Running head: VALUES AND FEAR OF CRIME
Values and Fear of Crime
Silvia Russo Michele Roccato University of Torino
Author’s note:
Silvia Russo, Department of Psychology, University of Torino, Via Verdi, 10, 10124 Torino, Italy. Telephone: ++390116702055, Fax: ++390116702061, E-mail: aivlisss@gmail.com
Michele Roccato, Department of Psychology, University of Torino, Via Verdi, 10, 10124 Torino, Italy. Telephone: ++390116702015, Fax: ++390116702061, E-mail: roccato@psych.unito.it
(corresponding author)
Values and Fear of Crime
Silvia Russo Michele Roccato
Department of Psychology, University of Torino, Italy
ABSTRACT
The feeling of unsafety, usually operationalised as fear of crime, is mainly studied by sociologists and criminologists, who rely upon sociological or socio-demographic variables to account for fear variations. However, some psycho-social variables may be effectively used to predict such fear. In this chapter we examined, in a psycho-social perspective, whether individual values priorities exert
influence on personal fear of crime (or concrete fear) and concern about crime as a social problem (or abstract fear), performing a secondary analysis of the data gathered by the Observatory of North-West on a representative sample of the Italian population over 18 (N = 1,667). We tested two structural equation models, both aimed at predicting concrete and abstract fear of crime using values among our independent variables. In the first one we used Schwartz’s approach to values (assessing values in terms of Openness to change vs. Conservatism and Self-enhancement vs. Self-transcendence), and in the second one we used Inglehart’s approach (assessing values in terms of Materialism vs. Post-materialism). Besides individual values, we took into account four sets of independent variables: socio-demographic characteristics, victimization experiences, political variables, and mass media exposure. Findings from previous research on fear of crime were partially confirmed. Concrete fear of crime was positively influenced by direct and indirect victimization and size of area of residence. Abstract fear of crime was positively influenced by direct victimization, and negatively influenced by
education, size of area of residence, and National TV news watching. As concerns the innovative part of our research, values influenced abstract fear as well as concrete fear. When we operationalised values using Schwartz’s model, both concrete and abstract fear of crime were negatively influenced by Openness to change. When we operationalised values using Inglehart’s model, Materialism exerted a positive influence on concrete fear of crime and Post-materialism exerted a negative influence on abstract fear of crime. Our two models accounted for almost the same proportion of variance of the two dependent variables, and the effects our two operationalisations of values exerted on fear were
reasonably similar in magnitude. Limits and possible developments of this research are discussed.
INTRODUCTION
The entire human history has been permeated by the feeling of unsafety. This feeling takes today new shapes and becomes more and more relevant since it may compromise chances for people and institutions to face the present and to outline the future [Amerio, 1999]. As a matter of fact, it can cause several consequences that may involve a variety of human life levels: intra-psychic consequences (anger, aggressiveness, anxiety states, disempowerment), behavioural consequences (refusal of social participation, constraints imposed on one’s own life), and social ones (delegitimisation of outgroups, reduction of cohesion and solidarity, and even development of an “ideology of safety” capable to turn the legitimate demand for living in safe communities into an attempt to make the most violent racist and xenophobic behaviours acceptable) [Amerio, 2004]. Thus, it is far from surprising that feeling of unsafety diffusion led researchers to try to learn more about this topic. However, its intrinsic
complexity, its prismatic nature and some survey difficulties brought several authors to radically simplify their object of study, which often became restricted to micro-crime psychological reactions.
Recent contributions on this matter pointed out that crime-related fear is composed of at least two basic dimensions, a cognitive one and an affective one [Ferraro, 1995; Kanan & Pruitt, 2002; Rountree & Land, 1996]. The former, risk perception, is a generic evaluation of the probability of being
victimised, while the latter, fear of crime, is a specific emotional response of anxiety to crime. The literature on fear of crime showed the relevance of another distinction, originally suggested by
Furstenberg [1971], that between personal fear of crime and concern about crime as a social problem. In particular, personal fear (sometimes labelled as concrete fear) refers to a feeling of dread or anxiety about personal safety or about personal possessions preservation; it may be a response to both a real and a potential risk. On the other hand, concern about crime as a social problem (sometimes labelled as abstract fear) is a state of dread about crime spread in one’s own country more than about personal safety [Nardi, 2003; Roché, 1993].
Recent research showed a weak correlation between abstract fear and concrete fear. Moreover, these two fears are predicted by different set of variables [Amerio & Roccato, 2005; Roché, 1993]. Since concrete fear concerns people’s everyday life, during the last thirty years researchers into fear of crime mainly focused their attention on this topic, while just a few of them tried to broaden the
knowledge about predictors of abstract fear.
The body of literature about concrete fear of crime gives high relevance to physical and social vulnerability [Killias & Clerici, 2000; Pantazis, 2000], originally assessed using socio-demographic variables. Research highlighted that the most fearful social categories are: (a) young and old people [Amerio & Roccato, 2005]; (b) women, particularly if fear of sexual assault is not controlled for [Ferraro, 1995]; (c) low socio-economic status groups, e.g. poor people [Kanan & Pruitt, 2002; Pantazis, 2000] and poorly educated ones [Austin, Woolever & Baba, 1994; Kennedy & Silverman, 1985]; and (d) people living in urban centres, because in large towns social and physical incivilities on the one hand [Miceli, Roccato & Rosato, 2004; Sartori, 2003], and crime on the other hand [Nardi, 2003] are particularly spread. Another predictor of concrete fear highly regarded is victimization, usually considered in two ways: as an experience of undergoing an offence (direct victimization) and as a psychological consequence of an offence involving people of one’s social network (indirect or vicarious victimization) [Skogan & Maxfield, 1981]. Both the experiences foster concrete fear of crime, while they usually don’t exert any influence on abstract fear [Nardi, 2003; Roccato, 2007].
However, only a few psycho-social studies were performed trying to predict concrete fear. Van der Wurff, Van Staalduinen and Stringer [1989] identified four psychosocial variables predicting fear of crime: (a) attractivity (the perception of being an attractive target for criminal acts); (b) evil intent (the degree to which a person ascribes criminal intents to another individual or to a given group); (c) power (the sense of self-assurance and the feeling of control relative to potential crime threats); and (d) criminalisable space (the perception that a given situation can lead to personal victimization). More
recently, Zani, Cicognani, and Albanesi [2001] have shown that fear of crime is promoted by personal involvement in negative, even though not necessarily criminal, experiences in the area of residence. Finally, Amerio and Roccato [2005] showed that the socio-demographic and victimization variables traditionally used to predict fear of crime may be fruitfully integrated with some psychosocial variables (mainly distrust of others and ethnic prejudice) which express the ways of perceiving, representing and evaluating one’s own social world. However, the main predictors of fear of crime remained the socio-demographic and victimization variables.
On the contrary, the literature shows that the psycho-social variables, combined with the mass-mediatic ones, are more effective in predicting concern about crime as a social problem than the socio-demographic and the victimization ones. For what concerns abstract fear, originally studied in its relation with mass media exposure [Tyler, 1980; Tyler & Cook, 1984], the main predictors, in the Italian contest, turned out to be participants’ political position (to prefer right parties) and their frequency of watching television; other relations that came out involved education and interest in politics, that exerted a negative influence on abstract fear [Amerio & Roccato, 2005; Nardi, 2003; Roccato, 2007]. As a whole, as some authors underscored, variables that foster concern about crime as a social problem are those pertaining to the way of perceiving, representing and evaluating one’s own social contest [Amerio, Gattino, & Roccato, 2004; Nardi, 2003]. However, these general statements on perception and representation relevance were seldom empirically examined in research focused on these aspects. In this chapter we tried to delve into this unexplored facet of crime-related fear predictors, studying the influences exerted on concrete and abstract fear of crime by cognitive representation of motivations and important goals in people lives, namely values priorities.
Besides a first contribution published by Rokeach in 1973, two theories on values have dominated values research in the last decades. The first one, developed by Shalom Schwartz [1992], describes values as an integrated system which particular structure frames a “quasi-circumplex” model. In this
model, ten universal values [Power, Achievement, Stimulation, Hedonism, Self-direction,
Universalism, Benevolence, Tradition, Conformity, and Security), characterised by a cross-culturally common meaning, may be represented by two principal dimensions: (a) Openness to
change/Conservatism, reflecting a conflict between emphasis on thought and action independence on the one hand and preference of traditional practice observance and stability and protection on the other hand; and (b) Self-enhancement/Self-transcendence, reflecting a conflict between the pursuit of
personal success and predominance on others on the one hand and acceptance of outgroups and commitment for their well-being on the other hand. Schwartz’s model has been widely used by social psychologists to study political attitudes [Barnea & Schwartz, 1998; Caprara, Schwartz, Capanna, Vecchione, & Barbaranelli, 2006; Mayton & Furnam, 1994], well-being and self-perception [Rice, 2006; Sagiv & Schwartz, 2000], decision processes [Feather, 1995; Rohan, 2000], intergroup relations [Sagiv & Schwartz, 1995; Schwartz, Struch, & Bilsky, 1990], and also adolescents’ macro-social worries and goals life [Martynowicz, 2006; Ojala, 2005].
The second theory, developed by Ronald Inglehart’s [1990], emphasizes a change in values priorities that took place in the last decades as a consequence of the economic development of the Western societies. In particular, Inglehart identified two latent dimensions that elucidate individual differences in values priorities: (a) the Traditional/Secular-rational dimension, reflecting the contrast between religious and traditional values that generally prevail in agrarian society on the one hand, and rational, bureaucratic and secular values that are predominant in industrialized societies on the other hand; and (b) the Survival/Self-expression dimension, involving the polarization between Materialistic and Post-materialistic values: The former emphasize economic and physical safety, while the latter emphasize subjective well-being and quality of life. Inglehart’s approach is very popular among sociologists and political scientists, especially among those interested in comparative studies [see for instance Galland & Lemel, 2006], while is less widely used by social psychologists.
The research undertaken by Schwartz, Sagiv, and Boehnke [2000] on the relation between values and worries is pretty interesting for our study. Schwartz and colleagues took into account two kinds of worries, differing one from the other for the object which they refer to: the self or people with whom one identifies himself (micro-worries) on the one hand, and objects external to the self, like society, world, and even the whole universe (macro-worries) on the other hand. In their research giving priority to Self-transcendence values positively correlated with macro-worries and negatively correlated with micro-worries, while giving priority to Self-enhancement values negatively correlated with macro-worries and positively correlated with micro-worries. Moreover, arraying of values along the Self-enhancement/Self-transcendence dimension was more effective than along the Openness to change/Conservatism dimension in predicting both kinds of worries; thus, according to Schwartz and colleagues concern for the self versus concern for the wider society and world should be considered as the critical factor discriminating micro- from macro-worries. As a whole, Schwartz and colleagues showed that values exert a stronger influence on macro worries rather than on micro-worries. They explained this gap considering that other variables are likely to affect micro-worries: Among them, directly experienced objective threats to personal welfare and individual differences in objective or subjective coping ability may reduce or neutralize the effects of values differences on micro-worries.
However, as in previous research, in their study Schwartz and colleagues did not deeply analyze the relation between values and fear of crime. Still, from a theoretical standpoint, it seems appropriate to try to predict fear of crime taking into account the analogy between macro-worries and abstract fear and micro-worries and concrete fear. That’s we tried to do in this study.
GOALS AND HYPOTHESIS
In this chapter we tried to empirically shed some light on the effects exerted by values on concrete and abstract fear, comparing the predictive power exerted by Schwartz’s [1992] and
Inglehart’s [1990] values operationalisations, statistically keeping under control the influence of the other variables the literature shows to influence fear of crime. We tried to answer three main research questions: (a) do the results found by sociological and criminological models hold when values are added to the models aimed at predicting concrete and abstract fear of crime?; (b) do values exert a significant influences on concrete and abstract fear of crime when the impact of the socio-demographic and victimization variables is partialled out?; and (c) which one, between Schwartz’s and Inglehart’s values models, is the best predictor of concrete and abstract fears of crime?
In general, since research showed that the way of perceiving, representing and evaluating the social environment is an effective predictor of abstract fear [Amerio, Gattino, & Roccato, 2004; Nardi, 2003], we hypothesized that, even controlling for the socio-demographic and victimization variables, values should exert a weak influence on concrete fear of crime and a stronger influence on concern about crime as a social problem.
In particular, based on Schwartz et al.’s [2000] results concerning micro- and macro-worries, we hypothesized that: (a) the Self-enhancement/Self-transcendence dimension should exert a stronger influence on concrete fear of crime than Openness to change/Conservatism; (b) adhesion to Self-enhancement values should exert a negative effect on abstract fear and a positive effect on
concrete fear, since a strong interest for the self may be in contrast with a strong concern for the wider society, but at the same time it may be consistent with a strong concern for personal welfare or with that of the closest ones; (c) adhesion to Openness to change values should negatively influence concrete fear, since people who value new, challenging, and uncertain outcomes are less concerned about uncertain personal consequences; and (d) adhesion to Conservatism values should positively influence abstract fear, since it includes values emphasizing acceptance of prevailing cultural and societal tradition and those who espouse these values may be especially sensitive to the threats to the accepted social order [Altemeyer, 1996].
As regards Inglehart’s [1990] theory, we hypothesized that Materialistic values should exert a positive influence on both concrete and abstract fear of crime, while Post-materialistic values should exert a negative influence on them: This is because Inglehart’s values reflect a polarization between an emphasis on economic and physical security and an emphasis on self-expression, subjective well-being, and quality of life concerns.
METHOD Data Set
We performed a secondary analysis on data gathered by the Observatory of North-West.1 The sample (N = 1,667), interviewed by mail, was representative of the Italian population over 18 according to the main socio-demographic (gender, age, education, geopolitical area of residence, and size of area of residence) and political (vote cast at the 2004 European election) variables.
Dependent variables
We assessed fear of crime using two four-categories items: “Think of micro-criminality. How would you define the situation regarding this problem in your area of residence?” and “Think of micro-criminality. How would you define the situation regarding this problem in Italy?”. As previously done [see Amerio & Roccato, 2005, 2007; Roccato, 2007], we considered these items as
operationalisations of concrete fear of crime and of concern about crime respectively. In order to use reasonably balanced dependent variables, we recoded our two indicators of fear into two dummy variables using different strategies. As concerns concrete fear, we assigned the code 0 to participants who considered the situation in their area of residence as just a little dangerous or not dangerous at all (N = 924), and the code 1 to those who considered the situation as quite or very dangerous (N = 743). As concerns abstract fear, we assigned the code 0 to those who considered the situation in Italy as not
1 The Observatory of North-West (www.nordovest.org), a research institute of the University of Turin, conducts surveys on the attitudes and behaviours of the Italian population three times a year using wide representative samples.
unsafe, or just a little, or quite unsafe (N = 758) and the code 1 to those who considered it as very unsafe (N = 908).2
Independent variables
Based on the findings of previous studies on fear of crime, we considered five families of independent variables.
Socio-demographic variables. We used gender (1 = woman), age, years of education, number of family members, size of area of residence, and geopolitical area of residence. The size of the area of residence was recoded into a dummy variable (1 = living in over 100,000 people towns). In the same way four dummies were created for geopolitical area of residence (North-Western Italy, North-Eastern Italy, Southern Italy, Italian Islands; residence in Central Italy was used as reference category).3
Victimization experiences. Seven variables referred to micro-criminal offences toward person or ownership (assault, vandalism, burglary in one’s own home, car theft, pick-pocketing and bag-snatching, robbery, and fraud) were available. We computed a synthetic index computing the number of direct victimization experiences occurred to the participant him/herself in the year preceding the survey and a synthetic index computing the number of indirect victimization experiences occurred in the year preceding the survey.4
Political variables. We used a four-categories item assessing interest in politics and a ten-categories item assessing political placement on the left-right axis (1 = extreme left, 10 = extreme right).
2 For abstract fear, we performed parallel analysis using the same coding system we used for concrete fear, assigning 0 to participants who considered the Italian situation as just a little unsafe or not unsafe at all (N = 67) and 1 to those who considered it as quite or very unsafe (N = 1600). Obtained results (available on request) were not satisfying since the dummy was closer to a constant more than a variable.
3 North-Western Italy included Piedmont, Lombardy, Liguria, and Aosta Valley; North-Eastern Italy included Trentino Alto Adige, Veneto, Friuli Venezia Giulia, and Emilia Romagna; Southern Italy included Abruzzo, Molise, Campania, Puglia, Basilicata, and Calabria; Italian Islands included Sicily and Sardinia; and Central Italy included Tuscany, Umbria, Marches, and Latium.
4 As Roccato [2007], we performed parallel analysis with analytic indexes for direct and indirect victimization. Results, available on request, were almost the same as those we chose to publish.
Mass media exposure. Two dummies were created on the basis of which channel news our respondents declared they had watched more frequently in the four months before the survey. The first one assessed preferential exposure to Rai news, while the second one assessed preferential exposure to Mediaset ones5 (we used exposure to minor channel news or no exposure to television news as
reference category). Moreover, as Amerio and Roccato [2005], we estimated the minutes of daily exposure to television, using Ricolfi’s [1994] approach.
Values orientation. As concerns Schwartz’s approach, we used a 21-item short version of the Italian adaptation of the Portrait Values Questionnaire [Capanna, Vecchione, & Schwartz, 2005]. Items underwent a confirmatory factor analysis. In order to get a model characterized by a satisfactory fit, χ²(32) = 167.052, p < .001, CFI = .954, TLI = .921, RMSEA = .050 (90% CL = .043, .058), we had to modify the original model. On the one hand, we had to drop 10 items, and on the other hand we had to introduce in the model a method factor, controlling for acquiescent response-set. The latter result was not unexpected at all, both for substantive reasons--as they are socially and individually desirable goals, it’s easy to forecast that, independent of their content, all the values will tend to be positively evaluated by participants: see Roccato, 2008]--and for methodological ones--the sample we used was much more educationally heterogeneous than those typically used in social-psychological research, and the
literature consistently shows that the less educated the sample, the more the probability of getting results distorted by response-sets [see Gasperoni & Giovani, 2002]. Figure 1 shows the final model (standardized regression weights, variances, and correlations are provided; all the parameters were significant at least with p < .05).
As concerns Inglehart’s approach, based on the 4 items used by Inglehart [1990] to assess participants’ adhesion to Materialistic and Post-materialistic values, we created two dummies: one for
5 Italian analogical television system is substantially a duopoly. On the one side the Rai network (with three TV channels, Rai1, Rai2, and Rai3), a public television service, traditionally controlled by Government, on the other side the Mediaset network (Canale5, Italia1, and Rete4), owned by Silvio Berlusconi, the Italian prime minister at present in charge. The minor and local channels have just a residual audience.
adhesion to Materialistic values and another one for adhesion to Post-materialistic values (we used adhesion to mixed values as reference category).
Data Analysis
We tried to answer our three questions by empirically testing two structural equations models. In the first one we used Schwartz’s operationalisation of values, and in the second one we used Inglehart’s one. All the other variables we used were the same in the two models. As it usually happens, we had to drop some variables, either because they did not exert a significant influence on fear of crime or because they their inclusion would have worsened the global fit of the models.
RESULTS
Figure 2 shows our model predicting personal fear of crime and concern about crime as a social problem using Schwartz’s model for operationalising values, modelled as latent variables (standardized regression weights, variances, and correlations are provided; all the parameters were significant at least with p < .05; the portion of the model dedicated to the confirmatory factor analysis of the Portrait Values Questionnaire, already presented in Figure 1, is absent for visual ease). The model accounted for a moderate proportion of variance of concrete fear (Adj. R² = .173) and of abstract fear (Adj. R² = . 145). Its fit, χ²(135) = 905.709, p< .001, CFI = .861, TLI = .825, RMSEA = .059, (90% CL = .055, .062) was reasonable. Results obtained in previous research were partially confirmed: The most effective variables for predicting concrete and abstract fear were direct victimization, which fostered both fears, and the size of the area of residence, which fostered concrete fear and negatively influenced abstract fear. Indirect victimization positively influenced concrete fear also. Adhesion to Openness to change values exerted a negative influence on both the dimensions of fear, even though the effect on concrete
fear was stronger than that on abstract fear. In spite of the large N, the effect of the Self-transcendence dimension, however, did not achieve statistical significance.
The variables exerting the strongest influence on abstract fear were those pertaining to the mass media domain: TV news exposure, to both the Rai and the Mediaset channels, in contrast with exposure to minor news channels and to no exposure to the TV news, exerted a strong negative influence on concern about crime as a social problem.
In order to improve the fit of our model, we had to free four correlations. Three of them were expected, either because of the structure of the variables available in the file (it was the case of the correlation between Rai and Mediaset news exposure and of that between direct and indirect victimization)6, or based on the literature (it was the case of the correlation between abstract and concrete fear, that was detected in previous research: see Amerio & Roccato, 2005, 2007]. The last one —that between the Openness to change and Self-transcendence factors was unexpected, as, according to Schwartz et al. [2000], such dimensions usually tend to be orthogonal.
Figure 3 shows the model aimed at predicting concrete and abstract fear using Inglehart’s [1990] model to operationalise values (standardised regression weights, variances, and correlations are
provided; all the parameters were significant at least with p < .05). As that built using Schwartz’s approach to values, this model accounted for a moderate proportion of variance of concrete (Adj. R² = . 168) and of abstract fear (Adj. R² = .160). Again, the model showed a reasonable fit, χ(31) = 371.883, p < .001, CFI = .863, TLI = .801, RMSEA = .081 (90% CL = .074, .089).
The model structure was the same as that of the model showed in Figure 2--with, of course, the exception of the Portrait Values Questionnaire latent factors being replaced by the two dummies
6 As concerns news exposure, it has been asked to respondents which was the news channel they watched more in the last four months before the survey; they could answer choosing a news channel from a list of National and local channels. Thus, the choice of a news channel automatically excluded the chance of choosing other channels. As concerns victimization, for each kind of offence respondents had three choices: the first one in the case the offence had involved themselves, the second one if the offence involved someone of their social network, and the third one if no one, neither the respondent nor people of his/her social network, had been involved in the last year before the survey. The choice of the first answer, however, did not let the chance of indicating an eventual experience of indirect victimization too.
assessing Materialistic and Post-materialistic values. The former exerted a positive effect on concrete fear, while the latter exerted a negative effect on abstract fear. Besides those variables already included in the previous model, we had to free the correlation between Materialistic and Post-materialistic values, because, by construction, these categories were mutually excluding.
As far as concerns direct and indirect victimization, education, size of the area of residence and news exposure, the direction and strength of the relations linking these variables to concrete and abstract fear were pretty analogous to those highlighted in the previous model.
CONCLUSION
A bulk of studies concerning the feeling of unsafety, mainly investigated in terms of fear of crime, has been published in the last decades. Just a few of these contributions approached the topic from a psycho-social standpoint, even if some authors underscored that psycho-social variables may contribute to explain different levels of fear of crime [see, for instance, Amerio & Roccato, 2005; Farrall, Bannister, Ditton, & Gilchrist, 2000; Van der Wurff, Van Staalduinen, & Stringer, 1989; Zani, Cicognani, & Albanesi, 2001]. We performed this research aiming at investigating the links between individual values priorities and crime-related fear. Structural equation modelling analysis, taking into account the variables traditionally used in research on fear of crime and those used to assess individual values priorities, allowed us to build two models including the most important predictors of concrete and abstract fear of crime in the Italian population. In the first one we modelled values as latent factors influencing our participants’ answers to the Portrait of Values Questionnaire short Italian version [Capanna, Vecchione, & Schwartz, 2005], while in the second one we used two dummies created from the four items developed by Inglehart [1990] to assess values. The other variables we included were the same in both the models; thus, we could compare the effects of two different values operationalisations.
As concerns our participants’ socio-demographic characteristics, only two variables have been kept in our final models: The size of the area of residence and years of education. The first one exerted a positive influence on concrete fear, just as in previous Italian research [see, Amerio & Roccato, 2007; Miceli, Roccato, & Rosato, 2004]. This variable may be considered as a proxy of the exposure to social and physical incivilities, since these disorders are likely to be more spread in large towns than in the smaller ones [Sartori, 2003]. Incivilities diffusion may contribute to increase (or even originate) concrete fear of crime since, in particular physical disorders, may represent a warning of an oncoming danger [Skogan, 1990]. Moreover, in large urban centres crime rates spread is tendentially higher than in small towns and this implies a higher probability of being a victim of an offence [Nardi, 2003]. Finally, in big cities human relations are more autonomous than those typical of rural areas: This may lead to a decrease in social support that usually exerts a moderating action on concrete fear [Lagrange, 1992].
We obtained analogous results to those gained in previous research also for what concerns the impact of the socio-demographic variables on concern about crime as a social problem [see Amerio & Roccato, 2005]: Both the size of the area of residence and the years of education negatively influenced abstract fear. This result may be interpreted considering people living in big towns as “socially
central”. An analogous interpretation may be done for the level of education [Barbagli & Maccelli, 1985]. Moreover, the exclusion of some socio-demographic variables we had to perform in order to improve our models’ fit was consistent with some results obtained in previous Italian research on fear of crime [see Miceli, Roccato, & Rosato, 2004], in which, among the socio-demographic variables taken into account the only one exerting a statically (and negative) significant influence on fear of crime was education.
To be a victim of an offence, as highlighted in the literature [Nardi, 2003; Roccato, 2007],
victimization experiences too. Even though in previous Italian research [Amerio & Roccato, 2005] direct victimization came out to be the most effective predictor of concrete fear, the relevant influence of indirect victimization is not inconsistent with the literature, in that it may be explained by the fact that, since it does not strongly push to use effective coping strategies, its impact on concrete fear may be sometimes at least the same of that of direct victimization [Hale, 1996].
However, differently from what previous Italian studies showed [Amerio & Roccato, 2005], people involved in direct victimization experiences were also more concerned about crime as a social problem. Thus, it seemed that, for our participants, victimization, besides influencing their perceptions, evaluations, and representations of their life-space, influenced the way they perceived, represented, and evaluated their social world also. Nonetheless, the variables exerting the strongest influence on concern about crime, just as underscored in previous research on this dimension of fear, were those pertaining to the mass-media domain. In particular, National TV news exposure had a strong impact on abstract fear, even though its direction was opposite to the one obtained in previous Italian research [Amerio & Roccato, 2005, 2007]: News exposure, to both Rai and Mediaset channel, reduced abstract fear.
However, it is important to point out that the variables we used in this study were different from the ones used in previous research. In Amerio and Roccato [2005, 2007] TV news exposure was measured as frequency of TV news watching, while in this research our participants were asked to report which TV news channel they used to watch more. Thus, our mass-mediatic variables may be considered as proxies of “social centrality”, just as education and size of the area of residence, we mentioned above, more than as variables actually assessing mass-media exposure. Future research performed using more satisfying mass-mediatic variables will be obviously welcome.
Thus, as a whole, even when values were added to the models aimed at predicting concrete and abstract fear of crime, many results found by sociological and criminological models still held. We now come to discuss the most innovative aspect of this research: the links between values priorities and
abstract and concrete fears. In general, values exerted a significant influence on both abstract and concrete fear, even when the impact of socio-demographic, victimization and mass-mediatic variables was partialled out.
As concerns the influence of values assessed with the Portrait Values Questionnaire short version, results partially confirmed our hypothesis based on Schwartz and colleagues’ [2000] research. In particular, our analysis revealed that the Self-enhancement/Self-transcendence dimension did not exert a significant influence on abstract and concrete fear: Although reflecting an opposition between values addressing the self and values addressing the whole community welfare, this dimension did not show any statistically significant relation that could have accounted for the similar polarization existing between concrete and abstract fear. This result may be due to the divergence existing between some of the contents of this dimension (especially power, achievement, benevolence, and universalism) and the particular fear we took into consideration. New research performed using a more analytic
operationalisation of Schwartz’s values will plausibly be fruitful.
On the other hand, a significant connection between Openness to change/Conservatism dimension and fear of crime did exist: People whose values priorities address Conservatism showed higher levels of both abstract and concrete fear. It is not surprising that giving importance to security, tradition and conformity may lead to an increased dread about crime, since crime and social disorder are some of the more dangerous threat that may exist for people pursuing these values [Duckitt, 2001]. As Schwartz and colleagues [2000] showed, adhesion to Conservatism values may foster self-constrained
behaviours, preference for order, and resistance to changes: All of them may allow people to avoid dangerous situations. It is reasonable that these ways of acting may constitute the basis for a vicious circle with fear, a paradoxical situation in which a fearful person enacts constrained behaviours leading him/her to be even more fearful [Liska, Sanchirico, & Reed, 1988]. A test of such hypothesis,
crime, at least in Italy, frequently includes fear of foreign people, very often stigmatized as criminals [Cima & Zambrino, 2008]. In this line of thinking, the positive influence of adhesion to Conservatism values on fear of crime may reflects the intolerance that people assuming conservative positions usually show [Nardi, 2003].
As concerns the second model, in which we used Inglehart approach to values,
Materialistic and Post-materialistic values exerted influence on both the dimensions of fear: As expected, traditional and survival values fostered concrete fear, while secular and self-expressive values negatively influenced concern about crime as a social problem. Thus, even if at first glance Inglehart’s and Schwartz’s theories are very different from each other, they brought to very similar results as concerns their relation with fear of crime. This is consistent with Inglehart [2006], who showed that, although based on different theories and strategies of measurement, the two approaches share many important aspects. Materialistic values stress physical and economic safety in the same way some facets of the Conservatism dimension do; thus, it is far from surprising that both exerted a
positive influence on concrete fear. On the other hand, Post-materialistic values stress free self-expression and quality of life, values priorities pretty close to Stimulation and Self-direction included in the Openness to change dimension; thus, it is far from surprising that both exerted a negative influence on concern about crime. Moreover, our two models accounted for almost the same proportion of variance of our dependent variables. Since it is reasonable to consider both the values operationalisations we used as satisfactory, it will be advisable, for future research, to chose one or the other based to the methodological characteristics of the survey and the sample in hand (e. g. preferring Inglehart’s items for samples with a low educational level; preferring Schwartz’s ones if interested in analytic assessment of values).
Altemeyer, B. (1996). The authoritarian specter. Cambridge, MA: Harvard University Press. Amerio, P. (1999). Le dimensioni della sicurezza [The dimensions of safety]. In P. Amerio (Ed.), Il
senso della sicurezza [The feeling of safety] (pp. 17-24). Milano: Unicopli.
Amerio, P. (2004). Problemi umani in comunità di massa [Human problems in mass communities]. Torino: Einaudi.
Amerio, P., Gattino, S., & Roccato, M. (2004). Insécurité et représentation du domaine politique [Unsafety and representation of the political world]. Psychologie et Société, 7, 89-114.
Amerio, P., & Roccato, M. (2005). A predictive model for psychological reactions to crime in Italy: An analysis of fear of crime and concern about crime as a social problem. Journal of Community and Applied Social Psychology, 15, 17-28.
Amerio, P., & Roccato, M. (2007). Psychological reactions to crime in Italy: 2002-2004. Journal of Community Psychology, 35, 91-102.
Austin, D. M., Woolever, C., & Baba, Y. (1994). Crime and safety-related concerns in a small community. American Journal of Criminal Justice, 19, 79-97.
Barbagli, M., & Maccelli, A. (1985). La partecipazione politica a Bologna [Political participation in Bologna]. Bologna: Il Mulino.
Barnea, M. F., & Schwartz, S. H. (1998). Values and voting. Political Psychology, 19, 17-40. Capanna, C., Vecchione, M., & Schwartz, S. H. (2005). La misura dei valori: Un contributo alla
validazione del Portrait Values Questionnaire su un campione italiano [Measuring values: A contribution to the validation of the Portrait Values Questionnaire on an Italian sample]. Bollettino di Psicologia Applicata, 246, 29-41.
Caprara, G., Schwartz, S. H., Capanna, C., Vecchione, M., & Barbaranelli, C. (2006). Personality and politics: Values, traits, and political choice. Political Psychology, 27, 1-27.
Cima, R., & Zambrino, S. [2008], Le priorità degli italiani [The priorities of the Italians]. In L. Ricolfi (Ed.), Ostaggi dello stato: Le origini politiche del declino e dell’insicurezza [Hostages of the state: The political origins of decline and of unsafety] (pp. 59-92). Milano: Guerini.
Duckitt, J. (2001). A dual-process cognitive-motivational theory of ideology and prejudice. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (Vol. 33, pp. 41-113). San Diego: Academic Press.
Farrall, S., Bannister, J., Ditton, J., & Gilchrist, E. (2000). Social psychology and the fear of crime: Re-examining a speculative model. British Journal of Criminology, 40, 339-413.
Feather, N. T. (1995). Values, valences, and choice: The influence of values on the perceived attractiveness and choice of alternatives. Journal of Personality and Social Psychology, 6, 1135-1151.
Ferraro, K. F. (1995). Fear of crime: Interpreting victimization risk. Albany, NY: State University of New York Press.
Furstenberg, F. F. (1971). Public reactions to crime in the streets. American Scholar, 40, 601-610. Galland, O., & Lemel, Y. (2006). Tradition-modernité: Un clivage persistant des sociétés européennes
[Tradition-modernity: A persistent cleavage in European societies]. Revue Française de Sociologie, 47, 687-724.
Gasperoni, G., & Giovani, F. (2002). L’inversione della polarità semantica nelle frasi delle scale Likert [The inversion of the semantic polarity in Likert scales items]. In A. Marradi & G. Gasperoni (Eds.), Costruire il dato 3 [Building the data 3] (pp. 123-190). Milano: Angeli.
Hale, C. (1996). Fear of crime: A review of the literature. International Review of Victimology, 4, 79-150.
Inglehart, R. (1990). Culture shift in advanced industrial societies. Princeton, NJ: Princeton University Press.
Inglehart, R. (2006). Mapping global values. Comparative Sociology, 5,115-136.
Lagrange, H. (1992). Appréhension et préoccupation sécuritaire [Worry and concern for safety]. Déviance et Société, 16, 1-29.
Liska, A. E., Sanchirico, A., & Reed, M. D. (1988). Fear of crime and constrained behaviour specifying and estimating a reciprocal effects model. Social Forces, 66, 827-837.
Kanan, J. W., & Pruitt, M. V. (2002). Modelling fear of crime and perceived victimization risk: The (in)significance of neighborhood integration. Sociological Inquiry, 72, 527-548.
Kennedy, L. W., & Silverman, R. A. (1985). Perception of social diversity and fear of crime. Environment and Behavior, 17, 275-295.
Killias, M., & Clerici, C. (2000). Different measures of vulnerability in their relation to different dimensions of fear of crime. British Journal of Criminology, 40, 437-450.
Martynowicz, E. (2006). Values accepted by youth. US-China Education Review, 3, 9-16.
Mayton, D. M., & Furham, A. (1994). Value underpinnings of antinuclear political activism: A cross-national study. Journal of Social Issues, 50, 117-128.
Miceli, R., Roccato, M., & Rosato, R. (2004). Fear of crime in Italy: Spread and determinants. Environment and Behavior, 36, 776-789.
Nardi, L. (2003). Il senso di insicurezza [The feeling of unsafety]. In M. Barbagli (Ed.), Rapporto sulla criminalità in Italia [Report on criminality in Italy] (pp. 525-554). Bologna: Il Mulino.
Ojala, M. (2005). Adolescents’ worries about environmental risks: Subjective well-being, values, and existential dimensions. Journal of Youth Studies, 8, 331-347.
Pantazis, C. (2000). “Fear of crime”, vulnerability and poverty. British Journal of Criminology, 40, 414-436.
Rice, G. (2006). Individual values, organizational context, and self-perceptions of employee creativity: Evidence from Egyptian organizations. Journal of Business Research, 59, 233-241.
Ricolfi, L. (1994). Elezioni e mass media: Quanti voti ha spostato la tv [Election and the mass media: The effects of televisions on the vote]. Il Mulino, 43, 1031-1044.
Roccato, M. (2007). Victimization and fear of crime. In K. T. Froeling (Ed.), Criminology: Research focus. New York: Nova Publishers.
Roccato, M. (2008). La rilevazione empirica dei valori [The empirical assessment of values]. Rassegna Italiana di Sociologia, 1, 39-60.
Roché, S. (1993). Le sentiment d’insécurité [The feeling of unsafety]. Paris: Presses Universitaires de France.
Rohan, M. J. (2000). A rose by any name? The values construct. Personality and Social Psychology Review, 4, 255-277.
Rokeach, M. (1973). The nature of human values. New York: Free Press.
Rountree, P. W., & Land, K. C. (1996). Perceived risk versus fear of crime: Empirical evidence of conceptually distinct reactions in survey data. Social Forces, 74, 1353-1376.
Sagiv, L., & Schwartz, S. H. (1995). Values priorities and readiness for out-group social contact. Journal of Personality and Social Psychology, 69, 437-448.
Sagiv, L., & Schwartz, S. H. (2000). Values priorities and subjective well-being: Direct relations and congruity effects. European Journal of Social Psychology, 30, 177-198.
Sartori, L. (2003). Degrado e paura per la criminalità [State of decay and fear of crime]. In M. Barbagli (Ed.), Rapporto sulla criminalità in Italia [Report on criminality in Italy] (pp. 489-524). Bologna: Il Mulino.
Schwartz, S. H. (1992). Universals in the content and structure of values: Theory and empirical tests in 20 countries. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 25, pp. 1-65). New York: Academic Press.
Schwartz, S. H., Sagiv, L., & Boehnke, K. (2000). Worries and values. Journal of Personality, 68, 309-346.
Schwartz, S. H., Struch, N., & Bilsky, W. (1990). Values and intergroup social motives: A study of Israeli and German students. Social Psychology Quarterly, 53, 185-198.
Skogan, W. (1990). Disorder and decline: Crime and the spiral of decay in American neighborhoods. New York: Free Press.
Skogan, W., & Maxfield, M. (1981). Coping with crime. Beverly Hills, CA: Sage.
Tyler, T. R. (1980). Impact of directly and indirectly experienced events: The origin of crime-related judgments and behaviors. Journal of Personality and Social Psychology, 39, 13-28.
Tyler, T. R., & Cook, F. L. (1984). The mass media and judgments of risk: Distinguishing impact on personal and societal level judgments. Journal of Personality and Social Psychology, 47, 693-708.
Van der Wurff, A., Van Staalduinen, L., & Stringer, P. (1989). Fear of crime in residential
environments: Testing a social psychological model. Journal of Social Psychology, 129, 141-160. Zani, B., Cicognani, E., & Albanesi, C. (2001). Adolescents’ sense of community and feeling of
unsafety in the urban environment. Journal of Community and Applied Social Psychology, 11, 475-489.
Figure captions
Figure 1. Confirmatory factor analysis of the Portrait Values Questionnaire. Figure 2. Predicting fear of crime using Schwartz’s approach to values. Figure 3. Predicting fear of crime using Inglehart’s approach to values.
Figure 1. Openness .16 Self enh. .34 .45 PVQ20 1 error 1.73 1 1 1 1 1 1 1 1 1 Response set .31 .4 9 1 PVQ14 PVQ1 PVQ5 PVQ6 PVQ21 PVQ2 PVQ4 PVQ8 PVQ17 PVQ19 error .54 error .94 error .93 error .91 error .70 error .82 error 1.08 error .84 error .77 error .58
Figure 2. abstract fear Openness .12 Self enh. .29 .45 concrete fear 1 1 size area residence error .19 1 1 1 1 .10 Rai
news Mediasetnews
1 1 -.76 -.42 -.40 .18 education direct victimiz. indirect victimiz. error 16.77 error .29 error 3.63 error .25 error .23 error .22 error .20
Figure 3. abstract fear concrete fear 1 1 size area residence error .19 1 1 1 1 .10 Rai
news Mediasetnews
1 1 -.76 -.41 -.38 .19 education direct victimiz. indirect victimiz. error 16.77 error .29 error 3.63 error .25 error .23 error .21 error .20 Post material. Material. error .08 error .22 1 1 -.22 -.17 .17
