Skip to main content

A Conditional Likelihood Model of the Relationship Between Officer Features and Rounds Discharged in Police Shootings



Assess whether the number of rounds fired in an officer-involved shooting is related to police officer features.


The data come from 55 member agencies in the Major Cities Chiefs Association. The full dataset describes 2574 officers involved in 1600 shootings between 2010 and 2018 but only incidents involving multiple officers provide information. Our final dataset included 317 shooting incidents involving 849 officers and 5026 rounds.


We match officers on the scene of a shooting incident and develop a conditional truncated Poisson model that eliminates confounding due to time, place, and environment. We use a permutation test to formally assess the strength of the relationship between officer features and shooting rate.


We find no officer feature strongly predicts shooting rate. Age at recruitment, age at the time of the shooting, and years of experience all had relative rates nearly equal to 1.0. There was no statistical relationship with an officer being female (p = 0.27), black (p = 0.64), or Hispanic (p = 0.39). Having prior involvement in shootings, prior force complaints, and special assignments appear to elevate the relative rate of shooting, but all confidence intervals included 1.0.


Officer features appear to have little or no relationship with shooting rate. These findings correspond with police scholars’ supposition that duty assignment may be more responsible for explaining differences in police use of force than individual officer characteristics. It contrasts with some prior research suggesting that officer race, age at recruitment, and prior performance affect shooting risk. In doing so, these results also lend support to theoretical frameworks emphasizing the role of organizational features and other system-level factors over individual-level explanations for police use of force. The proposed methodology addresses bias due to confounding, but demands a large number of shootings. Expanded participation in multi-agency data collections and including data on all non-shooting officers at the scene of the incident can increase precision.

This is a preview of subscription content, access via your institution.

Fig. 1


  1. The Major Cities Chiefs Association (MCCA) is a professional association of the sixty-nine largest law enforcement agencies in the United States, and nine largest in Canada. It provides a forum for executives to share ideas and experiences, as well as influence national public policy on law enforcement matters and encourage research.

  2. White and Klinger (2012) found that 25% of Philadelphia police shootings involved multiple shooting officers, Klinger (2004) found that 45% of shootings across 19 departments involved multiple shooters, and Klinger et al. (2015) reported that 22% of shooting in St. Louis involved multiple shooters. In our data, we find the percentage of shootings involving multiple officers to be in the same range.

  3. We account for the fact that we do not have data on non-shooters at the scene of these incidents. If we had simply assumed that the observed shot count was from a Poisson distribution, then computation would be trivial but would produce biased estimates. Instead, this methodology assumes the observed shot counts come from a truncated Poisson distribution, \(P\left( {Y = y | Y > 0} \right) = \frac{{e^{ - \theta } \theta^{y} }}{{y!\left( {1 - e^{ - \theta } } \right)}}\). This greatly complicates the computational effort needed to obtain estimates, but allows us to obtain consistent and asymptotically unbiased estimates. The absence of data on non-shooters decreases precision, but does not introduce bias when using a truncated Poisson.


  • Baker A (2007, March 17) 3 detectives are indicted in 50-shot killing in queens. New York Times

  • Braga AA, Cook PJ (2018) The association of firearm caliber with likelihood of death from gunshot injury in criminal assaults. JAMA Netw Open 1(3):e180833

    Article  Google Scholar 

  • Cascio WF (1977) Formal education and police officer performance. J Pol Sci Admin 5(1):89–96

    Google Scholar 

  • DeGue S, Fowler KA, Calkins C (2016) Deaths due to use of lethal force by law enforcement: findings from the national violent death reporting system, 17 U.S. States, 2009–2012. Am J Prev Med 51(5):S173–S187

    Article  Google Scholar 

  • Donner CM, Maskaly J, Piquero AR, Jennings WG (2017) Quick on the draw: assessing the relationship between low self-control and officer-involved police shootings. Pol Q 20(2):213–234

    Article  Google Scholar 

  • Fridell LA (2017) Explaining the disparity in results across studies assessing racial disparity in police use of force: a research note. Am J Crim Justice 42:502–513

    Article  Google Scholar 

  • Fyfe JJ (1978) Shots fired: an examination of New York City police firearms discharges. PhD dissertation, School of Criminal Justice, State University of New York at Albany

  • Fyfe JJ (1981) Who shoots? A look at officer race and police shooting. J Pol Sci Admin 9(4):367–382

    Google Scholar 

  • Fyfe JJ (1988) Police use of deadly force: research and reform. Justice Q 5(2):165–205

    Article  Google Scholar 

  • Geller WA, Karales KJ (1981) Shootings of and by Chicago Police: uncommon crises-part I: shootings by Chicago Police. J Crim Law Criminol 72(4):1813–1866

    Article  Google Scholar 

  • Goldstein H (1967) Police policy formulation: a proposal for improving police performance. Mich Law Rev 65(6):1123–1146

    Article  Google Scholar 

  • Good P (1994) Permutation tests: a practical guide to resampling methods for testing hypotheses. Springer, New York

    Book  Google Scholar 

  • Ho T (1994) Individual and situational determinants of the use of deadly force: a simulation. Am J Crim Justice 18(1):41–60

    Article  Google Scholar 

  • Kalbfleisch JD, Sprott DA (1973) Marginal and conditional likelihoods. Sankhyā Indian J Stat Ser A 35(3):311–328

    Google Scholar 

  • Klinger D (2004) Into the kill zone: a cop’s eye view of deadly force. Jossey-Bass, San Francisco

    Google Scholar 

  • Klinger D (2005) Social theory and the street cop: the case of deadly force. Ideas in American Policing, Number 7. Police Foundation, Washington

    Google Scholar 

  • Klinger D, Rosenfeld R, Isom D, Deckard M (2015) Race, crime, and the micro-ecology of deadly force. Criminol Public Policy 15:193–222

    Article  Google Scholar 

  • Liseo B (2015) Likelihoods that eliminate nuisance parameters. In: Wright J (ed) International encyclopedia of the social and behavioral sciences, vol 14, 2nd edn. Elsevier, Amsterdam, pp 120–124

    Chapter  Google Scholar 

  • McElvain JP, Kposowa AJ (2008) Police officer characteristics and the likelihood of using deadly force. Crim Justice Behav 35(4):505–521

    Article  Google Scholar 

  • Morrison GB (2006) Deadly force programs among larger U.S. Police Departments. Po Q 9(3):331–360

    Article  Google Scholar 

  • Nowacki JS (2015) Organizational-level police discretion: an application for police use of lethal force. Crime Delinq 61(5):643–668

    Article  Google Scholar 

  • Paoline EA, Terrill W (2007) Police education, experience, and the use of force. Crim Justice Behav 34(2):179–196

    Article  Google Scholar 

  • Paoline EA, Gau JM, Terrill W (2018) Race and the police use of force encounter in the United States. Br Cournal Criminol 58(1):54–74

    Google Scholar 

  • Perrow C (1984) Normal accidents: living with high-risk technologies. Basic Books, New York

    Google Scholar 

  • Reason J (1990) Human error. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • Reiss AJ (1972) Police brutality. In: Radzinowicz L, Wolfgang ME (eds) Crime and justice: the criminal in the arms of the law, vol 2, pp 293–308

  • Remington FJ (1965) The role of police in a democratic society. J Crim Law Criminol 56(3):361–365

    Article  Google Scholar 

  • Ridgeway G (2016) Officer risk factors associated with police shootings: a matched case-control study. Stat Public Policy 3(1):1–6

    Article  Google Scholar 

  • Riksheim EC, Chermak SM (1993) Causes of police behavior revisited. J Crim Justice 21(4):353–382

    Article  Google Scholar 

  • Sherman LW (1980) Causes of police behavior: the current state of quantitative research. J Res Crime Delinq 17(1):69–100

    Article  Google Scholar 

  • Sherman LW (2018) Reducing fatal police shootings as system crashes: research, theory, and practice. Ann Rev Criminol 1:421–449

    Article  Google Scholar 

  • Sherman LW, Cohn EG, Gartin PR, Hamilton EE, Rogan DP (1986) Citizens killed by big city police, 1970–1984. Crime Control Institute, Washington

    Google Scholar 

  • Sklansky DA (2006) Not your father’s police department: making sense of the new demographics of law enforcement. J Crim Law Criminol 96(3):1209

    Google Scholar 

  • State of Illinois vs. City of Chicago, Consent Decree, 17-cv-6260 (United State District Court for the Northern District of Illinois September 13, 2018)

  • Terrill W (2001) Police coercion: application of the force continuum. LFB Publishing, New York

    Google Scholar 

  • White M, Klinger D (2012) Contagious fire? An empirical assessment of the problem of multi-shooter, multi-shot deadly force incidents in police work. Crime Delinq 58(2):196–221

    Article  Google Scholar 

  • Zimring FE (2017) When police kill. Harvard University Press, Cambridge

    Book  Google Scholar 

Download references


The data used in this analysis was collected by the National Police Foundation with the assistance of the Major Cities Chiefs Association. Funding for the data collection was provided by Arnold Ventures. The research presented in this article represents a consensus of the authors and does not necessarily represent the official position or policies of the National Police Foundation, Major Cities Chiefs Association, or Arnold Ventures. The authors appreciate the assistance of Dr. Amber Askey who worked on developing the data collection during her time at NPF.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Greg Ridgeway.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A: Efficient Computation of the Conditional Log-Likelihood

Appendix A: Efficient Computation of the Conditional Log-Likelihood

A.1 Derivation of Recursive Computation

Estimating the maximum likelihood estimator requires repeated evaluation of the log conditional likelihood function shown in (11). Let

$$Q_{si} \left( \rho \right) = \frac{1}{\rho !}{ \exp }\left( {\left( {\rho - r_{si} } \right)\beta^{{\prime }} {\mathbf{x}}_{si} } \right)$$

In (13) we drop the \(r_{si} !\) that had appeared in (11) since these are constants that can factor out. Equation (10) demonstrates this most clearly as the 144 can factor out of the denominator.


$$A_{s} \left( {k,n_{r} } \right) = \mathop \sum \limits_{{\mathop \sum \limits_{i = k}^{{n_{s} }} \rho_{i} = n_{r} ,\rho_{i} > 0}} \mathop \prod \limits_{i = 1}^{{n_{s} }} Q_{si} \left( {\rho_{i} } \right)$$

where \(n_{s}\) is the number of officers involved in shooting s and \(n_{r} = \mathop \sum \limits_{i = 1}^{{n_{s} }} r_{si}\) is the total number of rounds fired in shooting s.

So that the log conditional likelihood is

$$\log L\left( \beta \right) = - \mathop \sum \limits_{s = 1}^{S} { \log }\left( {A_{s} \left( {1,n_{r} } \right)} \right)$$

We gain efficiency by factoring out \(Q_{s,1} \left( {\rho_{1} } \right)\) in (14) so that we can define \(A_{s} \left( {k,n_{r} } \right)\) recursively.

$$A_{s} \left( {1,n_{r} } \right) = \mathop \sum \limits_{{\rho_{1} = 1}}^{{n_{r} - n_{s} + 1}} \left[ {Q_{s,1} \left( {\rho_{1} } \right)\mathop \sum \limits_{{\mathop \sum \limits_{i = 2}^{{n_{s} }} \rho_{i} = n_{r} - \rho_{1} ,\rho_{i} > 0}} \mathop \prod \limits_{i = k + 1}^{{n_{s} }} Q_{si} \left( {\rho_{i} } \right)} \right]$$
$$= \mathop \sum \limits_{{\rho_{1} = 1}}^{{n_{r} - n_{s} + 1}} Q_{s,1} \left( {\rho_{1} } \right)A_{s} \left( {2,n_{r} - \rho_{1} } \right)$$

Essentially, this recursion works by setting \(\rho_{1}\) to a specific value in the summation, then summing over all the other combinations of \(\rho_{i}\)s that sum to \(n_{r} - \rho_{1}\) and multiplying the result by \(Q_{s,1} \left( {\rho_{1} } \right)\). Then the sum is over fewer officers. We can repeat the recursion for \(A_{s} \left( {2,n_{r} - \rho_{1} } \right)\), setting specific values for \(\rho_{2}\) and summing over the remaining \(\rho_{i}\)s that sum to \(n_{r} - \rho_{1} - \rho_{2}\). More generally

$$A_{s} \left( {k,n} \right) = \mathop \sum \limits_{{\rho_{k} = 1}}^{{n_{r} - n_{s} + k}} Q_{s,k} \left( {\rho_{k} } \right)A_{s} \left( {k + 1,n - \rho_{k} } \right)$$
$$A_{s} \left( {n_{s} ,n} \right) = Q_{{s,n_{s} }} \left( {\rho_{{n_{s} }} } \right)$$

A.2 Monte Carlo Estimation of \(A_{s} \left( {1,n_{r} } \right)\)

For shootings that are complex, such as when \(\left( {\begin{array}{*{20}c} {n_{r} - 1} \\ {n_{s} - 1} \\ \end{array} } \right) > 10^{8}\), the recursive computation in “Appendix A.1” is slow. We can approximate a shooting’s contribution to the log-likelihood using a Monte Carlo estimate of the expected value of a function of a draw from a multinomial distribution. From (13) and (14) we have

$$A_{s} \left( {1,n_{r} } \right) = \mathop \sum \limits_{{\mathop \sum \limits_{i = 1}^{{n_{s} }} \rho_{i} = n_{r} ,\rho_{i} > 0}} \mathop \prod \limits_{i = 1}^{{n_{s} }} \frac{1}{{\rho_{i} !}}{ \exp }\left( {\left( {\rho_{i} - r_{si} } \right)\beta^{{\prime }} {\mathbf{x}}_{si} } \right)$$

Rewrite the indices on the summation so that the \(\rho_{i}\) start at 0. In this way the \(\rho_{i}\) count how many rounds in excess of 1 the ith officer discharges.

$$A_{s} \left( {1,n_{r} } \right) = \mathop \sum \limits_{{\mathop \sum \limits_{i = 1}^{{n_{s} }} \rho_{i} = n_{r} - n_{s} ,\rho_{i} \ge 0}} \mathop \prod \limits_{i = 1}^{{n_{s} }} \frac{1}{{\left( {\rho_{i} + 1} \right)!}}{ \exp }\left( {\left( {\rho_{i} + 1 - r_{si} } \right)\beta^{{\prime }} {\mathbf{x}}_{si} } \right)$$
$$= \mathop \sum \limits_{{\mathop \sum \limits_{i = 1}^{{n_{s} }} \rho_{i} = n_{r} - n_{s} }} \mathop \prod \limits_{i = 1}^{{n_{s} }} \frac{1}{{\rho_{i} + 1}}\frac{1}{{\rho_{i} !}}\left( {{ \exp }\left( {\beta^{{\prime }} {\mathbf{x}}_{si} } \right)} \right)^{{\rho_{i} }} \left( {{ \exp }\left( {\beta^{{\prime }} {\mathbf{x}}_{si} } \right)} \right)^{{1 - r_{si} }}$$
$$= \frac{1}{{\left( {n_{r} - n_{s} } \right)!}}\mathop \prod \limits_{i = 1}^{{n_{s} }} \left( {{ \exp }\left( {\beta^{{\prime }} {\mathbf{x}}_{si} } \right)} \right)^{{1 - r_{si} }} \mathop \sum \limits_{{\mathop \sum \limits_{i = 1}^{{n_{s} }} \rho_{i} = n_{r} - n_{s} }} \left( {n_{r} - n_{s} } \right)!\mathop \prod \limits_{i = 1}^{{n_{s} }} \frac{1}{{\rho_{i} + 1}}\frac{1}{{\rho_{i} !}}\left( {{ \exp }\left( {\beta^{{\prime }} {\mathbf{x}}_{si} } \right)} \right)^{{\rho_{i} }}$$

The term inside the sum resembles the components of a multinomial distribution with extra \(1/\left( {\rho_{i} + 1} \right)\) terms. We need an additional term to normalize the \({ \exp }\left( {\beta '{\mathbf{x}}_{si} } \right)\) so that they sum to 1.

$$\begin{aligned} & \frac{1}{{\left( {n_{r} - n_{s} } \right)!}}\frac{{\left( {e^{{\beta^{{\prime }} {\mathbf{x}}_{s1} }} + \cdots + e^{{\beta^{{\prime }} {\mathbf{x}}_{{sn_{s} }} }} } \right)^{{n_{r} - n_{s} }} }}{{\left( {e^{{\beta^{{\prime }} {\mathbf{x}}_{s1} }} } \right)^{{r_{s1} - 1}} \cdots \left( {e^{{\beta^{{\prime }} {\mathbf{x}}_{{sn_{s} }} }} } \right)^{{r_{{sn_{s} }} - 1}} }} \\ & \quad \sum\limits_{{\sum\nolimits_{i = 1}^{{n_{s} }} {\uprho_{i} = n_{r} - n_{s} } }} {\frac{1}{{\rho_{1} + 1}} \cdots \frac{1}{{\rho_{{n_{s} }} + 1}}\left( {\begin{array}{*{20}c} {n_{r} - n_{s} } \\ {\uprho_{1} \cdots \uprho_{{n_{s} }} } \\ \end{array} } \right)\left( {\frac{{e^{{\beta^{{\prime }} {\mathbf{x}}_{s1} }} }}{{e^{{\beta^{{\prime }} {\mathbf{x}}_{s1} }} + \cdots + e^{{\beta^{{\prime }} {\mathbf{x}}_{{sn_{s} }} }} }}} \right)^{{\rho_{1} }} \cdots \left( {\frac{{e^{{\beta^{{\prime }} {\mathbf{x}}_{{sn_{s} }} }} }}{{e^{{\beta^{\prime}{\mathbf{x}}_{s1} }} + \cdots + e^{{\beta^{{\prime }} {\mathbf{x}}_{{sn_{s} }} }} }}} \right)^{{\rho_{{n_{s} }} }} } \\ \end{aligned}$$

Let \(p_{i} = e^{{\beta^{\prime}{\mathbf{x}}_{si} }} /\left( {e^{{\beta^{\prime}{\mathbf{x}}_{s1} }} + \cdots + e^{{\beta^{\prime}{\mathbf{x}}_{{sn_{s} }} }} } \right)\).

$$= \frac{1}{{\left( {n_{r} - n_{s} } \right)!}}\frac{1}{{p_{1}^{{r_{s1} - 1}} \cdots p_{{n_{s} }}^{{r_{{sn_{s} }} - 1}} }}\mathop \sum \limits_{{\mathop \sum \limits_{i = 1}^{{n_{s} }}\uprho_{i} = n_{r} - n_{s} }} \frac{1}{{\rho_{1} + 1}} \cdots \frac{1}{{\rho_{{n_{s} }} + 1}}\left( {\begin{array}{*{20}c} {n_{r} - n_{s} } \\ {\uprho_{1} \cdots \uprho_{{n_{s} }} } \\ \end{array} } \right)p_{1}^{{\rho_{1} }} \cdots p_{{n_{s} }}^{{\rho_{{n_{s} }} }}$$

If \(\varvec{\rho}\sim {\text{Multinomial}}\left( {n_{r} - n_{s} ,\varvec{p}} \right)\) then the sum in (25) is \(E\left( {\frac{1}{{\left( {\rho_{1} + 1} \right) \cdots \left( {\rho_{{n_{s} }} + 1} \right)}}} \right)\), which can be estimated via Monte Carlo draws from the multinomial distribution.

A.3 Computation

Here we provide R/C++ code for computing the negative log conditional likelihood. The analyst can insert negCLL() into an optimizer (like nlm()) to obtain maximum likelihood estimates of β.

figure a

Rights and permissions

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ridgeway, G., Cave, B., Grieco, J. et al. A Conditional Likelihood Model of the Relationship Between Officer Features and Rounds Discharged in Police Shootings. J Quant Criminol 37, 303–326 (2021).

Download citation

  • Published:

  • Issue Date:

  • DOI: