma - Surds and Roads Study Codbe Research & Analysis
Page 35
SeomW 6allat Tna Special Tax Sample
Second Ballot Test - Special Tax Sample
Qio. Nola drat you've heard a little After providing respondents with the wording of the proposed measure, the possible tax
more about the proposed measure, amounts, a list of projects that may be funded by the measure, as well as arguments both in
ler me read a summary of it again. favor and against the measure, respondents were once again asked whether they would vote
In order to pay for the repair and yes or no on the measure that would increase taxes to raise 20 million dollars to pay for the
maintenance o£local streets and
roads, shad lite City ofPetalu„ la repair and maintenance of local streets and roads throughout the City. Because respondents
increase local taxes to raise twenq, learn a great deal more about the proposed measure during the survey process, the survey
million dollars? simulates a campaign environment. Therefore, Question 10, also referred to as the second
ballot test is useful in assessing the way in which support for the proposed measure may
change once voters are presented with basic information and arguments concerning the
measure. The language used in the second ballot test is identical to that used in the first bal-
lot test to ensure that any change in support for the measure is due to lire information pro-
vided in the survey -- not due to subtle changes in the question wording.
Figure 9. Second Ballot Test
Definitely no
19.4%
Probably no
11.8%
DK INA
6.8
Probably yes
27.1
efinitely yes
34.9
As shown in Figure9, approximately 02 percent of respondents indicated that they would
support the measure at this point in the survey, 31 percent staled they would not support the
measure and seven percent were undecided or unwilling to state their opinion. These num-
bers represent an 11 percent increase in support from the first ballot test.
TablelB provides some insight into how certain respondents changed their minds about the
measure during the course of the interview. Approximately 95 percent of respondents who
indicated during the first ballot test that they would 'definitely' vote yes on the measure con-
tinued to support the measure at the second ballot test. On the other hand, 81 percent of
respondents who initially indicated that they would'definitely' vote no on the measure con-
tinued to oppose the measure at the second ballot test. In addition, the majority of respon-
dents who initially indicated (hat they did not know or had no answer (DK/NA) supported
the measure at the second ballot test.
City of Petaluma - Streets and Roads Smdy Codbe Research & Analysis
Page 36
Secmid I allul Test Special Tax Sample
Table 18. Second Ballot Test by First Ballot Test
Tables 19 through 22 show how the distribution of support for the measure varies across a
host of voter characteristics. The tables show that overall support at the second ballot lest
('definitely' and 'probably' yes) for the measure was greatest among those with an annual
household income of $60,000 to $89,999, those with a 'medium' propensity to vote, individ-
uals who live in dual Democratic households, females, and respondents who perceive the
streets' condition as 'fair'.
Table 19. Second Ballot Test by Income
( L Overall i i. Income
__
Under Wilito III $Bok to
$90k to
1$119k to
P %MIR $59k'- $B9k
� '$189k
� s149k.:
5158k +
r*
Base p 502 1H 70 741 123
64
51
R4F}, F �
--` !r
H H
�2=8�F
175 3 22 49
Definitely yes t 23 t
' G 349 (. 326 ;l 296% r 398%
28
143340
( 22
�' 42595
10
� 34.9%
Probablyes 136 22 % 25 11 34
Y 272. 31.11%_33.7% 27.495
N 15
233%
12
22.6%
6
19.8%
6
Probably no 59 5 a 10 13
-, 1
- i 117 � 779 r 13300 tOBo
6
� 9.4%
10
20--2%
3
11 B%
I 97 t 15 12 f 15
Definitely no t
12
ppp
8
183 l; 212 156:6 h 11.95
187`,6
144%
283%
_
34 y 5 F 6 I 12
II 3
1
h t
DK I NA
- fi.9 6 7 B. 196 1 10.1%
11l 5.3%
-
5.296 I
City of Pdalmna -Surds and Roads Study Godbe Researdi & Analysis
Page 37
Second 9allnl Test - Special Tax Sample
Table 20. Second Ballot Test by Propensity to Vote
[ ..._.,�.�. Overall %: Propensity to Vote
Table 21, Second Ballot Test by Household
j 3Overall
E Low Medium
II High.;l
Household,
Base 502
it9 116��
242 ;
'i
4, 175
36 52
82
(
Definitely you
r_24 ?L6-
30.2% 44.7%
34A%
502 100
100
^Probably yes f 13fi
27 2 / `.
35 32
29 6 2].5°
09 S
24.456
142
DeMilal es
YY
175 38
J'
E� 31
{
Probably no { 1159
1215 r 17
98%
_
Definitely no '.� 19.3 { 19 17
i v 3/ 159
61
37.1°�
_
34 #
DK I NA
11 '' S M1
3.9%
15
p6.49%
Table 21, Second Ballot Test by Household
j 3Overall
-W
Household,
Party Type
'i
pDemecralit
Demodral
[I (2+)
=Repb
� an (1)
(1)
Re '.
an (Z+), � -, Other
an(2+
Mixed )
502 100
100
;v 84
42
75 59
142
DeMilal es
YY
175 38
J'
E� 31
j s 15
I 20 15
56 I
`y.
1
349% 38.09,14
37.1°�
359%I
26.1/ 258%
39.6%
p[
G
136 f 29
29
9
6s 22 16
32
ProbablyYes
272/ h 295
Ij
33.4912
21.0%
290,fi 27.5%
22.6%
59 II
5
f7
10 F 10
14
Probably no
117 119
e 5.5%
17696
1386 Y 17.5%
10.196
Definitely no
{�E�
97 10
k. i6
_
�- ]
19 13
--.-_
32
_
VI
193% 104/
f� 18.5%
k 159%
t 25t 218%
22.9%
DK 1 NA
`F
34 j 10
5
44
4
7
j
p
,°
89 10.2%115.5%
9.6%
59% 7.3°%
4.9%
G j4l'aaluma- Strcels and Roads Study Godbe Research & Analysis
Page 38
Second Ballot rest Special Tax Sample
Table 22, Second Ballot Test by Gender and Condition of Petaluma's Streets
City of Petaluma - Streets and Roads Study Godbe Researdi & Analysis
Page 39
II Overal
Gentler r' Condition
of Petaluma's
5treels
h�Male
C
Female �EzceIle nt;
(� to Good`
Falr,
I poor Very
pcor
Base
�' 002
234 i 269 102
�) 122
121 156
s
P
L' r, t
1
—46
Definitely yes
I' 175
90 66 22
r.
35
71
,.
34.9/
P 3B3'o 31.9 212%
209
364e 45.696 I
I Probably yes -
136
272,E
49 S7 I, 32
21 19 32.5 - 31.1 0
50
410/
�=27 N 28
222% jl 17.696
Probably no
59
27
�' L 32 ( 23
I'
F 7
I! 17 12
I.,
I 117
117 �' 11.9/ 2.9%
61
If 13.6% 75%
33
Definitely no
i—
97
54 43 x 19
¢[
20
I
21 37
193�G
°31 16.0J 1B 690
t 1fi.4%
176% (i 23.49.
t(I
DK I NA
(_
L
� 34 690
Y
74 p 21 � fi
59 7.6 6 2%
9
r 7B6
� 10
929,14� 57%
City of Petaluma - Streets and Roads Study Godbe Researdi & Analysis
Page 39
Pork and Recreation Pacilides- special Tax Sample
Park and Recreation Facilities - Special
Tax Sample
pn. Some people have proposed that The final substantive section of the survey asked respondents to indicate their level of support
in addition to funding the repair for the allocation of funds for die maintenance and operation of park and recreation facili-
and maintenance of streets and ties in addition to the repair and maintenance of local streets and roads within the City of
roads, the measure should also fund Petaluma.
the maintenance and operation of
park and recreation facilities in the
City of Petaluma. Masked to vote Overall, 48 percent of likely November 2002 voters surveyed supported a measure to fund the
today, would you vote yes or no on maintenance and operation of park and recreation facilities in the City in addition to fund -
the measure if it ivould fund park ing the repair and maintenance of Petaluma streets and roads, 46 percent opposed, and six
and recreation facility maintenance percent were undecided or unwilling to state their opinion,
in addition to street, sidenvalk and
storm drain maintenance? Figure "10. Support for the Measure with Park and Recreation Facilities
Without Additional Revenue
rIA
Nr'alldy n.
iP
City of Petaluma -Streets and Roads Study Godbe Research G Analysis
Page 40
Park and Ren emimi Fat Rl6s - Special Tax Sample
Q12. In order to fund the additional Next, respondents were presented with the information that the funding for the additional
park and recreation projects, the park and recreation projects would require an additional tax increase of four million dollars,
total tax amount would have to be bringing the total tax amount to 24 million dollars. With this information, 41 percent of
increased by 4 muton dollars to a
total of 24 million dollars. Xnmving respondents indicated they would support the measure and 50 percent indicated they would
this, would you vote yes or no on the oppose the measure. The remaining nine percent of voters did not know or had no opinion.
measure?
Figure 11. Support for the Measure with Park and Recreation Facilities with
$4 Million in Additional Revenue
Diu
Definitely yes
i
ZZ
Ciyoffetaluna-Strectsand Ruads Sludy - Codbe Research &Anal}sir
Page 41
Ql. Now long have you lived at your
current residence?
QA.Before you took this survey,
would you say dial you had beard,
read or seen a lot of information,
some information, a little
information, or no information
about dhe condition ofPetaluma's
streets and roads?
Addiltueal acmogmphlc and Oehzvim'nl Measures - Special Tax Sasepe
Additional Demographic and Behavioral
Measures - Special Tax Sample
Figures 12 through 21 graphically present the demographic and behavioral information col-
lected in the surrey. Some of the information was gathered during the interview (e.g., length
at current residence, amount of prior information on streets, own commercial property or
apartments in the City, newspapers that are read, household income, gender), whereas other
information was collected from the California voter file (e.g., zip code of residence, age, par-
tisanship, household party type). Although the primary motivation for collecting the demo-
graphic and behavioral information was to provide a better insight into how responses to the
substantive questions of the survey vary across certain voter subgroups, the information is
also useful for better understanding the profile of the City of Petaluma's November 2002
electorate. As with all results discussed in this section of the report, these results reflect the
data as weighted for gender, age, party affiliation, and geography.
12. Length at Current Residence
15 -year
2).4',
amts yea
2o.ty,
m n year
las%
..cG Ihan 4 yens
32.8%
13. Amount of Information on Local Streets
Na Information
18.1"
A IItft lnfarma8on
19.5%
OK INA
0.fi %
^'otof Information
21M.
Some Inrortna(ion
33.3
City of Petaluma - Streets and Roads Study Godbe Research & Analysis
Page 42
Ad4ilictml oemogizphtc:uul Belmeioial Mus9r9s Special Tax Sample
QB. Do you own commercial Figure 14. Own Commercial Property or Apartments in the City
property or apartments in the City?
Cammemlel property
10%
Apartments
Beth
Neither
94.0%
QC. 177iatnewspapersdoyouread? Figure 15. Newspapers that are Read
Sona Foe9p.pvl Rees opmaeml
Peedooe Weekly l Argus Courier
San Pmneiseo Ceemelp
011e"
0o.a9ad le. nowseapel
fireewnoepenorte Jpuleml
Reluwtl
San Fmrio.-Ennminer
0%
ma lo% 39% 40yc 50% 60% 70%
QD.Towrapthings up, Iamgain, Figure 16. Household income
to read some income categories.
Please slop me when I reach the
category that Gest describes ownaReweud Under $40,000
your 4991
total household income. Coke 5179,990
4.0%
$159,9001n S169,999
1.11.6 540,000 to 559,999
$130,000 to 5149,999 14'7%
4,4'.:
S110,000 to 5129,999
6 0�:
590.000 1. $109,999 590,090 m 509,999
34.5',:
11.7%
City o(Pelaluma - Sti ee6 and Roads Said, Codbe Research y Analysis
Page 43
Addioiurntl Demographic and Behavioral Measures Special Tax Sample
QE. Gender idenh'fied by voice of Figure 17.
respondent.
Fomalc
51<%
mmu
46.6=%
City of Petaluma - Streets and Roads Study Godbe Research & Malvsis
Page 44
Additional Booag'aphic and Behavioral I,feasures Special Tax Sample
Information Gathered from Voter File
Figure 18. Zip Code of Residence
96954
se.0%
Not metal 1 &29 yetus
d i r
r
EO 64 yn,S
go
YdY52
City o(Pelalmna - Sueels and Roads Study Codbe Research d, Analyais
Page 45
AddiiimmI Demographic and Behavioral hlea.vures -Special Tax Sample
Figure 21. Household Party Type
Democrat &Other MIxe26E Democrat (1)
9.4°/. 19.9%
Democrat & Republican
16.9%
Other (2) - - Democrat (2+)
a.6% 16.1%
Other ll)
8.2% Republican (2+) Republican (l)
14.9% BA%
City o(Petaluma - Saveis and Roads Study Godbe Research & Analysis
Page 46
Issues of lmpanance- Assesswent
Issues of Importance - Assessment
Q2. I'm going to read a list of issues The first substantive question of Elie survey presented respondents with a series of issues fac-
Pacingyour community today and ing residents of the City of Petaluma and asked them to rate the importance of each issue.
for each one, please tell n,r how This set of questions not only provides insight into how important an issue is on a scale of
Important as feel the user is
o('ex6amely
to
you, using a simportance, it also provides a relative ranking among die issues. Participants' responses
imponant', 5reryimportant� were coded using the following scale: 'not at all important' = 0, 'somewhat important' _
somewhaumportam', or 'not at all +1, 'very important' = +2, and 'extremely important' =+3. The aggregate responses to
important'. each item are presented below in the form of a mean, which is simply a summary statistic
obtained by taking the overall average of the response codes for the entire sample. A mean of
+2, for example, indicates that, overall, respondents felt that the issue was 'very important'.
To avoid a systematic position bias, the order in which the items were read to respondents
was randomized for each respondent.
As shown in Figure22, owners ranked 'Maintaining the water quality' (2.34) as the most
important issue tested, followed by 'Improving the quality of public education' (2.29),
'Repairing and maintaining local streets and roads' (2.14), 'Reducing traffic congestion'
(2.10), and'Prolecdng the environment' (2.04). It is worth noting (fiat all issues received an
average importance ranking of at least 'somewhat important', with many near or exceeding
life level of'very important'.
Figure 22. Issues of Importance
07b r.1.1r0lning anis, quality
02c I,npmvin8 In. d r.1d, of puallc ntlucation
021 RapalN,a and maintaining local shorts and roads
Oh Harrold, tm?c congostlon
02. Pmlocnn, Or, anvlmmnant
02i Prasarvlu, .,an apoco
02r Nalrolnln, pais and mcraandd f rallalae
02, Limind, ,,..h
021, Pmont.O., local is. in ..... as
Table23 shows the importance ratings assigned to each issue by respondent age and
Table24 shows the importance ratings by respondents' household party type. Overall, those
aged between 18 and 29 years shoved the highest importance ratings across the issues,
whereas respondents between the ages of 40 and 49 years displayed the lowest. Overall,
examining the importance ratings by respondents' household party type, ratings were high -
City of Petaluma -So-eels and Roads Study Godbe Research G Analizis
Page 47
Ismes of Iuiporlance- Assessment
est among respondents living in dual Democratic (2.06) or 'Mixed' (2.05) households and
lowest among respondents living in dual Republican (1.88) households.
Table 23. Issues of Importance by Age
�sr
Overaail
Age
-.�.�...��.#
18-29 l 36 39:„d 40 495964 65+�
years
years' f, years f years 11
years ? '
1
i ease
# 200 t 212
ZO 191 p. 2.01
2.08
«
f C12b Maintaining water
I 244 f;
246 214 [ 2.42
2.34
l quality.234
f I�
I
02c Improving the
[
! I
quality of public
p 229 § 278 (
2.30 220 - 222
2.30
education
_
j 021 Repairing and
f !
i maintaining local streets
l 2. 14 C 2.23 U
2.06 t 1'.92.26
2.32
and roads _
S I
a
t
�2e'.Retluc Ing 4. fl.
t "'
_„_
c P
210 217
2.08 a 199 2.19
2.10
.congestion
i
I rotecting the '
204 192
1'
r 01 2.06
2.00 ..-
2.04
envimnment
q_
t QZf Preserving open
! 1.90 210 p�
7.9"a 1 8 ! 1.8� 8
7.97
s�4
nl parks
! l'
[ 2 J
and recreation fa
antl recreation fadliges{_
1.84 1.99
2.02 1 79 1.79
!
1.78
_
I Q2g. Limiting growth
177 _t74
1.84 1.66 1.74 0_2.00
I tax
02h Preventingincreases
t-
1 586, 8 F
1.591.68
1.43 1.56
1.62
t 111.55 µ
City 6f Pcialuma - Sueels and Roads Study Godbe Researd & Analysis
Page 48
Issues of hupsnauceAssuswmu
Table 24. Issues of Importance by Household Party Type
n r n 1 rarry ,ype
��1 Qemocmt'Qemocmt6Re ublic Rem u611cA
!i
y ' 'i
[
�r- Base Y
'11) Ij (2+) ; an (1)
;an (2+)
1 Other Mixed
y�
200
{- _d
1 95 2 06, _�1 94
i BB
2.02 2 05
` 102b Maintaining water
239
_
_
fE
1(1
l quality
233 243 214
X�29al
7 2.33 3
i Q2c-Improving the F
i
-'-
quality of publ lc 219
�- 218 2.32 i 36
! 225
k 227 G 2.32 '
i _education
if I
li i1_t
a21 Repairing and
maintaining local streets 214
and roads
1192 it 2.14 I; 2.23
t
i 217
192 2.27
02e Reducing lmflic
cohgesllon Ij 2.10
195 I" 2.13 ! 2.21
I 2.16
193 1 2.13
666
q2a Protecting the +�-
!{ environment ' 204
_
�1 I:' 1
II 221 2.17 1.61
{ 1.68
t
r�'
L 2.24 2.02
C
02f Preserving open
190
1.97 2 .0 1 161
1.63
( 1.90
11
6
02dspace
old Maintaining parks 7.69
and recreation fa.lilies I!
1.91 3 1.94 f 193
�_ 1
i64
i .65 1.60
;
a2g Li_Itln9 9rawth �' 177
if i 67 76 i.5o
11.77 9 1.96
Q21h Preventing local tax'I�
1159
_7.76
11 1.32 1.56 II iso
_141
1.66
1.61 1 1.70
I Increases
'I
Q3. Generallyspeaking, would you Respondents were next asked to rate the condition of Petaluma's streets. Less than one per -
say that the condition of Petaluma's cent of those surveyed in the Assessment Sample felt that Petaluma's streets were in 'excel-
streetsis escellen4 good, lair, peer, lent' condition, 18 percent felt they were in 'good' condition, and 24 percent perceived the
or very poor? streets' condition as 'Fair'. The remaining 58 percent of respondents perceived the condition
of Petaluma's streets as 'poor' (2696) or'very poor' (3396).
23. Condition of Petaluma's Street
City of Petaluma - Sovelts and Roads Study Godhe Research A Analysis
Page 49
First Bill at Test - A5Sr5501c0I Smnple
Figure24 illustrates that among respondents overall, 50 percent would 'definitely' or'proba-
bly' support the proposed tax increase at this point in the survey. Approximately 41 percent of
respondents indicated that they would vote 'no' on the measure, whereas nine percent were
undecided or unwilling to state their vote choice.
As one might expect, support for the measure during the first ballot test was not uniform
across the various subgroups of voters identified in the survey. Tables 25 through 28 show
how the distribution of support for the measure varies across a host of voter characteristics.
The tables show that overall initial support ('definitely' and 'probably' yes) for the measure
is greatest among those with a'medium' propensity, to vote, respondents between the ages of
50 and 04, individuals who live in single Democratic households, females, and those who
perceive the streets' condition as 'fair'.
City of Petaluma - Streets and Roads Study Godbe Research & Analln'is
Page 50
First Ballot Test - Assessment Sample
Q4.1Jexr year, Petaluma voters may
One of the primary research objectives of this survey was to determine voters' support for a
be asked to tote on local ballot
measure that would raise taxes to pay for the repair and inaintenance of local streets and
measures. Let me read.you one of
roads throughout the City.
the proposals. In at to pay for the
repair and maintenance of local
streets and roads, shall the City of
Question 4, also referred to as the first ballot test, takes an early assessment of respondents'
Petaluma increase local taxes to
support for increasing taxes to raise 20 million dollars for the repair and maintenance of
raise twenty million dollars?lf
local streets and roads. The reasons for the strategic placement of the first ballot test early in
asked to vote today, Ivould you vote
the survey are twofold. First, at this point in the survey, the respondent has not been primed
yes or no on this measure?
with information about the ballot measure or project needs beyond what is presented in the
ballot language. This situation is analogous to a voter going to the polls with limited knowl-
edge about the measure in the absence clan information campaign. Secondly, the first bal-
lot test serves as a baseline from which to determine the impact of information items upon
support for the measure. Subsequent questions will assess support for the measure after pre-
senting respondents with additional information.
Figure 24. First Ballot Test
Figure24 illustrates that among respondents overall, 50 percent would 'definitely' or'proba-
bly' support the proposed tax increase at this point in the survey. Approximately 41 percent of
respondents indicated that they would vote 'no' on the measure, whereas nine percent were
undecided or unwilling to state their vote choice.
As one might expect, support for the measure during the first ballot test was not uniform
across the various subgroups of voters identified in the survey. Tables 25 through 28 show
how the distribution of support for the measure varies across a host of voter characteristics.
The tables show that overall initial support ('definitely' and 'probably' yes) for the measure
is greatest among those with a'medium' propensity, to vote, respondents between the ages of
50 and 04, individuals who live in single Democratic households, females, and those who
perceive the streets' condition as 'fair'.
City of Petaluma - Streets and Roads Study Godbe Research & Analln'is
Page 50
His] Ballot Test - Assessment Sample
Table 25. First Ballot Test by Propensity to Vote
overall b _'.. Propensity to Vote��-�
_ FLow ,-Medium High
'. Base 462 ate; 90
97 ami
252y
S.M. 462
� 117
1 22 71 89 12a dl 126 li 73
{{
Definitely yes yes �) 117 14
r254 �160/k2405.1301
23
76
9,6
Probably yes 114 .� 9
27
54
247 . 320
275%
213
Probably no 74 ie
13
a9
211
137%
15,%
i --
Definitely no a 115 -17
21
j fig
0 121.8%
i� 27 3
1195
OK I NA ! �f 41 �, 13
8
i 8 12.29 k 13.0/
10
80,4
Table 26. First Ballot Test
CRy ofPelalumn-Sireels and Roads 5ludy Codbe Research S Anafjsis
Page 51
Age
18-29[II. 30 39 :40-09 3. 50-64 .� fib+ °
I
years years i years years years
S.M. 462
� 117
1 22 71 89 12a dl 126 li 73
¢t Definitely es
y y 20,4 /
5 (' 25 30 36 17
20_5 i 28 2 24 .29,16 28.5_n 23.5%�
1146
Probably yes
247/
4 21 30 36 11
�t q
X25.0 23.0%X203 Ii 28.8%6 14,99%
Prbbabl 74
y no 161
5 9 1I 17 21 19
23.4% 10.6% 135 167y
115
�5 22 , 32 Ur 22—� 23
Definitely no
- U 20 09
f 24.4%124.3%.P 25.8%it 17 3°p 3Z 0°6
(
DKINA 41
PI 1 k 12 10 it A
I 88,6 i;
6.0 1' 73.9% a 61 �� 8.7%
CRy ofPelalumn-Sireels and Roads 5ludy Codbe Research S Anafjsis
Page 51
First Ballot Test- Assessmorm Sample
Table 27. First Ballot Test by Household Party Type
Overall
Table 26. First Ballot Test by Gender and Perceived Condition of Streets
overall
Household PartyType:.®�
Canditlon
Excellent
of Petaluma's
Demoorat°Democral Republic-IRepubllc•
Very
t
462 -
e
Male
246
6 Female
to Goad
213 66
j1) (2+) an (1) 1E an 12+)
` Other
1 Mixed
Bassa
Definitely yes r�i
'1
462 �N
117
[�t
62 5 116 7 31 69
10 0 36 ffi 5 it
( 45
139
4 40
Probably yes
25,4%t1
7141
t 247
159 32.6° 164i_� 244/
25 i 23 7 it 15
_16.9
12
26.696
31
i^ 60 N 17
280%20.295
37
34.0%
404 s 200° a 235/ +19
273,;
I( 22.3%
' 16.15.
74
i''
12
111990
1 t
I Probably no
Definitely nog"
13 76 5 12
6
t6
30
27.596
101
2106 15.49 16'% 17.790
179%
G 130°/
17
I 7.0%
115
11 25 f 6 20
14
39
t Definitely no
2509
_ 176Y B 21.5° 164/ 296%
322°�
281%
DK I NA ',.,N
by 41
3 112 fry 6 4
I 3
71
I
W...r
8481%
5.0% V 10A% 25.0% fi 490
5 7%
7.790
Table 26. First Ballot Test by Gender and Perceived Condition of Streets
City of Peraluma - Streets and Roads Study Godbe Researdi & Analysis
Page 52
overall
Gender
Canditlon
Excellent
of Petaluma's
Streels
'f
Very
Base
462 -
e
Male
246
6 Female
to Goad
213 66
Fair
I 109
Y;,�Poor
I 115
poor '
151
Definitely Yea $ :!'
II' 117
hhh�254%
�� 66 Ate.
267%
51 14
23.951. 160%
24
224%
31
I 26.6%
55 49
Probably yes _6
�
114
24 7°_Ii
�r 54
2t 9 /
i^ 60 N 17
280%20.295
37
34.0%
26
24A%
1-321%
6 31
20.5
�Probably na
' 16.15.
15 4 '.
(I 7
17 0%j 31.0%
12
111990
21
1 2 %
15
10.0%
Definitely nog"
115
25 04
72
29.0 %
pI 43 17
1 204%
➢
30
27.596
` 25
21.3%
94
29.3
_
A'. DK I NA41
(..m................s._........._._:
86%
17
I 7.0%
_
23 11
-06% t3.4%
6
53%
11
19.596
12
! 6.296
City of Peraluma - Streets and Roads Study Godbe Researdi & Analysis
Page 52
Assessment Threshold - Aesesammii Sanipkc
Assessment Threshold - Assessment
Sample
Q5. It you knell, that the proposed One of the central objectives of the study was to estimate the maximum assessment rate at
measure would be an increased which a necessary percentage of voters would be willing to support the measure. To achieve
assessnieni in the annual properly this objective, GRA has developed a 'Dutch Auction' technique, which has been successful in
lax of , a'ould,pou vote yes predicting the appropriate rate for passing a tax measure. In a'Dutch Auction', the res on -
or no on this measure: P
dents are first presented with the highest proposed assessment amount then the next highest
amount, until the lowest amount is read. For each amount, respondents are asked whether
they would vole yes or no on the measure. Those who stated 'definitely yes' on an amount
were automatically coded as 'definitely yes' for all smaller amounts.
Each bar in Figure25 represents one assessment threshold and each segment of the bar rep-
resents a voting position on the measure at a particular threshold. As shown in the figure, 51
percent of voters would be supportive of the measure at the highest threshold ($80). As the
assessment rates decreased, the percentage of voters who would support the measure
increased. Wesee that if voters heard that property taxes were to be increased by $40 peryear,
68 percent of the electorate would be willing to support the measure.
Figure 25. Assessment Threshold
$80
$70
$60
$50
$40
0% 20% 40% 60% 80%
o Definitely yes
M Probably yes
Cil' of Petaluma-Sotets and Roads Study Godbe Researdi S Anal}'sis
Page 53
Spending Prof ects - AssessnlenI Semple
Spending Projects - Assessment Sample
QT. The money generated by the The ballot language used in Question 4 stated that the proposed measure would be used to
measure can be used to fund pay for the repair and maintenance of streets and roads within the City of Petaluma. The
various types ofprayectrin the Cuy of purpose of Question 7 was to provide more specific information about the types of projects
Petaluma. Ifyou know that some of that may be funded by the measure, as well as gau e the impact of this information upon
the money ivould be used to_, gauge P
rvould you be more or less Irkely m voters' likelihood of supporting the tax increase. Specifically, respondents were read a brief
voteyes on the measure, or does this description of a project that may be funded by the measure and were subsequently asked
information have no effect on your whether they would be more or less likely to support the measure given the information. The
vote? order in which each spending project was presented to the respondent was randomized to
avoid a position bias.
To ease interpretation of the results, responses were recoiled and averaged. Individual
responses of 'much more likely' were receded as +2, responses of `somewhat more likely'
were recoiled as +1, responses of'no effect' were assigned a value of 0, responses of'some-
what less likely' were recoiled as -1, and responses of'much less likely' were receded as -2.
Figure 26. Proposed Spending Projects
U?L RgOir W,d i1,0inNi�1 „civtom nwJ alreL
arz Fn � .roa
mn nnlR mm,n.Irmee.a,�o..ln�
07, RlplhvL 1. 1-11 ue Aiulunu &rJ....
U'f Ral`�I[ m[I mp4.c: a <bwilks
m. RnnX, n nn !nr •.n mnn„n,wm.
0/I, F,winv Ik.w, u,N L.0 iL
0:I rm Ml bl"neh,slbllhy
01 P vXn n. �-ixn n nn, II,n PNnlnma RNn�
For analyses of this nature, GRA looks for mean score of 1.00 or above which indicates that
voters, as a group, would be at least 'somewhat more likely' to support the measure given
that it would fund the project. Three of the projects tested achieved this threshold: 'Resurface
major streets' (1.04), 'Repair and maintain neighborhood streets' (1.03), and 'Fix potholes'
(1.01). Two of the remaining projects, although they did not meet this threshold, were quite
close. 'Help reduce traffic congestion' (0.97) and 'Repair and maintain major streets in Pet-
aluma, such as Petaluma Boulevard, McDowell Boulevard, East Washington Street and Ely
Boulevard South' (0.95) each exceeded a mean of 0.90. The remaining five projects had
means ranging from 0.60 to 0.32.
City of Petaluma - Streets and Roads Study Godbe Rescardi & dnalysis
Page 54
511nnding Projects - Assessment 5aniple
For the interested reader, Tables 29 and 30 show how the impact of the information items
upon support for the measure varied by voters' responses at the first ballot test and their
household party type.
Table 29. Proposed Spending Projects by First Ballot Test
City of Petaluma - 9mts and Roads Study Godbe Roseerdi S Analysis
Page 55
Overall k
q4 First ballot
test
4
Definitely
H Probably ' Probably
yc
`Definitely
vl
DK NA
I -----I
w Yes -I,
'yes 11 no
no
j
Base ,',
J 0.75 !� 1.15
1 1.03 F 0.61
0.04
_
J 1.13
�
0.07
y� 1.24
07g Resurface major
streets -
1.04 C 169
F 138 0.89
_
j
1
g7bRepalrand
�I
maintain neighborhood
1 03 1.53
(I 11.47 0.76
0.16
1.30
streetsi
it
u
_ q7c Fix potholes
'i 1.01 �_ 1.42
1.31p 00 78
0.25
1.47
q7d Helpreduce traffic
0.97 1.25
F
1 19 # 1.02
0.32
1.25
_ Congestion
II
ng
FWe Repair streets
such as Petaluma
I 0.95 1.54
1 40 0.73
-0.12
1.44
_ Blvd....
!
07t Repair Repair and replace
060 0.91 4
P 0.80 0.51
0.00
0.94
sidewalks '
I � 1
j
iWe Repair and replace
0.59 0.79 ����i---0.8---3---
0.53
0.02
1.08
storm drains
d
r r
k
q7h Provide walkways
(((((( and tralis
i 0.57 I�I 0.80
r
0.79 t 0.47
-0.02
1.11
CIT Provide bike
accessibility
78 45 0
0 . I'
� � �)
0 ..
18 73 0
�
pp
-. N
17 0
q
0.91
071 Provide access.
along the Petaluma
0.32 0.71
0.40 0.20
-0.14 p
0.56
-River
r ii
City of Petaluma - 9mts and Roads Study Godbe Roseerdi S Analysis
Page 55
Spending Projects - AssesslnenI Sample
Table 30. Proposed Spending Projects by Household Party Type
City o(Pelaluma - Slnels and Roads Sludy Godbe Research & Analysis
Page 56
Overall ^, _ -, Household Party Type
a
Sj Democrat Oe moprpu611cen
at Re
� Other Mixed
fl �2+) 0 (i) (2+)
-
Base
015 0.72 y 0.89 ; 0.62 0.53 _ry.
0.72 080 f
( 079 Resurface majorr,[
i streets �.k
'
1.04 083 1.21 1.06 ! 097
k 0.66 1 1-14 6
_ _
Q7b Repair and TF
_ _ --
`I
maintain neighborhood : G
1.03 3 0 86 5 114 [ 0.Be f 0.92
0 92 1.15 E
streets [
f
j 07c Fix potholes F`7_1
01 _-0-89-11- 11 i 02 � 0.9-5,
1.03 E
07d Halp reduce imPoce
purification C
j '
0 97 1 04 11 1.10 1 1.07 0.73
'7�
0.89 @ 0.95
t77a Repair streets
E SuchNas Petaluma W.eM
eNd
0.90. 0 B8 0.99 W� 0.98 1.05
�
e� r
0,65 0.98
j l]7f Repair and replace9
sidewalks
0.60 0 55 0.69 r 0.27 0.43
k 0.61 0,66
PPP iii
107e Repair and replaca'
0 59 0.50 0 81 0.20 ( 0.40
O.fiO 0 6
� � storm dralns_g
h Provide
�
_y
walkways
0.57 0.76 0 �l
08 45 78 0
0
,t
£ 0 66 0.63.63
t{ and trails �3
.
A ti j
. I€
@ -.
_ e.._....,.,,._,...._......���
t rsCl7i.. Provide hike t [
( accessibility 0.45 0.44 0.74 f 0.35 0 10
,-....
t
F O.fi6 0.41
( o71 Provide access
f{
( along_ the Petaluma ,.,
0.32 037 2 0.45 0.02 0.13
4 0.43 0.46
_- River !
City o(Pelaluma - Slnels and Roads Sludy Godbe Research & Analysis
Page 56
Argwnenls Snppor ing the Meamre -Assess mem Sample
Arguments Supporting the Measure -
Assessment Sample
Q8. In the comtngmonths, you may Ballot measures do not succeed or fail in a political vacuum. Proponents of a measure will
hear supporters of the street and present arguments to try to persuade voters to support a measure, just as opponents will
roads measure we've been present arguments to achieve the opposite effect. The objective of Question 8 was thus to
Supportersing, talk about me proposal, present respondents with arguments in favor of the measure to identil' the impact of the
Supporlerssaydtal_ Ater Y P
hearing this, Ivoukiyou be more or information items upon their support for the proposed measure. Arguments in opposition to
less likely to vote yes on the measure, the measure were also presented and will be discussed in the next section. Il is important to
or does this information have no note, however, that the order in which respondents received Questions 8 and 9 was random -
Oki on your vote? ized so that respondents were not consistently presented with favorable arguments first (or
last).
Again, voters' responses were receded using the -2 to +2 scale, with -2 representing 'much
less likely' and +2 representing 'much more likely' to support the proposed measure. An
overall mean score was then derived for each argument by averaging all recoiled responses.
Thus, for example, a mean score of +1 indicates that respondents, as a group, were'some-
what more likely' to support the measure upon hearing the argument.
Figure 27. Arguments Supporting the Measure
Cela Lunovr we wml mvrp papvnmve Iv N
OOf Cvmmillev wlll h¢ fvrmvtl Iv unsure 5'¢ ¢pent pmpetly
06. Pululumv ¢Impls M1a1v Eevn nealvclaJ Ivr 30 years
COh MTC mlutl Polpluma'c m-1. v¢ wvr¢I In Lvunl,
Oah Aszv¢mnon!cammurclal prvpvJY v e.e, wlll pay as wvn
00, As v ¢vlva We lvurlz. will pvy v prop.. vl n,c movzum
0016atl rvvtls curl SM. per year In car tl¢map¢
00c Buzb,v¢¢ cpmmunlly ¢uppvriz lhv mv¢¢wo
OOu CHy may by Ivrcetl Iv l.k. S's Irvm wt,v, prvOmm¢
The most compelling positive argument tested was 'We need to act now. The longer we wait,
the more expensive it will be to make repairs and the worse the streets will get' (0.91), fol-
lowed by 'A Citizen's Oversight Committee will be formed to ensure that the funds are spent
properly' (0,86), 'The streets of Petaluma have been neglected for over 30 years, and they are
now in desperate need of repair' (0.83), and 'Metropolitan Transportation Commission
(N1TC) rated Petaluma's streets as the worst in the nine county Bay Area' (0.79).
City of Petaluma - Streets and Roods Study Godbe Research S Analysis
Page 57
Amgeaenls Supporting Ow Measure-Assessumnl sample
Table 32 shows the mean scores assigned to each argument by respondents' gender as well
as by owners who were swing votersih. Overall, women were more likely to be positively influ-
enced, across the arguments than their male counterparts. Of the swing voters, those chano
ing their vote from 'negative to positive' rated the positive arguments more favorably than
did the'positive to negative' swing voters. Table32 shows the mean scores assigned to each
argument by respondents' age. Overall, those between the ages of 50 and 64 years, rated the
positive arguments more favorably than did respondents in the other age groups.
Table 31, Arguments Supporting the Measure by Gender and Swing Voters
"'Respondents who changed their vole train probably no, definitely no, or don't Imuw/no answer to definitelyyes or pinba-
bly yes were labeled 'negative to positive' swing voters. Alternately, individuals who changed their vote from definitely or
probably yes to probably no, definitely no, or don't ]mow/no answer were labeled'positive to negative' swing voters.
City of Pelaluma - Streets and Roads Study Codbe Research E Analysis
Page 58
Overall- Gentler it Swing'Voters
t
�' (z Negative
R f Male i Female itc Posltive
Posltive tot
Ne ative
l Base
A 07070 [ 063 07BF, 09
0.56 t
OBdLonger we wait
more expensive tofix
0.91 0.87 0.96 1.26
0.65
oaf Committebe
will be�
farmed to ensure 5's
m 0.86 172 0 i 02 1,12
078 i
E spent properly
&a Petaluma streets
j!
have been neglected. for
0.83 0.80 086 1 04
0.59
30 years
1� [1
I CIBb MTC rated
E Petaluma's streets as
€C worst In County.
&& +
0.79 0.71 O.B7 1.03
deh Assessment
commercial property,
0.77 i� 0.72 0.82 1.OB
p 0.63 Ft
owners will pay.as wait
QBg As a sales lax,,
tourists will pay a.
portion of the measure
0.65 t 0.60 j 0.77 0.96
l
0.41
I QW Bad reads cost $Jso
7 0.49 0.39 0.61 f 0.77
per year In car damage
(i
PF P
08e Busmess community
I' N
0.49 O42 b 058 i 074
-0.54
(I 0.25
supports the measure
r We City may be forced
�t!
I to take V. from other
0,49 1 0.44 0.56 0.46
0.39
r. programs
f I
"'Respondents who changed their vole train probably no, definitely no, or don't Imuw/no answer to definitelyyes or pinba-
bly yes were labeled 'negative to positive' swing voters. Alternately, individuals who changed their vote from definitely or
probably yes to probably no, definitely no, or don't ]mow/no answer were labeled'positive to negative' swing voters.
City of Pelaluma - Streets and Roads Study Codbe Research E Analysis
Page 58
Ajgumetes Supporling the hfensere - Assessment Smnple
Table 32. Arguments Supporting the Measure by Age
City offelalume - St -eels and Roads Sludp Godbe Research d An4,sis
Page 59
OveraII'j
'. Age
[�I30
18 29 li 39 j. 40-09 i 65++
-Jd
15054.:
4 years 4 years °. years years years
Base y
1 070
_
X 044 f 066 0Y71 0.79 �. 065
OBd 1-00981 we waif
more expenslve to flx
091
e
057 h 074 090 L 1.00 0.89
Q8f Committee will be
farmed to ensure $'s
spent properly
0.86
Ml 0.74 0.B6 0,94 0.91 t
L + e
OR. Petaluma
r i
streets"
i�
have been neglected for
0.B3
4 0.54 0.63 0.87 0.90 0,98
70 Years
We MTC mad
Petaluma s streets
worstin County
0 79
h
N L'
I 0 45 (r( 0+84 k 0.75 p�4S 0.90 0.78
L V
Asse
QW) commercial
commarclal property
property
I M7
1
0.53 it 090 1 0.76 �� 0.86 li 0.55
ownersjwlll pay as well
Q89 As a sales tax
tourists will pay a
portion of the measure
k 0.68
i
--
t 0.15 0,80 0.62 1 0.76 11 0.64
i qBl Bad roads cost 350-r
0.49
r aafi
0.35 OA3 0.46 0.65 0.33
per year In car damage
Y
Oft Business community
i�
049 Gi
r L
0.31 0.40 0046 ! 0.57 0.60
_supports the measure
QBe Clly may be forced
L to take .6 s from other'.
0.49
! 0145 0 49 062 „ 0.52 a 0.10
r kl 'y I
-programs .
jj
City offelalume - St -eels and Roads Sludp Godbe Research d An4,sis
Page 59
Q9. In the conungmonths, you may
hear opponents of the street and
I ads measure we've been
discussing tally about the measine.
Opponents say chat . Aller
hearing this, would you be more or
less likely to vote yes on the measure,
or does this information have no
effect on your vote?
Argmneme Opposing the Weasure - Assessimml sample
Arguments Opposing the Measure -
Assessment Sample
Whereas Question 8 presented respondents with arguments that were designed to elicit sup-
port for the measure, Question 9 presented respondents with arguments that were designed
to decrease support for the measure. The responses were recoded and averaged in the same
manner as Question 8,
Figure 26. Arguments Opposing the Measure
W LTu4 am vua, [.o M1 ph
oz Nmcum wnl wl iv¢c crougn rmm:
Ola V/eu MulE Lv pW,v111,p all Npbpp Ilk, edu.ibu
Gey Wnnl Lirlw M lb,,J... txy fnr Ib4
GM Canny 4 vvevvbbin+
oX C Iy will mnmxn. emvnw
t A A, 1.1 W av ZD yluh 0 hirh:r lart.
Despite presenting respondents with arguments that were designed to reduce their likelihood
of supporting the measure, one of the arguments tested actually resulted in a minimal net
increase in support for the measure and one of the items resulted in virtually no change in
support. The arguments that'If the measure passes it would commit the City to as many as
20 years of higher taxes' (425), 'The City will mismanage the money' (417), and 'The
County is considering a sales tax that would fund local street and road repairs, so we
shouldn't double -tax ourselves by supporting the City's measure' (416), 'It's not the voters
fault that the roads were not maintained for the past 30 years, so its not fair to make them
pay for it through this measure' (405), and 'We should be spending money on more press-
ing issues such as improving education' (404) reduced support for the measure, more than
a negligible amount, among respondents as a whole.
For the interested reader, Tables 33 and 34 show the mean scores assigned to each argument
by respondents' prior information on streets and their partisanship.
City ol'Peoduma - Slims and Roads Study Godbe Research & Analysis
Page 60
Argunienls Opposing the hinaeure - Assessment Sample
Table 33. Arguments Opposing the Measure by Amount of Information on
Table 34. Arguments Opposing the Measure by Party
4 Overall P
Information
an Streets
_ v_
�Republlcan
Other or
I I A lot of
F
� Same
`. A little ��j
'. No
_ _
J 0.09 004 0.18
nformation
1.=. f r
In io rmatIan
Informatlan�informatlon
-
II 0.09 t 005
it 009
-014
�.JBasev
raise enough fundsr-�
001 001 004
I 0.26
�0.14
Q96 Taxes are already
0.02 0.91
0.02
g.m
-O.D1 r
:'.too high
j.
Cog Its: not fair to make -u
�j
q9e Measure will not
raise :enough funds
t -0.01 008
pp
tl 0.03
K@, -0.05
-0.07
-
considering sales tax, nd
_i
-0.23
Q9a'Weshould 16.!
Issues
4
'
Q9c Citywill mismanage
p '
spending like
_ education
�- -0.04 0.01
-0.08
; 0.01
-O.tO
Q9g Its not fair to make
the voters pay for this
-0'05 0 12
0.04
0.13
{'�, 0.12
Qgd County is
considering sales tax no
I
-0.16 -0.09
L' -0.04
-0.41
-0.25
double -fax
1
i
Q9c City will mismanage
017 6 012
-0.20 �A
-0.14
_Il
:money
-
-0.23
Q9f As many as 20 yearsj',
1
of higher taxes
0,�a 5 II 0.10
P 0 24
g -029
-0.44 t
Table 34. Arguments Opposing the Measure by Party
City ofPefaluma-.Strecis and Roads Study Godbe Research S eInal}5is
Page 61
t Overall G - Parity
_ v_
�Republlcan
Other or
Democrat
-: ors
Base_
_ _
J 0.09 004 0.18
-0.12
49h Taxes are already
F ryry
( too high -
002-� OOfi H- 000
0.12
q9e Measure will not
raise enough fundsr-�
001 001 004
I 0.26
q9e We should be
F
spending on Issues him
-0.04 I 0.04 0.23
0.02
education.
Cog Its: not fair to make -u
-0.05 I' 0.04 p -0.20
-0.13
the Voters pay forthis ',�
%
A
Q9d County Is -.
-
considering sales tax, nd
-0:16:] -0.12 -0.17
-0.23
double -tax
Q9c Citywill mismanage
p '
mane y..._
-0.17 'j -0.10' -0.30-0.11
oaf As many as 20 years
_
-0.25 -0.2t -0.33
p -0.23
of higher. taxes
I 1
City ofPefaluma-.Strecis and Roads Study Godbe Research S eInal}5is
Page 61
Second Ballot Test Assessment Sample
Second Ballot Test - Assessment
Sample
Qlo. Now that you've heard a little After providing respondents with the wording of the proposed measure, the possible tax
more about the proposed measure, amounts, a list of projects that may be funded by the measure, as well as arguments both in
let me read a summary of it again. favor and against the measure, respondents were once again asked whether they would vote
In orders pay for al repair and yes or no on the measure chat would increase taxes to raise 20 million dollars to pay for the
maintenance ol'local streets and P )
roads, shall the Cigv of Petaluma repair and maintenance of local streets and roads throughout the City. Because respondents
increase local taxes to raise nvemy learn a great deal more about the proposed measure during the survey process, the survey
million dollars? simcdates a campaign environment. Therefore, Question 10, also referred to as the second
ballot test is useful in assessing the way in which support for the proposed measure may
change once voters are presented with basic information and arguments concerning the
measure. The language used in the second ballot test is identical to that used in the first bal-
lot test to ensure that any change in support for the measure is due to the information pro-
vided in the survey -- not due to subtle changes in the question wording.
Figure 29. Second Ballot Test
As shown in Figure29, approximately 61 percent of respondents indicated that they would
support the measure at this point in the survey, 33 percent stated they would not support the
measure and seven percent were undecided or unwilling to stale their opinion. These num-
bers represent an 11 percent increase in support from the first ballot test.
Table35 provides some insight into how certain respondents changed their minds about file
measure during the course of the interview. Approximately 96 percent of respondents who
indicated during the first ballot test that they would 'definitely' vote yes on the measure con-
tinued to support the measure at the second ballot test. On the other hand, 78 percent of
respondents who initially indicated that they would 'definitely' vote no on the measure con-
tinued to oppose the measure at the second ballot test. In addition, the majority of respon-
dents who initially indicated that they did not know or had no answer (DK/NA) supported
the measure at the second ballot test.
Ciq, of Pelaluma - Streets and Roads Study Goode Researdi & Analysis
Page 62
Second Onllul Tea Assessnienl $maple
Table 35. Second Ballot Test by First Ballot Test
Tables 36 through 39 show how the distribution of support for the measure varies across a
host of voter characteristics. Looking only at columns of voters that contain at least H
respondents due to the inherent risks of generalizing the results for subcategories that have
fewer respondents, the tables show that overall support at the second ballot test ('definitely'
and 'probably' yes) for the measure was greatest among those with an annual household
income of 990,000 to 9109,999, those with a'medium' propensity to vote, females, individu-
als who live in dual Democratic households, and respondents who are between 50 and 64
years of age.
Table 36. Second Ballot Test by Income
Dverall
qA Flrst hallot test
Income
IL�Deflnitely ProbabIyf, Prohe6ly,DeflnitelYr__f/NA
i
Base
� 462
¢
l' 117 yeses J no no
Is 117 114 74 115
$90k to
r 41
Deflnitely yes
j
I 16
35G
lig 98 19 4 �. 4
637 428 6.0,"1 34
7
$09k
$109k
h ii
p 16.1%
l Probably yes
119
E 14 49 20 15
20.
71
257/
E 122 f 431% 205/ r 1330
f 49.3%
Probably no
53 8�a
3 6 32 11
r
2
24
115
! 2.3% 5.4%426 921
€
45%
^1
Doflnitely no
_
i 96
212 %
2 1 13 g 00
1 a% q 0.6% 176% 689/
2
5.995
'.DK /,NA
31 ��
I 9 t g 6
G
to
30
24.995
66
�+ - 8.1 9 7.3,5 . 5350
24.1%
Tables 36 through 39 show how the distribution of support for the measure varies across a
host of voter characteristics. Looking only at columns of voters that contain at least H
respondents due to the inherent risks of generalizing the results for subcategories that have
fewer respondents, the tables show that overall support at the second ballot test ('definitely'
and 'probably' yes) for the measure was greatest among those with an annual household
income of 990,000 to 9109,999, those with a'medium' propensity to vote, females, individu-
als who live in dual Democratic households, and respondents who are between 50 and 64
years of age.
Table 36. Second Ballot Test by Income
Overall
lye
Income
i
Under $qOk to
$6 k to
'
$90k to
$110k to
'$40k� $59k`
$09k
$109k
�I $149k:
$150k+.
' Base462
�Deflnilely
42 51N,W
X116 ��
71
5.5.
32
yes'
e5
162
9 17
47
29
24
X13
35.0%
�� 21.3% 33.7%
40.1%
41.295
42.9%
392%
Probably yes
-1
j 119
25.7%
13 13
� 30.2% ? 28 5%
30
24.995
I 21
29.3-1.
11
20.1%
6
18396
Probably no
I 53
1155tl
l 7 q 5
156 107,1
10
fl 99%
5
�%
DeOnitely no
98
i 12 12
� i6
13
t-,'�4.
2126
v
N 277 23.9,0
�I 153%
176%
(
l -0K I NA
L............ ,_,.
31
1 66 5
2 3
5 295 9 5.3%
13
f 10.995
4
5 595 �i
city arhoduma - Streels and Roads So dy Godbe Researd R Analysis
Page 63
Second Ballot Test -Assessment Sample
Table 37. Second Ballot Test by Propensity to Vote and Gender
i, Overall 11 Propensity to Vale g...; Gender 7
[ Low Wedlp.P�High
II -- Male yFemalel
y}mom
Daae �' 462 90 97
.,m
5Z ° 246 213
Yh
YnJ �yyes � f
`�`
162 25 43
Definitely yes €
t � 35 0°° It 29.0% � 44.3%_
73
( Bs Bo I34.1%
� 34 9-,. � 35 9% .� 34 1 % j
Probably yes i 119 24
57 F 264 6 22.7
v3 5 t fiG j
250ia (, 21.1 y 31.09'.
53 14 15
Probably no Fb
r 2 r 26 27
I
1/ 5 p 15 5 15.7%
9,09,.S 10 4 ll 12.7%
De((nilely
64 65 33
Democrat
212 ¢ 163 c 12.1
252 2fi0% 15696
I DKINA
1
31 € 9 5
( 66
I F
1fi !
�56� 16 14 i
06 6.6% e
Table 38. Second Ballot Test by Household
Overall
r(
Household Party Type
I._
9uamocrat
Democrat
RepubllolRepubllc•
pg
t
14)
I� (2+i
an(l)
an(2+)�
.Other
Mixed
8
Base d 482
z
1
62
110 a 31
69
45
139
162
e Definitely yes
t
22
18
1
1 10%
56
350/
350
` 369%
33111 p
25.8
240
40.4%
a 119
20
II 35
7 i
16
119
Probably yes
25.7%
31 B /
[I 29.9%
8
e
�°
225% I
2a.5%
24.0%
i
20.]y p
1 Prdbabl no 'd 533
y
7
° 8
fi
5 �.
11
9
13
11.596
�.
It.
j Definitely no 98
q
105
10
71%
20
10.1%
5 (j
155%
19
205%
11
�'
1
r±
95%
34
! rt 212
153
r 170Y Ir
16A%
269
245%
?t
24596
31_.__5.
.II .8
,tr
4 N...
4
�3
�I
7
❑K/NA 6 %
F
7fii°
7.1%
12096
6A%6296
4.9/
A@
SR,
t
City o(Pclaluma - Streets and Roads Study Godbe Researd & Analysis
Page 64
Second Ballot Test - Assessment Sample
Table 39. Second Ballot Test by Age
Cit, of Pelaluma - S'tn:ets and Roads Study Codbe Research & Analysis
Pao 65
Age291
Ny,
9,_. 40- 49Fr
60-64I
Base
sets
482 22
yea
ji years years
89 �j 125
L years
128
I years
73
�Definitely
16'
pp
yes
tt350
B
xr339
l 35 4752
31.Q,`373
x41.390
tG
212%
yes
719 F 7
�257oF31B
k 1B j 31
�-207a 250I2a2a
it 32
2,5
20 l
271%
—Probably
Probablyno -.
53 I' 3
2
t5
r— Ay
I t0 @ t
�tt 0�6 @S 141 1
G 14
11.19
� 10.7%
Oopnilel no
y
9B 5
12�`�2iB222`,6�167
20 23H
/
25
33.6%
DK I NA
31 6
660 6
6 60
70°0 4B
% i
7.4%6
Cit, of Pelaluma - S'tn:ets and Roads Study Codbe Research & Analysis
Pao 65
Pork and lin Latin II Pacililies-Assessment Sample
Park and Recreation Facilities -
Assessment Sample
qu. Some people have proposed that
The final substantive section of the survey asked respondents to indicate their level of support
in addition to funding the repair
for the allocation of funds for the maintenance and operation of park and recreation facili-
and maintenance of streets and
ties in addition to the repair and maintenance of local streets and roads within the City of
roads, the measure should also fund
Petaluma.
the maintenance and operation of
park ano recreation facilities in the
Ciq, of Petaluma. If asked to vote
Overall, 44 percent of property owners surveyed supported a measure to fund the mainte-
today, would you vote yes or no on
nance and operation of park and recreation facilities in the City in addition to funding the
the measure if it would fund pard-
repair and maintenance of Petaluma streets and roads, 50 percent opposed, and six percent
and recreation facility maintenance
were undecided or unwilling to state their opinion.
in addition to street, sidewalk and
storm drain maintenance?
Figure 30. Support for the Measure with Park and Recreation Facilities
Without Additional Revenue
City ofPelalume - Streets and Roads Study Codbe Research & dnalysis
Page 66
Park and Recrnnlion Fidlllies - Assessmem Sanyle
012. In order to fund the additional Next, respondents were presented with the information that the funding for the additional
park and recreation projects, the park and recreation projects would require an additional tax increase of four million dollars,
total tae amount would have to be bringing the total tax amount to 24 million dollars. With this information, 40 percent of
increased oft4nilllion dollars. Knowing by minion dollars to a
totalotalof 24 respondents indicated they would support the measure and 52 percent indicated they would
this, would you rate yes or no on the not support the measure. The remaining nine percent of voters did not know or had no opin-
measure? ion.
Figure 31. Support for the Measure with Park and Recreation Facilities with
$4 Million in Additional Revenue
DKINA 00initely y..
all,-
r
City of Petaluma-So'eets and Roads Study Codbe Researdi R Analysis
Page 67
Additional eenaignphic and aehaviur l hleasuin-Awi5sniew Sanlple
Additional Demographic and Behavioral
Measures - Assessment Sample
Figures 32 through 41 graphically present the demographic and behavioral information col-
lected in the survey. Some of the information was gathered during the interview (e.g., length
at current residence, amount of prior information on streets, own commercial property or
apartments in the City, newspapers that are read, household income, gender), whereas other
information was collected from the California voter rile (e.g., zip code of residence, age, par-
tisanship, household party type). Although the primary motivation for collecting the demo-
graphic and behavioral information was to provide a better insight into how responses to the
substantive questions of the survey vary across certain voter subgroups, the information is
also useful for better understanding the profile of the City of Petaluma's property owner vot-
ers. As with all results discussed in this section of the report, these results reflect the data as
weighted for gender, age, party affiliation, and geography.
Ql. Flmvlongheveyoulivedatyour Figure 32. Length at Current Residence
current residence?
QA. Before you took this survey,
would you say that you had heard,
read or seen a lot of information,
some Information, a little
information, or no information
about the condition of Petaluma's
streets and roads?
Ratead
0.71 Less hon four years
Ffteen or more years202%
74.9
Four years to less than eight years
9.5%
Eight years to less than fifteen years
21.7%
33. Perceived Condition of Petaluma's Streets and Roads
No Information
1,
A
,^
A little Informa nin
1810°%
OKINA
0.9%
Some Infarmatlon
35.5%
I of mmrmanan
29.7%
City of Peoduma - Streets and Roads Study Godbe Research & Analysis
Page 68
Additional Nina raphic and Bell:ntural Wasurns Assessoenl Sample
QI7. Do you own commercial Figure 34, Own Commercial Property or Apartments in the City
properly or apartments in the City?
Oomme1c1.1 pr.p.hy
3.0%
Ap to ente
RO.d.d E.th
Wither
939;:
QC. Wlial newspapersdoyou read? Figure 35. Newspapers that are Read
QD. To limp things up, I am gaing
to read some income categories.
Please stop me when 1 reach the
category that hest describes your
total household income.
se,w He;em,.mr.:o:: w,e.am
vm�w,,,. weeuun:n,ac,,,nm
aw.�
ua:m N.o�n ano 4murwl
meam.u.M...s
Pnl n,a
36. Household Income
owaaamdn.d
x9.0
m.r$nu,000
s.0%
$150p00 10 $169,999
3.0%
5130,0001.5149,099
SA%
$110,0001. $129,99
6.5%
Under $40.000
9.1%
$90,00010 5109,999
15A
4W: ,IP.. 6P.'.
000 On 459,999
11.t:i
10,0901. $09,999
25,45:
City of Petaluma - Streets and Roads Study Godbe Researdi & Analysis
Page 69
Addllional Demographic and behavioral Aleasmes Assessmem Sample
QE. Gender identified by voice of Figure 37. Gender
respondent.
rJ
�)rq
r
4 rr.7,ar,
ix��..„,.a+' r„f9
ill
Gily o(Pelalurna-Streels and Roads Study Godbe Research & Analysis
Page 70
AddiIjumI 0emugrapIdc mid BeIinim'aI Meesires ASWS9 IUllf Sample
Information Gathered from Voter File
Figure 38. Zip Code of Residence
94954
60.4/
39.
18-29 years
Natcoded 5.0%
Ow
yeare
40.
94952
39.6
City ofPooduma - Streets and Roads Stndj, - Godbe Research d Analysis
Page 71
Addidu ml Demographic end Sehavlornl Menem - Assessmeal Sample
Figure 41. Household Party Type
0 .... 1x1&olhur MI -d Dpmpcmt (t)
z7% 1'BY. 13.4%
oumoml6 Rupubllcnn
10.4%
o�mo=rpi 121)
35.i5:
omu, 12)
as:
oma, ltd
e'�° Benubbcan izq Pnyubllil
tC9°A 4}°/.7%
City of Petaluma -Streets and Roads Study Godhe Research & Analysis
Page 72