118 The Romanian Journal of Society and Politics

Toma BUREAN and Raluca POPP

THE IDEOLOGICAL MAPPING OF

POLITICAL PARTIES IN ROMANIA

ABSTRACT
What are the main dimensions of competition between political parties
in Romania? Is there a match of the electorate policy preferences with those
of political parties? This paper analyzes what are the main dimensions of
competition between parties in Romania by looking at the issue preferences of the
party in the electorate. We use data obtained from the voting advice application
(VAA) applied in 2012 (votulmeu.com) in Romania. Finally we compare our
results with Chapel Hill expert party mappings and claim that VAA data are
a reliable source for finding out the programmatic peferences of party supporters.

Toma BUREAN
Babes-Bolyai University
burean@fspac.ro
Raluca POPP
University of Exeter
R.Popp@exeter.ac.uk

KEYWORDS

VAA
party mappings
Romania
online survey
left-right

Introduction
The analysis of the ideology of political parties and voters are important topics
for political scientists (Germann et al. 2012). The ideological space occupied by political
parties is analyzed by researchers with the help of candidates, politicians (Kitschelt et
al. 1999, Hix and Crombez 2005) or experts (Benoit and Laver 2006). Others use the
programs of political parties (Budge et al. 2001). Party mappings on the left and right
dimensions improve the chances for citizens to cast a correct vote. They provide a
heuristic device that assist citizens in expressing preferences on issues (Fuchs and
Klingemann 1989, Inglehart and Klingemann 1976). There is little research (Germann
et al. 2012, Marian and King 2014) that surveys the ideological preferences of the party
in the electorate (Key 1964) or that compares the ideological preferences of citizens
to the ideological positions of political parties (Huber and Powell 1994). This article
explores the potential of a new method of gathering data that can be used to explore
the political preferences of citizens. VAA (Voting Advice Applications) data allows
the comparison of the ideological positions of citizens with the positions of political
parties. It is also a tool that helps voters to make informed choices (Alvarez et al. 2014,
Ruusuvirta and Rosema 2009, Wall et al. 2014). VAAs have become popular in many
European countries (Garzia and Marschall 2012). The application asks users to express
preferences on 30 issues that are considered important for the upcoming elections. After

Toma BUREAN and Raluca POPP 119

completing the survey, the online tool ranks the political parties running in elections
according to their compatibility with the users preferences. The advantages of VAAs
for party mapping are that it is a cost efficient tool of data gathering, it secures more
anonimity to respondents than traditional or telephone surveys and includes a higher
number of issues than regular surveys (Garzia et al. 2014). The main disadvantage is
that users are self selected. This article explores the usefulness of VAA in Romania. It
claims that despite the problems that arise from self selection, the application generates
party mappings that are comparable with other methods of assessing the ideological
space of political parties.
The paper is organized as follows. First, we briefly account for the studies that use
VAA for party mapping and political behavior. Secondly, we present the approaches
used in party mappings in general with a focus on studies that used VAAs, followed
by a discussion on party mappings in Romania. Next we describe the Romanian VAA
and the methodology, followed by a presentation of the results and a discussion on the
implications of the findings. This paper will use the VAA (Voting Advice Application)
data collected from the application votulmeu.com to build party maps based on the
preferences of citizens and political parties.
VAA Studies
The rationale of a voting advice application is to provide an online tool that
helps citizens in making an informed vote choice at elections. The application matches
the users preferences to the stances of political parties. VAAs is now widely used in
Europe (Germann et al. 2012). The popularity of this application is motivated by the
increasing number of floating voters that need guidance during the electoral campaign
(Mair 2008). Researchers use these data to test hypotheses regarding the quality of
advice it gives (Alvarez et al. 2014) and the biases it encompasses (Gemenis 2013), as a
determinant of vote choice (Andreadis et al. 2013) or as a measurement of congruence
between citizens and candidates or political parties (Fivaz et al. 2014).
The number of citizens that respond to voting advice applications varies
substantially. For example in the Netherlands in the 2006 elections there were 4,6
million users, in Germany in 2009 there were 6,7 million (12% of the electorate),
Belgium (13% of the electorate), and Switzerland (more than 20% of the electorate)
had 1 million users each (Garzia 2010, Walgrave et al. 2008). In the Czech Republic
at the first direct presidential elections organized in 2013, over 1 million users used
KohoVolit.eu the Czech VAA. With the exception of the Czech Republic, in several
other Eastern European countries VAAs are not very popular. In the 2009 European
Elections in Poland there were 31389 users, in Lithuania 1836 and in Latvia 974. In
Romania in 2009 there were 1678 respondents VAA, Bulgaria 6319 and Hungary
6622 (Trechsel 2010). The Romanian 2012 VAA used in this paper had 16107 users,
representing 0, 08% of the electorate. Compared to the Romanian European elections
VAA from 2009 (1678 users) (Trechsel 2010) the number of respondents increased
substantially.
The research that uses VAAs targets its design, validity and its effects on vote
choice and political representation (Marschall and Garzia 2014, Triga et al. 2012).
Studies that inquire about the quality of the match given by VAAs (Lefevere and

120 The Romanian Journal of Society and Politics

Walgrave 2012) warn about the potential of the application to give the wrong advices
because of the way the statements are phrased. Researchers focus on the effect of
statement selection (Baka et al. 2012, Gemenis 2013) and they issue warnings about
the distortions and the manipulative potential of the application (Nuytemans et al.
2010). Secondly there are studies that measure the impact of VAAs on the party
choice (Marschall and Garzia 2014, Triga et al. 2012). VAAs can be considered as an
additional tool of the electoral campaign that helps voters to make informed choices.
That is why it could be considered as an instrument of political influence (arkolu
et al. 2012, Wall et al. 2014). There is an abundant literature on the effects of political
campaigning on political preferences. Hansen (2011) identifies six stories explaining
the impact of electoral campaigns. First citizens can learn from campaigns what are
their duties and choices (civic learning model) (Freedman et al. 2004 and Moore, 1987),
secondly citizens have predispositions that are simply reinforced through campaigns
(minimal effects model) (Berelson, Lazarsfeld and McPhee, 1954, Schmitt-Beck 2007:
753), thirdly political campaigns can set the issues according to which citizens judge
candidates (priming studies) (Iyengar and Simon, 1993) and influence party choice,
fourthly citizens acquire and digest information obtained from political campaigns and
make decisions according to the type of information they remember (memory based
models) (Althaus 2003, Zaller, 1992). The fifth model sees voters as continuously
ingesting information (Lodge et al. 1995) from political discourses and adjusting or
reinforcing their preferences. And finally, we have the model of heuristics (Iyengar,
1990, Lau and Redlawsk, 2001, Sniderman et al. 1991) that considers voters as investing
minimal efforts to obtain good quality or enough information to make a choice between
candidates. Studies that use VAA embraced the models of campaign influence on vote
choice (Fivaz and Nadig 2012, Marschall and Schultze 2012, Pianzola and Ladner
2011, Ruusuvirta and Rosema 2009). Some compared vote switching of VAA users
to non-users. The results showed that VAA users are more likely to resort to vote
switching (Andreadis, Pianzola et al. 2013) than non-VAA users. This finding confirms
the civic learning model in which the VAA is viewed as a political learning device. Wall
et al. (2011) used an experimental design to present the Dutch VAA as more of a tool
for reinforcing pre-existing preferences of citizens. They found confirmation of the
minimal effects model of VAA influence on vote in electoral campaigns.
Some accentuate the use of VAA as an alternative for measuring congruence
(Alvarez et al. 2014) or mapping the preferences of the party in the electorate (Garzia
and Marschall 2012, Gemenis 2013). The mapping of political parties could replace
expert surveys or party manifesto party mappings (Mendez and Wheatley 2014). VAAs
are easy to organize, the number of respondents is high and in many countries the data
obtained is representative (Mendez and Wheatley 2014). Research that includes VAA
party mappings focus on accuracy and the match with party mappings from manifesto
data or surveys (Gemenis 2014, Mendez and Wheatley 2014, Wheatley et al. 2014). The
results indicated that VAAs can be used especially in countries in which the number
of users is high and approximating the populations socio demographic characteristics
(Gemenis 2013, Mendez and Wheatley 2014). Although VAAs may accurately measure
party positions, these cannot take into account valence issues and can be manipulated
by candidates who might choose to have positions close to as many voters as possible
in order to get a higher score (Wagner and Ruusuvirta, 2010). VAA respondents

Toma BUREAN and Raluca POPP 121

are unlike the national population. They are well educated, young and interested in
politics. Where possible, the problem of representativeness was partially solved by
giving weight to under-represented groups (Wheatley et al. 2014).
Wheatley et al. (2014) mapped political parties in Scotland with results from an
experiment in which they compared the party supporters preferences on VAA with
a face to face survey. They discovered remarkable similarities between the political
preferences of online users and the surveyed population. Germann et al. (2012) build
party mappings with the Smartvote VAA application for the Swiss parliamentary
elections of 2007, and found comparable results with mappings from survey data.
Our paper belongs to this stream of studies. We explore and identify the patterns
yielding from the political preferences expressed by users of the Votulmeu (My Vote)
in Romania and compare our results with expert surveys.
Ideological space and party mappings
Researchers map parties most often according to expert surveys, interviews
with representatives or party activists (Kitschelt et al. 1999, Grecu 2003, Markowski
and Tucker 2010) and party manifestoes (Benoit and Laver 2006, Budge et al. 2001,
Klingemann et al. 2006, Laver 2001). The VAA presents certain advantages compared
to them: they are cost effective, the number of respondents is high and they offer a
platform that allows the comparison of preferences of citizens with parties. On the
other hand respondents are self-selected and candidates can manipulate their position
so that the VAA will advise users to vote for the candidates own party (Germann et
al. 2012).
For Romania, and for most of post-communist new democracies in Eastern
Europe, Herbert Kitschelt (1992) identifies two issue dimensions on which most
parties compete. One source of competition between political parties is viewed as
originating from the allocation of material resources in society. On the one hand there
are political parties that support the unstructured allocation while other parties will
support a redistributive system. Secondly the competition among parties stems from
the way rights should be distributed in society. One camp supports a universal, equal
opportunities view of how these rights should be allocated. The other camp agrees
that certain groups should enjoy more rights than others. Others topics identified
by Kitschelt (1992) were regime divide or the communist/anticommunist divide,
national-cosmopolitan and ethnic divide. Evans and Whitefield (1993) and Whitefield
(2002) confirm Kitschelts findings. Whitefield (2002) claims that despite the diversity
of cleavages, the dimensions of competition in Eastern Europe turned out to be not
very different from the Western European political systems. The importance of ethnic
diversity (attitudes toward Hungarians or Russians) remains one specific dimension
of competition in some countries from Eastern Europe (Latvia, Estonia, Slovakia, or
Romania). Kitschelt (1992) uses mass surveys and expert placement of parties when
building the party mappings for The Czech Republic, Slovakia, Poland, Hungary,
Bulgaria and Romania. For Romania he maps parties on the distributive versus
spontaneous market allocation axis and the authoritarian versus libertarian axis. Grecu
(2003), who uses expert survey data from 2002 presents similar findings to discover,
in Romania, an economic dimension and a pro versus anti attitude towards minorities.

122 The Romanian Journal of Society and Politics

Later studies confirm the existence of two dimensions of competition among

parties in Romania (Benoit and Laver 2006). Fesnic (2008) builds party mappings
using preferences of the Romanian electorate from 1993 until 2006 and identifies a
dimension labelled left-right referring to what Hooghe et al. (2010) called economic
left right. The second dimension identified is labeled by Fesnic New Politics referring
to the authoritarian/libertarian preferences.
European integration is an issue that gains relevance in Eastern Europe starting
with the negotiations to join the European Union. The Chapel Hill expert surveys of
2006 and 2010 focus on this topic, revealing parties as having differentiated positions
in Eastern Europe (Hooghe et al. 2010 and Bakker et al. 2012). In Poland the choice
for a Eurosceptic party in 2002 predicted an anti EU attitude in the referendum,
overshadowing socio-demographic determinants. The result is surprising for a political
system characterized by low party loyalties (Markowski and Tucker, 2010). Considering
these findings we hypothesize that with the help of the 2012 Romanian VAA data we will identify
dimensions of competition similar to other party mappings for Romania. The economic left right
dimension, GAL-TAN and attitudes toward EU are the issues on which Romanian parties would
be differentiated.
Methodology
Votul meu is the VAA that went online one month before the Romanian
parliamentary elections that took place on December 9, 2012. The users had to express
their preferences on thirty statements that were considered to be the most important
for that particular electoral campaign. The possible answers were: Completely Agree,
Agree, Neither Agree nor Disagree, Disagree and Completely Disagree.
One extra option presented to respondents was Dont Know. The positions of the
political parties in Romania were coded on the same 30 statements by country experts.
The application matched the responses of each applicant with the score obtained by
the party through the expert coding. Pre-electoral surveys predicted that four relevant
parties and coalitions would enter the parliament. These were The Social Democratic
Party (PSD-Partidul Social Democrat) and The National Liberal Party (PNL- Partidul
Naional Liberal) united in a coalition called The Social Liberal Union. They were
joined by two smaller parliamentary parties such as the Conservative Party (PC-Partidul
Conservator) and The National Union for the Progress of Romania (UNPR-Uniunea
Nationala pentru Progresul Romniei). Both parties had less than 2% of electoral
support yet they had parliamentary representation in the 2008-2012 legislature and
gained seats in the 2012 parliamentary elections. The Democratic Liberal Party (PDL
- Partidul Democrat Liberal) was also part of an electoral coalition named The Right
Romania Alliance (ARD- Alianta Romania Dreapta) that included the Civic Force
(FC - Forta Civica) and the National Peasant Party whos few representatives gained
parliamentary representation. The Peoples Party Dan Diaconescu (PP-DD- Partidul
Poporului Dan Diaconescu) surprisingly won 47 seats (Romanian Electoral Office
2012) in the lower chamber of representatives with a populist electoral message
(Gherghina and Miscoiu 2014). Finally. the Democratic Alliance of Hungarians in
Romania (UDMR- Uniunea Democrat Maghiar din Romnia) gained 18 seats in the
lower chamber of representatives (Romanian Electoral Office 2012).

Toma BUREAN and Raluca POPP 123

In addition to the preferences on policies users were asked questions on party

identification, voting behavior and left and right. Votul Meu contained 18484
respondents. Not all of them completed all the questions in the survey or had valid
answers. After applying various filtering procedure 16107 respondents answers were
kept.

Figure 1. Positions of political parties according to expert surveys used in the VAA Votulmeu at the 2012
parliamentary elections. These positions are matched with the users preferences.

Gemenis (2013) identifies a deductive and inductive method of finding issue

dimensions. A deductive method refers to the selection of issues to be introduced
in the factor analysis that match pre-determined dimensions, while the latter method
is employed to find dimensionality in all issues on which users express preferences.
We find the inductive method more convincing as our purpose is to explore the
preferences of the VAA online users and find dimensionality in all their responses. For
the inductive method the identification of dimensions is done by factor analysis with
Varimax rotation. To identify the groupings of statements that qualified for theoretical
constructs, we selected factors with Cronbach alpha higher than 0.7. As a further test
we made use of Mokken scaling to test the validity of our constructs (Wheatley et al.
2014).
In order to identify the positions of parties on the ideological maps we have
selected the respondents that felt close to a party and they intend to vote for that party.
Thus we followed the selection method used by Wheatley et al. (2014) and Germann
et al. (2012). These restrictions reduced the number of respondents to 6259. Next we
compared the Chapel Hill expert surveys of 2010 and 2006 to our party mappings.
These surveys contain identical questions with the 2012 Romanian VAA for the
GAL-TAN and economic dimensions although the coding method was different. We
recoded the VAA data to match the Chapel Hill coding.

124 The Romanian Journal of Society and Politics

Data analysis and results

In 2012, Romania registered the lowest Internet penetration in the European
Union. Therefore it is no surprise that the data generated by Votul Meu suffered to
a greater extent from representation concerns as other VAAs. 78% of the Romanian
VAA respondents have graduate studies, and 78% come from urban areas. A 2012
face to face survey showed that 13% of the population has graduate studies and half
of the population of Romania lives in rural areas. The population that was surveyed
differs in terms of political preferences as well. The majority prefers center right wing
parties. Moreover according to the VAA users the center right wing parties would win
the parliamentary election. Thus our data is not nationally representative. It is more
likely to be representing the urban and citizens with more years of education. On the
other hand Zaller (1992) and later Althaus (2003) showed that the better off category
of population might also be the most influential voice on representatives. Secondly
this group is ideal to survey because the chances to discover patterns of ideological
thinking are increased.
Before engaging into the analysis of preferences we follow Converses (1964)
warnings regarding the pursuit of patterns of ideological thinking. When discussing
about people having belief systems we have to take into account the number of nonresponses. Converse (1964) noted that the number of no opinion in surveys is quite
high leaving the expression of preferences to a few citizens. In the Romanian VAA the
number of non-responses on issues is low. It ranges from 0.6% (89) to 7.5% (1190).
Party supporters represent 38% (6177) of the respondents in our survey. The level
non response of party supporters ranged from 0.5% (31) to 4.6% (282). Next, in order
to map the preferences of the party in the electorate (Key, 1964) we used exploratory
factor analysis (Table 1).
Table 1. Factor analysis of 30 statements on important Romanian issues.
Nr.

1
2
3
4
5
6

Issue

Austerity measures/
European integration

GAL-TAN

Left-right

The state should intervene to lower the

price of basic foods.
The Romanian state should allow the
Canadian company Gabriel Resources to
continue operations at Rosia Montana.
The poorest citizens should pay less
income tax.
Pension funds should be nationalized to fill
shortages in the state pension fund.
External loans from institutions such as the
IMF are a good solution to crisis situations.

-.302

-.119

.684

,608

,123

-.179

-,214

-,011

.680

-,128

,039

.645

,662

,063

-.091

-,331

-,131

.437

Foreign private investors' access to natural

resources contracts in Romania should be
banned.

Toma BUREAN and Raluca POPP 125

7

It is acceptable for the government to cut

salaries in the public sector during the crisis
to keep the budget deficit at a low level.

,619

,043

-.297

Parliament should remain bicameral.

-,586

,033

.048

Romania should become a parliamentary

republic.
All proven collaborators of the Securitate
should be denied access to public
functions.

-,578

,112

.084

,158

-,050

-.085

Ethnically-based parties should be banned.

Romania should restore the monarchy.
The Bucharest authorities have too much
power to decide on matters in the rest of
the country.
Romania should cede more decisionmaking power to the EU.
Romania should improve its relationship
with Russia.
Romania should pursue reunification with
Moldova
International partners (such as the U.S.
or EU) have the right to interfere in the
internal affairs of Romania when they feel
there is a threat to democracy.
Homosexuals should have the right to
express themselves freely in public.

-,080
-,256

-,398
,122

-.094
-.069

,064

,244

.000

,615

,224

-.230

-,213

-,062

.070

,272

-,280

.147

,714

,154

-.082

,113

,161

-.073

The church has too much influence in

Romanian society.
The death penalty should be reintroduced
for heinous crimes.
The law should give priority to employees
against their employers.
Women should have the right to decide
whether to remain pregnant.
Between kindergarden and first grade,
children should go to a preparatory class
at school.
Small rural hospitals should be closed, and
the state should invest in larger regional
hospitals.
Hungarian politicians should cooperate
more with the authorities in Budapest
than those from Bucharest to meet the
objectives of the Hungarian minority in
Romania.

-,004

,042

-.032

,045

-,160

.218

-,113

-,054

.692

,058

-,058

.000

,576

,080

-.027

,594

,100

-.267

,121

,604

.074

10
11
12
13
14
15
16
17

18
19
20
21
22
23
24
25

126 The Romanian Journal of Society and Politics

26
27
28
29
30

All Hungarians should be obliged to learn

Romanian.
A territorial reform should include the
creation of an autonomous Hungarian
region.
Cultural autonomy for minorities would
constitute a threat to Romania.
Minorities should have the right to
education exclusively in the mother tongue,
including at university.
The Roma ethnic population of cities
should live isolated from the rest of the
community.

-,064

-,583

,262

,073

,784

-,012

-,158

-,466

,294

,137

,702

-.035

-,005

-,012

.062

The analysis yields eight dimensions that had an Eigen values larger than a unity.
We selected the issues with the loading coefficients higher than 0,6 and identified
three dimensions with Cronbach alpha 0,781, 0,616 and 0.712 showing robust internal
consistency.. The three dimensions are labeled in Table 1. In order to provide an
additional test of the robustness of our findings we use Mokken scaling. Mokken
scaling shows whether the identified dimensions have components that are scalable.
The indicator of scalability H shows that the scale has moderate strength (between .3
and .5 - see Table 2).
Table 2 Mokken scales with scalability coefficient (Hi) on issues that load as important on the factor
analysis
Item

Question

Scales
European
integration
(0.449)

1
2
3
4
5
7
14
17

The state should intervene to lower the price of basic

foods.
The Romanian state should allow the Canadian
company Gabriel Resources to continue operations at
Rosia Montana.
The poorest citizens should pay less income tax.
Pension funds should be nationalized to fill shortages in
the state pension fund.
External loans from institutions such as the IMF are a
good solution to crisis situations.
It is acceptable for the government to cut salaries in the
public sector during the crisis to keep the budget deficit
at a low level.
Romania should cede more decision-making power to
the EU.
International partners (such as the U.S. or EU) have the
right to interfere in the internal affairs of Romania when
they feel there is a threat to democracy.

LeftGAL-TAN Right
(0.403)
(0.423)
0.458

0.396
0.418
0.416
0.438
0.437
0.476
0.495

Toma BUREAN and Raluca POPP 127

21
25
27
29

The law should give priority to employees against their

employers.
Hungarian politicians should cooperate more with
the authorities in Budapest than those from Bucharest
to meet the objectives of the Hungarian minority in
Romania.
A territorial reform should include the creation of an
autonomous Hungarian region.
Minorities should have the right to education
exclusively in the mother tongue, including at
university.

0.394
0.313
0.464
0.426

The first dimension loaded on a mixture of economic policies and preferences

on the influence of IMF and EU. It includes environmental concerns, public fund
cuts, surrendering decision-making to EU and accepting IMF and EU as protectors of
democracy. This is a dimension that reflects the main electoral campaign topics. On
the 6th of July 2012, president Basescu was suspended by the Parliament. A referendum
was organized in order to support this decision. The referendum failed to reach the
50% turnout required for the validation. The EU, IMF and Venice Commission
issued several warnings to Prime Minister Victor Ponta and interim president Crin
Antonescu not to influence the results of the referendum and respect the decisions
of the Constitutional Courts. The other source for this dimension is the role IMF had
on the austerity measures imposed by the government of Prime Minister Emil Boc.
The most radical measure was the 25% wage cuts of each public sector employee. We
would call this the European integration and austerity measures dimension. We suspect
that it is a transient topic that is connected to the context of the 2012 referendum for
the dismissal of the president, the economic crisis and the tense relationship with IMF
and EU.
The second dimension covers preferences on issues referring to the Hungarian
community, the cultural autonomy for all minorities and education in mother tongue.
This issue is similar to the libertarian-cosmopolitan politics versus authoritarian
particularist politics dimension identified by Kitschelts(1992) party mappings.
Following Marks et al. (2006) and Wheatley et al. (2014) we call this dimension GreenAlternative Libertarian / Traditional-Authoritarian Nationalist GAL/TAN.
The third dimension includes preferences that refer to issues similar to the
economic left-right dimension. States intervention to lower the price of basic foods,
protection of the public pensions system and states protection of employees are topics
that divide citizens on the economic left-right scale. We consider that the last two
dimensions are more stable and explain party positions and shifts over time. Next, we
mapped parties by calculating the average score a group of party supporters receives
on the three dimensions and mapped the scores on pairs of dimensions.

128 The Romanian Journal of Society and Politics

Figure 2. Party mappings on three dimensions: European integration, GAL-TAN, economic left-right (Positions
represent the average position of party supporters recoded so that variation will range from 0 to 1.)
European integration -0 means opposition to European integration
GAL-TAN GAL means green, alternative, liberal and TAN means traditional, authoritarian, nationalist
Left-Right left is a preference for state intervention in the market and support for social services and right
symbolizes a minimal state and privatization of social services

The results show that the four groups of supporters of parliamentary parties
have distinct positions on the identified dimensions. Relative to other parties, the PSD
supporters have center-left preferences on the economic left-right dimension. The
party that is the most to the left is the populist PP-DD. PNL is considered a liberal

Toma BUREAN and Raluca POPP 129

pro-market party (Fesnic, 2008). The party mappings show, however, that supporters
are left wing, against the influence of IMF and EU and foreign investors. On leftright axis, this partys position is center-left confirming the partys ideological shift.
The liberal party joined a coalition with the social-democrats in 2011 and formed the
government together with PSD in 2012. Previous party mappings placed this party as
the most right wing as a pro-market and libertarian party (Grecu, 2003, Fesnic, 2008).
PDL is the party that is a former splinter from PDSR (PDSR - The Party of
Social Democracy from Romania). The party was in government from 2008 until May
2012 when the Ungureanu government received a vote of no confidence. PDL had
to implement harsh economic reforms following EU and IMF recommendations.
These included a 25% cut in public sector wages and the cancellation of all monetary
premiums. Its government initiated several controversial reforms in education, health
and public administration. The party mappings show this party as center-right on the
economic left-right axis and the strongest supporter of IMF and EU intervention in
the economy or countrys decision making process. This is not a surprise. Earlier in
2012, the dismissal of the president, supported by PDL, raised serious concerns about
the democratic consolidation and the separation of powers in the country. EU officials,
the Venice Commission and the US Ambassador criticized the USL government for
pressuring the Constitutional Court to issue a decision through which the dismissal
of the president would be permanent. These reactions generated criticisms from USL
leaders who accused The European Union and The United States of interfering with
Romanian national decision making. This was an unprecedented type of discourse
coming from a country in which citizens and parties showed solid support for the EU
and the United States. PDL encouraged its supporters not to turn out at the referendum.
UDMR supporters are centrist on European integration and the economic left-right
axes. The party is set apart from the other parties on GAL-TAN. UDMR is the party
that represents the ethnic Hungarians. Its main goal is to ensure the representation
of its minority thus it is the most open to rights accorded to the Hungarian minority.
Given its focus on Hungarian minority rights, UDMR was able to be in alliance with
both left and right wing parties from 1996 until 2012.
PNL, PSD and PPDD have similar positions on the austerity measures and the
interference of EU and IMF in national decision making. These parties were especially
vocal against the declarations of Angela Merkel, the chancellor of Germany, and other
officials that criticized Romanian officials with tampering with the independence of
the Constitutional Court during the dismissal of Traian Basescu. The PDL electorate
supports the austerity measures and the pro-market policies. PDLs economic policies
in 2011 and 2012 included austerity measures such as wage and benefit cuts, taxation
of pensions higher than 250 euro and increased VAT tax from 19% to 24%. It seems
that a part of the Romanian electorate supports these measures.
Besides the 30 statements, the users were asked to place themselves on a GALTAN and economic scales. We compared the positions of party supporters on the
scales identified in the factor analysis with the answers on the GAL-TAN and the
economic axes. It seems that supporters from all parties view themselves far more
social liberal when they are asked to express a preference between being a social liberal
or a conservative traditionalist regarding minority rights. On the economic left-right
the most surprising position is that of PNL supporters. They view themselves as

130 The Romanian Journal of Society and Politics

supporters of market economy when asked to express a preference on the left-right

axis but are at odds with the policy preferences they support (Fig. 2) and with the
position of their party (Fig. 1). While the expert party mapping of PNL is closely
matched by the mapping of policy preferences, the positioning of party supporters on
the left-right axis does not reveal such a shift. Party supporters see themselves as right
wing social liberals at odds with the policy preferences they support. One reason for
such a mismatch might lie in the lacking capacity of citizens to use the economic left
right and GAL-TAN (Marian and King 2014) axes as a proxy for identifying parties
closed to them. Marian and King (2014) observed that since 1990 no more than 40%
of citizens place themselves on the left-right scale, although data from 2012 shows that
this percentage increased.
We do not have surveys from 2012 applied in Romania to test the external
validity of our findings. However we use the Chapel Hill expert survey from 2010 and
2006 to compare the positions of PSD, PNL, PDL and UDMR with our findings. The
expert data of the Chapel Hill survey confirms the ideological positions of political
parties and party supporters on the GAL-TAN and economic left-right.

Figure 3. The ideological positions of political parties in Romania. For VAA data positions represent the
average position of party supporters recoded so that variation will range from 0 to 1, The Chapel Hill data
represent average party positions according to country experts
VAA 2012 preferences compiled from the aggregation of issue stances of party supporters
VAA 2012 positions of parties on GAL-TAN and economic axes according to the expert survey

Chapel Hill expert survey 2010

Chapel Hill expert survey 2006 GAL means green alternative
liberal and TAN- traditional, authoritarian, nationalist

The positions of political parties by experts are similar with the party supporters
preferences on the economy and GAL-TAN axes. PDL and PNL are viewed as center
right wing parties. On GAL-TAN, PNL and UDMR are the most liberal followed by
PDL and PSD. PNL has a different position than in our party mappings but this is
possibly due to its recent ideological shift towards the left. The obvious discrepancy
lies in comparing the preferences of citizens based on issues on GAL-TAN. Although
most party supporters view themselves as social liberal, in a similar vein to how experts

Toma BUREAN and Raluca POPP 131

see parties, the party supporters preferences on specific issues that target acceptance
of Hungarians reveal a far more conservative view. The party mappings in the expert
surveys match closely the party supporters positions on the left right and GAL-TAN
adding to the validity of our results.
Conclusions
VAA applications provide a unique opportunity for social scientists to map, test
and analyze the programmatic thinking of citizens. Despite its limitations, VAA data
are easy to obtain, the issues cover several domains, and the number of respondents
is high. In this paper VAA Votulmeu.com was used to test whether we can draw
valid party mappings of party supporters preferences. The data obtained from the
Romanian VAA is not representative of the whole population as it is confronted
with problems of self-selection and it is biased towards citizens who use the internet,
the highly educated living in urban areas. The biases are larger for Romania due to
its low number of internet users (40% in 2012), the smallest figure in the European
Union. There were above 16000 respondents that completed the survey. The number
of respondents was small in comparison with other VAAs applied in Europe but it
was ten times larger than a VAA applied in Romania in 2009. The party mappings of
preferences revealed that there were three important dimensions of competition. The
first included a mixture of economic austerity measures and attitudes on foreign policy.
This is a dimension that was particularly salient in 2011 and 2012. At the beginning
of 2011, Basescu, announced the implementation of harsh austerity measures such as
the raise of VAT from 19 to 24% and 25% wage cuts in the public sector. The IMF
recommended some of these measures. Later, in 2012, the president was suspended
by a majority vote in parliament. EU officials reacted to the attempts of the Ponta
government to avoid constitutional rulings and dismiss the president. The mingling
of international institutions affected the opinions of citizens. The second dimension
identified was GAL-TAN with left wing parties being more conservative than the
right wing ones. Thirdly the economic left-right dimension covered issues that have
to do with providing protection for the pension system and protecting employees
from employers. The mappings confirmed previous studies on the analysis of party
programmatic competition in Romania. The validity of the results was tested by
comparing the expert positioning of political parties in Romania by the Chapel Hill
survey. While the preferences on the economic dimension were similar, on GALTAN issues party supporters had more conservative preferences. At the same time,
they viewed themselves as more social liberal. The expert surveys mappings and the
mapping of party supporters preferences on the left-right and GAL-TAN, with the
help of the VAA, showed remarkable similarity. To conclude, we find that VAA for
Romania is a valid and cost effective resource for party mappings.
References
Althaus, S. L., Collective Preferences in Democratic Politics Opinion Surveys and the Will of
the People. American Government, Politics and Pol, Cambridge, Cambridge University Press,

132 The Romanian Journal of Society and Politics

2003.
Alvarez M.R., I. Levin, A.H. Trechsel, K. Vassil. Voting advice applications:
How useful and for whom? Journal of Information Technology & Politics, vol. 11 no.1,
2014, pp.82101
Andreadis, I., J. Pianzola and D. Garzia. Measuring the impact of Voting
Advice Applications on Vote Choice paper presented at the 2nd European Conference
on Comparative Electoral Research, Rhodes, Greece, 2013, http://true-european-voter.eu/
node/149 (accessed June 13, 2013).
Baka A., Figgou L., Triga, V. Neither agree, nor disagree: a critical analysis of
the middle answer category in Voting Advice Applications. Int. J. Electronic Governance,
Vol. 5, Nos. 3/4, 2012, pp. 264-268.
Bakker, R, C. de Vries, E. Edwards, L. Hooghe, S. Jolly, G. Marks, J. Polk, J.
Rovny, M. Steenbergen, and M. A. Vachudova. Measuring Party Positions in Europe:
The Chapel Hill Expert Survey Trend File, 19992010. Party Politics, 2012, 1-15.
Berelson, B. R., P. F. Lazarsfeld, and W. N. McPhee. Voting - A study of opinion
formation in a presidential campaign, Chicago, The University of Chicago Press, 1954.
2006

Benoit, K. and M. Laver. Party Policy in Modern Democracies, London, Routledge,

Budge I, H. D. Klingemann, A. Volkens, J. Bara , E. Tanenbaum, Fording RC,

Hearl DJ, Kim H-M,McDonald M and Mendez S. Mapping Policy Preferences: Estimates
for Parties, Electors, and Governments 19451998, Oxford, Oxford University Press, 2001
Burean, T. and Bdescu G. Voices of Discontent: Student Protest Participation
in Romania. Journal of Communist and PostCommunist Studies vol. 47, nos.3-4, 2014, pp.
385-397.
arkolu A., Vitiello T., Moral M. Voter advice applications in practice: answers
to some key questions from Turkey. Int. J. of Electronic Governance, Vol.5, No.3/4, 2012,
pp.298 - 317.
Converse, P. E. The Nature of Belief Systems in Mass Publics in D. E. Apter
(ed.) Ideology and Discontent, New York, Free Press, 1964, pp.206-261.
Evans, G., and S. Whitefield. Identifying the Bases of Party Competition in
Eastern Europe. British Journal of Political Science 23, no. 4, 1993, 52148.
Fesnic, F.. Welfare Chauvinism East and West: Romania and France. PhD
Thesis, University of Illinois, 2013.
Fivaz J., Louwerse T., Schwarz D. Keeping Promises: Voting Advice Applications
and Political Representation, in D. Garzia and S. Marschall (eds.) Matching Voters with

Toma BUREAN and Raluca POPP 133

Parties and Candidates: Voting Advice Applications in a Comparative Perspective, Colchester,

ECPR Press, 2014
Fivaz J. and G. Nadig. Impact of Voting Advice Applications (VAAs) on Voter
Turnout and Their Potential Use for Civic Education Policy&Internet vol. 2 no. 4,
2012, pp. 167-200.
Freedman, P., M. Franz, and K. Goldstein. Campaign Advertising and
Democratic Citizenship American Journal of Political Science 48, no. 4, 2004, pp. 72341.
Fuchs D., H.D. Klingemann. The Left-Right Schema in M.K. Jennings et al.
(eds.) Continuities in Political Action: A Longitudinal Study of Political Orientations in Three
Western Democracies, Berlin, Walter de Gruyter, 1989.
Garzia D., S. Marschall. Voting advice applications under review: the state of
research International Journal of Electronic Governance, vol. 5, no.3, 2012, pp. 203222
Garzia D., A.H. Trechsel, K. Vassil, E. Dinas Indirect campaigning: Past,
present and future of voting advice applications in Grofman et al. (eds.) The Internet
and Democracy in Global Perspective, Springer, 2014, 2541.
Garzia, D. The effects of VAAs on users voting behavior: an overview in Voting Advice
Applications in Europe: The State of the Art. Scriptaweb, 2010.
Gherghina, S., S. Miscoiu. A Rising Populist Star: The Emergence and
Development of PPDD in Romania. Debatte: Journal of Contemporary Central and Eastern
Europe vol. 22, no.2, 2014, pp. 181-197.
Gemenis, K. Estimating parties policy positions through voting advice
applications: Some methodological considerations Acta Politica vol. 47, no. 3, 2013,
pp. 268-295.
Germann M., M. Fernando, S. Uwe and J. Wheatley. Exploiting Smartvote Data
for the Ideological Mapping of Swiss Political Parties Paper presented for the XXVI
Convegno SISP, Universita Roma Tre, 13-15 September, 2012.
Grecu, R., A. Raduca, R. Nastase, A. Todor. Mapping Policy Preferences of
Romanian Political Parties Research paper Invisible College, 2003. http://razvan.grecu.
tripod.com/papers/mpp.pdf (accessed 1 July 2013).
Hansen, K. M. The Effect of Political Campaigns. Prepared for XV NOPSA
Conference on Political Communication: Antecedents, Contents and Effects of Political information
6-8 august 2008, http://tinyurl.com/kxrc8rz, (1 July 2013)
Hix S., Crombez C. Extracting Ideal Points from actors preferences in the EU
constitutional negotiation. European Union Politics vol. 6, no.3, 2005, pp. 353-376.
Hooghe, L., R. Bakker, A. Brigevich, C. De Vries, E. Edwards, G. Marks, J.

134 The Romanian Journal of Society and Politics

Rovny, M. Steenbergen, and M. Vachudova. Reliability and Validity of the 2002 and
2006 Chapel
Hill Expert Surveys on Party Positioning, European Journal of Political Research,
vol.49, no. 5, 2010, pp. 687703.
Huber, J. D., and G. B. Powell Jr. Congruence between Citizens and Policymakers
in Two Visions of Liberal Democracy World Politics vol. 46, no. 3, 1994, pp. 291326.
Inglehart R. and H.D. Klingemann. Party Identification, Ideological Preference
and the Left-Right Dimension among Western Mass Publics in I. Budge, I. Crewe and
D. Farlie (eds) Party Identification and Beyond, London, Wiley, 1976
Iyengar, S. Shortcuts to Political Knowledge: The Role of Selective Attention
and Accessibility in J. A. Ferejohn and J.H. Kuklinski (eds.) Information and Democratic
Processes, Urbana and Chicago, University of Illinois Press, 1990, pp. 16085.
Iyengar, S. and A. Simon.News Coverage of the Gulf Crisis and Public Opinion
A Study of Agenda-Setting, Priming, and Framing. Communication Research 20, no.
3:1993, pp. 36583.
Jiglau, G. and S. Gherghina The Ideological Institutionalization of the Romanian
Party System in Romanian Journal of Political Science vol. 11, no.1 , 2011, http://www.sar.
org.ro/polsci/?p=605 (accessed July 2, 2013)
Key, Jr. V.O. Politics, Parties and Pressure Groups 5th ed. New York, Thomas Y.
Crowell Company, 1964.
Kitschelt, H., Z. Mansfeldova, R. Markowski and G. Toka. Post communist party
systems: Competition, Representation and Inter-Party Cooperation, Cambridge, Cambridge
University Press, 1999
Kitschelt, H. The Formation of Party Systems in East Central Europe Politics &
Society vol. 20, no. 1, 1992, pp.750.
Klingemann, H., A. Volkens, J. Bara, I. Budge, and M. D. McDonald. Mapping
Policy Preferences II:Estimates for Parties, Electors, and Governments in Eastern Europe, European
Union, and OECD 1990-2003: Estimates for Parties, Electors, and Governments in Eastern
Europe, European Union, and OECD 1990-2003, Oxford, Oxford University Press, 2006.
Lau, R., and D. P. Redlawsk Advantages and Disadvantages of Cognitive
Heuristics in Political Decision Making American Journal of Political Science vol. 45, no.
4, 2001, pp. 95171.
Laver, M. Why should we estimate the policy positions of political actors?
in Michael Laver (ed.) Estimating the policy position of political actors, London: Routledge,

Toma BUREAN and Raluca POPP 135

2001, pp. 39.

Lefevere, J. and S. Waalgrave. A perfect match? The impact of statement
selection on VAAs ability to match voters and parties Working Paper MM, 2012,
http://www.m2p.be/publications/1351531800.pdf (accessed June 13, 2013)
Lodge, M., M.R. Steenbergen, and S. Brau. The Responsive Voter: Campaign
Information and the Dynamics of Candidate Evaluation. The American Political Science
Review vol. 89, no. 2, 1995, pp. 30926.
Mair, P. Democracies in D. Caramani (ed), Comparative Politics, Oxford, Oxford
University Press, 2008, pp.108-132.
Marian, C. and King R.F. Perceiving the line: The Relevance of the Left-Right
Ideological Dimension for Voter Preferences in Romania Expert Electoral no. 2, 2014,
pp.3-18.
Markowski, R. and J. Tucker. Euroscepticism and the Emergence of Political
Parties in Poland Party Politics, no. 16 , 2010, pp. 523-548.
Marks, G., L. Hooghe, M. Nelson, and E. Edwards. Party Competition and
European Integration in the East and West Different Structure, Same Causality
Comparative Political Studies vol. 39, no. 2, 2006, pp. 15575.
Marschall S., Garzia D. Voting advice applications in a comparative perspective:
an introduction in D. Garzia and S. Marschall (eds.) Matching Voters with Parties and
Candidates: Voting Advice Applications in a Comparative Perspective, Colchester, ECPR Press,
2014.
Mendez F. and Wheatley J. Using vaagenerated data for mapping partisan
supporters in the ideological space in matching voters to parties and candidates in S.
Marshall and D. Garzia (eds.): Matching Voters with Parties and Candidates: Voting Advice
Applications in a Comparative Perspective, Colchester, ECPR Press, 2014.
Moore, D. W. Political campaigns and the Knowledge Gap Hypothesis
in Public Opinion Quarterly vol. 51, no. 2, 1987, pp. 186-200. Nuytemans, M., S.
Waalgrave, K. Deschouver. Do the Vote Test: The Belgian Voting Aid Applications
in L. Cedroni and D. Garzia (eds.) Voting Advice Applications in Europe. The State of the
Art, Napoli, Scripta Web, 2010.
Pianzola J. and Ladner A. Tackling Self-Selection into Treatment and SelfSelection into the Sample Biases in VAA Research. 6th ECPR General Conference,
University of Iceland, Reykjavik, August 25th-27th, 2011.
Romanian Electoral Office. The Final Results of the Parliamentary elections
for the Chamber of Deputies, 2012, http://www.becparlamentare2012.ro/A-

136 The Romanian Journal of Society and Politics

DOCUMENTE/rezultate%20finale/Anexa%208A.pdf , (7 March 2015).

Ruusuvirta O., Rosema M. Do online vote selectors influence electoral
participation and the direction of the vote in ECPR General Conference, September 2009.
Schmitt-Beck, R. New modes of campaigning in R.J. Dalton and
H.D.Klingemann (eds.) The Oxford Handbook of Political Behavior, Oxford, Oxford
University Press, 2007.
Sniderman, P. M., R. A. Brody, and P. E. Tetlock. Reasoning and Choice: Explorations
in Political Psychology. Reprint edition, Cambridge, Cambridge University Press, 1993
Trechsel A. EU-Profiler: user profiles in the 2009 European
Elections,
2010-10,
http://data.eui.eu/RSCAS/EUDO/euprofiler/10025
UNF:5:WjqNsl3mv+qPCzmEm7AWow==
European
University
Institute
[Distributor] V5 [Version], (accessed 1 June 2014)
Wagner, M. and O. Ruusuvirta. Faulty Recommendations? Party Positions in
Online Voting Applications Paper presented at APSA Annual Meeting Toronto, 3-6
September, 2009
Walgrave S, P. van Aelst, M. Nuytemans Do the vote test: the electoral effects
of a popular vote advice application at the 2004 Belgian elections. Acta Politica, vol. 43,
no. 1, 2008, pp. 5070.
Wall, M., A. Krouwel and T. Vitiello. Sealing the deal: Assessing the impact
of an online voter advice application on user voting behavior in the 2010 legislative
elections Presented at the ELECDEM Amsterdam Workshop March, 2011, http://
tinyurl.com/m7g7fpk (accessed, July 2013)
Wall M., Krouwel A., Vitiello T. Do voters follow the recommendations of
voter advice application websites? A study of the effects of kieskompas.nl on its
users vote choices in the 2010 Dutch legislative elections Party Politics, vol. 20, no. 3,
pp. 416428.
Wheatley, J., C. Carman, F. Mendez, and J. Mitchell. The Dimensionality of the
Scottish Political Space: Results from an Experiment on the 2011 Holyrood Elections
Party Politics, vol. 20, no. 6, 2014, pp. 864-878
Whitefield, S. Political Cleavages and Post-Communist Politics Annual Review
of Political Science 5, no. 1, 2002, pp. 181200.