Trust in Institutional Actors across 22 Countries. Examining Political, Science, and Media Trust Around the World

Abstract

Social trust has long attracted the interest of researchers across different disciplines. Most of previous studies rely on single-country data and consider only one dimension of social trust at a time (e.g., trust in science, the media or political institutions). This research extends a framework developed by the Global Trust Inventory (GTI) by discussing several dimensions of social trust, while simultaneously analyzing how trust in institutions varies across societies. Drawing on an online panel survey collected in 22 countries (N = 22,033), we examine cross-country differences in social trust—including government trust, trust in governing bodies, security, and knowledge producers. Additionally, this paper fills a gap in current literature by including a measure of trust in the media. Findings are discussed in the context of comparing emerging and developed countries based on the Human Development Index.

Full text

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 237
How to cite this article in bibliographies / References
H Gil de Zúñiga, A Ardèvol-Abreu, T Diehl, M Gómez Patiño, J H Liu (2019): “Trust in
Institutional Actors across 22 Countries Examining Political, Science, and Media Trust Around
the World”. Revista Latina de Comunicación Social, 74, pp. 237 to 262.
http://www.revistalatinacs.org/074paper/1329/12en.html
DOI: 10.4185/RLCS-2019-1329en
Trust in Institutional Actors across 22
Countries. Examining Political, Science,
and Media Trust Around the World
Homero Gil de Zúñiga [CV] Department of Communication, Media Innovation Lab. University of
Vienna, Austria. homero.gil.de.zuniga@univie.ac.at
Alberto Ardèvol-Abreu [CV] Departamento de Psicología Cognitiva, Social y Organizacional.
Universidad de La Laguna, España. aardevol@ull.es
Trevor Diehl [CV] Department of Broadcast and Cinematic Arts. Central Michigan University,
Estados Unidos. diehl1th@cmich.edu
María Gómez Patiño [CV] Departamento de Lingüística General e Hispánica. Universidad de
Zaragoza, España. mariagp@unizar.es
James H. Liu [CV] Centre for Applied Cross-Cultural Research. Massey University, Nueva
Zelanda. j.h.liu@massey.ac.nz
Abstract
Social trust has long attracted the interest of researchers across different disciplines. Most of previous
studies rely on single-country data and consider only one dimension of social trust at a time (e.g., trust
in science, the media or political institutions). This research extends a framework developed by the
Global Trust Inventory (GTI) by discussing several dimensions of social trust, while simultaneously
analyzing how trust in institutions varies across societies. Drawing on an online panel survey collected
in 22 countries (N = 22,033), we examine cross-country differences in social trust—including
government trust, trust in governing bodies, security, and knowledge producers. Additionally, this
paper fills a gap in current literature by including a measure of trust in the media. Findings are
discussed in the context of comparing emerging and developed countries based on the Human
Development Index.
Keywords
Cross-cultural, political trust, trust in the media, trust in science, social trust, trust in institutional actors.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 238
Contents
1. Introduction. 2. Literature review. 2.1. Political trust: specific and diffuse support. 2.2. Trust in
scientific institutions. 2.3. Trust in the media. 3. Research question. 4. Method. 4.1. Sample. 4.2.
Measures. 4.3. Analyses. 5. Results. 5.1. Political trust: trust in government, trust in governing bodies,
and trust in security institutions. 5.2. Trust in knowledge producers. 5.3. Trust in the media. 6.
Discussion and conclusion. 7. References. 8. Appendix.
Writen in English by_the Authors
1. Introduction
As societies become more democratized, persuasion tends to replace coercion whenever possible, and
trust becomes a key element in almost every sphere of public life (Levi, 1998). Without a certain
amount of political trust, citizens will not empower their governments through elections and
participation, the media will not perform as ‘watchdog’ for the public’s interest, and scientists will not
be provided with the resources they require to produce and disseminate, knowledge. Broadly defined,
trust is the favorable expectation an individual hold about the positive outcomes when interacting with
another individual, group, or institution (Coleman, 1990; Tsfati, 2003). Trust is a prerequisite for basic
human interactions, including partnership and marriage, patient-health care provider relationships, or
economic exchanges (Harris, Skogrand, & Hatch, 2008; Lorenz, 1999; Luhmann, 2000; Tsfati &
Cappella, 2003).
Over the last decades, trust has attracted the interest of scholars across many disciplines. Research
revolving trust is therefore extensive and comprises many different areas and dimensions. Thus,
political trust (Boix & Svolik, 2013; Catterberg & Moreno, 2005; Cook & Gronke, 2005); trust in
science and knowledge producers (Achterber, de Koster, & van der Waal, 2015; Allum, Sturgis,
Tabourazi, & Brunton-Smith, 2008; Aupers, 2012); and trust in the media (Hovland & Weiss, 1951;
Kohring & Matthes, 2007; Tsfati & Capella, 2003) are all sub-dimensions of trust of particular interest
to social and political science researchers. However, most of previous studies rely on single-country
data (most often in the United States and a few other Western democracies) and examine one—or
maximum two—dimensions of trust at a time. This purpose of this paper is to: a) examine several
dimensions of social trust, and b) move beyond single-country studies to simultaneously analyze how
these different sub-dimensions vary in different societies. We do so by adopting a more comprehensive
framework of trust that includes ‘institutional actors,’ organizations whose work involves producing
information, services and rules (i.e., institutional products) that will affect the rest of actors in the
society (Furusten, 2013). Drawing from 4 factors identified by the Global Trust Inventory (Liu,
Milojev, Gil de Zúñiga, & Zhang, 2018)—trust in government, trust in governing bodies, trust in
security institutions, and trust in knowledge producers—, as well as a measure of trust in the media,
we analyze the results of a large data set collected in 22 societies (N = 22,033). We found important
differences in levels of social trust among societies, uncovering certain patterns and clustering
according to the stage of social development of the countries, measured through the Human
Development Index (HDI) [1].

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 239
2. Literature review
2.1. Political Trust: Specific and Diffuse Support
Traditionally, scholars have defined political trust as a basic evaluative orientation toward the
government, based on the match between people’s expectations and subsequent perceptions of
performance (Hetherington, 1998; Miller, 1974). From a normative perspective, a well-functioning
democracy requires a certain level of political trust, since it legitimates the acts of governing
institutions and allows for more effective governance with lower levels of coercion (Gamson, 1968;
Hetherington, 1998; Levy, 1998). However, discrepancies about the dimensionality, measurement,
antecedents, and effects of political (dis)trust still persist.
Political trust taps into feelings towards the government as a whole, so that “it likely affects
assessments of the government’s component parts, namely, incumbent and institutions, at the same
time” (Hetherington, 1998, p. 791). This double orientation of political trust (toward incumbents—
specific support— and toward the political system and institutions —diffuse support—) is crucial to
properly assess the implications of the generalized drop in levels of social trust around the world in
the last decades (Bennet, Rhine, Flickinger, & Bennet, 1999; Catterberg & Moreno, 2005; Easton,
1965; Hetherington, 1998). On the one hand, a steady diffuse support decline may be viewed as an
indicator of political alienation with potential to deter people from public participation (Miller, 1974;
Putnam, 2000; 2002). On the other, eroded specific support may suggest the emergence of a more
critical and politically sophisticated citizenry that would maintain a ‘vigilant skepticism’ (Cook &
Gronke, 2005; Hardin, 1999).
Correlates of political trust found in previous studies—both at the micro and macro levels—lend
support to the theoretical and empirical complexity of this construct. Thus, individual well-being,
social capital, political interest, external efficacy, country’s economic situation, and level of
congressional/ presidential approval are positive predictors of political trust (Citrin & Green, 1986;
Hetherington, 1999; Catterberg & Moreno, 2006). Conversely, political radicalism, post-materialism,
and corruption permissiveness are negatively related to political trust (Catterberg & Moreno, 2006).
Some studies have explored political trust cross-culturally using survey data from different countries
with different levels of democratic consolidation (Catterberg & Moreno, 2006). Consistently with
‘cultural theories’ (Citrin, 1974; Ulsaner, 2002), the relationship between democratic development and
political trust has been found to be curvilinear in nature. That is, the arrival of democracy is typically
characterized by increased levels of political trust, which tend to decline again after some decades of
citizens’ unmet expectations. According to Catterberg and Moreno, this ‘post-honeymoon’
disillusionment period results from the fact that “aspirations of civic, political, and economic rights”
are not achieved in many cases, resulting in citizens’ skepticism (2006, p. 33). In highly developed
societies, however, relatively low levels of political trust are explained by ‘post-materialistic’ values
and citizens’ increased performance demands (Catterberg & Moreno, 2006; Inglehart, 1997).
2.2. Trust in Scientific Institutions
Attitudes of trust towards knowledge producers, or towards science in general, have been an area of
increasing academic interest since the early eighties, particularly in the U.S. and Great Britain (Evans
& Durant, 1989; Miller, 1983; Ziman, 1991). Perhaps one of the most noteworthy of these attempts is
the successive European Union’s Eurobarometer series (in 1989, 1992, 2001, and 2005) focused on
‘public understanding of science’ (Pardo & Calvo, 2002). These measures—and subsequent studies
based on them—have however met with some criticism based of the “lack of theory” in the formulation

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 240
and selection of the items and the low reliability and internal consistency of the attitudinal scales (Pardo
& Calvo, 2002, p. 167)
Comparative studies based on survey data suggest that knowledge producers have not been entirely
immune to the negative trends in the public’s level of social trust (Allum et al., 2008; Aupers, 2012;
Inglehart, 1997). While the majority of the population tends to support science and scientific research,
some societies have experienced growing distrust in “scientific authorities, the knowledge they
produce and the (technical) solutions they propose” (Aupers, 2012, p. 26). Existing research in this
area is however limited to only a handful of western democracies, and it would be therefore necessary
to extend these analyses to other societies, particularly in “Africa, Asia and the middle east” (Allum et
al., 2008, p. 52).
When considering science, scientific principles and methods, on the one hand, and scientific
institutions (including scientists), on the other, are assessed differently (Achterberg et al., 2015).
According to this strand of research, only scientific institutions, and not scientific principles and
methods, would be facing a crisis of confidence (Achterberg et al., 2014; Millstone & Zwanenberg,
2000). For this reason, in this paper we only consider the institutional dimension of science. By doing
so, we answer the call made by Bauer et al. (1994) for further research to explore public perceptions
of this specific dimension of science.
2.3. Trust in the Media
News media are a main source of information about the political and social world. In democratic
contexts, individuals and societies place their trust in the media with the expectation that they will
serve as a watchdog for the public interest (Dyck & Zingales, 2002; Habermas, 1989). The media are
not only entrusted with specific tasks such as filtering, selecting, and communicating ‘objective,’ ‘bias-
free’ information, but they are also expected to contribute to democratic stability by fostering
deliberation, negotiation, and collective decision-making (Farnsworth & Lichter, 2007; Schudson,
1978).
Because of these important links between media, public opinion, and democracy, trust in the media—
and the related concept of media credibility—has received the attention from sociology and
communication researchers since the 1950s. Throughout these seven decades of studies on the topic,
researchers have found a variety of effects of (dis)trust in news media. Thus, the level of trust in the
source of information is directly related to the persuasive power of messages, and mediates both
agenda-setting and priming effects (Druckman, 2001; Miller & Krosnick, 2000). At a behavioral level,
(dis)trust in the media has been found to affect media consumption patterns, since audiences tend to
get information from sources they trust and to avoid exposure to sources they do not consider reliable
(Ardèvol-Abreu & Gil de Zúñiga, 2016; Tsfati & Cappella, 2003).
Despite the considerable progress made in the understanding of media trust over the past decades,
important conceptual and methodological gaps persist. First, as Kohring and Matthes (2007) have
pointed out, we still lack an explicit and comprehensive theory of trust in the media, which results in
different and sometimes inconsistent measures of the construct. Second, research on media trust
originates almost entirely on the United States, so that we lack a cross-country perspective (Tsfati &
Ariely, 2014). Previous findings on media trust are thus hardly amenable to generalization to other
democratic societies, let alone non-democratic contexts. This study is designed to fill some of these
gaps by relying on a cross-national exploration of trust in the news media.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 241
3. Research Question
The objective of this study is twofold. First, it is aimed at ‘grasping together’ (Sibley & Liu, 2013)
different types of social trust to help theorize trust as a global system of meaning (Liu et al., 2018).
Also, we investigate possible differences in the level of trust in institutional actors across countries.
Based on the literature reviewed above, and in the light of these objectives, we ask a research question:
RQ: What are the levels of trust in different institutional actors —trust in government, trust in
governing bodies, trust in security institutions, trust in knowledge producers (e.g., science), and trust
in the media— across different societies?
4. Methods
4.1. Sample
Data for this study comes from an online panel survey collected in 22 countries from the Americas,
Asia, Europe and South Africa. The survey is a part of an international project conducted by a
partnership between research groups based in Europe and New Zealand. Items were translated for each
country by a large group of participating scholars, employing either back-translation with a team
approach (Behling & Law, 2000; Thato, Hanna, & Rodcumdee, 2005) or the committee approach
(Brislin, 1980). Survey administration was performed from September 14-24, 2015. AC Nielsen
curated a massive pool of potential respondents across 22 countries – over 10 million. Nielsen used
stratified quota sampling techniques to create samples whose demographics closely matched those
reported by official census agencies (Callegaro et al., 2014) [2].
The largest sample size was collected in Brazil (N = 1,224), and the smallest in India (N = 409) as it’s
only representative of New Delhi; (mean sample size, for all countries: M =1,136; SD = 238). Overall
cooperation rate was relatively high, averaging 77% across the panel (AAPOR, 2011; CR3). Since
Nielsen partners with companies that employ a combination of panel and probability-based sampling
methods, the limitations of web-only survey designs are minimized (AAPOR, 2011; Bosnjak, Das, &
Lynn, 2016; see Appendix for details).
4.2. Measures
Trust in government: This measure attempted to capture information about respondents’ ‘specific
support’ for incumbents; different from the broader regime-based ‘diffuse support’ (Miller, 1974;
Putnam, 2000; 2002). Building on the World Values Survey and other previous approaches (Bennett
et al., 1999; Catterberg & Moreno, 2006; Citrin, 1974), trust in government was measured with three
items, based on responses to the following prompt: “Please rate your feelings of trust towards the
following people and organizations, where 1 = do not trust at all, and 7 = trust completely: ‘national
government;’ ‘local government;’ and ‘your president or prime minister’ (Cronbach’s α = .87; M =
2.19; SD = 1.08; Table 3 for a detailed breakdown by country).
Trust in governing bodies: This variable taps respondents’ more general attitudes toward the political
regime, or their diffuse support for the system, irrespective of whether they trust current rulers or not
(Cook & Gronke, 2005; Hardin, 1999). This index included four items concerning respondents’ level
of trust towards the following actors: ‘the judiciary (courts),’ ‘government surveillance agencies,’ ‘the
tax system,’ and ‘election outcomes in your country’ (Cronbach’s α = .84; M =3.21; SD = 1.41; Table
3).

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 242
Trust in security institutions: Similar to trust in governing bodies, this dimension of social trust also
relates to the public’s attitudes toward regime-level institutions —and therefore to diffuse support
(Hetherington, 1998; Liu et al., 2018). Respondents were provided the same prompt as for trust in
government, and they rated their feelings of trust towards ‘police’ and ‘the military in your country’
(Cronbach’s α = .74; M = 3.84; SD = 1.54; Table 3).
Trust in knowledge producers. Using the same prompt, trust in knowledge producers was measured
with an index based on the respondents’ feelings of trust towards ‘scientists’ and ‘universities’ (2 items
averaged scale, Cronbach’s α = .87 M =4.43, SD = 1.42; Table 3).
Trust in the media. Building in previous measures (Jackob, 2010; Jones, 2004; Moy, McCluskey,
McCoy, & Spratt, 2004), media trust was measured with a three-item scale. Respondents were asked
about their feelings of trust towards ‘news from mainstream news media (e.g., newspapers, TV);’
‘news from alternative news media (e.g., blogs, citizen journalism);’ and ‘news from social media’
(Cronbach’s α = .77; M =3.51; SD = 1.12; Table 3).
4.3. Analysis
Based on a previously introduced model of social trust, Global Trust Inventory (Liu, et al., 2018), we
combined the 14 items of social trust into additive, averaged item indices—as described in the methods
section—to gather descriptive statistics (Table 3). Reliability testing was performed on each construct
(Cronbach’s α scale testing for multi-item indices, KR-20 parallel testing for the two-item indexes).
We then compared the mean score on each construct using t-tests against the overall mean score for
all countries, what we call the ‘grand mean’ (Table 4). Further, we employed post-hoc ANOVA tests,
using the Bonferroni procedure, to test differences between each country (Table 4 and Table 5). Results
are reported in clusters of countries based on the United Nations Human Development Index (HDI)
(Tables 2, 3, 4, and 5). Clusters analysis creates groups of countries that shown similar pattern of
response to a variable of interest, or a set of them (Human Development Index).
5. Results
Tables 2-5 present detailed descriptive statistics (disaggregated by country) for each sub-dimension of
trust. Countries were grouped in four clusters based on their Human Development Index1 (HDI)
(United Nations Development Program, 2015), from ‘highest’ to ‘medium’ HDI. Two-step cluster
analyses based on the Euclidean distance showed that a four-cluster solution was a good fit for the data
(average silhouette measure of cohesion and separation = 0.7; ratio of sizes = 2.5). As Table 1 shows,
the clusters differ in size, with the larger cluster representing the countries with the ‘highest’ HDI:
Germany, United States, New Zealand, United Kingdom, Korea, Japan, Taiwan, Spain, Italy, and
Estonia (N = 10, 45.5% of the countries). The second cluster includes four countries (18.2% of the
cases) with a ‘very high’ HDI: Poland, Argentina, Chile, and Russia. Four countries (18.2%) with a
‘high’ HDI form the third cluster: Turkey, Brazil, Ukraine, and China. Finally, the ‘medium HDI’
cluster (N = 4, 18.2% of the countries) includes Indonesia, Philippines, South Africa, and India.
To further test the internal consistency of the different dimensions of trust across countries, Cronbach’s
α coefficients were calculated for every sub-dimension of trust in every country. As Table 3 shows,
the five sub-dimensions of trust in institutional actors are reasonably consistent across the 22 countries
in the sample (alphas range from .59 - .94; Table 3).

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 243
5.1. Political trust: trust in government, trust in governing bodies, and trust in security
institutions
The first three columns in Tables 2, 3, 4, and 5 show the results for political trust, with some
dimensions more connected to trust in specific incumbents (specific support, i.e. trust in government)
and others more connected to trust in the political system and institutions as a whole (diffuse support,
i.e. trust in governing bodies and trust in security institutions). Interestingly enough, the average levels
of trust across the countries are consistently higher for governing bodies and security institutions
(diffuse support) than for government (specific support). There are however a few exceptions to this
pattern: Russia, Ukraine, China, and Indonesia show higher levels of trust in government (MR = 3.41;
MU = 2.43; MCh = 4.05; MI = 3.50, respectively) than in governing bodies (MR = 3.03; MU = 2.30; MCh
= 3.63; MI = 3.31, respectively). On the contrary, security institutions (police and military) score higher
than government without any exception (Table 3). This means that, at the aggregate level, people tend
to show lower levels of trust in incumbent-level government and personalities than in regime-level
institutions—the latter only indirectly linked to the government in office. This finding provides
additional support for the theoretical distinction between incumbent and regime-based trust (Bennet et
al., 1999; Easton, 1965; Hetherington, 1999).
Government trust ranged between 1.88 (Brazil, slightly below 2 = ‘trust a little’) and 4.93 (India, 5 =
‘trust significantly’). For their part, the levels of trust in governing bodies were between a minimum
of 2.30 (in Ukraine) and maximum of 5.03 (again, in India). Similarly, trust in security institutions
reaches its highest level in India (5.25) and its lowest value in South Africa (2.85) (Tables 2 and 3).
Table 4 shows detailed t-tests for single-country differences with the grand mean (government trust,
M = 2.92; trust in governing bodies, M = 3.21; trust in security institutions, M = 3.84). Estonia, Russia,
China, and India showed the higher levels of trust in government within their respective clusters; while
Spain, Poland, Brazil, and South Africa scored the lowest within their groups (Tables 4; Table 5).
Similarly, Estonia, Russia, China, and India peaked in trust in governing bodies within their clusters,
while Taiwan, Argentina, Ukraine, and South Africa scored significantly lower than the rest of the
countries in their clusters. Concerning the third sub-dimension of political trust—trust in security
institutions—, maximum values within clusters were found in Estonia, Chile, Turkey, and India.
Conversely, Taiwan, Argentina, Brazil, and South Africa showed the lowest levels of trust in police
and the military within clusters.
Overall, results suggest a non-linear relationship between levels of human development and political
trust (including government, governing bodies, and security institutions). As shown in Table 1, and
considering clusters of countries, average levels of political trust (government, governing bodies, and
security institutions) are relatively high at each end of the HDI (i.e., clusters 1 and 4). Thus,
respondents living in countries with the ‘highest’ (cluster 1) or a ‘medium’ (cluster 4) HDI tend to trust
more their government and institutions, compared to respondents from countries in clusters 2 and 3
(with a ‘very high’ and ‘high’ HDI).
5.2. Trust in knowledge producers
Concerning cross-country differences in trust in scientists and universities, the mean values ranged
from 3.41 (Taiwan. above 3 = ‘trust in some ways,’ and below 4 = ‘trust moderately’) to 5.60 (India,
above 5 = ‘trust significantly,’ and below 6 = trust a lot) (Tables 2 and 3). Note that, as with trust in
the media (below), Taiwan and India obtained the lowest and highest mean values for trust in
knowledge producers, respectively (Table 4 for more detailed comparisons with the ‘grand mean,’ M
= 4.43). Between-group differences within clusters are also significant for science trust (Tables 4 and

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 244
5). Thus, the mean levels of trust in knowledge producers in Estonia (M = 5.06, significantly higher
than the rest of countries grouped in first cluster but New Zealand), Argentina (M = 5.11), Turkey (M
= 4.72) and India (M = 5.60) peaked in their respective clusters. On the contrary, mean levels of trust
in scientists and universities in Taiwan (M = 3.41), Poland (M = 4.14), Brazil (M = 4.27, significantly
lower than the rest of countries within the third cluster but China), and South Africa (M = 4.10) were
found to be the lowest within their respective clusters. Trust in knowledge producers is inversely
related to the HDI at the aggregate level, showing a maximum value (M = 4.72) for the fourth cluster
(lower HDI) and a minimum (M = 4.32) for the cluster of ‘highest’ HDI countries (Table 1).
5.3. Trust in the media
Trust in the media ranged from 2.63 (Taiwan) (between 2 = ‘trust a little,’ and 3 = ‘trust in some
ways’) to 5.13 (India) (above 5 = ‘trust significantly’). Table 4 shows more detailed results at the
country level, specifying countries that scored above or below the ‘grand mean’ (M = 3.51). Post-hoc
ANOVA comparisons (Tables 4 and 5) show significant between-groups differences within clusters.
The minimum scores within clusters (significantly lower than any other country within their respective
clusters) were found for Taiwan (M = 2.63), Poland (M =3.40), and South Africa (M = 3.54). At the
other end of the index, Chile (M = 3.86) and India (M = 5.13) showed the higher levels of media trust
within clusters. Table 1 compares aggregated values of trust between clusters. Similar to trust in
knowledge producers, media trust reaches a maximum (M = 4.14) in the fourth cluster, formed by
those countries with a lower HDI. At the opposite end, the cluster including those countries with a
higher HDI scores the lowest in trust in the media (M = 3.33). These figures are also suggestive of an
inverse association between HDI and media trust.
6. Discussion and conclusion
In this overview, we summarize past research from different fields and perspectives, fostering the
ground for a multidimensional, internationally valid measurement of trust in institutional actors. Thus,
this study employed a five-dimensional model of trust in institutional actors. Our approach is in line
with concerns raised about the double orientation of political trust: a) toward the incumbents and their
current policy-making (specific support), and b) toward other more stable institutions and elements of
the political regime (diffuse support) (Bennet et al., 1999; Easton, 1965; Hetherington, 1999). In other
words, citizens’ attitudes toward politics —and, more specifically, their levels of political trust—
cannot be studied as a homogeneous block. When asked about their levels of political trust, people
tend to clearly distinguish between the government and leaders in power (specific support) and other
more ‘incumbent independent’ institutions (the judiciary, military, or surveillance agencies) (diffuse
support). In our sample, the levels of diffuse support are consistently higher than the levels of specific
support across countries. It should be noted that our proposed five-dimensional model of social trust
works reasonably well in nearly all countries of the sample. Cronbach’s alphas for all sub-dimensions
are acceptable (and sometimes good or excellent) across countries, with very few exceptions.
Overall, knowledge producers (scientific institutions and scientists) are the most trusted actors across
societies. Although some studies have called attention to declining levels of trust in science (for
example, Allum et al., 2008; Aupers, 2012), our findings indicate that people tend to particularly trust
in science, even when asking specifically about scientific institutions and not about scientific
principles and methods (see Achterberg et al., 2015). Without exceptions, respondents rate universities
and scientists as the most trustable institutional actors, with figures above 4 (trust moderately) in most
countries, and even above 5 (trust significantly) in some of them (Italy, Estonia, Argentina, and India).
On the other end of the spectrum, government is often the least trusted social actor, ranking below

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 245
governing bodies, security institutions, and the media. This finding is however inconsistent across
societies and in some of the most populated countries in our sample (United States, Russia, and China)
people trust more the government than the media (the least trusted institution in United States and
Russia). From a normative perspective, these results should be viewed with some concern. In
democratic societies, the media are entrusted with the responsibility of serving as a watchdog for the
public interest and to scrutinize the movements of all three branches of government (executive,
legislative, and judicative). A media system in which citizens do not place their trust in will be hardly
able to watch over any authority or institution.
Consistent with previous theories and findings (Catterberg & Moreno, 2006; Inglehart, 1997), our
results suggest that both cultural and institutional explanations may lie behind the different levels of
social trust across societies. Cultural theories argue that advanced societies have witnessed the
emergence of ‘post-materialist’ values, according to which citizens place greater demands on
government and institutions (Inglehart, 1997). Post-materialists “place less emphasis on economic
growth and more emphasis on the non-economic quality of life” (Inglehart, 1997, p. 375), resulting in
loss of respect for authority and social trust (Tsfati & Ariely, 2014). However, these losses in social
trust are not necessarily negative, since they may be suggestive of the emergence of a ‘vigilant
skepticism’ by a more critical and politically sophisticated citizenry (Cook & Gronke, 2005; Hardin,
1999). In line with these ideas, our study shows that countries with lower HDI tend to score relatively
high in all dimensions of social trust. Conversely, countries with a higher HDI show relatively lower
levels of social trust.
Nonetheless, this pattern is not perfect, and results suggest that variables at the macro level—other
than post-materialist values—drive social trust patterns in these data. Thus, political trust (including
trust in government, governing bodies, and trust in security institutions) tends to show higher values
in countries in the first cluster than in those in the second and third cluster. One complementary
explanation for this could be the so-called ‘post-honeymoon’ effect (Catterberg & Moreno, 2006).
While the arrival of democracy usually results in enhanced levels of social trust (especially trust in
government), this trend commonly reverses after some years or decades of citizens’ unmet
expectations. In many cases, social and political institutions are not in a position to meet the
tremendous expectations of citizens regarding civic, political, and economic rights, resulting in lower
levels of social trust (Catterberg & Moreno, 2006). Thus, performance of social institutions, and not
only cultural values, do matter in explaining cross-country differences in trust in institutional actors.
The findings of this study have to be interpreted with caution due to a number of caveats and limitations
to consider. First, we focused on the development of a multi-dimensional model of social trust and
thus we did not include any predictor or outcome variable. Therefore, our assessments about the role
of ‘post-materialist values’ or ‘post-honeymoon effects’ are only post-hoc speculations. Further studies
should use our proposed scale of social trust to better explore its antecedents and outcomes both at the
micro and macro levels. To this end, future research should conduct multi-level analyses considering
not only individual attributes (demographics, sociopolitical antecedents, news media use…) but also
cultural values (e.g. post-materialism), and differences in institutions’ performance that may better
predict trust in institutional actors across countries. Another qualification comes from the use of an
online survey. Although in most countries our samples are comparable to the National Census in terms
of age, sex, education, and income (see Appendix), participants were not randomly selected from the
general population, but from an opt-in panel. Finally, in two cases (South Africa and India), our
samples are not representative of the whole country, but only of the most populated city (Johannesburg
and Delhi, respectively).

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 246
Despite these limitations, this study makes theoretical and empirical contributions to the study of social
trust. In brief, we tested a multidimensional model of social trust—which includes trust in government,
trust in governing bodies, trust in security institutions, trust in knowledge producers, and trust in the
media—across different societies. The model’s sub-dimensions were robust across countries, showing
more than acceptable reliability estimates in most cases. We also found important differences in social
trust in different societies, which we tried to explain (in a post-hoc way) with the help of cultural and
institutional theories.
Notes
[1] The Human Development Index (HDI) was created by the United Nations Development
Programme (UNDP) as an indicator to show the well-being of a country’s people, aside from using
economic growth alone as an indicator of how ‘well’ a country is faring, since a country’s economic
growth can often come at a great cost of the well-being of its citizens. The HDI indices include life
expectancy at birth, expected years of schooling, mean years of schooling, and gross national income
(GNI) per capita (UNDP, 2015).
[2] It is important to note that countries with higher levels of Internet penetration are less problematic
for web-only designs (Mohorko, Leeuw, Hox, 2013). Therefore, in countries with higher levels of
income inequality, results should be interpreted with caution.
7. References
Achterberg, P., de Koster, W. y van der Waal, J. (2015). A science confidence gap: Education, trust
in scientific methods, and trust in scientific institutions in the United States, 2014.
Public Understanding of Science, 26(6), 704-720. doi: 10.1177/0963662515617367
Allum, N., Sturgis, P., Tabourazi, D. y Brunton-Smith, I. (2008). Science knowledge and attitudes
across cultures: a meta-analysis. Public Understanding of Science, 17(1), 35-54. doi:
10.1177/0963662506070159
American Association of Public Opinion Research (2011). Standard definitions: Final dispositions of
case codes and outcome rates for surveys. Recuperado de:
http://aapor.org/Content/NavigationMenu/AboutAAPOR/StandardsampEthics/Standardefinitions/Sta
ndardDefinitions2011.pdf
Ardèvol-Abreu, A., y Gil de Zúñiga, H. (2016). Effects of editorial media bias perception and media
trust on the use of traditional, citizen, and social media news. Journalism & Mass Communication
Quarterly, 94(3), 703-724. doi: 10.1177/1077699016654684.
Aupers, S. (2012). ‘Trust no one’: Modernization, paranoia and conspiracy culture. European
Journal of Communication 27(1): 22-34. doi: 10.1177/0267323111433566
Bauer, M., Durant, J. y Evans, G. (1994). European public perceptions of science. International
Journal of Public Opinion Research, 6(2), 163-186. doi: 10.1093/ijpor/6.2.163

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 247
Behling, O. y Law, K. S. (2000). Translating questionnaires and other research instruments:
problems and solutions. Thousand Oaks, CA: SAGE.
Bennett, S. E., Rhine, S. L., Flickinger, R. S. y Bennett, L. M. (1999). ‘Video malaise’ revisited:
Public trust in the media and government. Harvard International Journal of Press/Politics, 1999,
4(4), 8-23. doi: 10.1177/1081180X9900400402
Boix, C. y Svolik, M. W. (2013). The foundations of limited authoritarian government: Institutions,
commitment, and power-sharing in dictatorships. The Journal of Politics, 75(2), 300-316. doi:
10.1017/S0022381613000029
Bosnjak, M., Das, M. y Lynn, P. (2016). Methods for probability-based online and mixed-mode
panels selected: recent trends and future perspectives. Social Science Computer Review, 34(1), 3-7.
doi: 10.1177/0894439315579246.
Brislin, R. W. (1980). Translation and content analysis of oral and written materials. In Harry C.
Triandis & John W. Berry (Eds.), Handbook of cross-cultural psychology: Vol. 2. Methodology (pp.
389-444). Boston, MA: Allyn and Bacon.
Callegaro, M., Baker, R. P., Bethlehem, J., Göritz, A. S., Krosnick, J. A. y Lavrakas, P. J. (Eds.)
(2014). Online panel research: a data quality perspective. Sussex, UK: John Wiley & Sons.
Catterberg, G. y Moreno, A. (2006). The individual bases of political trust: Trends in new
and established democracies. International Journal of Public Opinion Research, 18(1), 31-48
Citrin, J. (1974). Comment: the political relevance of trust in government. American Political
Science Review, 68(3), 973-988. doi: 10.2307/1959141
Coleman, J. S. (1990). Foundations of social choice theory. Cambridge, MA: Harvard University
Press.
Cook, T. E. y Gronke, P. (2005). The skeptical American: Revisiting the meanings of trust in
government and confidence in institutions. Journal of Politics, 67(3), 784-803. doi: 10.1111/j.1468-
2508.2005.00339.x
Dautrich, K. y Hartley, T. H. (1999). How the news media fail American voters. Causes,
consequences, and remedies. New York: Columbia University Press.
Directorate General of Budget, Accounting and Statistics (Executive Yuan, Taiwan) (2016).
Composite index and related indicators. Recuperado de
http://eng.stat.gov.tw/ct.asp?xItem=25280&ctNode=6032&mp=5
Druckman, J. N. (2001). On the limits of framing: Who can frame? Journal of Politics, 63, 1041-
1066. doi: 10.1111/0022-3816.00100
Durant, J. R., Evans, G. A. y Thomas, G. P. (1989). The public understanding of science. Nature,
340, 11-14. doi:10.1038/340011a0.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 248
Dyck, A. y Zingales, L. (2002). The Corporate Governance Role of the Media, in The Right to Tell-
The Role of Mass Media in Economic Development, (pp. 101-137). Washington, DC: The World
Bank Institute.
Easton, D. (1965). A systems analysis of political life. New York: Wiley.
Evans, G. A. y Durant, J. R. (1989). Understanding of science in Britain and the USA. In R. Jowell,
S. Witherspoon, & L. Brook (Eds.), British Social Attitudes, 6th report (pp. 105-129). Aldershot
(Hampshire): Gover.
Farnsworth, S. J. y Lichter, S. R. (2007). The nightly news nightmare: Television’s coverage of US
presidential elections, 1988-2004. Plymouth (UK): Rowman & Littlefield.
Furusten, S. (2013). Institutional Theory and Organizational Change: Edward Elgar Publishing,
Incorporated.
Gamson, W. A. (1968). Power and disconnect. Homewood, IL: Dorsey.
George, D. y Mallery, M. (2003). Using SPSS for Windows step by step: a simple guide and
reference. Boston, MA: Allyn & Bacon.
Habermas, J. (1989). The structural transformation of the public sphere. Cambridge: Polity Press
Hardin, R. (1999). Do we want trust in government? In M. E. Warren (Ed.), Democracy and trust,
(pp. 22-41). Cambridge: Cambridge University Press.
Harris, V. W., Skogrand, L. y Hatch, D. (2008). Role of friendship, trust, and love in strong latino
marriages. Marriage and Family Review, 44(4): 455-488. doi:10.1080/01494920802454041
Hetherington, M. J. (1998). The political relevance of political trust. American Political Science
Review, 92(4), 791-808. doi: 10.2307/2586304
Hetherington, M. J. (1999). The effect of political trust on the presidential vote, 1968-96. American
Political Science Review, 93(2), 311-326. doi: 10.2307/2585398
Inglehart, R. (1990). Culture shift in advanced industrial society. Princeton, NJ: Princeton University
Press.
Inglehart, R. (1997). Modernization and postmodernization: Cultural, economic, and political
change in 43 societies. Princeton, NJ: Princeton University Press.
Jackob, N. G. E. (2010). No alternatives? The relationship between perceived media dependency, use
of alternative information sources, and general trust in mass media. International Journal of
Communication, 4, 18, 589-606.
Jones, D. A. (2004). Why Americans don’t trust the media. A preliminary analysis. The Harvard
International Journal of Press/Politics, 9(2), 60-75. doi: 10.1177/1081180X04263461
Kohring, M. y Matthes, J. (2007). Trust in news media development and validation of a
multidimensional scale. Communication Research, 34(2), 231-252. doi: 10.1177/0093650206298071

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 249
Levi, M. (1998). A state of trust. In V. Braithwaite & M. Levi (Eds.), Trust and governance (pp. 77-
101). New York: Russell Sage Foundation.
Liu, Milojev, Gil de Zúñiga y Zhang, (2018). The Global Trust Inventory as a ‘Proxy Measure’ for
Social Capital: Measurement and Impact in 11 Democratic Societies. Journal of Cross-Cultural
Psychology.
Lorenz, E. (1999). Trust, contract and economic cooperation. Cambridge Journal of Economics,
23(3): 301-315. doi: 10.1093/cje/23.3.301
Luhmann, N. (2000). Familiarity, confidence, trust: problems and alternatives. In D. Gambetta (Ed.),
Trust: Making and breaking cooperative relations (pp. 94-107). Oxford: Basil Blackwell
Miller, A. H. (1974). Rejoinder to ‘Comment’ by Jack Citrin: political discontent or ritualism?
American Political Science Review, 68, 989-1001.
Miller, J. D. (1983): Scientific literacy: a conceptual and empirical review. Daedalus, 112(2), 29-48.
Miller, J. M. y Krosnick, J. A. (2000). News media impact on the ingredients of presidential
evaluations: Politically knowledgeable citizens are guided by a trusted source. American Journal of
Political Science, 44(2), 301-315. doi: 10.2307/2669312
Millstone, E. y van Zwanenberg, P. (2000). A crisis of trust: for science, scientists or for institutions?
Nature Medicine, 6(12), 1307-1308. doi:10.1038/82102
Mohorko, A., Leeuw, E. D. y Hox, J. (2013). Internet coverage and coverage bias in Europe:
Developments across countries and over time. Journal of Official Statistics, 29(4), 609-622. doi:
10.2478/jos-2013-0042.
Moy, P., McCluskey, M. R., McCoy, K. y Spratt, M. A. (2004). Political correlates of local news
media use. Journal of Communication, 54(3), 532-546. doi: 10.1111/j.1460-2466. 2004.tb02643.x
Pardo, R. y Calvo, F. (2002). Attitudes toward science among the European public: a methodological
analysis. Public Understanding of Science, 11(2), 155-195. doi: 10.1088/0963-6625/11/2/305
Putnam, R. (2000). Bowling alone: the collapse and revival of American community. New York:
Simon & Schuster.
Putnam, R. (Ed.) (2002). Democracies in flux. The evolution of social capital in contemporary
society. Oxford: Oxford University Press.
Schudson, M. (1978). Discovering the news. New York: Basic Books.
Sibley, C. G. y Liu, J. H. (2013). Relocating attitudes as components of representational profiles:
Mapping the epidemiology of bicultural policy attitudes using Latent Class Analysis. European
Journal of Social Psychology, 43, 160-174.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 250
Thato, S., Hanna, K. M. y Rodcumdee, B. (2005). Translation and validation of the condom self-
efficacy scale with Thai adolescents and young adults. Journal of Nursing Scholarship, 37(1), 36-40.
doi: 10.1111/j.1547-5069.2005. 00012.x
Tsfati, Y. (2003). Does audience skepticism of the media matter in agenda setting? Journal of
Broadcasting & Electronic Media, 47(2), 157-176.
Tsfati, Y. (2010). Online news exposure and trust in the mainstream media: Exploring possible
associations. American Behavioral Scientist, 54(1), 22-42. doi: 10.1177/0002764210376309
Tsfati, Y. y Ariely, G. (2014). Individual and Contextual Correlates of Trust in Media Across 44
Countries. Communication Research, 41(6), 760-782.
Tsfati, Y. y Cappella, J. N. (2003). Do people watch what they do not trust? Exploring the
association between news media skepticism and exposure. Communication Research, 30(5), 504-
529.
Programa de las Naciones Unidas para el Desarrollo (PNUD) (2015). Human development report.
Work for human development. Recuperado de
http://hdr.undp.org/sites/default/files/2015_human_development_report_1.pdf
Ziman, J. (1991). Public understanding of science. Science, Technology and Human Values, 16, (1),
99-105. doi: 10.1177/016224399101600106
Ziman, J. (1992). Not knowing, needing to know, and wanting to know. In B. V. Lewenstein (Ed.),
When science meets the public (pp. 13-20). Washington, DC: American Association for the
Advancement of Science.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 251
Tables
Table 1
Comparison of Clusters of Countries Based on their Human Development Index (HDI), Sizes and Means for Evaluation Fields
Cluster #
1
(N = 10)
2
(N = 4)
3
(N = 4)
4
(N = 4)
Mean HDI
.89
.83
.75
.66
Proportional Size
45.5%
18.2%
18.2%
18.2%
Evaluation fields
Trust in the Government
2.88
2.72
2.84
3.47
Trust in Governing Bodies
3.38
2.89
2.90
3.66
Trust in Security Institutions
4.02
3.54
3.68
3.89
Trust in Knowledge Producers
4.32
4.50
4.45
4.72
Trust in the Media
3.33
3.56
3.57
4.14

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 252
Table 2
Analysis of Variance (One-Way ANOVA) for Testing Differences in Trust Levels Between Countries According to their Human Development Index
Notes. Countries have been grouped according to their scores on the Human Development Index (United Nations, 2015). Highest (1-25) comprises
Germany, United States, New Zealand, United Kingdom, Korea, Japan, Taiwan, Spain, Italy, and Estonia. Very High (25-49) comprises Poland, Argentina,
Chile, and Russia. High (50-105) comprises Turkey, Brazil, Ukraine, and China. Medium (106-130) comprises Indonesia, Philippines, South Africa, and
India
HDI
Government Trust
Trust in Governing
Bodies
Trust in Security
Institutions
Trust in Knowledge
Producers
Trust in the Media
Highest (1-25)
F(9, 10414) = 106.96***
F(9, 10289) =
130.95***
F(9, 10376) =
144.79***
F(9, 10496) = 158.97***
F(9, 10483) =
105.00***
Very High (25-
49)
F(3, 4234) = 132.29***
F(3, 4132) = 24.98***
F(3, 4173) = 83.04***
F(3, 4266) = 98.35***
F(3, 4283) = 55.07***
High (50-105)
F(3, 4181) = 475.27***
F(3, 4121) = 199.42***
F(3, 4148) = 252.31***
F(3, 4218) = 17.60***
F(3, 4247) = 34.09***
Medium (106-
130)
F(3, 2767) = 338.36***
F(3, 2745) = 205.67***
F(3, 2801) = 209.79***
F(3, 2792) = 88.29***
F(3, 2814) = 169.48***

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 253
Table 3
Descriptive and Reliability Statistics for Dimensions of Trust
Government Trust Trust Gov. Bodies Trust Sec. Instit. Trust Knowledge Prod. Media Trust
M
SD
α
M
SD
α
M
SD
α
M
SD
α
M
SD
α
N
All
2.92
1.44
.87
3.21
1.41
.88
3.84
1.54
.74
4.43
1.42
.87
3.51
1.12
.77
21,781
Highest
Germany
3.20
1.51
.94
3.58
1.34
.83
3.98
1.48
.80
4.42
1.41
.92
3.45
1.12
.68
1,045
United States
2.96
1.31
.80
3.29
1.29
.85
4.49
1.44
.77
4.32
1.56
.86
3.03
1.09
.72
1,152
New Zealand
3.19
1.35
.85
3.97
1.32
.86
4.65
1.34
.74
4.59
1.28
.86
3.21
1.01
.71
1,149
United Kingdom
2.94
1.43
.91
3.66
1.41
.87
4.20
1.44
.69
4.44
1.33
.89
3.04
1.13
.73
1,058
Korea (South)
2.86
1.47
.89
3.02
1.34
.89
3.48
1.40
.74
3.80
1.27
.79
3.58
1.09
.80
921
Japan
2.90
1.16
.88
3.37
1.17
.88
3.81
1.28
.76
3.73
1.22
.92
3.33
0.95
.81
968
*Taiwan
2.37
1.13
.88
2.79
1.21
.91
3.01
1.24
.72
3.41
1.33
.84
2.63
1.02
.86
994
Spain
2.31
1.20
.85
3.03
1.31
.84
3.88
1.56
.78
4.88
1.41
.88
3.64
0.98
.70
1,009
Italy
2.41
1.33
.90
2.99
1.35
.87
4.02
1.62
.87
4.52
1.47
.86
3.68
1.09
.76
1,031
Estonia
3.62
1.17
.78
4.13
1.32
.88
4.69
1.24
.76
5.06
1.08
.88
3.61
0.86
.67
1,158
Very High
Poland
2.41
1.28
.88
2.98
1.32
.86
3.68
1.42
.73
4.14
1.43
.91
3.40
1.11
.77
1,052
Argentina
2.42
1.36
.83
2.59
1.24
.82
2.99
1.37
.69
5.11
1.37
.87
3.63
0.97
.71
1,139
Chile
2.66
1.23
.81
2.95
1.30
.82
3.97
1.56
.71
4.37
1.42
.80
3.86
0.97
.63
959
Russia
3.41
1.46
.86
3.03
1.42
.89
3.51
1.49
.69
4.38
1.45
.89
3.36
1.12
.79
1,131
High
Turkey
3.02
1.56
.88
3.12
1.57
.88
4.47
1.70
.73
4.72
1.46
.80
3.25
1.16
.76
938
Brazil
1.88
1.21
.88
2.54
1.28
.83
3.00
1.49
.70
4.27
1.60
.87
3.62
1.23
.79
1,083
Ukraine
2.43
1.26
.87
2.30
1.17
.88
3.07
1.40
.59
4.46
1.31
.87
3.72
1.06
.77
1,202
China
4.05
1.43
.87
3.63
1.46
.94
4.17
1.45
.78
4.37
1.36
.84
3.67
1.22
.84
997

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 254
Medium
Indonesia
3.50
1.26
.86
3.31
1.26
.91
3.81
1.25
.64
4.61
1.22
.86
3.84
0.98
.85
1,057
Philippines
3.43
1.22
.86
3.45
1.18
.88
3.65
1.35
--
4.57
1.22
.84
4.05
1.00
.80
1,032
South Africa (Joh.)
2.00
0.93
.64
2.86
1.24
.81
2.85
1.26
.73
4.10
1.45
.83
3.54
1.07
.75
381
India (Delhi)
4.93
1.32
.83
5.03
1.26
.87
5.25
1.11
.24
5.60
1.18
.81
5.13
1.09
.81
325
Notes. Countries have been grouped according to their scores on the Human Development Index (United Nations, 2015). * Not a member of the UN:
2015 HDI calculated by the Taiwanese government (Directorate General of Budget, Accounting and Statistics, 2016). All items measured on 7-point
scales, from 1 = ‘do not trust at all’ to 7 = ‘trust completely’. The index for trust in security institutions in Philippines was created with 1 variable
instead of 2, since the question about the respondents’ levels of trust in the military was not asked in this country.
Table 4
T-tests for differences between countries for each dimension of trust and overall (grand) means
Government Trust Trust Gov. Bodies Trust Sec. Instit. Trust Knowledge Prod. Media Trust
mcountry
mcountry -
M
Sig
mcountry
mcountry -
M
Sig
mcountry
mcountry -
M
Sig
mcountry
mcountry -
M
Sig
mcountry
mcountry -
M
Sig
Highest
Germany
3.20
0.286
+
3.58
0.380
+
3.98
0.135
4.42
-0.010
3.45
-0.060
United States
2.96
0.046
3.29
0.083
4.49
0.652
+
4.32
-0.111
3.03
-0.474
-
New Zealand
3.19
0.278
+
3.97
0.761
+
4.65
0.812
+
4.59
0.160
+
3.21
-0.301
-
United
Kingdom
2.94
0.025
3.66
0.451
+
4.20
0.356
+
4.44
0.003
3.04
-0.360
-
Korea (South)
2.86
-0.057
3.02
-0.189
-
3.48
-0.357
-
3.80
-0.631
-
3.58
0.069
Japan
2.90
-0.015
3.37
0.162
+
3.81
-0.029
3.73
-0.703
-
3.33
-0.178
-
*Taiwan
2.37
-0.909
-
2.79
-0.514
-
3.01
-0.834
-
3.41
-1.018
-
2.63
-0.876
-

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 255
Spain
2.31
-0.604
-
3.03
-0.172
-
3.88
0.041
4.88
0.450
+
3.64
0.126
+
Italy
2.41
-0.502
+
2.99
-0.220
+
4.02
0.188
4.52
0.084
3.68
0.165
+
Estonia
3.62
0.700
+
4.13
0.924
+
4.69
0.845
+
5.06
0.627
+
3.61
0.102
+
Very High
Poland
2.41
-0.507
-
2.98
-0.221
-
3.68
-0.166
-
4.14
-0.296
-
3.40
-0.114
-
Argentina
2.42
-0.495
-
2.59
-0.612
-
2.99
-0.850
-
5.11
0.677
+
3.63
0.117
+
Chile
2.66
-0.255
-
2.95
-0.253
-
3.97
0.136
4.37
-0.066
3.86
0.350
+
Russia
3.41
0.489
+
3.03
-0.173
-
3.51
-0.366
-
4.38
0.051
3.36
-0.150
-
High
Turkey
3.02
0.101
3.12
-0.091
4.47
0.624
+
4.72
0.282
+
3.25
-0.265
-
Brazil
1.88
-1.031
-
2.54
-0.666
-
3.00
-0.837
-
4.27
-0.163
-
3.62
0.103
+
Ukraine
2.43
-0.479
-
2.30
-0.903
-
3.07
-0.767
-
4.46
0.030
3.72
0.208
+
China
4.05
1.14
+
3.63
0.423
+
4.17
0.331
+
4.37
-0.066
3.67
0.157
+
Medium
Indonesia
3.50
0.584
+
3.31
0.103
3.81
-0.032
4.61
0.173
+
3.84
0.330
+
Philippines
3.43
0.515
+
3.45
0.249
+
3.65
-0.193
-
4.57
0.134
+
4.05
0.540
+
South Africa
(Joh.)
2.00
-0.909
-
2.86
-0.343
-
2.85
-0.990
-
4.10
-0.331
-
3.54
0.030
India (Delhi)
4.93
2.017
+
5.03
1.824
+
5.25
1.404
+
5.60
1.166
+
5.13
1.617
+
Notes. Countries have been grouped according to their scores on the Human Development Index (United Nations, 2015). * Not a member of the UN: 2015
HDI calculated by the Taiwanese government (Directorate General of Budget, Accounting and Statistics, 2016). The index for trust in security institutions
in Philippines was created with 1 variable instead of 2, since the question about the respondents’ levels of trust in the military was not asked in this country.
Bonferroni-corrected p-values for 22 comparisons (two-tailed). +: Mean value significantly higher that the grand mean at the level p < .05 or better. -:
Mean value significantly lower that the grand mean at the level p < .05 or better. N = 21,781.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 256
Table 5
Post-hoc Comparisons for Between-Groups Differences for Each Dimension of Trust.
Government Trust Trust Gov. Bodies Trust Sec. Instit. Trust Knowedge Prod. Media Trust
mcountry
Higher
than1
Lower
than1
mcountry
Higher
than1
Lower
than1
mcountry
Higher
than1
Lower
than1
mcountry
Higher
than1
Lower
than1
mcountry
Higher
than1
Lower
than1
Highest
Germany (g)
3.20
u e k j t s i
o
3.58
u k j t s i
n o
3.98
k t
u n e o
4.42
u k j t
s o
3.45
u n e t
s i o
United States (u)
2.96
t s i
g n o
3.29
k t s i
g n e o
4.49
g e k j t s i
o
4.32
k j t
n e i o
3.03
t
(all but t)
New Zealand (n)
3.19
u e k j t s i
o
3.97
g i j k s t e u
(none)
4.65
g e k j t s i
(none)
4.59
u k j t
s o
3.21
u t
g k s i o
U. K. (e)
2.94
t s i
g n o
3.66
u k j t s i
n o
4.20
g k j t s
u n o
4.44
k j t
s o
3.04
t
g k j s i o
Korea (k)
2.86
t s i
g n o
3.02
t
g n e j o
3.48
t
(all but t)
3.80
t
g u n e s i o
3.58
u n e j t
(none)
Japan (j)
2.90
t s i
g n o
3.37
k t s i
g n e o
3.81
k t
u n e i o
3.73
t
g u n e s i o
3.33
u e t
k s i o
*Taiwan (t)
2.37
(none)
g u n e k j o
2.79
(none)
(all)
3.01
(none)
(all)
3.41
(none)
(all)
2.63
(none)
(all)
Spain (s)
2.31
(none)
g u n e k j o
3.03
t
g u n e j o
3.88
k t
u n e o
4.88
(all but o)
(none)
3.64
g u n e j t
(none)
Italy (i)
2.41
(none)
g u n e k j o
2.99
t
g u n e j o
4.02
k t
u n o
4.52
u k j t
s o
3.68
g u n e j t
(none)
Estonia (o)
3.62
(all)
(none)
4.13
g u e k j t s i
(none)
4.69
(all but n)
(none)
5.06
(all but s)
(none)
3.61
g u n e j t
(none)
Very High
Poland (p)
2.41
(none)
c r
2.98
a
(none)
3.68
a r
c
4.14
(none)
(all)
3.40
(none)
a c
Argentina (a)
2.42
(none)
c r
2.59
(none)
(all)
2.99
(none)
(all)
5.11
(all)
(none)
3.63
p r
c
Chile (c)
2.66
p a
r
2.95
a
(none)
3.97
(all)
(none)
4.37
p
a
3.86
(all)
(none)
Russia (r)
3.41
(all)
(none)
3.03
a
(none)
3.51
a
p c
4.38
p
a
3.36
(none)
a c
High
Turkey (y)
3.02
b x
h
3.12
b x
h
4.47
(all)
(none)
4.72
(all)
(none)
3.25
(none)
(all)
Brazil (b)
1.88
(none)
(all)
2.54
x
y h
3.00
(none)
y h
4.27
(none)
y x
3.62
y
(none)
Ukraine (x)
2.43
b
y h
2.30
(none)
(all)
3.07
(none)
y h
4.46
b
y
3.72
y
(none)
China (h)
4.05
(all)
(none)
3.63
(all)
(none)
4.17
b x
y
4.37
(none)
y
3.67
y
(none)

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 257
Notes. Countries have been grouped according to their scores on the Human Development Index (United Nations, 2015). * Not a member of the UN:
2015 HDI calculated by the Taiwanese government (Directorate General of Budget, Accounting and Statistics, 2016). (1) Post-hoc Bonferroni adjusted
comparisons between-group. Subscripts indicate mean values significantly different at the level p < .05 or better. N = 21,781.
Appendix for Data Collection Demographics
Table 6
Demographic Breakdown by Age, Gender and Race for 22 Country Study versus Census Data*
Age Group
Gender
Race
18-24
25-34
35-44
45-64
65+
Female
Male
Asian
Black
White
1.
Argentina
15.2(17.3)
24(21.4)
20.8(17.6)
34.2(28.4)
5.8(15.3)
51.7(53.1)
48.3(46.9)
--
--
71.7
2.
Brazil
5.7(8.7)
29.4(15.7)
29.4(15.7)
20(13.5)
3.7(13)
49.8(51.4)
50.2(48.6)
1.7(.5)
12.6(7.9)
68.1(46.2)
3.
Chile
26.3(14.8)
30(21.1)
19.7(18.4)
20.7(32.1)
3.2(13.7)
51.3 (51)
48.7 (49)
--
--
--
4.
China
10.5(12.7)
31.5(14.9)
27.9(18.2)
27.2(24.3)
2.9(8.9)
44.4(48.8)
55.6(51.2)
--
--
--
5.
Estonia
11.1(9.7)
17.8(17.9)
15.1(17)
33(32.4)
22(23)
54.3(48.2)
50.6(45.7)
--
--
97.8(68.2)
6.
Germany
11(6.2)
26(15)
43.8(24.6)
8.3(5.1)
10.9(17)
53.9(51)
46.1(49)
--
--
--
Medium
Indonesia (y)
3.50
f
d
3.31
f
l d
3.81
l f
d
4.61
f
d
3.84
f
l d
Philippines (l)
3.43
f
d
3.45
y f
d
3.65
f
y d
4.57
f
d
4.05
y f
d
South Africa (f)
2.00
(none)
(all)
2.86
(none)
(all)
2.85
(none)
(all)
4.10
(none)
(all)
3.54
(none)
(all)
India (d)
4.93
(all)
(none)
5.03
(all)
(none)
5.25
(all)
(none)
5.60
(all)
(none)
5.13
(all)
(none)

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 258
7.
India
17.6(21.8)
41.5(27.6)
26.5(21.5)
14.1(22.9)
.3(6)
38(46.4)
62(53.5)
--
--
--
8.
Indonesia
19.1(12.5)
36.9(24.3)
26.2(21)
13(24.2)
.6(4.8)
59.6(49.9)
38.9(50.1)
76.2(40.2)
9.
Italy
10.9(7.1)
21.9(11.5)
27.9(15.1)
34.4(28.7)
5(21.9)
54.8(51.5)
44.2(48.5)
--
--
--
10.
Japan
4.1(5.9)
13.4(13.9)
26.7(17.8)
45(32)
10.9(30)
41.6(51.3)
57.1(48.7)
99.3(98.6)
11.
Korea
16.7(11.5)
24.4(16.1)
24.3(19.6)
31.7(36.8)
2.8(15.9)
46.7(46.2)
53.3(53.8)
--
--
--
12.
N. Zealand
7.1(9.4)
13.2(16.6)
15.2(18.6)
36.7(35.5)
24(19.7)
56(52.1)
43.2(47.8)
7.8(11.6)
--
77(75.1)
13.
Philippines
17.7(9.2)
35.3(16.1)
25.9(12.4)
15.8(15.9)
1.3(4.8)
49.7(61.2)
39(50.2)
--
--
--
14.
Poland
13.9(10.7)
21.4(19.6)
22.6(18.1)
34.1(33)
8(18.6)
54(52.3)
46(47.7)
--
--
--
15.
Russia
18(13.6)
24.2(19.7)
26(16.6)
28.6(34.3)
2.5(15.6)
50.2(53.8)
48.4(46.2)
--
--
--
16.
S. Africa
10.2(10.4)
31.5(17.6)
23.5(12.4)
28.4(15.6)
2.9(5.3)
61.2(51.3)
37.2(48.7)
--
15(88.1)
45.8(8.9)
17.
Spain
11.7(7.4)
21.9(14.9)
26.4(16.9)
36.8(25.6)
2.9(17.3)
51.7(50.6)
46.5(49.3)
--
--
--
18.
Taiwan
15.4(15.5)
30.6(17.7)
30.6(18.7)
22.6(34.1)
1(13.9)
49.2(50.1)
50.8(49.9)
--
--
--
19.
Turkey
20.
UK
4.3(8.7)
12.8(17.7)
17.6(16.9)
42.7(33.4)
22.6(23)
54.1(51.4)
45.9(48.6)
3.1(6.9)
1.2(2.9)
91.9(87.6)
22.
Ukraine
13(7.8)
38.6(19.8)
26.6(17.5)
14.8(25.5)
1(19)
44(54.8)
54.9(45.1)
--
--
86.1(83)
22.
US
8.4(9.9)
13.5(13.6)
14.8(12.8)
42.7(26.2)
20(15.5)
59.5(50.8)
40.5(49.2)
3.5(5)
5.8(12.6)
83.3(73.8)
*Note: Census data reported in parenthesis, based on official estimates. Dashes indicate demographics not directly comparable. See
below for notes.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 259
Table 7
Demographic Breakdown by Education, Homeownership and Marital Status for 22 Country Study versus Census Data*
Education
Homeownership
Marital Status
High School or
less
Some
College
College
Degree+
Graduate
Degree+
Own
Rent
Married
Divorced
Single
Widowed
1.
Argentina
54(85)
13.1(9.4)
26.7(5.7)
--
--
--
53.2(52.8)
12.2(10.6)
32.4(28)
2.2(8.5)
2.
Brazil
52.2(39.4)
47.8(60.5)
--
--
--
--
--
--
--
--
3.
Chile
22.8(80.6)
44.2(12)
33(16.6)
--
62.1(80.6)
37.9(19.4)
44.7(44.3)
8.3(3.1)
46.3(47.2)
.7(5.4)
4.
China
9.3(15)
23(5.5)
58.7(3.7)
7.6(.3)
88.9(85.4)
11.1(11.9)
76.2(71.3)
1.4(1.4)
21.8(21.6)
.6(5.7)
5.
Estonia
44.6(64)
16.5(9.4)
14.5(7.8)
24.2(17.2)
--
-
--
--
--
--
6.
Germany
60.9(85.3)
--
7.2(1.3)
31.8(14.5)
44.1(41.3)
55.9(48.6)
54.5(54.8)
19.5(8.5)
21.4(28.2)
4.6(8.5)
7.
India
6.5(75.4)
4.3(10.6)
89.2(16.8)
--
--
--
72.3(50.42)
3.7(.3)
24(49.2)
--
8.
Indonesia
25.7(41.6)
13.1(29.2)
53.9(18.2)
4.7(10.9)
--
--
--
--
--
--
9.
Italy
52(49.7)
--
31.2(13.5)
--
79.3(72)
20.7(18)
56(48.4)
5.1(2.2)
37.5(41.9)
1.4(7.5)
10.
Japan
44.3(62.3)
14.4(16.4)
33.9(19.5)
7.4(1.8)
--
--
--
--
--
--
11.
Korea
31.8(56.5)
11.6(14.3)
56.6(29.3)
--
59.4(53.8)
40.6(46.2)
51.5(60.8)
2.5(4.2)
45(26.9)
--
12.
N. Zealand
33.5(38.2)
28.3(8.2)
24.4(12.1)
13.7(5.7)
--
--
--
--
--
--
13.
Philippines
5.5(7.1)
--
70.2(3.5)
--
66(61.6)
34(12.1)
50.3(45.3)
4.3(1.2)
43.2(43.5)
2.2(4.2)
14.
Poland
48.8(79.4)
15.4(7.6)
35.8(13)
--
80.5(83.5)
19.5(16.5)
67(57.7)
7.5(5)
22(27.8)
3.5(9.5)

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 260
15.
Russia
25.4(64)
10.6(4.2)
63.9(30.9)
3.5(1)
--
--
56.2(49.7)
6.3(8.3)
18.4(20.7)
--
16.
S. Africa
32.6(87.6)
--
45.3(12.1)
--
--
--
--
--
--
--
17.
Spain
18.6(46)
44.1(22.1)
37(31.9)
--
77.7(79.7)
21.4(20.3)
62.4(54.6)
6.4(5.2)
29.6(32.4)
1.3(7.6)
18.
Taiwan
21.9(57)
18.2(12.2)
46.1(24.6)
13.8(6.3)
70.1(84)
29.9(16)
41.6(51.1)
4(7.9)
50.6(34.7)
.3(6.3)
19.
Turkey
20.
UK
30.2(29.3)
31.9(20.5)
38(27)
--
65.1(64.8)
35.2(34.8)
48.5(41.5)
11.6(6.6)
31.7(46.4)
3.7(5.2)
22.
Ukraine
13.7(56.5)
--
31(20.7)
61.7(14.6)
--
--
--
--
--
--
22.
US
22.8(40.8)
33.5(29.1)
28.3(18.7)
15.4(11.4)
67.9(63.1)
32.1(36.9)
50.9(47.7)
12.9(11)
33.3(27)
(5.9)5.9
*Note: Census data reported in parenthesis, based on official estimates. Dashes indicate demographics not directly comparable. See below
for notes.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 261
8. Appendix
Footnotes on Demographic Breakdown of Country Studies
1.
Argentina
2014 World Values Survey. Other Race = Mestizo. Yearly income reported versus Pew 2013: $19,999 or less 73.3(31.7); 20,000 – 49,000
21.3(40.3); 50,000-99,000 4.8(19); 100,000 or more .6(9).
2.
Brazil
2013 Brazilian Census data. Numbers for age groups 15-19, 20-29, 30-39, 40-49, 50-59, and 60+. *Census numbers for Brown/Indigenous (45.3%)
categories were not recorded in the first wave, and were instead asked in the study as Latino (7.8%); Other = American and Pacific Islander.
Language in the census differs from the study on race and education items: High School = High School or less, Some College = High School +.
The Brazilian Census the information available is related to the level that people are studying at the moment. Yearly income categories reported
as: less than R$50,000 52.1(79.6); R$50, 000-100,000 16.3(6.2); R$100,000+ 13.2(3.1).
3.
Chile
2015 population estimates based on INE data.
4.
China
2010 Chinese Census made by China’s Office for National Statistics.
5.
Estonia
2015 population estimates for age and gender; 2011 for ethnicity and citizenship, 2014 for education levels. White = Estonian (official estimates
report Russian as 26.1% versus 1% in the study.
6.
Germany
2014 Satista estimates. Age categories are 18-24, 25-39, 40-59, 60-64, and 65+.
7.
India
2011 Delhi population estimates. Some College = 12-year Intermediate education.
8.
Indonesia
2010 BPS estimates. Asian = Java
9.
Italy
2015 ISTAT estimates.
10.
Japan
2010-2014 Japanese Census Estimates. Asian = Japanese; Other = Korean, Chinese, or Other. Yearly income categories reported as: 1.5 million
yen or less 13.3(10.6); 1.5-3.5 million 28.5(24.3); 3.5-7 million 31.7(38); 7-11 million 18(17.8); over 11 million 8.3(9.3).
11.
Korea
2015 population statistics from 2015 resident registration at the Ministry of Government Administration and Home Affairs; 2012 Korea Housing
Survey; and 2010 census.
12.
New Zealand
2013 NZ census. In age groups 18- 24 = 20-24. White= European; Other = Maori 4.8(12) and Pacific 1.5(5.7). Yearly income categories reported
as: $50,000 or less 41.7(32.9); 50,001-150,000 34.7(40.97); over 150,000 2.8(10.1).
13.
Philippines
2015 population estimates. In age groups 18- 24 = 20-24.
14.
Poland
Population estimates for 2011 and 2014 by GUS or Eurostat 2012.
15.
Russia
2010 census estimates.

RLCS, Revista Latina de Comunicación Social, 74 – Pages 237 to 262
[Funded Research] | DOI:10.4185/RLCS-2019-1329-12en |ISSN 1138-5820 | Year 2019
http://www.revistalatinacs.org/074paper/1329/12en.html Pages 262
Census data reported in parenthesis, based on official estimates. Dashes indicate demographics not directly comparable.
___________________________________________________________________
How to cite this article in bibliographies / References
H Gil de Zúñiga, A Ardèvol-Abreu, T Diehl, M Gómez Patiño, J H Liu (2019): “Trust in Institutional Actors across 22 Countries Examining Political, Science, and
Media Trust Around the World”. Revista Latina de Comunicación Social, 74, pp. 237 to 262.
http://www.revistalatinacs.org/074paper/1329/12en.html
DOI: 10.4185/RLCS-2019-1329en
Article received on 30 November 2018. Accepted on 21 January.
Published on 12 February 2019
16.
South Africa
2011 Census in Brief (Statistics South Africa) and Household Income and Expenditure Patterns in South Africa, 2011 (UNISA). Yearly income
categories reported as: Poor (R0-R54,344) 12.2(9.9); Low middle class (R54,345-R151,727) 13(18.7); Emerging middle class ( R151,278-
R363,930) 30.2(22.4); Realized middle class 14.3(17.7); Upper middle class 4.7(10.7); Emerging affluent or Affluent 5.5(20.6).
17.
Spain
2011 Population Census made by the Spanish Statistical Office (INE); 2011 European Union Statistics in Income and Living Conditions (EU-
SILC); 2011 Labor Force Survey (EPA).
18.
Taiwan
2014 Department of Statistics, Ministry of Interior.
19.
Turkey
20.
UK
2014 UK Census (ONS) estimates for age, homeownership and marital status, otherwise 2011 Census data is used.
21.
Ukraine
2001 Official census data. White = Ukrainian; Russian = 10.9(17.3).
22.
US
2014 U.S. Census American Community Survey (1-Year Estimates); Census asks about Hispanic (16.9%) ethnicity in a separate question, the study
offered Latino (5.1%) as an exclusive option in a single race item.