Do Amoral Familism and Political Distrust Really Affect North

Abstract

Especially in southern Italy, Banfield’s amoral familism is considered an obstacle to the formation of associations and growth of political participation. This article discusses Banfield’s concept, showing that it has been vulgarized merely as familism and, in particular, demonstrates that Banfield intended amoral familism to be understood in terms of political distrust. We investigated whether amoral familism or political distrust, operationalized as an individual difference variable, mediated the relationships between personality traits, personal values, and conventional and unconventional political acts, controlling for differences in political attitude. We recruited 405 participants, distributed across north, central, and southern Italy, to complete a questionnaire on political participation that also assessed Big Five personality factors, values, sociability and political attitude (expertise, interest, self-efficacy), and a new scale assessing amoral familism as a form of political distrust. Regression analyses were used to identify the best predictors of political acts, then structural equation modeling was used to test a model of political participation. Like political attitudes, familism mediated the relationships between personality traits, especially “openness to experience” and “taking conventional and unconventional political acts.” However, our data do not confirm the stereotype that northern and southern Italians differ in their tendency to amoral familism as defined by Banfield.

Keywords

cultural psychology ethnic identity amoral familism political distrust political participation and social capital personality and politics urbanization and familism

Through associations, political participation is thought be the cornerstone of social development. Specifically, in political science, connectedness and associative life are used as predictors of educational welfare, safe and productive neighborhoods, economic prosperity, health and happiness, as well as being recognized as factors that shape political life and democratic rules (e.g., Putnam, 2000). Increasing connectedness may bring about the preconditions for democracy and democratization (e.g., Christakis & Fowler, 2009; Foschi & Lauriola, 2014; Putnam, 2000).

Since Levinson (1958), a great deal of research has investigated the interplay between personality, sociocultural opportunities, and attitudes as determinants of political behavior (for a recent review, see Gallego & Oberski, 2012). Amoral familism, because it is related to sociocultural factors, is considered a critical impediment to social and political development (Alesina & Giuliano, 2010, 2011; Alesina & Ichino, 2009). The construct of amoral familism originated in Banfield’s (1958) studies showing that the social and political backwardness of southern Italian regions in the late 1950s was related to culturally shared amoral familistic tendencies. Banfield viewed amoral familism as a reaction to the lack of institutions and political life in economically depressed territories. Banfield argued that amoral familism led to political and civic disengagement and hence to greater poverty, high rates of illiteracy, conservatism, disorganization of institutions, and a general sense of fatalism. Banfield claimed that amoral familists’ first aim was to maximize the advantage of his or her nuclear family, positing that this general rule was also followed by others. The amoral familist is not confident that the democratic structure based on rules and institutions serves the family’s interest; he or she does not trust the public, only family members.

Banfield’s concept of amoral familism is more specific than the familism concept currently used in social sciences and psychology. In fact, a significant body of research has been developing in recent years concerning familism in a cultural, sociological, and evolutionary framework (especially in Latin American cultures, Eastern cultures, Asia and even Africa). This has prioritized an approach that considers familism as a cultural trait reflected in social values, socialization goals, developmental goals, protective factors for family violence, suicide, school dropout, and psychopathology (e.g., see Figueredo et al., 2001; Keeler, Siegel, & Alvaro, 2014; Stein et al., 2014).

Banfield (1958) also expanded on his explanatory model to account for the underdevelopment and socioeconomic hardship seen in America’s ghettos. This increased the popularity of the concept of amoral familism in various countries and academic fields, from political sociology to social psychology. Over the years, however, Banfield’s concept of amoral familism has been vulgarized as mere familism. Actually, for Banfield, the problem is the sense of political distrust associated with familism that makes it “amoral” and “maintains” the backwardness of societies, as he considered those of the southern Mediterranean. According to Banfield (1958), the political distrust was the way in which amoral familism shaped a variety of political behaviors. It must be emphasized that despite its potential relevance to research, political distrust is not yet viewed as a central concept of political psychology and, to the best of our knowledge, there are no scales for assessing political distrust (see also Schyns & Koop, 2010).

The concept of amoral familism has attracted a lot of criticism in Italy, on the grounds that other factors (e.g., cultural, historical, institutional, and socioeconomic factors) can explain the underdevelopment of the southern regions just as well as a lack of collective action by the inhabitants (for a review, see Ferragina, 2009, 2011). Despite this criticism, amoral familism is still a popular concept in political economy (e.g., Alesina & Giuliano, 2010, 2011).

Personality and Political Participation

In psychology, neither political distrust nor amoral familism have been considered as individual-level variables that might undermine one’s willingness to join formal and informal political associations. However there is, in psychology, a growing consensus that traits, cognitive styles, values, and motives are fundamental personality variables that affect political behavior and political participation (e.g., Foschi & Lauriola, 2014; Mondak, 2010; Vecchione & Caprara, 2009).

There has been a resurgence of interest in connecting research on personality and individual differences to social phenomena in a robust way. In particular, personality traits seem to be important for processes of exposure, selection, evocation, and manipulation of the environments (cf. Buss, 2009). In political psychology, the five-factor model has provided a framework for studying political behavior, and the Big Five personality traits are currently used to describe political orientation and participation in politics (Mondak, 2010). One of the most frequently debated hypotheses is that extraversion is more highly correlated with “virtually all aspects of group-based political participation” (Mondak & Halperin, 2008, p. 344) than other personality factor; however, recent studies have revealed that openness to experience also predicts political participation (Foschi & Lauriola, 2014; Mondak, 2010) and that the contribution of the two traits is similar (Vecchione & Caprara, 2009).

The associations between personality traits and political participation are often indirect, that is to say, they are mediated by a variety of factors such as political interest, values, sociability, and self-efficacy. These indirect associations have been formally tested through multivariate mediation analyses using structural equation modeling (Foschi & Lauriola, 2014; Gallego & Oberski, 2012; Vecchione & Caprara, 2009; Vecchione et al., 2015). It now seems to be generally accepted that openness and extraversion influence political behavior. However, none of the social aspects of extraversion and openness assessed by existing personality inventories captures political sociability, or amoral familism/political distrust, as reflected in participation in well-structured groups with collective goals (Foschi & Lauriola, 2014).

In view of its empirical importance, amoral familism/political distrust seems a good candidate for a political attitude that can account for distal effects of personality traits on a variety of political behaviors. To the best of our knowledge, however, there are no reliable, psychometrically sound indices of this variable at the individual level, and the construct has not been operationalized in a way that reflects Banfield’s (1958) conceptualization and its implications (see “Method” section; Alesina & Giuliano, 2010, 2011; Ferragina, 2011).

Amoral Familism, Culture, and Personality

More than other attitudes, amoral familism has been recently used to highlight cultural differences between northerners and southerners in Italy (Alesina & Ichino, 2009; Ferragina, 2009, 2011). As part of research on “national character,” there is a long-standing debate about the division in the Italian personality. From a historical point of view, northerners are typically described as orderly, conscientious, collectivist, gregarious, sensitive to community needs, and trusting in institutions whereas southerners are described as having a rebellious, individualist, and familistic character and lacking interest in pursuit of collective goals (Cimino & Foschi, 2014).

The possibility that there are differences of intelligence or personality underlying subcultures within a nation or territory still arouses the interest of researchers, who are reluctant to abandon the idea that there can be different “characters” within a nation. For these researchers, perceptions of national character may contain a kernel of truth or they may reflect real individual differences in intelligence and personality. In this context, it is worth mentioning the “differentialist” thesis of the psychologist Lynn, who thought he could demonstrate an intelligence gap between the Italians of the south and those of the north (and between Mediterranean populations and those of northern Europe), which—in his opinion—accounted for parallel differences in income, education, child mortality, and height (as a result of nutritional differences) (cf. Lynn, 2010a, 2010b, 2011). Other researchers have criticized the differentialist thesis, and have demonstrated in large samples (including Italian samples) that the national character is unrelated to the average personality profile of members of a culture and, at best, represents a stereotyped view of a specific culture (e.g., Allik, 2012; McCrae, Terracciano, Realo, & Allik, 2007; Terracciano et al., 2005). Moreover, some Italian psychologists have criticized the logic of the differentialist approach underlying Lynn’s research, and they have presented data which appear to definitively lay to rest the idea that cultural, economic, and social differences are derived from inherited mental features (cf. for example, Felice & Giugliano, 2011; D’Amico, Cardaci, Di Nuovo, & Naglieri, 2012; Robinson, Saggino, & Tommasi, 2011).

Aims and Scope

This study examines amoral familism/political distrust as a potential mediator of the relationship between personality and political participation. Specifically, we assessed amoral familism as an individual-level variable which was included in a holistic model in which personality traits were distal predictors of conventional and unconventional political acts, via a set of specific political attitudes. Furthermore, as this study was based on a sample of Italian participants recruited from different parts of the country, we were also able to explore whether self-ratings of amoral familism were different in northerners and southerners.

Method

Participants and Procedure

The participants were 405 Italians (53% women, 47% men); aged from 18 to 77 years (M age = 38.25 years; SD = 15.26); living in rural zones or villages (i.e., city size < 2,000 inhabitants; 13%,), towns (i.e., city size from 2,000 to 10,000 inhabitants; 19%), cities (i.e., city size from 10,000 to 100,000 inhabitants; 20%), and metropolitan areas (i.e., city size > 100,000 inhabitants; 48%); and who had been born and raised in the northern (19%), central (43%), and southern (38%) regions of the country. Participants were volunteers and were recruited by four undergraduate research assistants. Each assistant collected data from approximately 100 participants. The average time taken to complete the questionnaire was about 1 hr. The data collection procedures complied with Italian regulations on privacy and ethical standards in psychological research.

Measures

Data were collected via a structured questionnaire which included the instruments described below.

Amoral familism/political distrust

Banfield (1958) maintained that the amoral familist’s first aim was to maximize the advantage of his or her nuclear family, in the belief that this was a generalized practice. Banfield used multiple methods (observations, projective tests, interviews, etc.) to show that the amoral familist was not confident that the democratic structure based on rules and institutions served his or her family’s interest. Importantly, Banfield expanded on such a general rule, characterizing “amoral familism” in terms of 17 logical implications and consequences. It is worth noting that Banfield never included “interest in the nuclear family or neighbors” among this set of logical implications and consequences, focusing instead on distrust and disengagement from one’s community and its institutions (pp. 85-104).

We wanted to develop a reliable, self-report scale that directly taps Banfield’s (1958) amoral familism/political distrust construct, so we generated items based on Banfield’s descriptions of 17 behavioral manifestations of amoral familism and the social attitudes (concerning politics and the common good) characteristic of amoral familism. A scale consisting of 17 items to which responses were given using a 7-point Likert-type scale ranging from 1 (totally disagree) to 7 (totally agree), was piloted in sample of 120 volunteers. We carried out exploratory factor analysis on these data to enable us to create a one-dimensional scale based on items that loaded highly on the first unrotated factor. Five items with factor loadings greater than .50 were selected: “Authorities who claim to act in the public interest often act exclusively in their own interest” (first logical implication); “Public employees usually do only the bare minimum” (fifth logical implication); “If they are not punished for doing so people are likely to break the law to further their own interests” (sixth logical implication); “Unless they are closely regulated, civil servants tend to be dishonest” (seventh logical implication); “Public administration only works by using a strong hand” (eighth logical implication).¹ These items reflect distrust of public administration and other people, an attitude that Banfield considered characteristic of people with an amoral familistic attitude. We also carried out preliminary confirmatory factor analysis (CFA) on the data from this five-item scale as part of the main study, assuming a single latent variable. The model had an excellent fit to the data, χ²(5) = 1.50, p < .91, non-normed fit index (NNFI) = 1.00, comparative fit index (CFI) = 1.00, root mean square error of approximation (RMSEA) = .000 [.000, .026]), with factor loadings ranging from .57 to .88 and a high reliability coefficient (i.e., ω = .94).

For comparative purposes, we also measured amoral familism using a single item: “Do you agree with the statement that ‘My family’s interest should come before any other interest’?” When this item was included in the CFA, the model’s fit remained good but the additional item had a weaker factor loading (λ = .28) than the items tapping into distrust and disengagement.

Big Five Inventory (BFI; 44-item inventory)

The Italian version of the BFI was used to assess extraversion (α = .81), neuroticism (α = .78), conscientiousness (α = .77), agreeableness (α = .70), and openness to experience (α = .77). The 5-point response scale ranged from 1 (strongly disagree) to 5 (strongly agree; Fossati, Borroni, Marchione, & Maffei, 2010).

Portrait Value Questionnaire (PVQ; 40-item inventory)

We measured basic personal values with the 40-item PVQ which assesses conformity (α = .65), tradition (α = .60), benevolence (α = .70), universalism (α = .75), self-direction (α = .62), stimulation (α = .74), hedonism (α = .77), achievement (α = .83), power (α = .79), and security (α = .75). The PVQ consists of 40 short verbal portraits; respondents are required to indicate how strongly they resemble the person described in each using a 6-point Likert-type scale ranging from 1 (not like me at all) to 6 (very much like me; Schwartz, 2006).

Political Expertise (11-item scale)

Political expertise was measured using items regarded as valid indicators of expert-novice differences in political cognition and awareness. Eleven items were used, ranging from 0 to 1 (sample item: Who is Angela Merkel [the German Chancellor at the time of data collection]?; α = .89; cf. Foschi & Lauriola, 2014).

Perceived Political Self-Efficacy (PPSE)

The 10-item PPSE scale was used to assess participants’ evaluations of their capability to carry out political behaviors. Responses were given on a 5-point Likert-type scale ranging from 1 (not at all) to 5 (completely; sample item: “Please rate how confident you are in your ability to execute the specific action or behavior described . . . State your own political opinion openly, even in clearly hostile settings”; α = .90; Caprara, Vecchione, Capanna, & Mebane, 2009; Vecchione & Caprara, 2009).

Political interest

The three 2006 American National Election Studies Political Interest items (α = .94) were administered to participants to assess their interest in and attention to national government and politics (sample item: “How interested are you in information about what’s going on in government and politics?”).

Political and Associative Sociability (PASS; 8-item scale)

The PASS was used to assess political sociability (α = .86; sample item: “I enjoy situations in which I can debate my political views”) and associative sociability (α = .83; sample item: “Thinking about meetings with others is very pleasant for me”). In keeping with the principle that the more specific and targeted a measure, the more likely it is to predict a relevant behavior in a similar specific domain (i.e., principle of correspondence or compatibility; Ajzen, 2012, pp. 444-445), this two-factor scale assessed specific aspects of political participation related to sociability, such as spending free time in associations. Responses were given on a 7-point Likert-type scale ranging from 1 (totally disagree) to 7 (totally agree; for validation, see Foschi & Lauriola, 2014).

Political participation

Political engagement was measured by European Social Survey Political Participation seven-item scale (Jowell & The Central Co-Ordinating Team, 2003, 2005). More specifically, participants were asked to report the frequency with which they had carried out various low- and high-cost political acts during the past 12 months; they were asked about four conventional acts (sample item: “contacted a politician, government or local government official”) and three unconventional acts (sample item: “boycotted certain products”; de Rooij, 2012). Responses were given on a 5-point Likert-type scale ranging from 1 (never) to 5 (frequently). An average political participation score (α = .83) was obtained for each participant. Foschi and Lauriola (2014) showed that this average score predicted decisions to donate to one’s favorite political party or association in a dictator game situation. Average score on this scale seems to be a reliable proxy for political participation behaviors.

Sociodemographic information

Participants also completed a brief questionnaire to collect sociodemographic information, namely, age, gender, educational level (elementary or middle school, high school, higher education), urbanization (living in rural zones or village of <2,000 inhabitants; town of 2,000 to 10,000 inhabitants; city of 10,000 to 100,000 inhabitants; metropolitan area >100,000 inhabitants), income (Euros per annum, low: <15,000; low-medium: >30,000: medium-high: >70,000; high: >70,000), and region of upbringing (north, center, south), as well as other scales unrelated to the purpose of this study.

Statistical Analyses

After exploration of regression models, a structural equation model with two exogenous latent variables (openness to experience and extraversion) and six endogenous variables (amoral familism, political interest, political self-efficacy, political sociability, associative sociability, and political participation) was fitted to the data using EQS 6.1 (Figure 1). All latent variables except political participation were represented by a three-item parcel to provide more stable parameter estimates and proper solutions of model fit (e.g., Little, Cunningham, Shahar, & Widaman, 2002). Items thought to tap a given latent variable were first submitted to exploratory factor analysis and then sorted in decreasing order of loading on the first unrotated factor. Next, ranked items were systematically assigned to each parcel (e.g., Items 1, 4, and 7 to Parcel 1; 2, 5, and 8 to Parcel 2; Items 3, 6, and 9 to Parcel 3) to equalize the “true” and “error” variance components between parcels. As in a previous research, political participation was represented by two parcels based on items related to political acts and civic engagement (see Foschi & Lauriola, 2014). These two constructs—politic and civic engagement—are also proxies for bridging social capital (Putnam, 2000). The maximum-likelihood (ML) robust method was used to estimate model parameters, as the data violated the multivariate normal distribution assumption (Mardia’s coefficient = 11.89). Model fit was assessed with robust versions of the CFI, NNFI, and RMSEA. CFI and NNFI values greater than .95 were assumed to indicate good fit, and RMSEA values less than .06 were assumed to indicate a small approximation error.

Figure 1.

Structural model of political participation.

Results

Descriptive Analyses

First, we carried out a descriptive analysis that explored average differences between participants born and brought up in the north, center, and south of Italy on the measures used in this study. There were differences in age, urbanization, income, achievement, political self-efficacy, and political participation.

Although these differences were statistically significant, the effect sizes were in the null-small range for all variables except urbanization (see Table 1). There were no region of upbringing-related differences in amoral familism/political distrust, familism (single-item control).

Table 1.

Means Differences Between Participants That Were Born and Brought Up in the North, Center, and South of Italy.

Measure	North	Center	South	η²
Measure	M (SD)	M (SD)	M (SD)	η²
1. Age	42.2 (15.1)	37.9 (15.7)	36.7 (14.6)	.02*
2. Urbanization	3.08 (1.03)	3.38 (0.98)	2.58 (1.09)	.11**
3. Income	2.50 (1.02)	2.33 (0.79)	2.03 (0.85)	.04**
4. Achievement	3.29 (0.92)	3.52 (1.06)	3.72 (1.13)	.02*
5. Political self-efficacy	2.18 (0.75)	2.43 (0.85)	2.60 (0.89)	.03**
6. Political participation	1.78 (0.70)	2.14 (0.93)	2.08 (1.01)	.02**

Note. N = 405. Non significant measures have been omitted.

p < .05. **p < .01.

Second, we explored the zero-order correlations of political participation and amoral familism/political distrust. Political participation was negatively correlated with the familism control item (r = −.25), Amoral Familism/Political Distrust scale score (r = −.22), security (r = −.18), neuroticism (r = −.15), and conformity (r = −.14) and positively correlated with political self-efficacy (r = .62), political sociability (r = .54), political interest (r = .54), associative sociability (r = .25), openness (r = .24), stimulation (r = .19), political expertise (r = .19), universalism (r = .17), self-direction (r = .15), benevolence (r = .15), and extraversion (r = .11).

As expected, both Amoral Familism/Political Distrust scale score and the familism control item were negatively associated with political participation. Most notably, in our view, both indicators of amoral familism were negatively correlated with openness (r = −.14 and r = −.11, respectively) as well as having medium-large correlations with conservative values (security: r = .35, r = .37, respectively; conformity: r = .32, r = .38, respectively; tradition: r = .26, r = .38, respectively) and small-medium negative correlations with political factors (political expertise: r = −.23, r = −.12; political self-efficacy: r = −.11, r = −.13; political interest: r = −.21, r = −.27; political sociability: r = −.16, r = −.27, respectively).

Regression Analyses

A hierarchical regression model was used to identify the best predictors of political participation and to assess how the interplay of personality traits and political factors might have strengthened or weakened specific associations between each variable and political participation (Table 2). This analysis also served as a preliminary to structural equation modeling as the analysis of multivariate effects could reveal mediation paths that could be formally included in a structural model of political participation.

Table 2.

Hierarchical Multiple Regression Analysis to Assess How the Interplay of Personality Traits, Political Attitudes, and Values Predicts Political Participation.

Predictor	Political participation
Predictor	Step 1	Step 2	Step 3	Step 4	Step 5
1. Age	−.19***	−.18***	−.18***	−.17**	−.11*
2. Female gender	−.13*	−.11*	−.12*	−.11*	−.06
3. Southern latitude	.15**	.16**	.17***	.16**	.07
4. Urbanization	.14**	.13*	.11*	.09	.09*
5. Income	.00	.03	.02	.02	−.03
6. Education	−.00	−.02	−.03	−.04	−.00
7. Extraversion		.00	.02	.02	−.02
8. Neuroticism		−.11*	−.10	−.09	−.05
9. Openness to experience		.21***	.18***	.17**	.09*
10. Conscientiousness		−.13*	−.11*	−.09	−.09*
11. Agreeableness		.10*	.08	.08	.00
12. Amoral familism/political distrust			−.19**	−.15**	−.11**
13. Familism (single item)				−.13**	−.00
14. Associative sociability					.07
15. Political sociability					.14**
16. Political self-efficacy					.36***
17. Political expertise					−.10*
18. Political interest					.27***
R ²		.08***	.031***	.015**	.34***
Model R²	.08***	.16***	.19***	.20***	.55***

Note. N = 405. Entries are standardized regression coefficients.

p < .05. **p < .01. ***p < .001.

In Step 1 (see Table 2), we tested the overall effect of a block of sociodemographic variables that were entered together. Increasing age and female gender negatively predicted political participation, whereas a more urban residential location—urbanization—and a variable we named “southern latitude” (coded as follows according to region of birth and upbringing: 1 = northern Italy; 2 = central Italy; 3 = southern Italy) positively predicted political participation. Income and education were not associated with political participation. Overall, sociodemographic variables accounted for 8% of variance in political participation. In Step 2, we added the effects of the Big Five personality factors, which significantly increased the proportion of variance political participation explained by the model (Δr² = .08). Openness to experience was the best personality predictor of political participation; the other Big Five constructs were at most marginally associated with political participation when entered into the model together. Interestingly, when controlling for variance in demographic variables and other Big Five constructs, extraversion was not a predictor of political participation, while conscientiousness became a negative predictor. The regression weights for sociodemographic variables did not change substantially in this step, indicating that the two predictor blocks were relatively independent and that multivariate suppression effects for personality variables were due to correlations among the Big Five factors.² In Step 3, we entered amoral familism/political distrust score. This variable added 3.1% to the proportion of political participation variance explained, representing a small-medium effect. In Step 4, we entered the single-item familism indicator, which added 1.5% to the proportion of political participation variance explained. Despite the association between the single-item familism indicator and political participation, the effect of amoral familism/political distrust score also remained significant. Finally, we entered political factors (Step 5). These variables added 34% to the proportion of variance in political participation explained by the model, more than was accounted for by sociodemographic variables, personality traits, and familism variables put together (20%). In descending order, the best predictors of political participation were political self-efficacy, political interest, and political sociability. These findings suggested that the effects of other predictors might be mediated by these aspects of political participation. It should be emphasized that in Step 5, amoral familism/political distrust and openness to experience remained predictors of political participation.

As amoral familism has traditionally been studied from a sociological perspective, we investigated how this variable was related to sociodemographic variables. Table 3 shows that urbanization (i.e., living in rural, towns, cities, or metropolitan zones of the country) was a negative predictor of Amoral Familism/Political Distrust scale score and the single-item familism indicator. As a follow-up, we contrasted people who lived in rural areas (village or hamlet <2,000 inhabitants) with those living in larger urban areas with respect to amoral familism/political distrust (B = .17, t = 3.26, p < .01) and the single-item familism indicator (B = .12, t = 2.37, p < .05). Other contrasts between towns, cities, and metropolitan areas were non-significant. In fact, living in a small village (<2,000 inhabitants) was the single best predictor of both amoral familism indicators. Southern latitude (vs. other parts of Italy) did not predict any of the amoral familism measures, despite the fact that this variable was marginally significant (B = .11, t = 2.18, p < .03) in the univariate regression analysis. Last, as Table 3 shows, older age predicted the single-item familism indicator. Together, sociodemographic variables accounted for less than 5% of the variance in amoral familism/political distrust.

Table 3.

Multiple Regression Analysis of Amoral Familism Related to Sociodemographic Variables.

Predictor	AF/PD (scale)	Familism (single item)
1. Age	−.02	.15**
2. Female gender	−.07	.06
3. Southern latitude	.07	−.04
4. Urbanization	−.15**	−.15**
5. Income	−.02	.00
6. Education	−.06	−.07

Model R²	.04	.05

Note. N = 405. Entries are standardized regression coefficients. AF/PD = amoral familism/political distrust.

p < .01.

Structural Equation Modeling

The model depicted in Figure 1, with extraversion and openness as distal predictors of political participation via their effect on political factors, amoral familism/political distrust, and sociability, was acceptable and the fit indices approached the thresholds for good fit (χ² = 470.67, df = 207, p <.0001, CFI = .95, NNFI = .94, RMSEA = .056, RMSEA 95% confidence interval [CI] = [.049, .063]). As expected, all the parcels had a high and significant loadings on the latent variables that they were supposed to measure (λ > .59 except for Extraversion Parcel 3, λ = .53).

Openness to experience and extraversion did not have a direct association with political participation (Figure 1). However, both had positive effects on political self-efficacy and associative sociability. Interestingly, political self-efficacy was more strongly linked to openness than extraversion (.179 vs. .143, respectively). Conversely, associative sociability was more strongly linked to extraversion than openness (.180 vs. .148, respectively). Openness to experience also had a negative effect on amoral familism/political distrust, which was negatively linked with political interest and political participation. Taken together, these findings indicate that the effect of personality traits on political participation was fully mediated by the set of variables included in the model.³

General Discussion

Our findings support the hypothesis that the associations between personality traits and political participation are mediated by political factors. Specifically, openness to experience played a major role as a positive precursor of political participation. On one hand, its effect, in terms of increased political participation, was mediated by increased political self-efficacy and increased political sociability. On the other hand, a longer chain of mediators also linked higher openness to experience with greater political participation, involving decreased amoral familism/political distrust and increased political interest. Like openness to experience, extraversion was positively associated with political participation via political self-efficacy and sociability. Unlike openness to experience, extraversion was more strongly associated with political sociability than with political self-efficacy and was not associated with amoral familism/political distrust or political interest. Both extraversion and openness to experience appear to dispose people to political participation, but via different mechanisms.

At a more general level, our findings are consistent with the view that personality traits are linked to important personal and social outcomes by various aspects of political participation (Gallego & Oberski, 2012). Extravert individuals were more likely to spend their free time pursuing collective goals with others, whereas open-minded people were more interested in politics and reported greater political self-efficacy. Open-minded individuals were also less inclined to value family interests over the general social interest. We measured this latter tendency using an Amoral Familism/Political Distrust scale that was a reliable and effective direct and indirect negative predictor of participation in politics.

The single most important contribution of our research is the finding that amoral familism/political distrust mediates the relationship between openness and political participation. To date, empirical studies of amoral familism have not provided a psychologically meaningful indicator for this construct at the individual level (Alesina & Giuliano, 2010, 2011; Ferragina, 2009, 2011). By contrast, following Banfield’s (1958) logical implications, our operative definition of amoral familism portrayed one’s personal sense of distrust toward public administration and governance of public good. In other words, in our study, amoral familism was not defined simply in terms of interest in the nuclear family or close neighbors, but reflected endorsement of psychological justifications for amoral behaviors (e.g., looking after one’s own interest at the expense of the common good) and a belief that politicians and administrators―or just others in general―are amoral (e.g., “Authorities who claim to act in public interest often act exclusively in their own interest”; “If they are not punished for doing so people are likely to break the law to further their own interest”). Measured in this way amoral familism generated data that were consistent with Banfield’s claims that familist people are unwilling to act collectively in pursuit of the common good.

Whereas the amoral familism notion has been long criticized as a main cause for underdevelopment of southern Italy (for a review, see Ferragina, 2009, 2011), personality studies of national character were motivated by the expectation that northerners and southerners would differ with respect to sensitivity to community needs (Cimino & Foschi, 2014; on stereotypes in differential studies see also McCrae et al., 2007; Terracciano et al., 2005). Our data not only show that southern participants reported higher rates of conventional and unconventional political acts than northern participants but also reveal that no significant differences emerged on both the Amoral Familism/Political Distrust scale or on its single-item indicator of familism.⁴ Structural equation modeling show that amoral familism/political distrust has a meditational role, so its effect on political participation cannot be merely ascribed to general geographical differences (e.g., North vs. South or West vs. East). If any, geopolitical differences in political participation and familism might be more associated with city size, acculturation, and social change than with simple geography as we found differences in political participation between residents of rural areas and those living in urban and metropolitan zones (cf. Table 2 and 3; on theory of city size, see Batty, 2013; see also Greenfield, 2016).

Final Remarks

Before concluding, it is important to recognize some of the limitations of our study. First of all, as some scholars have pointed out, mediation analyses with cross-sectional data can sometimes provide biased estimates. In addition, our conclusions are based on data collected using a non-representative sampling method and so, although our sample was large, generalization of our findings to the general population is not warranted. Nevertheless our findings, in particular, the large effect size of the models reported here may have practical implications for future research on political participation using normative national samples. Of course, more research is needed, particularly cross-cultural research and experimental studies capable of demonstrating a causal relationship between personality and political participation and between political trust/distrust and the building of social capital (Chen, Stanton, Gong, Fang, & Li, 2009; Schyns & Koop, 2010). Despite these limitations, however, our study shows that a model of political participation that includes a comprehensive set of mediators provides useful insights into the psychological processes relating personality traits to observable political behaviors. To conclude, Banfield’s theory on amoral familism/political distrust may be extended to other domains, such as institutional distrust of minorities in the United States and in politically unstable areas of Latin America. Furthermore, it is worth exploring the connections between Banfield’s concept of amoral familism/political distrust and Putnam’s distinction between bonding and bridging social capital (Putnam, 2000).

Footnotes

Acknowledgements

The authors thank their students for their assistance with data collection.

Authors’ Note

This article updates material presented at the ISPP Annual Scientific Meeting, Rome, July 4-7, 2014. Both authors discussed the content of this article. Renato Foschi developed the theoretical framework and the epistemological research hypotheses; Marco Lauriola devised the methodology and analyzed the data.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Ajzen

(2012). The theory of planned behavior. In Lange

P. A. M.

Kruglanski

A. W.

Higgins

E. T.

(Eds.), Handbook of theories of social psychology (Vol. 1, pp. 438-459). London, England: Sage.

Alesina

Giuliano

(2010). The power of the family. Journal of Economic Growth, 15, 93-125.

Alesina

Giuliano

(2011). Family ties and political participation. Journal of the European Economic Association, 9, 817-839.

Alesina

Ichino

(2009). L’Italia fatta in casa. Indagine sulla vera ricchezza degli italiani [Homemade Italy. Investigation into the real wealth of the Italians]. Milano, Italy: Mondadori.

Allik

(2012). National differences in personality. Personality and Individual Differences, 53, 114-117.

Banfield

(1958). The moral basis of a backward society. New York, NY: Free Press.

Batty

M. C.

(2013). A theory of city size. Science, 340, 1418-1419.

Buss

D. M.

(2009). Adaptive problems define situations: An evolutionary formulation of person-situation interactions. Journal of Research in Personality, 43, 241-242.

Caprara

G. V.

Vecchione

Capanna

Mebane

(2009). Perceived political self-efficacy: Theory, assessment, and applications. European Journal of Social Psychology, 39, 1002–1020. doi:10.1002/ejsp.604

10.

Chen

Stanton

Gong

Fang

(2009). Personal Social Capital Scale: An instrument for health and behavioral research. Health Education Research, 24, 306-317.

11.

Christakis

N. A.

Fowler

J. H.

(2009). Connected: The surprising power of our social networks and how they shape our lives. New York, NY: Hachette Book Group.

12.

Cimino

Foschi

(2014). Northerners versus southerners: Italian anthropology and psychology faced with the “southern question.” History of Psychology, 17, 282-295.

13.

D’Amico

Cardaci

Di Nuovo

Naglieri

J. A.

(2012). Differences in achievement not in intelligence in the north and south of Italy: Comments on [Lynn, 2010a] and [Lynn, 2010b]. Learning and Individual Differences, 22, 128-132.

14.

de Rooij

E. A.

(2012). Patterns of immigrant political participation: Explaining differences in types of political participation between immigrants and the majority population in Western Europe. European Sociological Review, 28, 455-481.

15.

Felice

Giugliano

(2011). Myth and reality: A response to Lynn on the determinants of Italy’s North–South imbalances. Intelligence, 39, 1-6.

16.

Ferragina

(2009). The never-ending debate about the moral basis of a backward society: Banfield and amoral familism. Journal of the Anthropological Society of Oxford, 2, 141-160.

17.

Ferragina

(2011). Il fantasma di Banfield: Una verifica empirica della teoria del familismo amorale [The ghost of Banfield: An empirical test of the theory of amoral familism]. Stato e Mercato, 92, 283-312.

18.

Figueredo

A. J.

Corral-Verdugo

Frías-Armenta

Bachar

K. J.

White

McNeill

P. L.

. . . Castell-Ruíz

I. P.

(2001). Blood, solidarity, status, and honor: The sexual balance of power and spousal abuse in Sonora, Mexico. Evolution & Human Behavior, 22, 295-328.

19.

Foschi

Lauriola

(2014). Does sociability predict civic involvement and political participation? Journal of Personality and Social Psychology, 106, 339-357.

20.

Fossati

Borroni

Marchione

Maffei

(2010). The Big Five Inventory (BFI). Reliability and validity of its Italian translation in three independent nonclinical samples. European Journal of Psychological Assessment, 27, 50-58.

21.

Gallego

Oberski

(2012). Personality and political participation: The mediation hypothesis. Political Behavior, 34, 425-451.

22.

Greenfield

P. M.

(2016). Social change, cultural evolution, and human development. Current Opinion in Psychology, 8, 84-92.

23.

Jowell

, The Central Co-Ordinating Team. (2003). European Social Survey 2002/2003: Technical report. London, England: City University, Centre for Comparative Social Surveys.

24.

Jowell

The Central Co-ordinating Team. (2005). European Social Survey 2004/2005: Technical report. London, England: City University, Centre for Comparative Social Surveys.

25.

Keeler

A. R.

Siegel

J. T.

Alvaro

E. M.

(2014). Depression and help seeking among Mexican–Americans: The mediating role of familism. Journal of Immigrant and Minority Health, 16, 1225-1231.

26.

Levinson

D. J.

(1958). The relevance of personality for political participation. Public Opinion Quarterly, 22, 3-10.

27.

Little

T. D.

Cunningham

W. A.

Shahar

Widaman

K. F.

(2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9, 151-173.

28.

Lynn

(2010a). In Italy, North–South differences in IQ predict differences in income, education, infant mortality, stature, and literacy. Intelligence, 38, 93-100.

29.

Lynn

(2010b). IQ differences between the north and south of Italy: A reply to Beraldo and Cornoldi, Belacchi, Giofre, Martini, and Tressoldi. Intelligence, 38, 451-455.

30.

Lynn

(2011). IQs in Italy are higher in the north: A reply to Felice and Giugliano. Intelligence, 40, 255-259.

31.

McCrae

R. R.

Terracciano

Realo

Allik

(2007). Climatic warmth and national wealth: Some culture-level determinants of national character stereotypes. European Journal of Personality, 21, 953-976.

32.

Mondak

J. J.

(2010). Personality and the foundations of political behavior. New York, NY: Cambridge University Press.

33.

Mondak

J. J.

Halperin

K. D.

(2008). A framework for the study of personality and political behavior. British Journal of Political Science, 38, 335-362.

34.

Putnam

R. D.

(2000). Bowling alone: The collapse and revival of American community. New York, NY: Simon & Schuster.

35.

Robinson

Saggino

Tommasi

(2011). The case against Lynn’s doctrine that population IQ determines levels of socio-economic development and public health status. Journal of Public Mental Health, 10, 178-189.

36.

Schwartz

S. H.

(2006). Basic human values: Theory, measurement, and applications. Revue Française de Sociologie, 47, 249-288.

37.

Schyns

Koop

(2010). Political distrust and social capital in Europe and the USA. Social Indicators Research, 96, 145-167.

38.

Stein

G. L.

Cupito

A. M.

Mendez

J. L.

Prandoni

Huq

Westerberg

(2014). Familism through a developmental lens. Journal of Latina/o Psychology, 2, 224-250.

39.

Taylor

A. B.

MacKinnon

D. P.

Tein

J. Y.

(2008). Tests of the three-path mediated effect. Organizational Research Methods, 11, 241-269.

40.

Terracciano

Abdel-Khalek

A. M.

Adam

Adamovova

Ahn

C. K.

Ahn

H. N.

Meshcheriakov

(2005). National character does not reflect mean personality trait levels in 49 cultures. Science, 310, 96-100.

41.

Vecchione

Caprara

G. V.

(2009). Personality determinants of political participation: The contribution of traits and self-efficacy beliefs. Personality and Individual Differences, 46, 487-492.

42.

Vecchione

Schwartz

S. H.

Caprara

G. V.

Schoen

Cieciuch

Silvester

. . . Alessandri

(2015). Personal values and political activism: A cross-national study. British Journal of Psychology, 106, 84-106.

Do Amoral Familism and Political Distrust Really Affect North–South Differences in Italy?

Abstract

Keywords

Personality and Political Participation

Amoral Familism, Culture, and Personality

Aims and Scope

Method

Participants and Procedure

Measures

Amoral familism/political distrust

Big Five Inventory (BFI; 44-item inventory)

Portrait Value Questionnaire (PVQ; 40-item inventory)

Political Expertise (11-item scale)

Perceived Political Self-Efficacy (PPSE)

Political interest

Political and Associative Sociability (PASS; 8-item scale)

Political participation

Sociodemographic information

Statistical Analyses

Results

Descriptive Analyses

Regression Analyses

Structural Equation Modeling

General Discussion

Final Remarks

Footnotes

Acknowledgements

Authors’ Note

Declaration of Conflicting Interests

Funding

Notes

References