Explainable machine learning for predicting homicide clearance in the United States

Campedelli, Gian Maria

doi:10.1016/j.jcrimjus.2022.101898

Cited by 9 publications

(15 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although conflicting evidence has been found with respect to discretionary action by criminal investigators, there is nonetheless a wealth of studies that have shown the association between extra-legal factors and clearance, specifically those pertaining to victims' age, gender, and ethnicity (for example, Addington, 2006;Alderden and Lavery, 2007;Campedelli, 2022;Jiao, 2007;Lee, 2005;Litwin and Xu, 2007). At the same time, several studies have reported a significant correlation between the homicide clearance rate and factors related to the criminal event and its context such as the weapon used, the time and location of the homicide, and the motivations behind the homicide (for example, Alderden and Lavery, 2007;Carter and Carter, 2016;Keel et al, 2009;Rydberg and Pizarro, 2014).…”

Section: Discussionmentioning

confidence: 99%

“…Understanding of the factors that influence the resolution of homicide cases is still limited in several respects. First, much of the literature on homicide clearance has focused on the United Statesmore generally, on English-speaking countriesand provides partially contrasting results with respect to the relevance of multiple factors (for example, Addington, 2007;Alderden and Lavery, 2007;Braga and Dusseault, 2018;Campedelli, 2022;Hawk and Dabney, 2014;Hough et al, 2019;Jiao, 2007;Lee, 2005;Litwin, 2004;Mancik et al, 2018;Ousey and Lee, 2010;Pizarro et al, 2020;Puckett and Lundman, 2003;Roberts, 2007;Roberts andLyons, 2009, 2011). Second, few studies have efficaciously considered and operationalized the interaction of victims' characteristics, such as victim's socio-economic condition, that might trigger discretionary actions by police forces, and non-discretionary factors, such as weapon used to commit the homicide (Petersen, 2017;Vaughn, 2020).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Homicide clearance: Discretionary and non-discretionary factors

Guerette

Persurich

2023

Preprint

View full text Add to dashboard Cite

Previous studies have produced mixed findings regarding discretionary and non-discretionary factors associated with the likelihood of homicides being cleared. Performing Pearson's χ 2 test, logistic regressions, and random forest analyses on all homicide cases that occurred in Italy in 2014, we assess if factors pertaining to the discretionary domainthat is, nationality, age, sex, socioeconomic status of the victim, previous convictionsand non-discretionary factorsfor example, weapon, location, circumstanceare correlated with the outcome of the investigation. The empirical analysis does not identify relations between victims' nationality, socioeconomic status, previous criminal records and homicide clearance. On the other hand, homicides whose victim is male are less likely to be cleared. While high media coverage correlates with a higher clearance rate, low media coverage is not correlated with the homicide case remaining unsolved. Finally, especially in Southern Italy where mafia organizations are more entrenched, homicides committed in a criminal environment have a lower chance of being cleared. With respect to Italy, the results indicate a lack of support for the discretionary perspective that argues' that police may use discretion in clearing homicide cases according to the sociodemographic characteristics of victims.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Homicide clearance: Discretionary and non-discretionary factors

Guerette

Persurich

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Based on the available evidence, the hypothesis is that the higher the number of victims, the higher the likelihood of evidence leading to clearance. (Addington, 2007;Lee, 2005;Sturup et ( Alderden & Lavery, 2007;Avdija et al, 2022;Campedelli, 2022;Lee, 2005;Puckett & Lundman, 2003;Regoeczi et al, 2008;Riedel & Rinehart, 1996;Wolfgang, 1958) Victim's Age…”

Section: Countmentioning

confidence: 99%

“…More importantly, despite the salience of the question, we lack reliable estimates mapping the racial disparities in homicide clearance to date. Notwithstanding a rich production of studies directly or indirectly addressing the issue of race effects in homicide clearance (Addington, 2006;Campedelli, 2022;Cardarelli & Cavanagh, 1994;Lee, 2005;Paintsil, 2022;Puckett & Lundman, 2003;Regoeczi et al, 2008;Riedel & Rinehart, 1996;Roberts & Lyons, 2011;Roberts & Smith, 2023;Vaughn, 2020), no comprehensive and credible empirical quantification of this extremely policy-relevant divide is available. Although precious in contributing to knowledge on the broader issue of homicide clearance research, previous studies have faced two critical limitations due to data availability constraints.…”

Section: Introductionmentioning

confidence: 99%

Homicides involving Black victims are less likely to be cleared in the United States

Campedelli

2024

Criminology

Self Cite

View full text Add to dashboard Cite

Does a victim's race explain variation in the likelihood of homicide clearance? Attempts to address this issue date back to the 1970s. Yet, despite its theoretical and policy relevance, we lack a comprehensive and clear empirical answer to this critical question. Here, I causally focus on this problem by investigating racial disparity in homicide clearance in the United States, exploiting two sources covering the 1991–2020 period: the Murder Accountability Project data set (N = 522,278) and the National Incident‐Based Reporting System data set (N = 98,677). I primarily analyze these sources by employing exact matching to achieve perfect covariate balance and subsequently isolate the effect of race on the probability of clearance. For comparative purposes, I also use regression adjustment without matching obtaining complementary estimates. I demonstrate that the likelihood of clearance is 3.4 to 4.8 percent lower for homicides involving Black victims, depending on the sampling and estimation approach. In addition, I empirically show that this race effect is slightly higher for males and that racial disparity has moderately but significantly increased over time. These findings contribute to the extensive amount of evidence on discrimination affecting Black individuals in the administration of justice in the United States, calling for structural efforts to reduce this divide.

show abstract

“…Especially in contexts other than the United States, much remains to be understood about the relationship between specific features of murders and their investigations and the chances of murder cases being cleared (Brookman et al, 2019;Liem et al, 2019;Pizarro et al, 2020;Regoeczi et al, 2020;Vaughn, 2020). Related to this, the literature does not give a final answer on the explanatory power of the principal conceptual perspectives on the explanations of homicide clearances: the discretionary and the nondiscretionary perspective (Campedelli, 2022;Liem et al, 2019). Accordingly, this study explores the association between the clearance of homicide cases in Italy and factors related to discretionary (Black, 1976;Paternoster, 1984;Peterson and Hagan, 1984) and non-discretionary (Gottfredson and Hindelang, 1979;Klinger, 1997;Roberts, 2007) perspectives.…”

Section: Current Studymentioning

confidence: 99%

Homicide clearance: Discretionary and non-discretionary factors

Aziani

Persurich

2022

European Journal of Criminology

View full text Add to dashboard Cite

Previous studies have produced mixed findings regarding discretionary and non-discretionary factors associated with the likelihood of homicides being cleared. Performing Pearson’s χ2 test, logistic regressions, and random forest analyses on all homicide cases that occurred in Italy in 2014, we assess if factors pertaining to the discretionary domain – that is, nationality, age, sex, socioeconomic status of the victim, previous convictions – and non-discretionary factors – for example, weapon, location, circumstance – are correlated with the outcome of the investigation. The empirical analysis does not identify relations between victims’ nationality, socioeconomic status, previous criminal records and homicide clearance. On the other hand, homicides whose victim is male are less likely to be cleared. While high media coverage correlates with a higher clearance rate, low media coverage is not correlated with the homicide case remaining unsolved. Finally, especially in Southern Italy where mafia organizations are more entrenched, homicides committed in a criminal environment have a lower chance of being cleared. With respect to Italy, the results indicate a lack of support for the discretionary perspective that argues’ that police may use discretion in clearing homicide cases according to the sociodemographic characteristics of victims.

show abstract

Explainable machine learning for predicting homicide clearance in the United States

Cited by 9 publications

References 75 publications

Homicide clearance: Discretionary and non-discretionary factors

Homicide clearance: Discretionary and non-discretionary factors

Homicides involving Black victims are less likely to be cleared in the United States

Homicide clearance: Discretionary and non-discretionary factors

Contact Info

Product

Resources

About