School Cheating and Social Capital

Abstract

In this paper we investigate the correlation between cheating in school tests and social capital in the context of Italian standardized tests administered by Invalsi. Given that such tests are low-stakes, we find natural to interpret the strong opposition showed by teachers and students (sometimes resulting in explicit boycotting of the tests) as signs of low trust towards the central education authority. Indeed, we find a robust negative correlation between cheating and social capital. We also confirm previous results that have documented that: (i) the presence of an external examiner strongly reduces cheating (Bertoni, Brunello, and Rocco, 2012); and (ii) in less homogeneus classes cheating is more difficult (Bertoni, Brunello, and Rocco, 2012; Lucifora and Tonello, 2012). Furthermore, we show that cheating is instead positively correlated with measures of particularistic social values: such measures (as opposed to universalistic social values) are signs of strong and close links that end up facilitating collusive behaviours.

Measuring the Performance of Local Governments: Evidence from Mystery Calls

“…25 Paccagnella and Sestito (2014) show that this measure is negatively correlated to social capital, i.e. cheating scores are indicators of the scarcity of (universalistic) measures of social capital in the municipality.…”

“…measure of social capital that we use (i.e. the cheating scores) refers to statistical indicators of opportunistic behavior (on the part of students and/or teachers) in primary schools that occurred during the administration of the Italian national schools assessment tests (INVALSI National Assessment Program, 5th grade students) (Guiso et al, 2013, Paccagnella andSestito, 2014). 25 We then use average crime rates of organized crime offenses 26 and corruption at the municipal level, derived from the SDI.…”

“…, and the In-house technology dummy variable are defined in Appendix A A. The Lack of employees dummy variable indicates whether in the municipalities there are less employees than the number of employees defined by the law; the Social capital (lack of ) variable refers to statistical indicators (cheating scores) of cheating and opportunistic behaviors in primary schools that occurred during the administration of the Italian national assessment tests (INVALSI) (Guiso et al, 2013, Paccagnella andSestito, 2014); Organized crime and Corruption are crime rates per 10,000 total population (average over the period [2004][2005][2006][2007][2008][2009][2010][2011]; Population density is the ratio between the number of residents and the municipality area (squared Km); Firms pc is the number of per-capita registered firms in the municipality; the Morning dummy variable takes the value of 1 if the interview was started in the morning (0 otherwise); Btw 5,000 and 15,000 inhab. is a dummy variable which takes value 1 for municipalities in the population range; a similar definition occurs for the other population ranges (the excluded category are municipalities below 5,000 inhabitants).…”

Law Enforcement, Social Control and Organized Crime: Evidence from Local Government Dismissals in Italy

“…In the Figure we make use of the measure of civicness taken from cheating in standardized test scores in primary schoolsPaccagnella and Sestito 2014). The results are robust to the use of alternative measures of civic capital extensively used in the literature, such as the number of non-for profit organizations and electoral participation (seeAppendix Table A6).…”

The More Educated, the More Engaged? An Analysis of Social Capital and Education

“…However, we also endorse the need, argued by De Blasio et al (2014), to better understand social behaviour by taking into account the dichotomy between universalism and particularism. In the same vein but unlike Putnam's network-based distinction between bonding and bridging social capital, this dichotomy cuts across both the network-based and the values-related dimensions of social capital (Paccagnella & Sestito, 2014). As in the case of De Blasio et al, (2014), to measure both universalism and particularism across different social capital dimensions we rely on the definition proposed by Baurmann (2007), who points out how networks should be developed and maintained to make social capital useful for society and the democracy as a whole: "A group is all the more particularistic, the more its networks, its norms of reciprocity and trust and its aims are confined to the members of the group, whereas a group is all the more universalistic, the more its networks, its norms of reciprocity and trust and its aims transgress the confines of the group and encompass other citizens and groups in a society" (Baurmann, 2007: 173).…”