Contents lists available at ScienceDirect
Government Information Quarterly
journal homepage: www.elsevier.com/locate/govinf
Does the freedom of information law increase transparency at the local
level? Evidence from a field experiment
Peter Spáča
, Petr Vodab,⁎
, Jozef Zagrapanc
a
Department of Political Science, Faculty of Social Studies, Masaryk University, Jostova 10, Brno 602 00, Czech Republic
b
International Institute of Political Science, Jostova 10, Brno 602 00, Czech Republic
c
Sociologický ústav SAV, Slovenská akadémia vied, Klemensova 19, Bratislava 813 64, Slovakia
1. Introduction
Transparency is an important attribute of public administration (cf.
Bauhr & Grimes, 2014; Heald, 2006; Worthy, John, & Vannoni, 2017).
In general, information availability facilitates public accountability (cf.
Meijer, 2014; Olsen, 2015). Transparency increases control over government,
as it allows society to consider information that is otherwise
kept secret. Scholars agree that transparency generally enhances public
acceptance of political decisions, reduces corruption and provides more
legitimacy to democratic governments (de Fine Licht, 2014; Hood,
2006; Meijer, Hart, & Worthy, 2015). In addition, transparency may
stimulate the performance of political representatives, as the disclosure
of their mistakes and failures can eventually reduce their support (cf.
Grimmelikhuijsen & Meijer, 2012; James & Moseley, 2014). On the
contrary, Fenster (2015) challenges the simplified understanding of
transparency and concludes that for a functional state a balance between
transparency and secrecy is necessary.
Furthermore, Fenster (2015: 150) argues that “transparency is best
understood as a theory of communication that excessively simplifies
and thus is blind to the complexities of the contemporary state, government
information, and the public”. From this perspective it is important
to study less abstract phenomena related to transparency. This
article aims to study one such aspect of the concept of transparency –
the compliance of officials to FOI legislation. This legislation provides
citizens with the right to ask for information and binds public officials
to respond to such demands. However, in the end, the requests are
processed by officers. Therefore, how the requests are processed is just
as important as the implementation of FOI laws.
The first FOI acts were enacted in Sweden in 1766. However, the
mass wave of adoption of such law began in 1966, when the USA
adopted their FOIA. According to Hazell and Worthy (2010), by 2010
around ninety countries had adopted this legislation, while another fifty
were preparing such measures. In 2017, the number of countries with
FOI acts was higher than 115 (Freedominfo, 2017). Hence, over time
FOI acts have become a rather regular feature in countries worldwide
(cf. Walby & Larsen, 2012; Worthy, 2017).
It is crucial to examine to what extent FOI laws are followed in
practical terms. In other words, for FOI acts to fulfil their role in providing
access to information to the public, the compliance of public
officials is a necessary condition. In a growing number of countries,
research on the effectiveness of such laws has already been conducted
(cf. Cherry & McMenemy, 2013; Michener, 2015; Michener & Worthy,
2015). These studies mostly cover either Western democracies or developing
countries. One of our aims is to contribute to this state of the
art by analyzing Slovakia - a relatively new democracy belonging to a
group of post-communist countries in Central Europe. So far, this
group's compliance with FOI has not been examined by researchers. A
profile of Slovakia provides the opportunity to carry out research of FOI
compliance in a new context uncovered by the existing literature.
Transparency and its consequences do not work free of context, but
they depend on the institutional and political factors of political systems
(cf. Cucciniello, Porumbescu, & Grimmelikhuijsen, 2017). This
explains why, under some constellations, transparency reinforces the
trust of society in public institutions, while in others such effects do not
appear (cf. de Fine Licht, 2011; Welch, Hinnant, & Moon, 2005). This is
yet another reason to conduct studies in new contexts.
Researchers agree that central policies are not always followed by
local agencies (cf. Berliner, 2017; Ross & Whittaker, 2009). We thus
examine whether FOI laws assist the public in their demand for information
at the local level. More precisely, we aim to answer how the
FOI act improves the chances of receiving information from municipalities
and how this effectiveness is affected by the characteristics of
municipalities. In doing so, we divided the municipalities into three
groups. The control group received a baseline request, while two experimental
groups received the same request with an additional paragraph,
either including a moral appeal or a reference to FOI act. We
found that FOI requests led to a higher responsiveness from local authorities,
while the moral appeal had no effect. This relation is already
known in the current literature. What is new is that the effect of the
content of the requests is moderated by the capacity of the requested
authority. While the FOI requests led to a substantially higher response
rate by public authorities in less populated municipalities, in bigger
cities, which tend to have more staff, this effect is diminished and the
form of the request became less relevant. Finally, we also found a
https://doi.org/10.1016/j.giq.2018.05.003
Received 23 January 2018; Received in revised form 27 March 2018; Accepted 7 May 2018
⁎
Corresponding author.
E-mail address: pvoda@apps.fss.muni.cz (P. Voda).
Government Information Quarterly 35 (2018) 408–417
Available online 11 May 2018
0740-624X/ © 2018 Elsevier Inc. All rights reserved.
T
substantial effect of several contextual variables. Language barrier decreases
the response rate of local authorities from areas with a predominant
ethnic minority, whereas the probability of response is higher
in municipalities led by a female or independent mayor compared to
municipalities with a male or partisan mayor.
The paper is organized as follows. The first part outlines the theoretical
background of FOI as a tool for fostering higher levels of
transparency in public administration and discusses various forms of
information requests. The second part focuses on contextual factors
with potential impact on responsiveness of local authorities. After that,
the design of our experiment is presented. The fourth part presents the
data and methods. Next, the results of our study are shown and the final
part concludes our findings.
2. Legal backing and moral appeals of information requests
In general, FOI laws enhance the public right to demand information.
They normally enforce proactive transparency, i.e. they require
public authorities to publish information about their procedures and
outcomes, and also provide grounds for information requests addressed
to such authorities (Worthy et al., 2017). With the spread of FOI acts,
the research focus on this field has also increased. In recent years a
rising amount of literature has been responding to the demand for a
better understanding of FOI regimes and institutional factors affecting
their impact on society, press, political institutions and their relations
(cf. Cherry & McMenemy, 2013; Meijer et al., 2015; Nam, 2012;
Worthy, 2010).
Although the existing research provides important knowledge about
the overall performance of FOI legislation, its outcomes are mainly
based on studies conducted at the national level and on cross-national
comparisons. On the other hand, studies of transparency at the local
level are rather underdeveloped (cf. Berliner, 2017). The local level of
governance is of vital importance though, as citizens are more closely
related to local issues and data. As Piotrowski and Van Ryzin (2007)
found in their analysis of a national survey in the USA, the public is not
indifferent to the level of transparency on the local level. Other research
on local openness has been primarily devoted to measuring the transparency
of local institutions. To capture this, some authors (cf. da Cruz,
Tavares, Marques, Jorge, & de Sousa, 2016; Esteller-Moré & Otero,
2012; Jorge, Sá, Pattaro, & Lourenço, 2011; Pina, Torres, & Royo, 2007)
mainly use the amount and content of data provided by municipalities
via official channels, i.e. websites or annual reports, to propose indices
of specific segments of local transparency. Although such tools enable
the assessment of the level of openness on the local level and provide
conditions for wider comparative research, they all remain focused on
proactive transparency. Hence, despite the importance, these indices do
not map how local institutions react to FOI requests.
In regards to information requests, the compliance of officials to FOI
legislation is the crucial aspect of the concept of transparency. Although
the FOI acts are adopted by national parliaments, it is the individual
level of analysis where the actual mechanism of requesting information
takes place. While FOI legislation may prescribe precise rules for the
process of handling information requests, it is always an individual
public official who responds to and processes such demands from the
public. Hence, from the perspective of those applying for information,
the implementation of FOI laws is at least as important as their adoption.
In other words, ‘freedom of information laws are useful to the
extent that they are followed’ (Cuillier, 2010, 203). Given that the
central aims of public policies are not always adopted by all offices and
agencies (cf. Berliner, 2017; Peters, 2014), the question is what affects
the probability that those seeking information will be successful in their
effort.
The compliance of public officials when asked for information may
vary substantially. According to the works of Roberts (1998) and Snell
(2002), compliance may range from the cooperative disposition of officials
towards information requests and the requesters to the noncompliant
behavior that includes deliberate non-recording of information
in order to avoid future requests or the removal of information
from the requested data files. While failure to comply with FOI laws
might be case dependent, existing research does provide some guidance
as to why it occurs. According to studies in India, lower compliance
with FOI laws is caused by a lack of resources, insufficient training of
civil servants or their unawareness of the respective legislation (Raag/
NCPRI, 2009, 2014). Similarly, in their study of Brazil, Michener,
Moncau, and Velasco (2016) identified bureaucratic, legalistic and
technical barriers that lead to a difference between provisions of the
FOI act and its actual implementation.
More importantly, however, previous studies agree that the content
of requests influences the responsiveness of the receiving party and they
mostly discuss requests that include either legal grounds or moral appeals.
In respect to the former, the previous research indicates that
empowering requests with legal backing supports compliance of the
receiving party (cf. Fellner, Sausgruber, & Traxler, 2013; Fiorio, Iacus,
& Santoro, 2013). This approach is based on the presumption that the
request is handled in a more responsive way if it contains some sort of
threat (cf. Cuillier, 2010). In respect to information requests, this role is
played by the FOI act itself as it provides a stronger footing for the
requester in relation to the public authorities. Requests containing a
direct reference to the FOI act also place some pressure on public authorities
to supply the requested information, as they underscore that
the applicants have some leverage over the bureaucracy (cf. Peisakhin
& Pinto, 2010).
Although the literature is rather scarce in this regard, it supports the
positive effect requests with legal backing have on the response rate of
public authorities. Peisakhin and Pinto (2010) conducted a field experiment
on access to ration cards for people living in slums in India.
They randomly assigned participants into four groups that used different
approaches to obtain a ration card. The groups used approaches
which included FOI requests, bribery, and NGO support, while the
control group applied for the card in the standard manner. As the results
showed, the group using FOI requests had a substantially higher
probability of receiving the card than the control group and they almost
equaled the effectiveness of the bribery option. Peisakhin (2012) later
replicated the experiment, now on voting registration, with very similar
results. Similarly, Worthy et al. (2017) analyzed parish councils in
England and their responsiveness to information requests. Using e-mail,
they sent letters to more than four thousand parishes and asked them to
send and publish their organizational chart. One half of the parishes
were selected as the treatment group and received an FOI request, while
the other half, being the control group, received a less formal ask. The
authors found that FOI requests led to a higher level of compliance
among parishes, with more than double the response rate when compared
to the control group.
Another way to ensure compliance of the receiving party is through
moral appeals. This approach does not rely on intimidation but instead
on increasing trust on the part of the subject who is asked (cf. Cuillier,
2010). This type of request also aims to persuade the receiving party
that their compliance is a rightful act which contributes to some higher
good. This motive is shown in a paper by Torgler (2004), who ran a
controlled field experiment at the communal level in Switzerland. In
cooperation with the local tax administration, the author sent letters
containing normative appeals to a randomly selected group of citizens.
In particular, the letter stressed the role of taxes in keeping the region
attractive for its inhabitants and also referred to the sense of responsibility
of Swiss citizens.
In contrast to requests using legal threats, the effect of moral appeals
is found to be rather mixed. Experimental studies from the field of
economy showed only limited effect of normative appeals (cf.
Blumenthal, Christian, Slemrod, & Smith, 2001; Torgler, 2004) although
here, public authorities were the party sending the requests and
not the ones receiving them. A closely related study was provided by
Cuillier (2010). In his experiment, school agencies were requested to
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
409
provide superintendent contracts and here three versions of letters were
delivered: friendly, threatening and neutral. The friendly letter led to
slightly higher helpfulness of the school agencies but without increasing
the response rate. On the other hand, the threatening letter had a
substantially higher amount of responses.
Altogether, the existing literature shows that the content of requests
affects compliance behavior, with legal standing and moral appeals
having different effects in this matter. Based on the existing research we
formulate the following two hypotheses. In case of the moral appeal,
the hypothesis is rather tentative as the results of previous research on
this matter are mixed:
H1. FOI requests lead to a higher response rate from public authorities
than requests sent without reference to an FOI act.
H2. Requests containing a moral appeal lead to a higher response rate
from public authorities than requests sent without such reference.
3. Staff, language and political leadership at the local level as
possible predictors of responses
The previous section presented theoretical works dealing with the
effect of content of information requests on the responsiveness of public
authorities. However, there are other factors which impact the compliance
of public bodies besides whether the subjects appeal to legal or
moral grounds in their requests. So far, researchers have given less
attention to features of the local public authorities and the institutional
settings in which they operate. This chapter enlists three such variables,
namely ethnic composition and language, political leadership and size
of staff.
A number of studies, including cross-national works on public surveys,
found that people belonging to ethnic minorities respond at
substantially lower rates. Among the reasons for this, language barrier
is identified as the primary cause of their non-responses (cf. Feskens,
2009; Harzing, 1997; Sheldon, Graham, Pothecary, & Rasul, 2007).
Therefore, when asked to participate in a survey in the language of the
majority, members of ethnic minorities with less knowledge of the
language might find themselves to be unable to respond properly.
Darch and Underwood (2005) claim that the FOI act needs general
preconditions to be effective – citizens “literate and able to understand
the dominant discourse of power, both literally and metaphorically”.
From this point of view, a different language does not only lead to
problems with simply understanding the request, but also to problems
of legitimacy between requested and requesting. Language and ethnic
differences might affect the behavior of public authorities as well, although
more literature is needed here. In his experimental study,
Broockman (2013) found that political representatives from minorities
reflect interests concerning their own minority to a higher extent, even
if they provide them with no electoral gains. His study was later replicated
by Pilet, Reimink, and Schrobiltgen (2015) who tested the
responsiveness of candidates in local elections in Brussels, however
without confirming Broockman's results. In respect to information requests,
the language issue might be relevant if the request is handled by
a person with lesser knowledge of the language of a request. This might
occur in areas with a predominance of ethnic minorities, as representatives
of these minorities are elected to local offices and hired as
their staff to a greater extent. Based on the theoretical findings, this
scenario should lead to a lower response rate by local authorities.
H3. A higher share of ethnic minorities is negatively associated with the
response rate of local governments to information requests.
Previous research points to differences in the level of transparency
between municipalities led by mayors of different personalities. Fox
and Schuhmann (1999) argue that female mayors promote citizen
participation and communication to a greater extent than their male
counterparts. However, Araujo and Tejedo-Romero (2016) did not
confirm this hypothesis in their study about transparency in Spanish
municipalities. There is more agreement on the role of the party
ideology of a mayor. Albalate del Sol (2013) confirmed the assumption
that left-wing governments are more sensitive to providing transparency
than right-wing parties. The same finding was replicated by
Araujo and Tejedo-Romero (2016). However, these scholars differ in
their reasoning for this relationship. According to Piotrowski and Van
Ryzin (2007), both left-wing and right-wing parties are interested in
transparency. Moreover, we argue that partisans and independents
differ in their performance in local politics. Partisans view local politics
as a temporary stage of their careers before being promoted by their
parties to regional or central politics. In contrast to this, independents
typically lack such a perspective and rather maximize their performance
on the local level (cf. Gendźwiłł, 2012; Reiser & Holtmann,
2008). The higher motivation of independents for better performance
also stems from the character of local rallies. While partisans are backed
by their parties and their resources, independents rely only on their
performance in office (cf. Nagatomi, 2013). Their election support is
also less secure as, unlike voting for parties, it is less customary and
heuristic (cf. Madison et al. 2014). Based on the previous findings, local
independents are expected to have better performance than partisans
and this also includes responsiveness to external challenges such as
information requests, as this may also affect their record in their con-
stituency.
H4. Local governments led by a female have a higher response rate to
information requests than local governments led by a male.
H5. Local governments led by a leftist mayor have a higher response
rate to information requests than local governments led by a right-wing
mayor.
H6. Local governments led by an independent have a higher response
rate to information requests than local governments led by a partisan.
The final influential variable covers the capacities of public bodies.
In the recent period, local governments are facing greater expectations
of their overall performance (Raudeliūnienė & Meidutė-Kavaliauskienė,
2014). To accomplish this, public bodies need enough resources and
staff (cf. Lewis & Wood, 2012). In terms of local governments, more
populated towns are given more resources (cf. Styles & Tennyson,
2007) and are able to hire more personnel, which gives them an advantage
over less populated municipalities. Existing literature confirms
that cities with greater populations do more in terms of proactive
transparency and provide more access to information (cf. Bearfield &
Bowman, 2017; Pina et al., 2007; Sičáková-Beblavá, Kollárik, &
Sloboda, 2016). As handling information requests falls under the
overall performance of public bodies, the theoretical expectations lead
us to formulate the following hypothesis:
H7. Local governments with more staff will have a higher response rate
to information requests.
Unlike the language barrier and non-partisan leadership on the local
level, staff size can be expected not only to affect the general compliance
of public authorities but also to moderate the effect of various
versions of information requests on the response rate of these offices.
Under the existing demand for overall performance (Raudeliūnienė &
Meidutė-Kavaliauskienė, 2014), local bodies with lesser capacities
might find themselves with a need to prioritize their activities. Regarding
information requests, local governments with limited staff
might be more willing to answer requests with legal backing. The
reason for this is quite straightforward. Requests under an FOI act
provide some legal ground for the requester who, in case of a nonresponse,
may appeal and eventually ask the courts for help (cf.
Peisakhin & Pinto, 2010). By not answering these information requests,
understaffed public bodies risk facing legal processes that require additional
resources, which they already lack. On the other hand, information
requests without a proper legal backing, such as those with a
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
410
moral appeal, can be ignored more easily, as they do not result in any
similar legal processing. This should account mostly for local governments
with lesser personnel as bodies with more staff often including
specialized legal personnel do not need to concern about consequences
to not answering to a request backed by FOI act. Hence, for bigger
municipalities, the difference between the response rates to various
versions of the information requests should be smaller.
H8. Effect of different versions of information requests on response rate
of local authorities will be moderated by capacities of these public
bodies.
4. Experimental design and procedure
Following the trends in other Central and Eastern European countries,
Slovakia adopted its FOI law in 2000. The law itself is quite extensive
as it obliges a wide variety of actors, including state authorities,
regions and local governments, to provide information to applicants
upon request. Users can apply for information via regular or electronic
mail and they have to specify the content of the information and the
preferred method of delivery. The FOI law requires applicants to provide
their full name, address and phone number with their requests.
To test whether and how local governments respond to FOI requests,
we designed a field experiment. Using code in R, we divided our sample
– all municipalities in Slovakia – into three groups at random and sent
them different types of emails. We requested information about the
results for both elected and not elected candidates in local elections
held in 2010 and 2014. This was information which we needed for our
other ongoing research on ballot order effect in systems using alphabetical
ranking of candidates. The information we asked for was publicly
available only for the ten largest cities in Slovakia. Only the data
for elected candidates was available for all other Slovak municipalities.
Before we conducted our study, we contacted the two central bodies
which manage elections in Slovakia − the Ministry of Interior and the
Statistical Office of the Slovak Republic − to provide us with this information.
Both institutions informed us, however, that only municipalities
possessed complete electoral results from both years.
We designed three types of mail. In the first version, which we call
baseline, we introduced our research team and our focus on local elections.
We explained that we came across a problem with the lack of
availability of complete results of local elections and we mentioned our
earlier communication with the central institutions, which had informed
us that only municipalities possessed such data. Then, we asked
the municipality to send us the 2010 and 2014 lists of local election
candidates, both elected and not elected, and the number of votes these
candidates received. We asked for their response to be delivered by
electronic mail, and we also provided our address and phone number as
required by the FOI law. This letter was sent to the control group. The
two remaining versions of our letter were sent to experimental groups.
These letters were identical to the baseline version, however each
contained a short additional paragraph. Although we are aware that the
wording of the first request may appear to be an appeal to professionalism,
we decided to call it a “moral appeal” as this is line with previous
research studies and similar studies; the “legal” variant is thus labelled
because we refer to the exact law according to which the municipalities
should act.
To be specific, the moral appeal version was intended to create a
positive reaction by stressing the value and relevance of their cooperation.
Its final paragraph highlighted that sending us the requested
information would support our research, research that is intended to
increase society's understanding of the local government and also to
enhance the quality of functioning of local government in general. On
the other hand, the legal version included a final paragraph in which we
informed the recipients that our request was based on the FOI law that
obliges municipalities to provide information, given that it possesses
the respective data. Full versions of all three types of letters can be
found in the appendix.
Previous research has identified several requester characteristics
which affect the probability of response. Requesters belonging to vulnerable
social groups seem to be at a disadvantage (OSJI, 2006),
whereas requests from institutionally aligned persons (Michener et al.,
2016) had higher rates of obtaining the requested information than
other civilians. Also, the gender of the requester showed some effect, as
Michener et al. (2016) found some advantages given to men. On the
other hand, an experimental study in Mexico showed no discrimination
of requesters with common names compared to seemingly influential
requesters (Lagunes and Pocasangre, 2017). From this perspective, the
way how the letter is signed can influence the response rate. However,
all of our letters are signed in same way.
Our population consisted of all 2928 Slovak municipalities, which
enabled us to avoid any potential sampling bias. We downloaded their
email addresses from a publicly accessible database (http://www.eobce.sk/),
checked and confirmed that every municipality had its own
specific email account. On Monday 12.10.2015, after the random division,
we sent out the mails manually, without any robots or programs
which might have made them look like spam. Even though the law
states that institutions should reply to FOI requests within eight
working days with the option of doubling this time limit, for the purposes
of this study, we report and work with all the answers we received
within a month.
Furthermore, in our analysis we include variables on partisanship of
mayors, ethnic composition and size of municipality. Slovakia uses a
first-past-the-post electoral system to elect mayors; therefore, unlike in
a parliamentary election (where a party-list proportional representation
system, with one nationwide constituency, is used), voters can vote for
individual candidates. Slovakia therefore represents a case in which we
can study the mediating effects of partisanship, given that some municipalities
are run by partisan mayors (backed by different parties or
coalitions) and some by independents. In the latest election in 2014, no
party was as successful as independents, who won in almost 38% of
municipalities. In addition to this, Slovakia provides an opportunity to
study language barriers, given the variation in the ethnic composition
of municipalities. The Hungarian minority constitutes around 10% of
the Slovak population and is concentrated predominantly in the
southern part of the country. Also, since Slovakia is a relatively small
country, it enables us to examine all the municipalities in the country.
Moreover, there is substantial variance in the size of municipalities,
which allows us to also examine the role of size.
One of the most important things to consider when designing an
experiment is its ethics (e.g. Morton & Williams, 2010). We consider
that our experiment possesses no threat to experimental ethics, since we
included all the information required by law in the requests and we
used no deception. We also did not ask for information we did not need,
but for electoral results we were going to use in our ongoing research,
and we would also make it all publicly available once we were finished
coding it. We signed the e-mails with our real names and we made it
clear that the information received would be used for research purposes.
Most importantly, the two institutions expected to have the information
we requested – the Ministry of Interior and the Statistical
Office of the Slovak Republic – informed us that we should ask the
municipalities to provide it.
5. Data and methods
Our dataset uses the municipalities that gave us some sort of response
as well as those that did not. The data as well as replication files
are included in the on-line appendix. The dependent variable is the
response and it is coded into two categories. Value one is given to
municipalities that sent us at least some information. These include
both responses containing complete data for both local elections we
were interested in as well as partially complete responses that included
data for at least one of the two elections. In the event that a
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
411
municipality sent a link, we accepted this as a response containing some
information if the link led directly to the requested data or if it was
accompanied with guidelines leading to this result. Since the goal is to
study the effectiveness of FOI requests, all other municipalities are
given the value of zero on the dependent variable. These are municipalities
that did not reply at all or they replied but without sending any
data on candidates. 102 municipalities in our dataset had non-functioning
e-mail addresses in the database; these generated automatic
responses which are also coded as non-response. Table 1 presents the
coding scheme of the dependent variable.
A crucial independent variable in our analysis is request type, with
three categories associated with the above presented versions. The
“control” category is comprised of municipalities who received a
baseline letter, the “legal” variant is comprised of municipalities who
received the letter with the paragraph referring to FOI law and the third
category, the “moral appeal” variant, is comprised of municipalities
that received a letter with the moral paragraph. Each category covers
one third of municipalities. In the analysis, the “control group” is the
reference category and the results of analysis are interpreted as the
differences between the legal and moral variants as compared to control
group.
Furthermore, our research includes variables to control for the
mediating effects of partisanship and gender of mayors and language
barrier. Therefore, the variable of political background of mayors and
the share of Hungarian national minority are covered in the data. The
variable on the partisanship of mayors is categorical. It distinguishes
five types of mayors according the party which nominated them – left,
right-wing, coalition, independents and others. SMER mayors are categorized
as leftists, and the category of right-wing mayors includes
those nominated by SaS, SDKÚ-DS, KDH, SNS, SIEŤ and MOST-HÍD.
When a mayor is nominated by a coalition of parties including leftist
and right-wing parties simultaneously or if there is some other party,
we put these mayors in the category “coalition”. Gender is a dichotomous
variable, where female mayors are coded as 1 and male as 0. In
the case of the Hungarian minority variable, we differentiate between
four categories based on the municipal share of Hungarians.
The final variable covers the capacities of public bodies. Given that
no data source maps the exact number of staff in each Slovak municipality,
we use the size of the local population as a proxy. Recent study
has shown that bigger municipalities possess more resources and tend
to have a greater number of staff. While the smallest villages in Slovakia
have less than five employees, as population rises, staff size increases
substantially up to several tens of clerks (cf. Gecíková & Papcunová,
2013). Population size is measured by number inhabitants according to
the 2011 Census (Statistical Office of Slovak Republic, 2011;see
Table 2). We use a logged version of this variable, given that many
studies on various political issues (Blais & Dobrzynska, 1998; Boix,
1999; Buhaug & Lujala, 2005; Gowa & Mansfield, 1993; Hegre &
Sambanis, 2006; Reitan, Gustafsson, & Blekesaune, 2015; Remmer,
2004) have shown that population as an explanatory factor has a nonlinear
effect. Table 3 shows the descriptive statistics for the independent
and control variables. In the case of categorical variables,
the table indicates the proportion of cases in which the given category
occurs in our sample.
However, there is some information we do not have. Although we
sent requests to all municipalities, we do not know who actually responded
to us, or in the case of non-response, who decided not to send
the requested information. Even more fundamentally, we do not know
whether the non-response was intentional or just accidental.
Unfortunately, to gain such information, we would have to approach
this question with completely different research, and this would be
hardly useful for solving our main question about the effect of using FOI
in asking for information.
Binomial logistic regression is employed to analyze the data and
evaluate the effect of form of request on the probability of response.
This method is chosen because of the nature of the dependent variable
and the structure of data. This kind of analysis evaluates the influence
of individual characteristics connected to the cases (in this analysis,
municipalities) on the chance a certain event will happen or not. The
following model is constructed to analyze the relations between the
chance of getting a response and a set of independent variables which
includes an experimental group and control variables.
6. Results
The first part of the analysis deals with the description of the dependent
variable. Table 4 describes the ratio between the type of response
and how it correlates with different types of requests. Our requests
were thus effectively sent to 2826 municipalities. Out of these,
834 sent us at least some of the information requested about local
candidates, or attempted to send us the information by providing us
with a link. On the other hand, 2070 municipalities did not reply at all
and 25 municipalities sent a response without any data. Thus, a total of
2095 municipalities did not respond with any information. This means
that we received answers with some information from almost one third
Table 1
Coding scheme of the dependent variable.
Value Types of responses
1 Reply containing data about at least one of the two local elections
Direct link or link with guidelines leading to such data
Reply only with publicly available data
0 No reply
Reply without any data
NA Unfunctional e-mail address
Table 2
Number of municipalities according to the population size.
Inhabitants Number of municipalities
< 199 392
200–499 742
500–999 765
1000–1999 574
2000–4999 283
5000–9999 61
10,000–19,999 34
20,000–49,999 29
50,000–99,999 8
Over 100,000 2
Table 3
Descriptive statistics of independent variables.
Min Max Mean Median SD
Type of letter
Control 0.00 1.00 0.34
Legal 0.00 1.00 0.33
Moral appeal 0.00 1.00 0.33
Party affiliation of mayor
Left 0.00 1.00 0.29
Right-wing 0.00 1.00 0.12
Coalition 0.00 1.00 0.13
Others 0.00 1.00 0.07
Independent 0.00 1.00 0.38
Gender of Mayor
Female 0.00 1.00 0.22
Hungarian minority
0–25% 0.00 1.00 0.84
25–49% 0.00 1.00 0.04
50–74% 0.00 1.00 0.06
Over 75% 0.00 1.00 0.06
LogPopulation 2.64 12.93 6.56 6.49 1.20
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
412
of all municipalities. Among the experimental groups, the highest
proportion of responses with some information was received in response
to requests containing reference to the FOI act; > 2/5 of municipalities
queried sent us the requested data. The control and moral
appeal variants received the information requested in only about 1/5 of
the cases and there is no significant difference found between the two
groups.
The response rate is relatively high compared to other similar studies.
The reason for this probably lies in the fact that we ask for information
which is generally publicly available. All results of other
elections (parliamentary, presidential, regional), including the names of
candidates, are published on the webpage of the Slovak Statistical
Office. Moreover, our requests mentioned our university affiliation and
academic titles, which has previously been found to be an advantage
(Lagunes and Pocasangre, 2017; Michener et al., 2016).
There is an interesting trend in responses over time (Fig. 1). We
received approximately one third of all answers by the end of the
second day after sending the requests and more than half of the replies
arrived within three days. Moreover, 610 out of all 834 received answers
were sent during the first week. The speed of the replies only
slightly differed in relation to the type of requests. Both the baseline
and legal versions had the same average response time. On the other
hand, responses to the moral variant took one day longer on average.
The results of the binomial logistic regression clearly support the
first hypothesis about the effect of FOI requests. E-mail requests referencing
FOI law have almost three times greater odds of being processed
than requests without such reference (see Model 1 in Table 5). However,
as we do not find any substantive difference in obtaining a response
between the control group and e-mails with moral incentives,
we cannot confirm the hypotheses that moral appeals lead to a higher
response rate from public authorities than requests sent without such
reference. The results also show that several of the control variables
proved to be significantly influential. Municipal population size is
found to be a strong predictor of receiving a response. Requests processed
in the largest towns have an approximately ten times greater
chance of being answered than the same request in the smallest municipalities.
Since the variable “size of municipality” is logged, we illustrate
the effect in a more understandable way below (see Fig. 2).
Moreover, size of municipality also affects the influence of the form
of request (see Model 2 in Table 5). In small municipalities, the chance
of FOI requests being processed is about seven times higher than other
forms of requests, but it is still much lower than in more populated
municipalities. These differences fade away with growing population
size and the probability of responding to a FOI request compared to
other versions is only slightly and insignificantly higher among municipalities
with > 20,000 people.
Regarding the personalities of mayors, we found a strong and significant
effect of gender. The chances of receiving an answer from
municipalities governed by a female mayor are one third higher than
from municipalities led by a male mayor. This result is in line with our
hypothesis H5. Another important variable is the party affiliation of
mayors. The odds ratios indicate that requests have a greater chance of
being processed in municipalities with independent mayors than in
municipalities with partisan mayors. This might indicate that, as expected,
independents are more oriented towards maximizing their
performance on the local level than their partisan counterparts. However,
we do not find any evidence about the differences between municipalities
led by leftist and right-wing mayors (see model 2 in
Table 5).
These results are opposite to those of Araujo and Tejedo-Romero
(2016) on the case of Spain. The missing effect of ideology in our study
may be the result of the lack of interest in transparency among both
leftist and right-wing parties and by the nature of parties and
Table 4
Replies to information requests.
Total Absolute N Relative
No info Some info No info Some info
Control 961 750 211 78.04 21.96
Legal 943 529 414 56.1 43.9
Moral appeal 922 713 209 77.33 22.67
Total 2826 2095 834 70.49 29.51
Fig. 1. Received responses on time axis.
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
413
competition on the local level, where parties do not articulate clear
ideological stances and where candidates are often supported by a very
heterogeneous group of parties, thus making it difficult to identify the
ideological position of mayors. On the contrary, the effect of gender
may stem from the fact that Slovakia is a younger democracy than
Spain. Female politicians in Slovakia (Bitušíková, 2005), and in the
Central-East European region generally (Clavero & Galligan, 2005;
LaFont, 2001), have to overcome greater barriers than in Spain and
therefore they may be more motivated to behave differently than their
male counterparts.
In the previous section we presented the odds ratios of the chances
of receiving a response. As the substantive meaning of the obtained
effect can be misleading when presented in this fashion, the odds ratios
are computed to probabilities. Table 6 shows us the predicted probabilities
for all combinations of categorical independent variables from
model 3 when the size of municipality is average (approximately 700
inhabitants). The effect of population size as the effect of the cardinal
variable is illustrated in the chart below. Values of response probability
are the highest when an FOI request is sent to municipalities which
have both a predominant Slovak population and an independent female
Table 5
Odds ratios from binary logistic regression.
Model 1 Model 2 Model 3
Odds ratio Conf. int. Odds ratio Conf. int. Odds ratio Conf. int.
2.5% 97.5% 2.5% 97.5% 2.5% 97.5%
Type of letter (control is ref. category)
Legal 2.82a
2.29 3.47 5.72a
1.70 19.52 5.65a
1.68 19.23
Moral appeal 1.02 0.81 1.27 1.42 0.37 5.42 1.36 0.36 5.17
Party affiliation of mayor
Reference left left partisans
Right 1.15 0.84 1.57 1.16 0.85 1.58
Others 0.96 0.64 1.44 0.97 0.65 1.45
Coalition 1.25 0.93 1.67 1.25 0.93 1.67
Independent 1.36b
1.09 1.70 1.36b
1.09 1.70 1.26a
1.05 1.50
Gender (“Male” is ref. category)
Female 1.33a
1.08 1.64 1.33a
1.08 1.64 1.33a
1.08 1.64
Hungarian minority (“0–25%” is ref. category)
25–49% 1.23 0.78 1.91 1.24 0.79 1.92 1.24 0.79 1.92
50–74% 1.38c
0.95 1.99 1.39c
0.95 2.00 1.37c
0.96 1.95
Over 75% 0.74 0.50 1.09 0.74 0.50 1.09 0.73c
0.50 1.05
LogPopulation 1.54 1.44 1.67 1.63a
1.42 1.87 1.64a
1.44 1.88
Legal ∗ LogPopulation 0.90 0.75 1.08 0.90 0.75 1.08
Moral ∗ LogPopulation 0.95 0.79 1.16 0.96 0.79 1.16
Intercept 0.01a
0.01 0.03 0.01a
0.01 0.02 0.01a
0.01 0.02
N = 2826.
a
Sig. on level 99%.
b
Sig. on level 95%.
c
Sig. on level 90%.
Fig. 2. Probabilities of receiving a response according to size of municipality.
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
414
mayor. More precisely, in such a case, the probability of receiving an
answer containing some information is about 50%. On the contrary,
other forms of requests sent to dominantly Hungarian municipalities
with male mayors only have a 15% probability of being positively
processed. The effect of the presence of a Hungarian minority is similar
to findings by Darch and Underwood (2005). They analyzed FOI requests
in South Africa and argued that language diversity may be one of
reasons for the rather poor performance of FOI legislation.
Fig. 2 illustrates that the evolution of response probability is dependent
on the size of municipality and type of request. The upper solid
line clearly indicates that FOI requests have a significantly higher
chance of being answered than other forms of requests in municipalities
with < 10,000 inhabitants (dotted and dashed lines). With rising population,
the confidence intervals start to overlap and differences become
insignificant. The chart also illustrates the non-linearity of the
relation between size of municipality and probability of obtaining a
response to a request. In the case of FOI requests, the probability rises
from 15 to 40% when population grows from around ten in the smallest
municipality to 1000, and a probability of 50% is reached in municipalities
with about 2000 inhabitants. When the population rises even
further, the probability rises gradually and, due to the low number of
cases of large municipalities in our data (see Table 3), the confidence
intervals become wider. With other types of requests, the rise in
probability among municipalities with < 1000 inhabitants is less steep
as compared to the legal variant, however then the slope becomes
steeper and the probabilities in the biggest municipalities reach almost
the same level as the legal variant.
7. Conclusion
This research provides insights into how local authorities function
and how responsive they are to information requests. Optimally, when
asked for identical information, public offices should respond to such
requests in the same manner. Our field experiment, encompassing all
2928 municipalities in Slovakia, contradicts this and it shows that the
form of information request is influential in this regard.
Our results prove that in respect to achieving a higher response rate,
FOI requests are superior to other forms of information requests (H1).
The conducted analysis shows that referring to FOI acts leads to several
times higher odds in receiving the requested information. On the other
hand, including a moral appeal in the information request has no effect
in this respect and the response rate to such requests was found to be
the same as in the control group (H2a). Hence, we confirm our first
hypothesis but we found no support for hypothesis two.
Until this point our findings are consistent with the existing literature
(cf. Cuillier, 2010; Fellner et al., 2013; Worthy et al., 2017).
However, our analysis improves the understanding of the effect of FOI
requests by including the moderation effect of population. A crucial
outcome is that differences between various forms of information requests
are related to the size of municipalities. With rising population,
the responsiveness for all versions of letters increases. This supports our
expectation that bigger municipalities, which hire more staff, are able
to deal with requests more effectively H7 Considerable differences were
found among the three versions of the letters when they were sent to
municipalities with smaller populations. Here, the differences are substantial
and significant. This leads to the conclusion that smaller localities
and their personnel are more likely to be intimidated by information
requests that are legally backed by FOI acts (H8). So, while
FOI requests lead to higher responsiveness in municipalities with
smaller populations, the form of request becomes less relevant with
rising population. This finding is essential, as it shows the public different
possibilities for asking for information from local authorities in
various local contexts.
Apart from this, we found that the probability of obtaining an answer
including at least some information is also affected by the ethnic
composition of municipalities and the political affiliation and gender of
mayors. In the case of Slovakia, the predominance of a minority from a
different language family decreased the success rate of information
requests potentially due to language barrier (H3). Such a finding mainly
applies for countries with ethnic compositions similar to that of Slovakia,
i.e. countries with relevant ethnic minorities concentrated in
specific areas. In regions which have a predominant minority population,
requests using the minority language may increase the response
rate of the authorities. Finally, we found that towns with independent
mayors responded more often (H6). This finding may be attributed to
the higher responsiveness of independents to react to non-partisan initiatives,
i.e. from voters, the media, the wider public etc. from below.
However, our results do not indicate any important differences between
Table 6
Probabilities of receiving a response with some information in different situations.
Type of letter Minority Mayor Gender Response probability Confidence interval
Control Hungarians < 75% Independent Male 0.22 0.19 0.26
Female 0.27 0.23 0.32
Other Male 0.18 0.16 0.21
Female 0.23 0.19 0.28
Hungarians > 75% Independent Male 0.17 0.12 0.24
Female 0.21 0.15 0.30
Other Male 0.14 0.10 0.20
Female 0.18 0.12 0.25
Moral appeal Hungarians < 75% Independent Male 0.23 0.19 0.26
Female 0.28 0.23 0.33
Other Male 0.19 0.16 0.22
Female 0.24 0.19 0.28
Hungarians > 75% Independent Male 0.17 0.12 0.24
Female 0.22 0.16 0.30
Other Male 0.14 0.10 0.20
Female 0.18 0.13 0.26
Legal Hungarians < 75% Independent Male 0.45 0.40 0.49
Female 0.52 0.46 0.58
Other Male 0.39 0.35 0.43
Female 0.46 0.41 0.52
Hungarians > 75% Independent Male 0.37 0.28 0.47
Female 0.44 0.34 0.54
Other Male 0.32 0.24 0.41
Female 0.38 0.29 0.49
Note: Population is set on 700.
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
415
mayors from parties holding different ideological positions. Finally,
contrary to previous research, we found some support for the idea that
municipalities with female mayors are more compliant with FOI laws.
This research represents an important contribution to the theoretical
debate about the effectiveness of FOI requests and their impact on
public accountability. Our field experiment shows that, in a country
which has adopted an FOI act, citizens should not consider this as an
automatic guarantee that public bodies operate in such a transparent
regime. On the contrary, when seeking information, citizens should be
encouraged to reference the FOI act in their information requests, at
least when local authorities are in play. What is more, inclusion of
moral appeals seems to be irrelevant, as they bring no increase in the
success rate of information requests and in fact even prolong the response
rate. These suggestions are vital, especially in localities with
smaller populations, as here the form of information request affects the
responsiveness to a greater extent as compared to highly populated
centers.
Declaration of interest
No author has any financial and personal relationships with other
people or organizations that could inappropriately influence our work.
Acknowledgement
This work was supported by the Czech Science Foundation (GA18-
15700S) and by Slovak Research and Development Agency (APVV-14-
0527).
Appendix A. Appendix
A.1. Baseline variant
Dear Madam/Sir,
We are a research team from […] and we are focusing on local
elections in Slovakia in 2010 and 2014. We however encountered a
problem, as both the Statistical Office of Slovak Republic and the
Ministry of Interior of Slovak Republic do not have data about results of
candidates in these elections. According to these two offices, only
municipalities possess such information.
In this regard, we would like to ask you to send us lists of candidates
(both elected and not elected) that competed in your town in local
elections and the number of votes they received. We need this data for
the two most recent local elections, held in 2010 and 2014. In case local
elections are held in more constituencies in your town, we kindly ask
you to send us data for all these constituencies. We ask these data to be
delivered via electronic mail.
Best regards,
A.2. Moral appeal variant
Dear Madam/Sir,
We are a research team from […] and we are focusing on local
elections in Slovakia in 2010 and 2014. We however encountered a
problem, as both the Statistical Office of Slovak Republic and the
Ministry of Interior of Slovak Republic do not have data about results of
candidates in these elections. According to these two offices, only
municipalities possess such information.
In this regard, we would like to ask you to send us lists of candidates
(both elected and not elected) that competed in your town in local
elections and the number of votes they received. We need this data for
the two most recent local elections, held in 2010 and 2014. In case local
elections are held in more constituencies in your town, we kindly ask
you to send us data for all these constituencies. We ask these data to be
delivered via electronic mail.
Your cooperation is very important, as our research will lead to
greater understanding of functioning of local government by the society.
Our work also aims to increase the quality of governance on the
local level.
Best regards,
A.3. Legal variant
Dear Madam/Sir,
We are a research team from […] and we are focusing on local
elections in Slovakia in 2010 and 2014. We however encountered a
problem, as both the Statistical Office of Slovak Republic and the
Ministry of Interior of Slovak Republic do not have data about results of
candidates in these elections. According to these two offices, only
municipalities possess such information.
In this regard, we would like to ask you to send us lists of candidates
(both elected and not elected) that competed in your town in local
elections and the number of votes they received. We need this data for
the two most recent local elections, held in 2010 and 2014. In case local
elections are held in more constituencies in your town, we kindly ask
you to send us data for all these constituencies. We ask these data to be
delivered via electronic mail.
We submit our request based on Law No. 211/2000C. of free access
to information that in its §2 defines municipalities as mandatory actors
and at the same time it obliges them to provide information they pos-
sess.
Best regards,
References
Araujo, J. F. F. E., & Tejedo-Romero, F. (2016). Local government transparency index:
Determinants of municipalities' rankings. International Journal of Public Sector
Management, 29(4), 327–347.
Bauhr, M., & Grimes, M. (2014). Indignation or resignation: The implications of transparency
for societal accountability. Governance, 27(2), 291–320.
Bearfield, D. A., & Bowman, A. O. M. (2017). Can you find it on the web? An assessment
of municipal e-government transparency. The American Review of Public
Administration, 47(2), 172–188.
Berliner, D. (2017). Sunlight or window dressing? Local government compliance with
South Africa's promotion of access to information act. Governance, 30(4), 641–661.
Bitušíková, A. (2005). (In)visible women in political life in Slovakia. Sociologický Časopis/
Czech Sociological Review, 41(6), 1005–1021.
Blais, A., & Dobrzynska, A. (1998). Turnout in electoral democracies. European Journal of
Political Research, 33(2), 239–261.
Blumenthal, M., Christian, C., Slemrod, J., & Smith, M. G. (2001). Do normative appeals
affect tax compliance? Evidence from a controlled experiment in Minnesota. National
Tax Journal, 54(1), 125–138.
Boix, C. (1999). Setting the rules of the game: The choice of electoral systems in advanced
democracies. American Political Science Review, 93(3), 609–624.
Broockman, D. E. (2013). Black politicians are more intrinsically motivated to advance
Blacks' interests: A field experiment manipulating political incentives. American
Journal of Political Science, 57(3), 521–536.
Buhaug, H., & Lujala, P. (2005). Accounting for scale: Measuring geography in quantitative
studies of civil war. Political Geography, 24(4), 399–418.
Cherry, M., & McMenemy, D. (2013). Freedom of information and ‘vexatious’
requests—The case of Scottish local government. Government Information Quarterly,
30(3), 257–266.
Clavero, S., & Galligan, Y. (2005). ‘A job in politics is not for women’: Analysing barriers
to Women's political representation in CEE. Sociologický Časopis/Czech Sociological
Review, 41(6), 979–1004.
da Cruz, N. F., Tavares, A. F., Marques, R. C., Jorge, S., & de Sousa, L. (2016). Measuring
local government transparency. Public Management Review, 18(6), 866–893.
Cucciniello, M., Porumbescu, G. A., & Grimmelikhuijsen, S. (2017). 25 years of transparency
research: Evidence and future directions. Public Administration Review, 77(1),
32–44.
Cuillier, D. (2010). Honey v. Vinegar: Testing compliance-gaining theories in the context
of freedom of information laws. Communication Law and Policy, 15(3), 203–229.
Darch, C., & Underwood, P. G. (2005). Freedom of information legislation, state compliance
and the discourse of knowledge: The south African experience. The
International Information & Library Review, 37(2), 77–86.
Esteller-Moré, A., & Otero, J. P. (2012). Fiscal transparency. (why) does your local government
respond? Public Management Review, 14(8), 1153–1173.
Fellner, G., Sausgruber, R., & Traxler, C. (2013). Testing enforcement strategies in the
field: Threat, moral appeal and social information. Journal of the European Economic
Association, 11(3), 634–660.
Fenster, M. (2015). Transparency in search of a theory. European Journal of Social Theory,
18(2), 150–167.
Feskens, R. C. (2009). Difficult groups in survey research and the development of tailor-made
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
416
approach strategies. Centraal Bureau voor de Statistiek; Utrecht University.
de Fine Licht, J. (2011). Do we really want to know? The potentially negative effect of
transparency in decision making on perceived legitimacy. Scandinavian Political
Studies, 34(3), 183–201.
de Fine Licht, J. (2014). Policy area as a potential moderator of transparency effects: An
experiment. Public Administration Review, 74(3), 361–371.
Fiorio, C. V., Iacus, S. M., & Santoro, A. (2013). Taxpaying response of small firms to an
increased probability of audit: some evidence from Italy. Biccoca Department of
Economics, Management and statistics; University of Milan.
Fox, R., & Schuhmann, R. (1999). Gender and local government: A comparison of women
and men city managers. Public Administration Review, 59(3), 231–242.
Freedominfo (2017). freedominfo.org: Chronological and Alphabetical lists of countries
with FOI regimes. www.freedominfo.org/?p=18223.
Gecíková, I., & Papcunová, V. (2013). The structure of personal facilities local government
offices in the Slovak Republic. International Colloquium on Regional Sciences (pp.
451–458). Brno: Masaryk University.
Gendźwiłł, A. (2012). Independent mayors and local lists in large polish cities: Towards a
non-partisan model of local government? Local Government Studies, 38(4), 501–518.
Gowa, J., & Mansfield, E. D. (1993). Power politics and international trade. American
Political Science Review, 87(2), 408–420.
Grimmelikhuijsen, S. G., & Meijer, A. J. (2012). Effects of transparency on the perceived
trustworthiness of a government organization: Evidence from an online experiment.
Journal of Public Administration Research and Theory, 24(1), 137–157.
Harzing, A. W. (1997). Response rates in international mail surveys: Results of a 22country
study. International Business Review, 6(6), 641–665.
Hazell, R., & Worthy, B. (2010). Assessing the performance of freedom of information.
Government Information Quarterly, 27(4), 352–359.
Heald, D. (2006). Varieties of transparency. In C. Hood, & D. Heald (Eds.). Transparency:
The key to better governance (pp. 25–43). Oxford: Oxford University Press.
Hegre, H., & Sambanis, N. (2006). Sensitivity analysis of empirical results on civil war
onset. Journal of Conflict Resolution, 50(4), 508–535.
Hood, C. (2006). Transparency in historical perspective. In C. Hood, & D. Heald (Eds.).
Transparency: The key to better governance? (pp. 3–23). Oxford: Oxford University
Press.
James, O., & Moseley, A. (2014). Does performance information about public services
affect Citizens' perceptions, satisfaction, and voice behaviour? Field experiments with
absolute and relative performance information. Public Administration, 92(2),
493–511.
Jorge, S., Sá, P. M., Pattaro, A. F., & Lourenço, R. P. (2011). Local government financial
transparency in Portugal and Italy: A comparative exploratory study on its determinants.
13th biennial CIGAR conference, Bridging public sector and non-profit sector ac-
counting.
LaFont, S. (2001). One step forward, two steps back: Women in the post-communist
states. Communist and Post-Communist Studies, 34(2), 203–220.
Lagunes, Paul, & Pocasangre, Oscar (2017). Dynamic transparency: An audit of Mexico's
freedom of information act. Inter-American development bank.
Lewis, D. E., & Wood, A. K. (2012). The paradox of agency responsiveness: A federal FOIA
experiment. Working Paper 06–2012. Nashville: Center for the Study of Democratic
Institutions.
Madison, T. P., Rold, M., & Honeycutt, J. (2014). How partisans differ from independents:
The imaginative functions of self-understanding, rehearsal, and relationship maintenance.
Imagination, Cognition and Personality, 34(2), 105–116.
Meijer, A. (2014). Transparency. In M. Bovens, R. E. Goodin, & T. Schillemans (Eds.). The
Oxford handbook of public accountability (pp. 507–524). Oxford: Oxford University
Press.
Meijer, A., Hart, P., & Worthy, B. (2015). Assessing government transparency: An interpretive
framework. Administration & Society, 50(4), 501–526.
Michener, G. (2015). Assessing freedom of information in Latin America a decade later:
Illuminating a transparency causal mechanism. Latin American Politics and Society,
57(3), 77–99.
Michener, G., Moncau, L. F., & Velasco, R. B. (2016). The Brazilian state and transparency:
Evaluating compliance with freedom of information. Rio de Janeiro: FGV.
Michener, G., & Worthy, B. (2015). The information-gathering matrix: A framework for
conceptualizing the use of freedom of information laws. Administration & Society,
50(4), 476–500. Advance online publication https://doi.org/10.1177/
0095399715590825.
Morton, R. B., & Williams, K. C. (2010). Experimental political science and the study of
causality: From nature to the lab. Cambridge: Cambridge University Press.
Nagatomi, K. (2013). Independent success in mayoral elections in England: A study of factors
contributing to the Candidates' success. New York: Edwin Mellen Press.
Nam, T. (2012). Freedom of information legislation and its impact on press freedom: A
cross-national study. Government Information Quarterly, 29(4), 521–531.
Olsen, J. P. (2015). Democratic order, autonomy, and accountability. Governance, 28(4),
425–440.
OSJI (2006). Transparency & silence: A survey of access to information laws and practices in
14 countries. Open Society Institute.
Peisakhin, L. (2012). Transparency and corruption: Evidence from India. The Journal of
Law and Economics, 55(1), 129–149.
Peisakhin, L., & Pinto, P. (2010). Is transparency an effective anti-corruption strategy?
Evidence from a field experiment in India. Regulation & Governance, 4(3), 261–280.
Peters, B. G. (2014). Accountability in public administration. In M. Bovens, R. E. Goodin,
& T. Schillemans (Eds.). The Oxford handbook of public accountability (pp. 211–225).
Oxford: Oxford University Press.
Pilet, J. B., Reimink, E., & Schrobiltgen, M. H. (2015). Vote-seeking vs. intrinsic representation
in multimember districts. Results of a field experiment held in Brussels. ECPR
General Conference 2015, Montreal.
Pina, V., Torres, L., & Royo, S. (2007). Are ICTs improving transparency and accountability
in the EU regional and local governments? An empirical study. Public
Administration, 85(2), 449–472.
Piotrowski, S. J., & Van Ryzin, G. G. (2007). Citizen attitudes toward transparency in local
government. The American Review of Public Administration, 37(3), 306–323.
Raudeliūnienė, J., & Meidutė-Kavaliauskienė, I. (2014). Analysis of factors motivating
human resources in public sector. Procedia-Social and Behavioral Sciences, 110,
719–726.
Reiser, M., & Holtmann, E. (Eds.). (2008). Farewell to the party model?: Independent local
lists in east and west European countries. Wiesbaden: VS Verlag für Sozialwissenschaft.
Reitan, M., Gustafsson, K., & Blekesaune, A. (2015). Do local government reforms result
in higher levels of Trust in Local Politicians? Local Government Studies, 41(1),
156–179.
Remmer, K. L. (2004). Does foreign aid promote the expansion of government? American
Journal of Political Science, 48(1), 77–92.
Right to Information Assessment and Analysis Group and Centre for Equity Studies (Raag/
CES) (2014). Peoples Monitoring of the RTI Regime in India 2011–13. New Delhi:
NCPRI.
Right to Information Assessment and Analysis Group and National Campaign for People's
Right to Information (Raag/NCPRI) (2009). Safeguarding the right to information –
Report of the People's RTI assessment 2008. New Delhi: NCPRI.
Roberts, A. (1998). Limited access: Assessing the health of Canada's freedom of information
laws. Working paperSchool of Policy Studies, Queen's University.
Ross, J., & Whittaker, P. (2009). Freedom of information: Is openness too expensive, too
difficult or too dangerous? Journal of Finance and Management in Public Services, 7(1),
53–70.
Sheldon, H., Graham, C., Pothecary, N., & Rasul, F. (2007). Increasing response rates
amongst black and minority ethnic and seldom heard groups. Oxford: Picker Institute
Europe.
Sičáková-Beblavá, E., Kollárik, M., & Sloboda, M. (2016). Exploring the determinants of
transparency of Slovak municipalities. NISPAcee Journal of Public Administration and
Policy, 9(2), 121–145.
Snell, R. (2002). FoI and the delivery of diminishing returns, or how spin-doctors and
journalists have mistreated a volatile reform. An Australian Review of Public Affairs,
2(3), 187–207.
Sol, D. A. (2013). The institutional, economic and social determinants of local government
transparency. Journal of Economic Policy Reform, 16(1), 90–107.
Statistical Office of Slovak Republic (2011). Census 2011. Bratislava: Statistical Office of
Slovak Republic.
Styles, A. K., & Tennyson, M. (2007). The accessibility of financial reporting of US municipalities
on the internet. Journal of Public Budgeting, Accounting & Financial
Management, 19(1), 56.
Torgler, B. (2004). Moral suasion: An alternative tax policy strategy? Evidence from a
controlled field experiment in Switzerland. Economics of Governance, 5(3), 235–253.
Walby, K., & Larsen, M. (2012). Access to information and freedom of information requests:
Neglected means of data production in the social sciences. Qualitative Inquiry,
18(1), 31–42.
Welch, E. W., Hinnant, C. C., & Moon, M. J. (2005). Linking citizen satisfaction with Egovernment
and Trust in Government. Journal of Public Administration Research and
Theory, 15(3), 371–391.
Worthy, B. (2010). More open but not more trusted? The effect of the freedom of information
act 2000 on the United Kingdom central government. Governance, 23(4),
561–582.
Worthy, B. (2017). The politics of freedom of information. How and why governments pass
laws that threaten their power. Manchester: Manchester University press.
Worthy, B., John, P., & Vannoni, M. (2017). Transparency at the parish pump: A field
experiment to measure the effectiveness of freedom of information requests in
England. Journal of Public Administration Research and Theory, 27(3), 485–500.
Peter Spáč is assistant professor at Department of Political Science at Masaryk University.
His research interest includes electoral systems, political parties and Slovak politics.
Petr Voda is a postdoctoral associate at the International Institute of Political Science,
Masaryk University. His research focuses on local government, electoral behavior and
electoral geography.
Jozef Zagrapan is a postdoctoral associate at the Institute for Sociology of the Slovak
Academy of Sciences, with current focus mainly on Slovakia in cross-national comparative
research.
P. Spáč et al. Government Information Quarterly 35 (2018) 408–417
417