The Discretion to Search

• Date: 01 May 2012
• Author: Charles F. Klahm,Robin S. Engel,Rob Tillyer
• Published date: 01 May 2012
• DOI: 10.1177/1043986211425721
• Subject Matter: Articles

Journal of Contemporary Criminal Justice

28(2) 184 –205

© 2012 SAGE Publications

Reprints and permission: http://www.

sagepub.com/journalsPermissions.nav

DOI: 10.1177/1043986211425721

http://ccj.sagepub.com

425721CCJ28210.1177/1043986211425721Tillyer

et al.Journal of Contemporary Criminal Justice

© 2012 SAGE Publications

Reprints and permission: http://www.

sagepub.com/journalsPermissions.nav

1University of Texas–San Antonio, San Antonio, TX, USA

2Wayne State University, Detroit, MI, USA

3University of Cincinnati, Cincinnati, OH, USA

Corresponding Author:

Rob Tillyer, Department of Criminal Justice, University of Texas–San Antonio, 501 W. Cesar E. Chavez Blvd.

San Antonio, TX 78207-4415, USA

Email: rob.tillyer@utsa.edu

The Discretion to Search:

A Multilevel Examination

of Driver Demographics

and Officer Characteristics

Rob Tillyer1, Charles F. Klahm IV2,

and Robin S. Engel3

Abstract

Understanding police decision making has been a priority for policing scholars since

the middle part of the 20th century. Recent emphasis has focused on examining the

decision to search drivers and vehicles during pedestrian and traffic stops. The current

study contributes to this body of literature by testing a series of hypotheses based on

Skolnick’s notion of “symbolic assailants” and Smith and Alpert’s social conditioning

model. Using data gathered from a large, Midwestern municipal jurisdiction over an

8-month period during 2005 and 2006, we estimate a series of hierarchical models

to assess the relationship between discretionary searches and driver, vehicle, stop,

and officer characteristics. Results indicate that specific driver groups including young,

Black males are more likely to be searched for discretionary reasons. This relationship

is further conditioned by officer assignment. Policy implications and suggestions for

future research are also discussed.

Keywords

police discretion, discretionar y searches, driver race/ethnicity, officer characteristics

Article

Tillyer et al. 185

Introduction

Concerns of racial/ethnic disparities in police encounters have been woven into the

fabric of research on officer decision making. Based on discretion studies and against

the backdrop of the civil rights movement of the 1960s, a body of research developed

that focused on explaining coercive outcomes of police–citizen interactions that could

be easily counted using official data sources (i.e., citations, arrests, use of force,

searches). This research examined, nearly exclusively, whether drivers’ characteristics

influenced police behavior (for review, see National Research Council, 2004). From

simple bivariate comparisons of police decisions and driver characteristics, this body

of research evolved into the use of multivariate statistical techniques in an effort to

isolate the effects of extralegal factors (e.g., drivers’ race/ethnicity, age, gender,

demeanor, etc.) on police decision making, after controlling for legal factors. After 5

decades of research examining police behavior, studies have demonstrated some

racial, ethnic, and gender characteristics that may, on occasion, influence officers to

take coercive actions against drivers, but the explanatory power of these extra-legal

factors is small compared with the influences of legal and other relevant factors

(National Research Council, 2004).

Most recently, officer decisions to search drivers and vehicles during pedestrian and

traffic stops have generated substantial research interest. This body of research generally

suggests that minority drivers are more likely to be searched compared with White driv-

ers (Engel & Johnson, 2006); however, there are several theoretical and methodological

limitations of this research that suggest a more detailed examination of officers’ decision

to search is required. First, many of the available studies do not separately consider man-

datory and discretionary searches. Mandatory searches are dictated by organizational

policy and do not allow an examination of the core issue—officer discretion. As a result,

analyses of all searches may reveal more about organizational responses toward driver

behavior than about officer decision making. Second, statistical analyses of searches

have potentially suffered from model misspecification. Little attention has been given to

examining specific groups of drivers, such as young, Black males and their experiences

with search activity. Analyses of other poststop outcomes (i.e., warnings and citations)

indicate that this specification is important to fully understand minority experiences dur-

ing traffic stops (Tillyer & Engel, forthcoming). Other variables not considered include

vehicle characteristics and driver demeanor, which have been linked with other poststop

outcomes, such as arrest (Engel, Klahm, & Tillyer, 2010). Third, examinations of officer

search behavior have been predominately executed using pooled variance models

(Alpert, Dunham, & Smith, 2007; Lundman, 2004; Moon & Corley, 2007). However,

this approach does not address the nested nature of traffic stop data that requires the use

of hierarchical models to disentangle encounter-level effects from officer or neighbor-

hood factors and has important implications for understanding the relationship between

officer characteristics and discretionary searches. Finally, consistent calls for the use of

theory have largely gone unheeded in past research on search and seizure activity (Engel,

Calnon, & Bernard, 2002; Tomaskovic-Devey, Mason, & Zingraff, 2004).