Toward A Communicative Theory of Government Performance
Alfred Tat-Kei Ho
Associate Professor
School of Public Affairs and Administration
University of Kansas
and
Jungho Park
Doctoral Candidate
School of Graduate School of Public and International Affairs
University of Pittsburgh
Paper presented at the Korea Association of Public Administration
June, 2012
Abstract
Citizen survey is an important tool for government officials to receive citizen
feedback and evaluate citizen satisfaction with public programs. However, it has an
implicit assumption: Citizens are well informed about the effort and accomplishments
of public services and can make a reasonable judgment on the service results. Using
the citizen survey data from the City of Tulsa in the U.S., we show that there is indeed
a significant relationship between the perception of being informed by the
government and citizens’ satisfaction with public service performance, and the
communication effect is especially important for services that have greater publicness
and transaction cost challenges. Based on the results, we discuss the importance of
public reporting in performance management and suggest how public communication
strategies may tailor toward different citizen profiles and needs.
Toward A Communicative Theory of Government Performance
Introduction
In recent years, how to engage the public and structure programs and policies to
tailor better to the public needs and priorities have been one of the core questions and
challenges in public administration (Thomas, 1995; Berman, 1997; Lukensmeyer and
Torres 2006). In engaging the public, user and citizen survey has become a common
tool used by government officials. The use of this tool is nothing new. In the
1970s, professional organizations such as the International City Management
Association and the Urban Institute already advocated its use to solicit citizen
feedback and prioritize program needs (Webb and Hatry, 1973). In recent decades,
the tool has gained even more importance because of the rise of customer-oriented
management in the public sector (Osborne and Gaebler, 1992), and the counter
response to new public management by emphasizing the need to have greater public
engagement (Frederickson, 1992; Denhardt, et al., 2009; Kelly 2005). National and
regional professional organizations in the U.S. have also contributed to the growing
usage of citizen and customer surveys by providing more user-friendly guidelines,
standardized citizen survey tools, and benchmarking services so that jurisdictions and
departments can compare citizen satisfaction results with peers more easily and
conveniently (Nayyar-Stone and Hatry 2010; National League of Cities, 2012).
As citizen or user surveys become more institutionalized in program evaluation
and policymaking, more questions have been raised about how citizens actually
process information and come to their evaluative judgment on public program
performance and satisfaction with their government. Many past studies have already
pointed out different factors that may influence citizens’ satisfaction ratings. For
example, the wording and structure of the survey questions may influence
significantly the responses of citizens (Miller and Miller 1991). Also, the sampling
methodology and the way a survey is conducted may affect the response rate and the
representativeness of the results (Miller and Miller, 1992). Furthermore, many
socio-economic and demographic factors, such as income, race, and neighborhood
characteristics, may influence citizens’ expectation and perception of service results
(Campbell, et al. 1976; Lineberry, 1977; McDougall and Bunce, 1984; Miller and
Miller, 1991; Kusow, Wilson and Martin, 1997; Van Ryzin, 2004, 2006; Licari, et al.
2005). Prior experiences and contacts with government officials may also shape
1
their views about the trustworthiness and performance of the government (Kelly and
Swindell, 2002a, 2002b; Brown, 2007). As a result, the same program performance
can be perceived very differently by citizens, leading to very different satisfaction
ratings of the services.
Hence, policymakers and program managers have been warned about the
subjective nature of citizen opinions and satisfaction ratings (Kelly 2005). It is
perception-based and is influenced heavily by many socio-economic and demographic
factors that can be unrelated to the actual program performance and are outside the
control of public managers. However, there is one factor that public managers can
have some degree of influence but have been overlooked by many past studies – how
to keep the public informed and educated about program priorities, challenges, and
results. Public managers tend to be skeptical about the public’s understanding of the
complexity and constraints of government administration (Berman, 1997; Milbrath,
1981; Melkers and Thomas, 1998; Bok, 2001). Even though there is an
understanding that effective communication with the public is important to help
reduce public cynicism toward the government and should be a critical component of
public administration (Garnett, 1992; Wheeler, 1994), public reporting has remained a
peripheral exercise and is often viewed as mundane paperwork by many governments.
In today’s fiscal environment, public communication and reporting is often the first
target of budget cut when there are so many competing needs for growingly scarce
resources.
The purpose of this paper is to reexamine the importance of information and
public communication in the context of performance management and explore
whether information availability, or at least the perception of information availability
and communication effectiveness, may change the public’s satisfaction with
governmental services. Citizen or user surveys assume implicitly that citizens are well
informed about the efforts and accomplishments of a service and can make a
reasonable judgment of the service results (Swindell and Kelly, 2000). If that is not
the case, either because citizens do not have personal experience of the services, or
citizens are poorly informed about the service results, then their satisfaction ratings
may indeed be biased and unhelpful to program management and planning since they
may be driven primarily by subjective feelings, prior beliefs, or ideological factors
that are unrelated to the actual characteristics of the delivery or the performance of the
services (Stipak, 1977; Melkers and Thomas, 1998; Kelly and Swindell, 2002b).
In this paper, we propose a “communicative theory” of government performance
2
by suggesting that information plays a critical link between government performance
and citizen satisfaction. We hypothesize that this relationship is especially important
for public services that have diffused social impact, are less frequently used by
citizens, and have greater problems of measurability by ordinary citizens. Using a
structural equation model and a citizen survey data from a U.S. southern city, we
confirm our hypothesis by showing that more informed citizens tend to have more
positive ratings of government services after controlling for their demographic and
socio-economic background, neighborhood factors, and their participation in
community affairs. Such relationship is especially obvious among services that
demonstrate greater publicness, such as city image building and public safety. Based
on our findings, we discuss the importance of public communication in today’s
political and media environment and the implications for future performance
management and citizen engagement research.
Government Performance, Satisfaction, and Information – an Extended Logic
Model
In recent years, both professional and academic communities in public
administration have begun to pay more attention to the roles of the public in
evaluating government performance (Ho and Coates 2004; GASB, 2005; Hawn and
Siegel, 2006). Whether one believes in the logic of new public management, which
views citizens as customers and advocates the need for the government to partner
more closely with the public (Osborne and Gaebler, 1992), or in the principle of
democratic governance, which emphasizes the roles of citizens not just as customers
but also as the owners of the government (King et al., 1998), citizens certainly have a
legitimate role to play to ensure the public accountability of the govenrment.
In the process of engaging citizens, citizen or user survey is a common tool used
by government officials to solicit citizen feedback. However, citizen opinions and
their perception of government performance do not happen in a political, social and
cultural vacuum. For example, past studies have shown that minority residents,
lower income householders, and residents in poorer neighborhoods tend to have lower
satisfaction with public services than others (Campbell, et al., 1976; McDougall and
Bunce, 1984; Lineberry, 1977; Miller and Miller 1991; Kusow, Wilson and Martin,
1997; Swindell and Kelly 2002a; Licari, et al. 2005). This may be caused by the fact
that these residents tend to have lower quality of life and so they may associate that
with the quality of public services received. It is also possible that these residents may
indeed receive poorer public services because the ruling regime may incentivize
3
government leaders to pay less attention to their needs (Stone, 1980).
In addition, citizens’ satisfaction with public services may be impacted by the
service context and their interactions with government officials. For example,
Brown(2007) differentiates public services by the degree of “captivity” and shows
that citizen satisfaction varies by types of services. In general, citizens are more
likely to rate “captive services,” such as police stops, or “monopolized client
services,” such as citizen call center, poorly if they have bad personal experiences in
interacting with the service provider. For services that are freely selected by
customer choices, such as parks and recreational programs, the finding is the opposite
and reflects a positive bias if citizens have good interactions with the service provider.
However, what many past studies have not explored much is the role of
information. All citizen surveys implicit assume that citizens have a sufficient level
of understanding and information about the quality of public services. However, in
reality, this may not be the case. Information problems and lack of understanding
about the services they receive can be especially challenging in public administration
because the operations and impact of many public services are complex, not easily
observable from the perspective of ordinary citizens, and may be highly remote to
their daily experiences.
For example, city planning, code inspection and
enforcement, and infrastructure planning and maintenance are technical in nature and
the benefits of these services are jointly consumed, non-excludable, and extended
over a long period of time. As a result, even though citizens benefit from these
services daily, they may not fully appreciate the value and challenges of these services
and how these services contribute to their quality of life.
Moreover, some public services are impersonal or may not be used by citizens
frequently. For example, many citizens may not have called the emergency response
number to ask for help from the police or fire department. So unless citizens receive
good information about the efforts and accomplishments of these services, citizen
ratings of these services, such as responsiveness, professionalism, and courtesy, will
have to base mostly on their subjective perception, which is often influenced by
citizens’ socio-economic and demographic background and the information from the
mass media.
Based on the above understanding, figure 1 provides an extended logic model of
government performance. In the traditional logic model, government managers
examine the relationships between input, output, and outcome and try to use
4
performance measurement and business process improvement tools to enhance the
cost-effectiveness and efficiency of services (Poister, 2003; Hatry, 2006). However,
this model can be extended to examine how service outcomes are linked to citizen
perception and satisfaction, and how the logical link is influenced by prior
experiences and expectations, interactions with government officials, and the
availability and quality of public information and communication. The link between
citizen satisfaction and program outcomes is important because this is where the
technical, administrative world of performance management meets with the messy,
chaotic world of politics, and where performance measurement and management may
get the attention of elected officials.
Figure 1. An Extended Logic Model of Government Performance
Input
Output
Outcome
Prior expectations
and experiences
Interaction
with officials
Socio-economic, demographic, and
neighborhood characteristics
Citizen
satisfaction
Being
informed
Service Nature
In this study, we hypothesize that citizens who are better informed are more
likely to rate public services positively. Also, we hypothesize that this relationship
varies according to the nature of public services.
H1: Controlling for other factors, citizens who are more informed are more likely to
rate public services satisfactorily.
H2: The impact of information on citizen satisfaction is stronger in public services
that are less observable, have greater publicness, and have more impersonal or
diffused impact than in services that have less publicness and more direct personal
impact.
5
The positive relationship between information effectiveness and service
satisfaction can be caused by several factors. First, being more informed may
simply overcome some of the information costs and uncertainty faced by citizens.
As the public understand the service value, priorities, challenges, and results better,
they may feel less uncertain about how their tax money is used and what value they
are getting. As a result, they may become less cynical about the government and
more positive about various services (Berman, 1997).
Second, the positive relationship can be caused by the “framing” effect (Iyengar,
1991). As citizens consume more information provided by the government about
government policies and program results, they may consciously or unconsciously
adopt the governmental perspective to frame the political, social, and administrative
“facts”. Since government-provided information is likely to be less critical of the
departments’ performance and may purposively highlight certain outcomes and results
to win public support (De Vries, 2004), citizens who are more informed by the
government media and information sources may be more likely to be co-opted by the
government and, as a result, feel more positive about the performance of various
public services.
Third, being more informed may simply be a reflection of the social and political
influence certain residents may enjoy. For example, wealthier residents may be
more able to afford time to learn about social and political events, find out more about
program performance related to their residential neighborhood, and have more access
to different sources of information. They may also participate more actively in
community affairs and have greater networking capacity to get more information.
As a result, these residents who are more informed may have greater access to the
political elite and have more influence over community decision-making. So they
may indeed receive more privileged access to public services and are more satisfied
with their outcomes. This is why our analysis in the later part of the paper has to
control for residents’ socio-economic status and their activeness in community
participation so that we can truly measure the information impact.
We hypothesize that this impact of information on citizen satisfaction should be
especially important for public services that are more technical and less observable by
ordinary citizens, and for services that have more diffused social impact and less
direct, personal benefits (see Figure 2). These services tend to have greater
transaction cost problems.
So unless citizens can somehow overcome the
6
information barriers and lack of understanding about these services, they may
appreciate less their challenges and value and under-rate their performance.
Figure 2. The Importance of Information and Service Types
Measurability/observability
Low
Publicness
Low
Service Results
Low
Personal
Experience with
High
High
High
♦crime investigation and control
♦ City image building/
beautification
♦ Code enforcement
♦ fire & EMS
Importance of
information
increases
♦Sewage & ♦ Traffic management
flood control
♦ Parks & Rec.
♦Water supply and quality control
♦ Street & sidewalk ♦Trash collection
maintenance
♦ Customer service /
complaint center services
Note: The measurability of different services is largely based on the transaction cost framework
suggested by Levin and Tadelis (2010).
Methodology and Data
We use the citizen survey results of the City of Tulsa, Oklahoma, to the test the
above hypotheses. The City of Tulsa is a mid-sized city in the U.S, with a
population of about 392,000. In 2010, the per-capita income was about $26,069,
which was about the same as the U.S. average ($27,334). A citizen survey was
conducted in December 2010-January 2011, which asked citizens to rate their
satisfaction with various public services. A total of 1,803 responses were received,
representing a response rate of 35.9 percent.1 Table 1 provides a summary of the
respondent characteristics. Compared to the 2010 U.S. Census data, the sample has
higher ratios of female, white, older, and wealthier respondents. Since the
proportion of older respondents is significantly greater than that of the population,
which is not too surprisingly for a mail survey, a sample weight based on age is used
to adjust for the survey statistics.
1
The survey was conducted by Shapard Research for the City of Tulsa.
survey results, please refer to Shapard Research (2011).
7
For more details of the
Table 1. Characteristics of the Tulsa Survey Respondents
Unweighed Sample
Weighed Sample
The 2010 Census
Sample size or
Population
1,803
1,800*
391,906
Female
60.7%
61.0%
51.3%
White
75.7%
68.5%
62.6%
Black
13.5%
15.4%
15.9%
Age older than 55
41.1%
32.5%
32.9%
76.9%
69.6%
54.4%
among the adult
population (older
than 20 years old)
Homeownership
rate
Note: * Three observations have no sample weights since the age information was missing in the
responses.
Based on the addresses of the respondents, all responses were also geo-coded by
geographical information system, and then merged with the 2010 census block group
data. Figure 3 is a map of the respondent geographic locations and their
neighborhood clusters. In addition, we use the census block group data to cluster the
demographic and housing characteristics of the respondents’ neighborhoods into three
classes of neighborhoods, which is used as a control variable for neighborhood
characteristics and, possibly, a control variable for any service quality differences due
to the socio-economic characters of neighborhoods.2
2
The neighborhood demographic and housing characteristics used in the cluster analysis include the
percentage of vacant housing, the percentage of renters in the household population, the number of
crimes per 1,000 residents, the number of assaults to police officers per 1,000 residents, the number of
serious assault cases per 1,000 residents, the number of sex offense cases per 1,000 residents, the
number of homicide, the number of burglary during daytime per 1,000 residents, the number of
burglary at night per 1,000 residents, and the number of property crimes per 1,000 residents.
8
Figure 3.
Geographical Locations of the Survey Respondents and their
Neighborhood Clusters
Table 4 summarizes the citizen perception of the quality of life, and their
satisfaction with different services. Over 70 percent of the Tulsa survey respondents
(over 70%) were satisfied or very satisfied with their quality of life and with the city
as a place to live. The majority were also satisfied with the city as a place to work
and with most city services. However, they were less satisfied with the city’s policy
direction. A possible contribution to the dissatisfaction was with the city’s traffic
flow management – only one third of the respondents were satisfied or very satisfied
due to significant road construction and traffic re-routing issues during the time of
the survey. Also, relative to other city services, the respondents were less satisfied
with the City’s public communication effectiveness and its efforts to keep citizens
informed.
9
Table 4. Summary Statistics of the Citizen Ratings of Selected Services
Very
dissatisfied
dissatisfied
Neutral /
don’t
know
20.4%
13.4%
Quality of life
3.4%
6.2%
The City as a place to
3.7%
5.8%
live
The city as a place to
7.1%
10.0%
18.7%
work
The
City
moving
11.1%
13.2%
27.1%
toward a right direction
Overall quality of city
5.3%
10.2%
12.7%
services
Public safety
8.5%
12.3%
13.6%
Police response time
12.0%
8.5%
25.7%
Fire department
1.2%
2.3%
13.8%
Parks and recreation
5.4%
10.6%
19.8%
Traffic flow of major
25.7%
25.9%
13.4%
streets
Trash collection
3.9%
4.4%
5.2%
Street cleanliness
7.3%
16.8%
16.6%
Sewage
2.3%
3.7%
16.7%
Water services
3.2%
3.9%
7.5%
City
communication
11.9%
17.2%
25.6%
with the public
City keeping citizens
9.4%
17.0%
21.5%
informed
Note: Sample size = 1800. All frequencies are weighted by age.
Satisfied
Very
Satisfied
51.5%
49.1%
18.4%
28.2%
46.9%
17.3%
36.3%
12.3%
53.3%
18.5%
40.8%
27.1%
35.7%
38.5%
28.3%
24.8%
26.7%
47.0%
25.7%
6.7%
31.8%
44.1%
35.5%
36.5%
34.5%
54.7%
15.1%
41.8%
48.9%
11.0%
38.2%
13.9%
A structural equation model (SEM) is used to examine whether public
information and communication effectiveness influences significantly the
respondents’ satisfaction with various services (see Figure 4). SEM allows us to
understand not only the relative significance of information to different services, but
also how different socio-economic and demographic characteristics of the respondents
and their neighborhood profiles influence the respondents’ feeling of being informed
and their service ratings, respectively. Past studies have already shown that the
socio-economic and demographic background of respondents may reflect their prior
expectations of various services and should therefore be controlled for since
expectation may significantly influence respondents’ satisfaction with services (Van
Ryzin, 2004, 2006). Respondents’ prior contacts with the police department and the
activeness of the respondents in participating in community affairs are also included
10
in the SEM to control for their prior contact experiences with the City.3
Figure 4. The Structural Equation Model to Test the Information Effect
Because there are too many questions that can be used to evaluate the
effectiveness and quality of different services, principal component analysis is used to
combine and standardize various service ratings. Table 5 explains the service
elements included in each of the principal components and their statistical validity.
Results
3
The activeness of community participation is a principal component of the following variables:
whether the respondents have contacted any departments besides the police, have contacted the
mayor’s office or the city council members, read the major local newspaper regularly, have visited the
City’s website, and have visited the city council website.
11
Because of the complexity of the model, the results of the SEM model are
presented in parts below. The Satorra-Bentler (S-B) scaled statistics is used to
examine the goodness of fit due to the violation of the multivariate normality (Satorra
and Bentler, 1988, 1994). The chi-square statistics (χ2(df=184)=369.48) confirms that
there is a significant difference between the observed and model covariance matrices.
Because our sample size is over 200, which tends to make the chi-square statistics
always significant, we also use a combination of other model fit indexes (CFI=0.970,
RMSEA=0.024, SRMR=0.019, TLI=0.928) to re-confirm that the hypothesized
conceptual model for this study fits well with the observed data (Hu & Bentler, 1999).
Table 6 shows the impact of the perception of being informed on respondents’
ratings of various services. All coefficients of the SEM are standardized for the ease
of comparison. In general, the pattern of impact follows our expectation – services
that are of lower measurability, less direct personal experiences, and higher publicness,
such as public safety and city image building, tend to have stronger information
impact than services that allow for more personal experiences and higher
observability, such as customer service, trash collection, and water services.
Table 7 examines what factors influence the respondents’ perception of being
informed. The results show that residents with more education tend to feel less
informed, indicating that highly educated residents may have higher expectation of
the City to engage and inform them and therefore tend to be less satisfied if
insufficient information is provided. Also, the perception of being informed
increases with age, implying that the younger generation needs more attention by the
government, or that older citizens can afford finding more time and ways to keep
them stay informed. Also, people who were contacted by the police tend to feel less
informed. We hypothesize that people who violated certain rules and legal
obligations are more likely to blame the government for failing to inform them
effectively about the legal requirements. As a result, they tend to feel less informed.
Controlling for other factors, we also find that hispanic residents feel less informed,
probably because of the language barriers. However, one should notice that the
impact of the age and racial factors is only marginally significant at the 10-percent
level. Education and police contact remain the most important driver of feeling
informed. Other factors, such as income, neighborhood status, and years of
residency, are not statistically significant. The implications of these findings on city
communication and engagement strategies are discussed later in the paper.
12
Table 5. Summary Statistics of the Principal Components of Public Services
Principal component Citizen Rating Variables (Respective Eigenvectors)
of Service Quality
Public safety
Satisfaction with public safety (0.59)
Police quality as a top three concern: yes = 1, no = 0 (-0.40)
Satisfaction with neighborhood police response time (0.56)
Satisfaction with traffic law enforcement (0.41)
Feeling safe
Feeling safe in neighborhood during daytime (0.58)
Feeling safe in neighborhood at night (0.64)
Feeling safe in downtown at night (0.50)
Fire and emergency Satisfaction with the fire department services (0.39)
medical
services Satisfaction with the fire department’s response time (0.48)
(EMS)
Satisfaction with the assistance provided by EMS staff (0.56)
Satisfaction with the EMS response time (0.56)
Parks and recreation Satisfaction with parks and recreation services (0.43)
Satisfaction with outdoor field facilities (0.49)
Satisfaction with parks grounds (0.54)
Satisfaction with parks indoor facilities (0.53)
Trash collection and Satisfaction with trash collection (0.35)
cleanliness
Satisfaction with street cleanliness (0.71)
Public area cleanliness as a priority concern: yes = 1, no = 0 (-0.62)
Neighborhood
Satisfaction with cleaning up properties in neighborhood (0.54)
clean-up issues
Satisfaction with mowing the lawn in abandoned properties (0.54)
Cleaning up properties as a priority concern: yes = 1, no = 0 (-0.45)
Mowing in neighborhood as a priority: yes = 1, no = 0 concern (0.54)
Traffic management Satisfaction with traffic law enforcement (0.46)
Satisfaction with traffic flow in major streets (0.69)
Traffic flow as a top three concern: yes = 1, no = 0 (-0.56)
13
Eigenvalue
(proportion of variation explained)
1.90
(0.48)
1.81
(0.60)
2.40
(0.60)
2.56
(0.51)
1.33
(0.44)
2.26
(0.56)
1.29
(0.43)
Street conditions
Satisfaction with major street conditions (0.56)
Satisfaction with neighborhood street conditions (0.49)
Satisfaction with street repair progress (0.40)
Street maintenance as a top three concern (-0.23)
Improving major street conditions as a priority concern: yes = 1, no = 0 (-0.34)
Improving neighborhood street condition as a priority concern: yes = 1, no = 0
(-0.35)
Water and sewage Satisfaction with water services (0.70)
services
Satisfaction with sewage services (0.69)
Water services as a top three concern: yes = 1, no = 0 (-0.20)
Feeling informed
Quality of the City’s communication (0.43)
Effectiveness of keeping citizens informed (0.59)
Satisfaction with information availability to the public (0.58)
Satisfaction with the City’s public TV channel (0.36)
Customer
service Customer service as a one of top three concern (0.71)
concerns
Dissatisfaction with city staff courtesy: yes = 1, no = 0 (0.71)
Non-police Contact Dissatisfaction with city staff responses: yes = 1, no = 0 (0.42)
concerns
Dissatisfaction with the contacts with the public works staff: yes = 1, no = 0 (0.52)
Dissatisfaction with the contacts with the permit office staff: yes = 1, no = 0 (0.53)
Dissatisfaction with the contacts with the code enforcement staff: yes = 1, no = 0
(0.53)
City image
Satisfaction with the city appearance (0.71)
Satisfaction with the city image (0.71)
14
2.01
(0.33)
1.64
(0.55)
2.14
(0.53)
1.31
(0.66)
3.26
(0.82)
1.60
(0.80)
Table 6.
Impact of Communication Effectiveness on Service Ratings
Standardized Coefficient
IV: Feeling Informed
DV: Service Rating Index
Estimate
S.E.
Est./S.E.
P-Value
City image (DV)
0.496
0.030
16.614
0.000
Parks & Rec. (DV)
0.458
0.031
14.559
0.000
Public Safety (DV)
0.419
0.031
13.329
0.000
Street conditions (DV)
0.346
0.031
11.313
0.000
Fire & EMS (DV)
0.324
0.032
10.072
0.000
0.289
0.031
9.316
0.000
Water & sewage (DV)
0.289
Mowing and code enforcement
(DV)
0.271
Trash collection (DV)
0.263
0.032
9.160
0.000
0.031
8.748
0.000
0.033
7.915
0.000
Feeling safe
Traffic management
(DV)
(DV)
0.239
0.032
7.404
0.000
Non-police contact concerns (DV)
-0.092
0.031
-3.006
0.003
Customer service concerns (DV)
-0.209
0.035
-6.039
0.000
Table 7. Factors Influencing the Perception of Being Informed
DV: Feeling Informed
Estimate
S.E.
Est./S.E.
P-Value
Female
0.011
0.030
0.373
0.709
AGE
0.056
0.033
1.714
0.086
EDUCATION
-0.097
0.033
-2.933
0.003
INCOME
0.007
0.034
0.199
0.842
RESIDENCE YEAR
-0.040
0.031
-1.296
0.195
NEIGHBORHODD
-0.029
0.029
-0.994
0.320
HOUSE OWNER
-0.032
0.034
-0.941
0.346
BLACK
-0.048
0.033
-1.445
0.149
HISPANIC
-0.052
0.030
-1.738
0.082
CONTACT POLICE
-0.005
0.035
-0.152
0.879
CONTACTBYPOLICE
-0.098
0.033
-2.982
0.003
15
Tables 8A-8L show the impact of different demographic and socio-economic
factors on the satisfaction of various services, after controlling for the information
effect. Our results show that residents in nicer neighborhoods tend to have lower
rating of the police; females tend to feel less safe and have lower satisfaction with
parks and recreational services, mowing and neighborhood code enforcement issues;
but African Americans tend to have a significantly stronger feeling of safety in the city.
These findings may reflect different expectations of city services and are consistent
with previous studies. A major positive factor on city service ratings, besides the
information factor, is education. It is positively related to most service ratings,
including public safety, traffic flow, street conditions, water and sewage services, city
cleanliness, parks and recreation. This may reflect that more educated citizens tend to
appreciate the challenges and value of public services more.
Our model also controls for activeness in community participation. To our
surprise, it shows negative impact on a few services. It increases significantly the
concerns of customer service and the concerns of non-police departmental contacts
with citizens. It also reduces the satisfaction of neighborhood mowing and clean-up
issues. The only exception is fire and emergency medical services, which is
positively related to participatory activeness. These findings may reflect the fact that
active citizens who read the local newspaper more, visit with their elected officials,
and actively seek out information on the governmental website are more likely to
have issues with some governmental services. As a result, they are less satisfied,
especially with neighborhood issues and customer services.
16
Table 8A.
Factors Influencing Public Safety Satisfaction
Standardized Coefficient
DV: Public safety satisfaction index
Estimate
S.E.
Est./S.E.
P-Value
Feeling informed
0.419
0.031
13.329
0.000
Participatory activeness
0.052
0.043
1.207
0.228
FEMALE
0.011
0.023
0.479
0.632
AGE
0.058
0.025
2.322
0.020
EDUCATION
0.065
0.026
2.538
0.011
INCOME
-0.006
0.026
-0.231
0.817
YEARS OF RESIDENCE
0.033
0.024
1.350
0.177
NEIGHBORHOOD
-0.040
0.023
-1.720
0.085
HOUSE OWNERSHIP
0.030
0.026
1.162
0.245
BLACK
-0.020
0.026
-0.788
0.430
HISPANIC
0.012
0.023
0.517
0.605
CONTACTED POLICE BEFORE
-0.053
0.031
-1.717
0.086
CONTACTED BY POLICE BEFORE
-0.101
0.028
-3.622
0.000
Table 8B. Factors Influencing Traffic Management Satisfaction
Standardized Coefficient
DV:
Traffic
satisfaction index
management
Estimate
S.E.
Est./S.E.
P-Value
FEELING INFORMED
0.289
0.031
9.316
0.000
PARTICIPATORY ACTIVENESS
0.002
0.039
0.042
0.967
FEMALE
-0.040
0.023
-1.714
0.087
AGE
-0.049
0.026
-1.889
0.059
EDUCATION
0.060
0.027
2.238
0.025
INCOME
0.002
0.027
0.062
0.951
YEARS OF RESIDENCE
-0.015
0.025
-0.620
0.535
NEIGHBORHOOD
-0.038
0.024
-1.622
0.105
HOUSE OWNERSHIP
0.011
0.026
0.432
0.666
BLACK
0.118
0.022
5.281
0.000
HISPANIC
0.047
0.022
2.128
0.033
CONTACTED POLICE BEFORE
0.014
0.030
0.458
0.647
CONTACTED BY POLICE BEFORE
-0.011
0.027
-0.420
0.675
17
Table 8C. Factors Influencing Street Condition Satisfaction
Standardized Coefficient
DV: Street condition satisfaction
Estimate
index
S.E.
Est./S.E.
P-Value
FEELING INFORMED
0.346
0.031
11.313
0.000
PARTICIPATORY ACTIVENESS
-0.036
0.036
-0.982
0.326
FEMALE
-0.017
0.023
-0.724
0.469
AGE
-0.002
0.025
-0.065
0.948
EDUCATION
0.062
0.026
2.343
0.019
INCOME
0.009
0.027
0.348
0.728
YEARS OF RESIDENCE
-0.018
0.025
-0.705
0.481
NEIGHBORHOOD
0.010
0.024
0.440
0.660
HOUSE OWNERSHIP
0.027
0.027
0.997
0.319
BLACK
0.001
0.026
0.052
0.959
HISPANIC
0.059
0.020
2.922
0.003
CONTACTED POLICE BEFORE
0.016
0.027
0.595
0.552
CONTACTED BY POLICE BEFORE
-0.065
0.026
-2.475
0.013
Table 8D. Factors Influencing Satisfaction of Fire and EMS Services
Standardized Coefficient
DV: fire & EMS satisfaction index
Estimate
S.E.
Est./S.E.
P-Value
FEELING INFORMED
0.324
0.032
10.072
0.000
PARTICIPATORY ACTIVENESS
0.123
0.045
2.763
0.006
FEMALE
-0.014
0.023
-0.598
0.550
AGE
0.066
0.026
2.575
0.010
EDUCATION
-0.003
0.027
-0.112
0.911
INCOME
-0.082
0.027
-2.987
0.003
YEARS OF RESIDENCE
0.113
0.025
4.503
0.000
NEIGHBORHOOD
0.019
0.023
0.814
0.415
HOUSE OWNERSHIP
-0.032
0.027
-1.173
0.241
BLACK
-0.016
0.024
-0.672
0.502
HISPANIC
-0.059
0.023
-2.493
0.013
CONTACTED POLICE BEFORE
-0.052
0.031
-1.649
0.099
CONTACTED BY POLICE BEFORE
0.028
0.028
1.016
0.309
18
Table 8E. Factors Influencing Satisfaction of Water and Sewage Services
Standardized Coefficient
DV: water & sewage satisfaction
Estimate
index
S.E.
Est./S.E.
P-Value
FEELING INFORMED
0.289
0.032
9.160
0.000
PARTICIPATORY ACTIVENESS
-0.014
0.042
-0.339
0.735
FEMALE
0.016
0.024
0.690
0.490
AGE
-0.048
0.026
-1.837
0.066
EDUCATION
0.066
0.027
2.439
0.015
INCOME
0.108
0.025
4.223
0.000
YEARS OF RESIDENCE
0.052
0.026
2.031
0.042
NEIGHBORHOOD
0.036
0.024
1.520
0.129
HOUSE OWNERSHIP
-0.006
0.027
-0.220
0.826
BLACK
-0.064
0.026
-2.470
0.014
HISPANIC
-0.045
0.022
-1.991
0.046
CONTACTED POLICE BEFORE
-0.017
0.028
-0.599
0.549
CONTACTED BY POLICE BEFORE
0.011
0.027
0.406
0.684
Table 8F. Factors Influencing Satisfaction of Trash collection and city cleanliness
Standardized Coefficient
DV: Cleanliness satisfaction index
Estimate
S.E.
Est./S.E.
P-Value
FEELING INFORMED
0.263
0.033
7.915
0.000
PARTICIPATORY ACTIVENESS
0.024
0.039
0.607
0.544
FEMALE
0.006
0.024
0.259
0.796
AGE
-0.010
0.027
-0.391
0.696
EDUCATION
0.049
0.027
1.854
0.064
INCOME
-0.015
0.027
-0.569
0.569
YEARS OF RESIDENCE
-0.020
0.023
-0.841
0.400
NEIGHBORHOOD
-0.009
0.023
-0.382
0.702
HOUSE OWNERSHIP
0.002
0.028
0.080
0.936
BLACK
-0.036
0.026
-1.419
0.156
HISPANIC
-0.026
0.022
-1.209
0.227
CONTACTED POLICE BEFORE
0.009
0.027
0.329
0.742
CONTACTED BY POLICE BEFORE
-0.048
0.027
-1.797
0.072
19
Table 8G. Factors Influencing Satisfaction of Parks and Recreation Services
Standardized Coefficient
DV: Parks & Recreation satisfaction
Estimate
index
S.E.
Est./S.E. P-Value
FEELING INFORMED
0.458
0.031
14.559
0.000
PARTICIPATORY ACTIVENESS
0.076
0.047
1.621
0.105
FEMALE
-0.052
0.022
-2.339
0.019
AGE
-0.020
0.025
-0.802
0.422
EDUCATION
0.071
0.026
2.742
0.006
INCOME
0.019
0.028
0.686
0.493
YEARS OF RESIDENCE
-0.012
0.023
-0.541
0.588
NEIGHBORHOOD
0.005
0.024
0.193
0.847
HOUSE OWNERSHIP
-0.012
0.027
-0.459
0.647
BLACK
-0.081
0.027
-3.025
0.002
HISPANIC
-0.050
0.022
-2.235
0.025
CONTACTED POLICE BEFORE
-0.005
0.030
-0.166
0.868
CONTACTED BY POLICE BEFORE
-0.038
0.028
-1.369
0.171
Table 8H. Factors Influencing Satisfaction with City Image
Standardized Coefficient
DV: City image satisfaction
Estimate
S.E.
Est./S.E. P-Value
FEELING INFORMED
0.496
0.030
16.614
0.000
PARTICIPATORY ACTIVENESS
-0.031
0.042
-0.738
0.461
FEMALE
0.010
0.022
0.448
0.654
AGE
-0.006
0.026
-0.218
0.828
EDUCATION
0.036
0.025
1.413
0.158
INCOME
-0.019
0.027
-0.718
0.472
YEARS OF RESIDENCE
-0.026
0.024
-1.087
0.277
NEIGHBORHOOD
0.004
0.022
0.185
0.853
HOUSE OWNERSHIP
0.010
0.026
0.371
0.711
BLACK
-0.035
0.024
-1.446
0.148
HISPANIC
0.015
0.018
0.809
0.419
CONTACTED POLICE BEFORE
0.026
0.029
0.896
0.370
CONTACTED BY POLICE BEFORE
-0.081
0.026
-3.107
0.002
20
21
Table 8I. Factors Influencing Public Feeling of Safety
Standardized Coefficient
DV: Index of Feeling Safe
Estimate
S.E.
Est./S.E. P-Value
FEELING INFORMED
0.239
0.032
7.404
0.000
PARTICIPATORY ACTIVENESS
0.045
0.041
1.102
0.271
FEMALE
-0.208
0.022
-9.533
0.000
AGE
-0.090
0.025
-3.561
0.000
EDUCATION
0.102
0.026
3.967
0.000
INCOME
0.193
0.025
7.668
0.000
YEARS OF RESIDENCE
-0.007
0.024
-0.290
0.772
NEIGHBORHOOD
-0.010
0.024
-0.413
0.679
HOUSE OWNERSHIP
0.025
0.026
0.979
0.328
BLACK
0.072
0.025
2.855
0.004
HISPANIC
0.012
0.026
0.462
0.644
CONTACTED POLICE BEFORE
-0.058
0.027
-2.132
0.033
CONTACTED BY POLICE BEFORE
-0.068
0.026
-2.641
0.008
Table 8J. Factors Influencing Customer Service Concerns
Standardized Coefficient
DV: Index of customer service
Estimate
concerns
S.E.
Est./S.E. P-Value
FEELING INFORMED
-0.209
0.035
-6.039
0.000
PARTICIPATORY ACTIVENESS
0.140
0.050
2.788
0.005
FEMALE
-0.005
0.023
-0.210
0.834
AGE
0.011
0.027
0.417
0.677
EDUCATION
-0.037
0.026
-1.408
0.159
INCOME
-0.005
0.027
-0.184
0.854
YEARS OF RESIDENCE
0.013
0.024
0.529
0.597
NEIGHBORHOOD
-0.022
0.022
-1.033
0.302
HOUSE OWNERSHIP
0.003
0.026
0.111
0.912
BLACK
0.048
0.026
1.822
0.068
HISPANIC
-0.002
0.021
-0.100
0.920
CONTACTED POLICE BEFORE
0.059
0.040
1.483
0.138
CONTACTED BY POLICE BEFORE
-0.009
0.027
-0.340
0.734
22
23
Table 8K. Factors Influencing Non-Police Contact Concerns
Standardized Coefficient
DV: Index of non-police contact
Estimate
concerns
S.E.
Est./S.E. P-Value
FEELING INFORMED
-0.092
0.031
-3.006
0.003
PARTICIPATORY ACTIVENESS
0.249
0.063
3.923
0.000
FEMALE
-0.032
0.024
-1.333
0.183
AGE
-0.014
0.024
-0.571
0.568
EDUCATION
-0.032
0.027
-1.189
0.234
INCOME
0.016
0.025
0.660
0.509
YEARS OF RESIDENCE
0.000
0.025
0.008
0.993
NEIGHBORHOOD
0.007
0.022
0.322
0.748
HOUSE OWNERSHIP
0.016
0.023
0.717
0.474
BLACK
0.009
0.022
0.405
0.686
HISPANIC
0.002
0.019
0.113
0.910
CONTACTED POLICE BEFORE
0.051
0.048
1.060
0.289
CONTACTED BY POLICE BEFORE
0.039
0.029
1.313
0.189
Table 8L.
Factors Influencing Satisfaction with Neighborhood Mowing and Code
Enforcement
Standardized Coefficient
DV: Index of non-police contact
Estimate
concerns
S.E.
Est./S.E. P-Value
FEELING INFORMED
0.271
0.031
8.748
0.000
PARTICIPATORY ACTIVENESS
-0.066
0.038
-1.738
0.082
FEMALE
-0.055
0.023
-2.376
0.017
AGE
-0.061
0.026
-2.388
0.017
EDUCATION
0.014
0.027
0.534
0.594
INCOME
0.029
0.027
1.075
0.282
YEARS OF RESIDENCE
-0.068
0.025
-2.677
0.007
NEIGHBORHOOD
0.040
0.024
1.711
0.087
HOUSE OWNERSHIP
-0.045
0.026
-1.715
0.086
BLACK
-0.027
0.025
-1.075
0.282
HISPANIC
0.001
0.021
0.036
0.971
CONTACTED POLICE BEFORE
-0.021
0.027
-0.779
0.436
CONTACTED BY POLICE BEFORE
-0.020
0.026
-0.748
0.454
24
25
Finally, we examine the factors that influence the respondents’ activeness in
community participation. The results in Table 9 show that people who are more
educated, have higher income, live in nicer neighborhood, or own a house are more
likely to be active participants. The findings are consistent with studies of urban
regime theories, which show that lower income residents face more socio-economic
constraints in public affairs involvement while residents of higher socio-economic
status tend to have greater voice in local policy decision-making (Stone, 1980).
Interestingly, the results also show that if residents have contacted the police or have
been contacted by the police, they are more likely to be engaged in public affairs.
Again, this confirms the fact that citizens who have issues with the government are
more likely to voice their concerns and participate in community affairs, and these
citizens tend to be less satisfied with government services, unless they are more
informed by the government.
Table 9. Factors Influencing Activeness in Community Participation
Standardized Coefficient
DV:
Index
of
activeness
in
Estmate
S.E.
Est./S.E. P-Value
FEMALE
-0.047
0.028
-1.682
0.092
AGE
0.044
0.030
1.458
0.145
EDUCATION
0.118
0.031
3.775
0.000
INCOME
0.101
0.031
3.239
0.001
YEARS OF RESIDENCE
0.051
0.026
1.961
0.050
NEIGHBORHOOD
0.060
0.027
2.251
0.024
HOUSE OWNERSHIP
0.079
0.026
2.987
0.003
BLACK
0.020
0.027
0.744
0.457
HISPANIC
0.004
0.015
0.261
0.794
CONTACTED POLICE BEFORE
0.353
0.051
6.992
0.000
CONTACTED BY POLICE BEFORE
0.176
0.033
5.424
0.000
community participation
Discussion
The findings above confirm our communicative theory of government
performance. The positive impact of information on citizen satisfaction is
26
significant across all services.
When compared with other socio-economic and
demographic factors, the impact is usually the most important positive factor that can
mitigate other negative factors, such as the impact of police contacts. We also find
that the information impact is particularly strong in services that have greater
publicness and transaction cost problems. The best examples are city image building
and public safety. This finding is understandable because in services like these, city
officials need to engage the public more proactively, explaining the logics behind their
policy actions, and showing what they have done and accomplished. It may also be
necessary to explain to the public what challenges they face in delivering these
services and how the City has tried to overcome these challenges through
governmental actions and public-private partnerships. Otherwise, citizens may not
appreciate fully what government officials have done for them and how the tax money
is used since the results of these services are not highly measurable and obvious to the
general public, and there are many external factors that may influence the outcomes of
these services.
Our findings also show another interesting dimension of public communication
and engagement – in general, residents who have been contacted by the police tend to
feel that they are less informed. Also, they are not only more negative about public
safety, but also negative about the city image, street conditions, and cleanliness. It is
interesting to notice that these services tend to have greater measurability problems
and publicness. As a result, citizens may rely more heavily on their own subjective
perception to judge the performance of these services. The findings reflect that there
is a certain degree of negative “spill-over” feeling of the police actions on other
programs and departments. If a resident is contacted by the police because of some
personal problems or law enforcement issues, he or she is likely to view the city
government as a whole more negatively, and this negative feeling is not limited only
to the police department but is likely to spread over other services and programs.
These findings have significant implications for local officials today. First, the
need to think about information and communication effectiveness is particularly
relevant and significant in today’s political and media environment. Citizens do not
lack information about public policies and the roles of the government. However,
the problem of information overload, the fragmentation of the mass media, and the
ideological biases in many media channels may actually hurt the public feeling that
they are truly informed about government performance. In response to these
problems, it is important for the government to take a proactive stand and
communicate with the public more. Otherwise, many citizens may receive only
27
biased or fragmented opinions from the mass media or through the internet, and may
not fully appreciate what various public services have done for the community. In
today’s fiscal environment, in which many government agencies have to cut back
many program spending, including public communication and reporting, the findings
here may deserve special attention of budgeters and policymakers.
Furthermore, our findings indicate that government officials need to be more
sensitive about the differences in informational needs among various profiles of
residents. For example, while the more educated tend to have more positive views
of public services, they also have higher expectations of being informed, and if the
latter is done poorly, the negative impact of poor communication may outweigh the
positive impact of education on various service satisfaction ratings. We also find that
residents who are younger or are Hispanic are more likely to feel less informed. This
may be because they face more time constraints because of work, or they may be less
interested in the traditional media commonly used by the government. Some
Hispanic residents may also have language barriers in understanding government
policies and public information.
Fortunately, the social media and mobile
communication technologies may offer new possibilities of communication with these
groups more cost-effectively and conveniently. So how to use different media
platform to engage different profiles of citizens may be a management challenge that
government officials should pay more attention to in the 21st century.
Finally, it is interesting to notice that prior contacts with the police and being
contacted by the police tend to have a significantly negative impact on citizens’
ratings of customer services and neighborhood code enforcement issues. These
residents are also more likely to participate in community affairs, such as contacting
the offices of elected officials and seeking governmental information more actively.
At the same time, those who have been contacted by the police are more likely to
blame the government and feel that they are less informed. All these findings
suggest that government officials need to think more creatively about how to
communicate with the residents who have been contacted by the police and to channel
their anger, disappointment, and grievances more positively. The police department
plays a key role in the public relations of a city government. They are indeed the
“public face” of the city and get the public and media attention most frequently. If
the police-citizen contacts are managed poorly, there may be potentially negative
spillover effects on other departments’ performance.
28
Conclusion
In the new era of communication and social media, in which the public is
bombed by many confusing, conflicting, unfair, and inaccurate information about
government policies and program results, government officials should no longer stay
passive and ignore the importance of communication with the public. Using a
citizen survey data from the City of Tulsa, this paper shows that effective public
communication has a direct and significant impact on citizen satisfaction with many
public services, even after controlling for their socio-economic, demographic, and
neighborhood background that may influence the public’s expectation and access to
services. The impact of communication effectiveness is especially strong for
services that have high publicness, more diffused personal impact, and less
measurability. For these services, the government needs to pay even more attention
to not just what they do and deliver, but also to what they say and how they let the
public know about what they have done. From this perspective, performance
management today should be more than just managing the cost-effectiveness and
efficiency of services. If policymakers want to connect service outcomes with
citizen satisfaction, and if program managers want to make performance measurement
more relevant in the political process, how to communicate more effectively with the
public about service outcomes and how to manage the public’s expectations and
satisfaction are also important tasks, especially in the era of information revolution we
face today.
While this paper has proven the critical but often overlooked linkage between
communication effectiveness and perceived government performance, many questions
about the relationship between communication, performance management, and citizen
satisfaction remain. For example, this study only uses data from a single city.
More future research is needed to reexamine the results to make sure that they are
generalizable in other political, cultural, and social settings. Also, this study does
not fully explore the internal dynamics between communication and citizen
satisfaction. For example, when citizens are more informed, do they trust the
government more, which lead to higher satisfaction? Or is it possible that citizens who
trust the government more are more likely to be informed and become more satisfied
with government services? Also, what media channels and communication
mechanisms are more effective in communicating with the public, and do these
mechanisms differ by the demographic and socio-economic profiles of citizens?
These are some of the questions that future performance management scholars may
29
continue to explore.
References
Berman, E. M. (1997) Dealing with cynical citizens. Public Administration Review
57(2): 105-112.
Bok, D. (2001) The Trouble with Government. Cambridge, MA: Harvard University
Press.
Brown, Trevor. 2007. “Coercion versus Choice: Citizen Evaluations of Public
Service Quality across Methods of Consumption.” Public Administration Review
67(3): 559-572.
Campbell, A., Converse, P., and Rogers, W. 1976. The Quality of American Life:
Perceptions, Evaluations and Satisfactions. New York: Russell Sage Foundation.
De Vries, M. S. (2004). Framing crises: Response patterns to explosions in
fireworks factories. Administration and Society 36(5): 594-614.
Denhardt, R. B. and Denhardt, J. V. 2000. The New Public Service: Serving Rather
than Steering. Public Administrative Review 60(6): 549-558.
Denhardt, J., Terry, L., Delacruz, E., and Andonoska, L. 2009. Barriers to citizen
Engagement in Developing Countries.
International Journal of Public
Administration 32: 1268-1288.
Frederickson, H. G. 1992. Painting bull’s eyes around bullet holes. Governing
12(10): 13.
Garnett, J. L. (1992). Communicating for Results in Government: A Strategic
Approach for Public Managers. San Francisco, CA: Jossey-Bass.
Governmental Accounting Standards Board (2005). Government Service Efforts and
Accomplishments Performance Reports: A Guide to Understanding. Norwalk, CT:
Author.
Hatry, H. P. (2006). Performance Measurement: Getting Results, 2nd edition.
Washington, D.C.: the Urban Institute Press.
Hawn, J., and Siegel, R. (2006). Communicating performance: Working with
Bellevue’s residents to develop meaningful measures. Government Finance Review
(June): 22-26.
Ho, A.T., and Coates, P. 2004. Citizen-initiated performance assessment: The initial
Iowa Experience. Public Performance and Management Review 27(3): 29-50.
Hu, L. T., & Bentler, P. M. (1999). Cutoff Criteria for Fit Indexes in Covariance
Structure Analysis: Conventional Criteria Versus New Alternative. Structural
Equation Modeling, 6(1), 1-55.
Kelly, J. 2005. The Dilemma of the Unsatisfied Customer in a Market Model of
Public Service. Public Administration Review 65(1): 76-84.
30
Iyengar, S. (1991).
Is Anyone Responsible? How Television Frames Political Issues.
Chicago: University of Chicago Press.
Kelly, Janet, and David Swindell. 2002a. Service Quality Variation across Urban
Space: First Steps toward a Model of Citizen Satisfaction. Journal of Urban Affairs
24(3): 271-288.
Kelly, Janet, and David Swindell. 2002b. “A Multiple-Indicator Approach to
Municipal Service Evaluation.” Public Administration Review 62(5): 610-621.
King, C. S., Stivers, C., and Box, R. (1998). Government is Us: Public
Administration in an Anti-Government Era. Thousand Oaks, CA: Sage.
Kusow, Abdi, M., Leon C. Wilson, and David E. Martin. 1997. “Determinants of
Citizen Satisfaction with the Police: The Effects of Residential Location.” Policing
20(4): 655-664.
Levin, J., nad Tadelis, S. (2010). Contracting for government services: Theory and
evidence from U.S. cities. Journal of Industrial Economics 58(3): 507-541.
Licari, M. J., William McLean, and Tom W. Rice. (2005). "The Condition of
Community Streets and Parks: A Comparison of Resident and Non-Resident
Evaluations." Public Administration Review 65(3): 360-368.
Lineberry, R. L. (1977). Equality and Urban Policy: The Distribution of Municipal
Services. Beverly Hills, CA: Sage.
Lukensmeyer, C., and Torres, L.H. (2006). Public Deliberation: A Manager’s Guide
to Citizen Engagement. Washington, D.C.: IBM Center for the Business of
Government.
McDougall, G., Bunce, H. 1984. Urban service distributions: Some answers to
neglected issues. Urban Affairs Quarterly 19(3): 355-71.
Melkers, J., and Thomas, J.C. (1998). What do administrators think citizens think?
Administrator predictions as an adjunct to citizen surveys. Public Administration
Review 58(4): 327-334.
Milbrath, L. W. (1981). Citizen surveys as citizen participation mechanisms.
Journal of Applied Behavioral Science 17(4): 478-496.
Miller, T. I., and Miller, M. A. (1991). Standards of excellence: U.S. residents’
evaluations of local government services. Public Administration Review 51(6):
503-514.
Miller, T. I., and Miller, M. A. (1992). Assessing excellence poorly: The bottom line
in local government. Journal of Policy Analysis and Management 11(4): 612-623.
National League of Cities. 2012. The National Citizen Survey: Getting Data from
Residents that You can Act On. Washington, D.C.: Author.
Nayyar-Stone, R., Hatry, H. 2010. Using Survey Information to Evaluate Citizen
Feedback for Public Service Decisions. Washington, D.C.: Urban Institute Center on
31
Development and Governance.
Osborne, D., and Gaebler, T. (1992) Reinventing Government: How the
Entrepreneurial Spirit is Transforming the Public Sector. MA: Addison-Wesley.
Poister, T. H. (2003).
Measuring Performance in Public and Nonprofit
Organizations. San Francisco, CA: Jossey-Bass.
Shapard Research (2011). City of Tulsa City Survey. Tulsa, OK: City of Tulsa.
Stipak, B. (1977). Attitudes and belief systems concerning urban services. Public
Opinion Quarterly 41(1): 41-55.
Satorra, A., & Bentler, P. M. (1988). Scaling corrections for chi-square statistics in
covariance structure analysis. Proceedings of the American Statistical Association,
308-313.
Satorra, A., & Bentler, P. M. (1994). Corrections to test statistics and standard errors
in covariance structure analysis. In A. von Eye & C. C. Clogg (Eds.), Latent variables
analysis: Applications for developmental research (pp. 399-419). Thousand Oaks, CA:
Sage.
Stone, C. (1980). Systemic power in community decision making: A restatement of
stratification theory. American Political Science Review 74(4): 978-990.
Swindell, D. and Kelly, J. M. (2000). Linking citizen satisfaction data to
performance measures: A preliminary evaluation.
Public Performance and
Management Review 24(1): 30-52.
Thomas, J.C. 1995. Public Participation in Public Decisions. San Francisco:
Jossey-Bass.
Van Ryzin, G. 2004. Expectations, Performance, and Citizen Satisfaction with
Urban Services. Journal of Policy Analysis and Management 23(3): 433-448.
Van Ryzin, G. 2006. Testing the expectancy disconfirmation model of citizen
satisfaction with local government. Journal of Public Administration Research and
Theory 16: 599-611.
Webb, K., Hatry, H. 1973. Obtaining Citizen Feedback: The Application of Citizen
Surveys to Local Governments. Washington, D.C.: the Urban Institute.
Wheeler, K. M., ed. (1994). Effective Communication.
Washington, D.C.:
International City Management Association.
32