{PSOCIAL}

Mexico ranked 59th out of 163 countries in the Positive Peace Index (PPI), “with a better overall score than both the global and the Central American and Caribbean regional averages” (Institute for Economics and Peace, 2018b, p. 41). When a country ranks higher in the PPI than the GPI (Global Peace Index), this indicates that it has the potential to improve its levels of peacefulness (Institute for Economics and Peace, 2018b); the present study seeks to contribute to this goal.

In the context of violence and low peace indexes in Mexico, collecting information that may help reduce these issues is essential. The CC and CE scales constitute useful tools in the Mexican context for the constant assessment and measurement of community processes that help strengthen solidarity, compliance with norms, participation, and autonomy in problem-solving. These positive community characteristics can contribute to decreasing violence and insecurity and, better still, promote peace.

Regarding the first objective of this study, the psychometric properties of the CC and CE scales display high levels of internal reliability with coefficients of .90 for the former and .91 for the latter, suggesting that these scales can be used with Mexican university students. These data are consistent with those collected previously in Colombia, where Cronbach’s alpha values for CC were found to be .91 (Ruiz, 2005), .93 (Ruiz, 2007a, 2007b), and .94 (Ruiz, 2010). In regard to CE, the level of internal reliability in this study is also similar to that reported in previous studies with values of .90 (Carroll & Reese, 2003, as cited in Carroll, Rosson, Zhou, 2005; Espinosa & Ruiz, 2012).

In Ruiz (2007a), it was suggested that a study with a larger sample be carried out to clarify the factorial structure of the CC construct, which might—as was hypothesized—yield a one-dimensional solution as a result. In Mexico, a larger sample than the one in previous studies (n = 1007) was made available. For example, in Ruiz (2007a), the sample size was n = 253. Sample sizes could explain the differences in factorial structure. Moreover, the strong direct correlations between CC and CE levels in the current study and others (Ruiz, 2010; Ruiz & Turcios, 2009; Ruiz-Pérez, 2019) raise another point: are CC and CE different constructs or can one be a component of the other? This question should be studied in future research since it goes beyond the purpose of the present study. Likewise, it is worth continuing to explore the relationship between CC and CE, and homicide rates.

As for the CE scale, the three-factor solution found in this study largely coincides with the four-factor solution found by Carroll et al. (2005), with a smaller sample (n = 146). In the case of the present article, with a sample almost six times larger, it follows that the structure should be simplified to three factors: care, cooperation, and autonomy, with the second factor from Carrol et al. (2005) being distributed among axes 1 and 2 of the present result. Moreover, the structure coincides with the factors previously identified in the model developed by Ruiz et al. (2012) within the Hispanic population.

Since CC surveys are generally used to identify and monitor cities that have made the construction of citizenship a key element of their public security policy, the measurements yielded by this type of constructs are useful for comparing communities and identifying areas where changes for the promotion of more peaceful, democratic societies can be implemented (Corpovisionarios, 2015; Mockus et al., 2012).

As a second objective of this research on the assessment of the association between CC, CE, and socioeconomic status, the results in this study indicate that, as in previous studies (Ruiz, 2010), the higher the socioeconomic status, the higher the levels of CC and CE. This fact requires the social scientists to continue shedding light on poverty and to continue intervening to narrow social gaps. It is important to point out that 53,400,000 people in Mexico (a third of the country’s population) live in poverty (CONEVAL, 2018). According to the Standardized World Income Inequality Database (2016) and the Global Wealth Report (2014), as cited in Esquivel (2015), Mexico is within the 25% comprising the most unequal countries, 21% of Mexico’s total income belongs to the wealthiest 1% in the country, while the richest 10% owns 64.4% of the country’s wealth. Whereas the number of millionaires in Mexico grew by 32% between 2007 and 2012, in the rest of the world, the number decreased by 0.3% (Esquivel, 2015). Even though this context permeates the levels of CC and CE in the country, the opposite effect can occur: if CC and CE are strengthened, inequality between Mexicans could decrease. This issue should be addressed in subsequent research.

Moreover, the fact that higher CC is reported for younger ages may indicate that efforts in civic and citizen education (Flórez, 2011) are bearing fruit in the upcoming generations of Mexicans. This is particularly true in the group of Mexican university students who were active students in the past decade. This hypothesis is valuable and hopeful and should therefore be addressed in future research within this line of inquiry.

In regard to the fact that men show a stronger sense of belonging than women, given the evidence of an inverse association between CC and CE and the perception of security (Ruiz, 2007a, 2007b; Ruiz et al., 2012; Valenzuela, 2012), the fact that women feel more insecure in public as well as private places (INEGI, 2018) could influence this result.

Lastly, with respect to the third objective of the present study, data confirm the significant inverse relationship between crime victimization, CC and CE that has been found in previous studies (Ruiz, 2007b, 2016). This finding is consistent with the result showing Chihuahua to be the city with the least CE, particularly as regards the autonomy factor, at least at the individual level, because the ICC1 of CC and CE failed to show a group effect. Chihuahua is ranked 26th in Mexico’s peace index, while Jalisco and Puebla are ranked 15th and 7th respectively (Institute for Economics and Peace, 2018b). Furthermore, according to data from the ENVIPE (INEGI, 2017), 74.2% of Chihuahua citizens feel insecure, as opposed to 65.7% in Jalisco, and 68.1% in Puebla. In keeping with this information, in an earlier study, significant differences were found in Chihuahua in comparison with Guadalajara and Puebla regarding victimization reports by university students (Chan et al., 2017). The results of this study showed a consistent pattern. There were more victimization reports by Chihuahua students, followed by Puebla students, in 15 out of 16 victimization indicators, with a higher victimization mean in the former state. In the same study, some aspects, such as sexual aggression and harassment, increased two-, three-, or five-fold in Chihuahua, in comparison with Puebla and Guadalajara.

Moreover, the correlations between scale scores and objective violence indicators (Citizen Council for Public Security and Criminal Justice, 2016) show that responses to the scales can be considered indirect indicators of generalized criminal violence and its effect on the social network. It is striking that, in Chihuahua, despite the low collective efficacy and a low autonomy factor, there was a higher score in the factor related to active participation in community issues. This could be explained as a collective response mechanism to the violence present in this context, and possibly the activity of Non-Governmental Organizations.

Finally, we would like to point out certain limitations of the present study. First, the samples are not representative of the cities or the country in which the research was conducted. This issue affects the external validity of the measurements of psychological constructs such as EC and CC. Caution must therefore be exercised in generalizing results when comparing cities, especially when the effect sizes found in each pair of cities are weak. However, it is important to recognize the usefulness and legitimacy of using samples consisting exclusively of university students to build indicators representative of regions of societies (Páez & Vergara, 2000). In this respect, the literature suggests finding theoretically coherent relations between psychosocial indicators at the aggregate level and macro-objective indicators, since people ultimately inhabit contexts and perceive social realities. For example, the pioneering efforts of Hofstede (1991, 2011) to culturally describe over 50 countries were based on surveys of IBM employees in each of these societies. It is important to consider the differences that could exist in the answers to these scales in different populations from those in this research. It is necessary to continue this line of research to include other population groups in specific communities.

Second, although the results of the present study show correlational relationships between CC, CE and objective violence indicators, these data must be observed with caution. These correlational results could be moderated by other variables. For example, factors such as the perception of the police have been found to play a greater role in crime reduction than collective efficacy (Xu et al., 2005). In futures studies, it would be important to consider this this of variables and their role in the relationship between CC, CE and victimization. Apropos of this, Sampson (2003) wrote that high collective efficacy has an effect on a society with lower levels of criminality, but that high criminality causes a low level of collective efficacy. The results of the association with indicators of violence and victimization warrant further research.

Just as Carroll et al. (2005) proposed, the CE measurement is important for recognizing a community’s capacities. If there are positive beliefs in CE, it is more likely that common goals will be pursued through collective actions, resources will be utilized better, and the community will be more persistent in coping with failure and producing quicker, more positive results for all its members. Conversely, high levels of insecurity or violence (such as organized crime) can hamper efforts to achieve community integration, collective efficacy and/or citizenship culture. In other words, when can collective efficacy reduce crime? Is it possible to build collective efficacy in communities with high levels of violence, not only in social environments with low or medium crime rates?

In this sense, the main limitation of this study is related to the cross-sectional design used. Ideally, the relationship between CC and CC with respect to crime should be tested longitudinally, especially through an intervention design that could determine whether an improvement in the constructs measured by these scales through the implementation of various programs reduces crime. Likewise, it would be useful to observe whether certain programs designed to fight crime have an effect on such constructs, or whether it is necessary to implement specific measures to improve citizen culture and collective effectiveness in addition to reducing crime itself.

The results of our study provide tools for assessing the levels and improvements of CC and CE, after any program or implementation strategy is put into effect, as well as their relationship with crime and victimization indicators. In addition, based on the factorial structure found in both scales, the proposed categories warrant further study and a more detailed analysis of indicators with respect to the two-way relationship between communities and security and peace.

[1] In 2008, the peace index was calculated for 143 countries. Between from 2017 and 2020, it was calculated for 163 countries.

[2] This data displays a dark figure of approximately 94%.