


































CSWR Spring 2022 




COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   3   

Law Enforcement 
and Restorative 
Circles: Impacts 
on Educational 
Achievement

GABRIEL BAUTISTA 
CAITLIN E. MELLO 
JENNIFER SONG 
RICHARD A. UNITE



4  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS

ABSTRACT

Connections have been observed between police involvement in school 
KPZJPWSPUL�HUK�[OL�\[PSPaH[PVU�VM�Z\ZWLUZPVUZ�HZ�W\UPZOTLU[��>OPSL�
proponents of school surveillance believe that police are necessary 
[V�WYV]PKL�ZHML[`�PU�ZJOVVSZ��LK\JH[PVU�HK]VJH[LZ�X\LZ[PVU�Z[\KLU[Z»�
WLYJLW[PVU�VM�ZHML[`�HUK�P[Z�LɈLJ[Z�VU�LK\JH[PVUHS�V\[JVTLZ��;OPZ�
HY[PJSL�L_HTPULZ�[OL�YLSH[PVUZOPW�IL[^LLU�WVSPJL�VɉJLY�WYLZLUJL�HUK�
certain educational outcomes, including student attendance, access to 
OPNOLY�LK\JH[PVU��Z[HUKHYKPaLK�[LZ[�ZJVYLZ��HUK�Z\ZWLUZPVU�YH[LZ��(SZV�
included in this analysis is an exploration of the relationships between 
these variables and classroom restorative circles used to manage 
JVUÅPJ[�HUK�ÄUK�HS[LYUH[P]L�ZVS\[PVUZ�[V�ZHML[ �̀�+VLZ�WVSPJL�WYLZLUJL�
OH]L�H�ZPNUPÄJHU[�PTWHJ[�\WVU�H[[LUKHUJL��HJJLZZ�[V�OPNOLY�LK\JH[PVU��
Z[HUKHYKPaLK�[LZ[�ZJVYLZ��HUK�Z\ZWLUZPVU�YH[LZ�MVY�Z[\KLU[Z&�+VLZ�[OL�
use of restorative circles at school, an alternative to traditional student 
KPZJPWSPUL��OH]L�H�ZPNUPÄJHU[�PTWHJ[�\WVU�H[[LUKHUJL��HJJLZZ�[V�OPNOLY�
LK\JH[PVU��Z[HUKHYKPaLK�[LZ[�ZJVYLZ��HUK�Z\ZWLUZPVU�YH[LZ�MVY�Z[\KLU[Z&�
;V�HKKYLZZ�[OLZL�PUX\PYPLZ��HU�6YKPUHY`�3LHZ[�:X\HYLZ��63:��YLNYLZZPVU�
analysis is used on both predictors with the School Survey on Crime and 
:HML[`�JVSSLJ[LK�I`�[OL�<�:��+LWHY[TLU[�VM�,K\JH[PVU��+6,���9LZ\S[Z�
KLTVUZ[YH[L�H�Z[H[PZ[PJHSS`�ZPNUPÄJHU[�YLSH[PVUZOPW�IL[^LLU�WVSPJL�VɉJLY�
presence and suspensions. Restorative circle use in schools has a 
ZPNUPÄJHU[�PTWHJ[�\WVU�KLJYLHZLK�ZJOVVS�Z\ZWLUZPVUZ�HUK�PUJYLHZLK�
Z[HUKHYKPaLK�[LZ[�ZJVYLZ��;V�JVUJS\KL�[OL�WHWLY��YLHS�^VYSK�PTWSPJH[PVUZ�
on school policy development are discussed.

Keywords: police in schools, suspension rates, educational outcomes, 
restorative justice circles, standardized test scores



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   5   

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE

S
chool surveillance and the presence of law enforcement 
in schools contribute to the school-to-prison pipeline, a 
JVUULJ[PVU�VM�Z`Z[LTZ�TLHU[�[V�JYPTPUHSPaL�WYLKVTPUHU[S`�
Black and Latine students at an early age (Nance, 2016). 

By removing students from the classroom through suspension or 
arrests, students' academic performance and chances of graduating 
are severely impacted, ultimately causing complications in other life 
domains (Nance, 2016). Education advocates propose restorative 
circles and social-emotional learning (SEL) as alternative solutions to 
promoting school safety and better educational outcomes (Hulvershorn 
& Mulholland, 2018). Researchers have explored the relationship 
between school safety practices and student academic performance in 
order to understand better the possible impact of school surveillance 
HUK�YLZ[VYH[P]L�JPYJSLZ��;OPZ�WHWLY�HUHS`aLZ�YLSL]HU[�KH[H�[V�L_WSVYL�
this correlation and inform policy addressing academic performance 
and safety for students, especially Black, Indigenous, and people of 
JVSVY��)076*��Z[\KLU[Z�KPZWYVWVY[PVUH[LS`�HɈLJ[LK�I`�Z\JO�KPZJPWSPUHY`�
practices. 

REVIEW OF EXTANT LITERATURE
SCHOOL POLICING AND PUNISHMENT 

Police have had a presence in U.S. public schools since the 1950s, but 
gained prominence following several school shootings in the 1990s. 
>P[O�PUJYLHZLK�M\UKPUN�MYVT�[OL��+LWHY[TLU[�VM�1\Z[PJL��+61���THU`�
ZJOVVSZ�OH]L�OPYLK�HKKP[PVUHS�WVSPJL�VɉJLYZ�[V�THPU[HPU�VYKLY�HUK�
protect students within classrooms (Coon & Travis, 2012; Weiler & Cray, 
2011). Many schools associate student safety with increased policing 
and the use of punishment to enforce school policy. While this method 
is prevalent, it does not always ensure student safety—especially when 
¸ZHML[`¹�PZ�KLÄULK�HZ�MYLL�VM�JHYJLYHS�MVYJL·HUK�P[�JHU�OH]L�ZPNUPÄJHU[�
academic repercussions. 

(�ZPKL�LɈLJ[�VM�PUJYLHZLK�WVSPJL�WYLZLUJL�PZ�PUJYLHZLK�UVU]PVSLU[�
VɈLUZLZ�HUK�L_JS\ZPVUHY`�W\UPZOTLU[��(������Z[\K`�I`�.V[[MYLKZVU�L[�



6  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

al. observed 33 public California middle and high schools that increased 
[OLPY�WVSPJL�WYLZLUJL�I`�OPYPUN�VɉJLYZ�[OYV\NO�[OL�+61»Z�*VTT\UP[`�
6YPLU[LK�7VSPJPUN�:LY]PJLZ�/PYPUN�7YVNYHT��0U�[OPZ�SVUNP[\KPUHS�HUHS`ZPZ��
the 33 schools were compared to 72 matched public schools within 
the state based on similar school characteristics, such as metropolitan 
status, percentage of students on free or reduced-priced lunch, and 
YH[L�VM�Z\ZWLUZPVUZ�MVY�[OL�`LHY�WYPVY�[V�HU�PUJYLHZL�PU�:96�Z[HɈ��
;OPZ�Z[\K`�\[PSPaLK�TVU[OS`�KH[H�[V�L]HS\H[L�[OL�PTWHJ[�VM�PUJYLHZLK�
WVSPJL�VɉJLYZ�VU�[OL�ZJOVVS�LU]PYVUTLU[�V]LY�[OYLL�TVU[OZ��-PUKPUNZ�
demonstrated a correlation between expanded police presence and 
PUJYLHZLK�PKLU[PÄLK�KY\N�HUK�^LHWVUZ�YLSH[LK�VɈLUZLZ�JVTWHYLK�[V�
ZJOVVSZ�[OH[�KPK�UV[�PUJYLHZL�WVSPJL�Z[HɈ��;OL�Z[\K`�HSZV�MV\UK�H�YPZL�
in disciplinary action within schools that had increased police presence 
(Gottfredson et al., 2020). With increased policing, students were more 
SPRLS`�[V�IL�YLMLYYLK�[V�SH^�LUMVYJLTLU[�MVY�UVU]PVSLU[�VɈLUZLZ��(Z�H�
YLZ\S[��W\UPZOLK�Z[\KLU[Z�OHK�WV[LU[PHS�SLNHS�Z`Z[LT�JVUZLX\LUJLZ�
at a younger age, increasing the risk of entering the school-to-prison 
pipeline. Furthermore, it is essential to acknowledge that punishments 
involving removal from school settings impact academic achievement 
(Tucker & Vance, 2016).

POLICING'S DISPROPORTIONATE IMPACT ON BLACK AND 
LATINE STUDENTS

While this study does not address race due to data set limitations, the 
implications of increased policing and school punishment cannot be 
ignored when it comes to young students of color, particularly Black and 
Latine students. Schools that primarily serve students of color are more 
likely to use strict surveillance measures than other schools, creating 
disproportionate suspension rates and arrests for students of color 
JVTWHYLK�[V�[OLPY�^OP[L�JSHZZTH[LZ��+6,���������;OL�<�:��+LWHY[TLU[�
VM�,K\JH[PVU�KH[H�ZUHWZOV[�VM�ZJOVVS�KPZJPWSPUL�PU������YLWVY[LK�[OH[�
“Black students are suspended and expelled at a rate three times 
NYLH[LY�[OHU�^OP[L�Z[\KLU[Z����+6,���������(�OPNOLY�Z\ZWLUZPVU�YH[L�
for students of color could cause a direct negative impact on their 
academic achievement. Tucker and Vance (2016) found that “school 
suspensions have been linked with academic disengagement, 

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   7   

KLSPUX\LUJ �̀�HUK�ZJOVVS�KYVWV\[Z¹��W�������0UJYLHZPUN�WVSPJL�WYLZLUJL�
PU�ZJOVVSZ�LX\H[LZ�[V�HU�PUJYLHZL�PU�W\UP[P]L�TLHZ\YLZ�Z\JO�HZ�
suspension, and could contribute to disparate impacts on Black and 
Latine students in U.S. public schools. 

Perceptions of safety are also important to consider, especially for 
Z[\KLU[Z�VM�JVSVY��1VOUZVU�L[�HS���������Z\Y]L`LK��������Z[\KLU[Z�
from 98 middle and high schools in Maryland to determine whether a 
correlation existed between observed security measures in school and 
Z[\KLU[Z»�WLYJLW[PVUZ�VM�ZHML[ �̀�LX\P[ �̀�HUK�Z\WWVY[��:LJ\YP[`�TLHZ\YLZ�
included the addition of both internal and external surveillance cameras, 
HZ�^LSS�HZ�WVSPJL�VɉJLYZ�PU�ZJOVVSZ��;OPZ�Z[\K`�MV\UK�[OH[�)SHJR�HUK�
Latine students felt less safe than their white counterparts in schools 
^P[O�TVYL�WVSPJL�VɉJLYZ"�OV^L]LY��HSS�Z[\KLU[Z�Z[PSS�OHK�HU�V]LYHSS�OPNOLY�
perception of school safety. Perceived school safety was low in schools 
with high numbers of low-income students of color, especially Black 
students (Johnson et al., 2018). Even student perception regarding 
police presences was disparate, as it did not always make Black and 
Latine students feel safer in school. Therefore, a new alternative for 
safety should be made that includes the safety of all students of color. 

SCHOOL SAFETY AND ACADEMIC PERFORMANCE 

*OHV[PJ�VY�\UZHML�ZJOVVS�LU]PYVUTLU[Z�Z[YVUNS`�PUÅ\LUJL�[Y\HUJ`�
�.HZL�L[�HS����������(Z�WYL]PV\ZS`�UV[LK��[OL�KLÄUP[PVU�VM�H�ZHML�ZJOVVS�
environment varies from student to student, and can be dependent 
on students’ identities. Students might feel unsafe if they consider 
school policies to be violating their right to privacy or perceive school 
surveillance as having disciplinary motives. Research has proven over 
the last twenty years that attendance in class and lectures correlates 
^P[O�OPNOLY�OPNO�ZJOVVS�.7(Z��Z[HUKHYKPaLK�[LZ[�ZJVYLZ��HUK�L]LU�
ÄUHS�NYHKLZ�PU�JVSSLNL��.\TW������"�9LUKSLTLU���������)HZLK�VU�[OPZ�
research, it can be presumed that suspensions often lead to lower 
NYHKLZ�ILJH\ZL�[OL`�YLZ\S[�PU�ZPNUPÄJHU[�TPZZLK�JSHZZ�[PTL�

Furthermore, Tanner-Smith and Fisher (2016) highlight the impact 
of visible school security measures, such as security guards, metal 
detectors, and cameras, on academic performance, attendance, and 

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE



8  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

postsecondary aspirations. The data for this study came from the 
following two extensive national surveys: the School Crime Supplement 
[V�[OL�5H[PVUHS�*YPTL�=PJ[PTPaH[PVU�:\Y]L`��5�$��������Z[\KLU[Z"�����
THSL������^OP[L��(NL�$��������HUK�[OL�:JOVVS�:\Y]L`�VU�*YPTL�HUK�
:HML[`��5�$��������ZJOVVSZ"�����THSL������^OP[L���9LZ\S[Z�PUKPJH[LK�
UV�JVUZPZ[LU[�ILULÄJPHS�LɈLJ[Z�VM�ZLJ\YP[`�TLHZ\YLZ�VU�HJHKLTPJ�
outcomes. Trends showed heavy surveillance had a moderately harmful 
impact on academic success in a small sample of high schools that 
contain a large low-income student population; however, visible security 
measures such as fences or cameras did not have any substantial 
WVZP[P]L�VY�ULNH[P]L�LɈLJ[�VU�T\S[PWSL�LK\JH[PVU�V\[JVTLZ��;HUULY�
Smith & Fisher, 2016). 

RESTORATIVE PRACTICE IN SCHOOL SETTINGS: 
AN ALTERNATIVE 

Restorative justice, an alternative to the deeply rooted colonialism 
and white supremacy present in carcerality, has roots in indigenous 
cultures around the world and encourages reparation and accountability 
for wrongdoing in a non-punitive way (Wachtel, 2016). Schools are 
ILNPUUPUN�[V�\[PSPaL�[OPZ�WYHJ[PJL�PUZ[LHK�VM�Z\ZWLUZPVUZ�HUK�KL[LU[PVUZ��
which remove students from the classroom. Restorative circles provide 
space for students and teachers to develop relationships, build 
JVTT\UP[ �̀�HUK�THUHNL�JVUÅPJ[�[OYV\NO�OVULZ[�HUK�JVUZ[Y\J[P]L�
JVTT\UPJH[PVU��6Y[LNH�L[�HS����������(�YLZ[VYH[P]L�JPYJSL�PZ�H�TLL[PUN�
between a student who has caused harm and individuals impacted by 
the student’s actions. Such circles are facilitated by a moderator and 
can include teachers, school administrators, parents, and classmates. 
Members of the circle take turns sharing how they felt harmed by the 
Z[\KLU[»Z�ILOH]PVY��;OL�Z[\KLU[�^OVZL�HJ[PVUZ�HYL�PU�X\LZ[PVU�PZ�NP]LU�H�
chance to discuss their decision-making, mitigate the harm they caused 
by taking accountability, and commit to changing their behavior to better 
meet the needs of their community. When performed in a group setting 
with a trained counselor, the restorative alternative enables everyone to 
be heard and has been used “to address truancy, bullying, disciplinary 
PZZ\LZ��HUK�PU[LYWLYZVUHS�JVUÅPJ[¹��6Y[LNH�L[�HS����������

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   9   

0U�[OLPY�Z[\K �̀�6Y[LNH�L[�HS���������L_WSVYLK�[OL�TP[PNH[PUN�V\[JVTLZ�VM�
restorative circles on high school student and teacher experiences. This 
Z[\K`�PUJS\KLK����Z[\KLU[Z�HUK����ZJOVVS�Z[HɈ�HUK�HKTPUPZ[YH[VYZ�MYVT�
a large city in the southeastern United States. Restorative circles were 
MHJPSP[H[LK�PU�YLZWVUZL�[V�HK]LYZL�ILOH]PVY�HUK�X\HSP[H[P]L�PU[LY]PL^Z�
^LYL�JVUK\J[LK�^P[O����VWLU�LUKLK�X\LZ[PVUZ�H[�[OL�PTTLKPH[L�LUK�VM�
LHJO�YLZ[VYH[P]L�JPYJSL��0U[LY]PL^�X\LZ[PVUZ�MVJ\ZLK�IV[O�VU�[OL�JVUÅPJ[�
HUK�[OL�YLZ[VYH[P]L�JPYJSL�L_WLYPLUJL�MVY�Z[\KLU[Z�HUK�ZJOVVS�Z[HɈ��;OPZ�
study produced the following favorable themes: supporting students 
in taking ownership and engaging with the reconciliation process, 
improving relationships, engaging in meaningful dialogue, preventing 
KLZ[Y\J[P]L�^H`Z�VM�LUNHNPUN�JVUÅPJ[��`PLSKPUN�IL[[LY�HJHKLTPJ�HUK�
social achievement, and interrupting the school-to-prison pipeline. 
There were two unfavorable outcomes discovered in the study, however: 
student frustration with the process and disappointment due to the 
inability for some students to be vulnerable in those spaces. The adults 
in this study believe comfort and trust could have been established 
PM�[OLYL�^HZ�¸[PTL�[V�I\PSK�YLSH[PVUZOPWZ�^P[O�[OL�MHJPSP[H[VYZ¹��6Y[LNH�
et al., 2016). Despite this, the overall impact implies that less punitive 
TL[OVKZ��Z\JO�HZ�\ZPUN�YLZ[VYH[P]L�JPYJSLZ�[V�UH]PNH[L�JVUÅPJ[Z�HUK�
ILOH]PVYHS�WYVISLTZ��LɈLJ[P]LS`�YLK\JL�ILOH]PVYHS�PZZ\LZ�HTVUN�
students.

RESEARCH QUESTIONS AND CONTRIBUTIONS 

6\Y�YLZLHYJO�X\LZ[PVUZ�PUX\PYLK�^OL[OLY�[OLYL�PZ�H�ULNH[P]L�JVYYLSH[PVU�
IL[^LLU�[OL�WYLZLUJL�VM�ZJOVVS�ZLJ\YP[`�VɉJLYZ�PU�ZJOVVSZ�HUK�Z[\KLU[�
HJHKLTPJ�HJOPL]LTLU[��:WLJPÄJHSS �̀�[OLZL�PUX\PYPLZ�L_HTPULK�^OL[OLY�
WVSPJL�VɉJLYZ�PU�ZJOVVSZ�OH]L�H�ZPNUPÄJHU[�PTWHJ[�\WVU�H[[LUKHUJL��
Z[\KLU[Z»�HJJLZZ�[V�OPNOLY�LK\JH[PVU��Z[HUKHYKPaLK�[LZ[�ZJVYLZ��HUK�
suspension rates. This analysis also sought to determine whether 
[OL�\ZL�VM�YLZ[VYH[P]L�JPYJSLZ�PU�ZJOVVSZ�OHZ�H�ZPNUPÄJHU[�PTWHJ[�\WVU�
H[[LUKHUJL��HJJLZZ�[V�OPNOLY�LK\JH[PVU��Z[HUKHYKPaLK�[LZ[�ZJVYLZ��HUK�
suspension rates for students. Finally, this statistical study explored 
implications for disciplinary alternatives that could prevent students 
from becoming involved in the criminal legal system and increase their 
access to the safety net of education.

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE



10  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

CONCEPTUAL FRAMEWORK 

A regression model was used to identify and explore the relationship 
IL[^LLU�ZJOVVS�Z\Y]LPSSHUJL�HUK�JVUÅPJ[�TLKPH[PVU��HUK�IL[^LLU�
Z[\KLU[�HJHKLTPJ�WLYMVYTHUJL�HUK�Z\ZWLUZPVU��6UL�PUKLWLUKLU[�
]HYPHISL�L_HTPULK�PZ�[OL�WYLZLUJL�VM�WVSPJL�VɉJLYZ�PU�ZJOVVS��
statistically controlling for crime at the school’s location and whether 
the school is in a metropolitan area. Student involvement in restorative 
JPYJSLZ�PZ�[OL�ZLJVUK�WYLKPJ[VY�HUHS`aLK��HM[LY�JVU[YVSSPUN�MVY�Z^VYU�
SH^�LUMVYJLTLU[�VɉJPHSZ»�WHY[PJPWH[PVU�PU�KPZJPWSPUL��[LHJOLY�JYPZPZ�
intervention training, and student SEL. Dependent variables include the 
average percentage of (1) daily attendance, (2) students likely to attend 
higher education after high school, (3) students who score below the 
��[O�WLYJLU[PSL�VU�Z[HUKHYKPaLK�[LZ[Z��HUK�����Z[\KLU[Z�YLJLP]PUN�V\[�VM�
school suspension with no continuing school services provided for the 
remainder of the punishment. Exploring these relationships will allow for 
better understanding of the connection between discipline and student 
performance. 

HYPOTHESES 

(M[LY�HUHS`aPUN�SP[LYH[\YL�HKKYLZZPUN�[^V�KPZJPWSPUHY`�VW[PVUZ�MVY�ZJOVVSZ�
and establishing the variables being tested, two hypotheses were 
MVYT\SH[LK��-PYZ[��WVSPJL�VɉJLY�WYLZLUJL�KVLZ�UV[�ZPNUPÄJHU[S`�YLSH[L�
directly to testing scores and access to higher education, but does have 
H�ZPNUPÄJHU[�ULNH[P]L�YLSH[PVUZOPW�^P[O�H[[LUKHUJL�HUK�Z\ZWLUZPVU�YH[LZ��
:LJVUK��[OL�\ZL�VM�YLZ[VYH[P]L�JPYJSLZ�PU�ZJOVVS�OHZ�H�ZPNUPÄJHU[�WVZP[P]L�
relationship with attendance, access to higher education, test scores, 
and a decrease in suspension rates. 

METHOD
DATA SOURCE AND SAMPLE CHARACTERISTICS 

This study used the National Center for Education Statistics’ (NCES) 
:JOVVS�:\Y]L`�VU�*YPTL�HUK�:HML[`��::6*:���H�UH[PVUHSS`�YLWYLZLU[H[P]L�
cross-sectional survey on U.S. K-12 public schools (Padgett et al., 
�������;OL�::6*:�L_HTPULZ�ZJOVVS�JYPTL�HUK�ZHML[`�PZZ\LZ��PUJS\KPUN�
school policies and programs, disciplinary practices, parent/community 

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   11   

PU]VS]LTLU[��ZJOVVS�ZLJ\YP[`�WYLZLUJL��HUK�Z[HɈ�[YHPUPUN��<ZPUN�::6*:�
2017-2018 data, the following analyses have examined the relationship 
between school academic achievement and the programs, procedures, 
and policies that schools implement to prevent and reduce crime. The 
data was collected between February 20, 2018 and July 18, 2018. 

METHOD OF ANALYSIS 

;̂ V�63:�YLNYLZZPVUZ�^LYL�\ZLK�[V�[LZ[�[OL�WYLKPJ[HIPSP[`�VM�H[[LUKHUJL��
HJJLZZ�[V�OPNOLY�LK\JH[PVU��Z[HUKHYKPaLK�[LZ[�ZJVYLZ��HUK�Z\ZWLUZPVU�
YH[LZ��VUL�IHZLK�VU�[OL�WYLKPJ[VY�VM�WVSPJL�VɉJLY�WYLZLUJL�VU�ZJOVVS�
grounds and another on the use of restorative justice circles as an 
alternative to discipline. 

RESULTS 

Table 1 addresses the descriptive statistics for each dependent and 
independent variable, without controlling for other factors. Tables 2 
and 3 display the regressions with controls to ensure a more accurate 
description of the correlation.

POLICE OFFICER PRESENCE

;OL�PUJS\KLK�YLNYLZZPVUZ�`PLSKLK�]HY`PUN�YLZ\S[Z��6U�H]LYHNL��^OLU�H�
WVSPJL�VɉJLY�^HZ�WYLZLU[�PU�ZJOVVSZ��[OLYL�^HZ�UV�ZPNUPÄJHU[�JOHUNL�PU�
H[[LUKHUJL�^P[O�H�WVPU[�LZ[PTH[L�VM���������������([[LUKHUJL�HUK�VɉJLY�
WYLZLUJL�PU�[OL�ZJOVVS�OHK�H�Z[H[PZ[PJHSS`�UVU�ZPNUPÄJHU[�JVYYLSH[PVU��
After controlling for crime rates in the school’s location and the school 
ZL[[PUN»Z�WVW\SH[PVU�KLUZP[ �̀�HZ�ZLLU�PU�;HISL������ ���VM�[OL�]HYPHUJL�PU�
H[[LUKHUJL�YH[LZ�^HZ�L_WSHPULK�I`�[OL�WYLZLUJL�VM�WVSPJL�VɉJLYZ�PU�[OL�
school setting, rate of crime in the area, and population density of the 
ZJOVVS�ULPNOIVYOVVK��([[LUKHUJL�OHK�H�Z[YVUN�Z[H[PZ[PJHSS`�ZPNUPÄJHU[�
positive correlation with a low level of crime in the school neighborhood 
�W#�������:JOVVSZ�^P[O�WVSPJL�VɉJLY�WYLZLUJL�OHK��VU�H]LYHNL������ �
(0.93) fewer students accessing higher education after controlling for 
crime rates in the school’s location and the school setting’s population 
KLUZP[ �̀�7VSPJL�VɉJLY�WYLZLUJL�PU�[OL�ZJOVVS�ZL[[PUN��[OL�HYLH»Z�JYPTL�
YH[L��HUK�[OL�ZJOVVS�ULPNOIVYOVVK»Z�WVW\SH[PVU�KLUZP[`�L_WSHPU��������

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE



12  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

of student variance in achieving higher education. Higher education 
HJOPL]LTLU[�HUK�VɉJLY�WYLZLUJL�PU�ZJOVVS�OHK�H�Z[H[PZ[PJHSS`�UVU�
ZPNUPÄJHU[�JVYYLSH[PVU��<[PSPaPUN�OPNO�JYPTL�YH[LZ�HZ�[OL�YLMLYLUJL�
group, attending higher education was positively correlated with low 
and moderate crime levels in the school neighborhood, with a strong 
Z[H[PZ[PJHS�ZPNUPÄJHUJL��W#�������>OLU�KLUZLS`�WVW\SH[LK�JP[PLZ�HYL�HSZV�
the reference group, students living in towns and rural communities were 
SLZZ�SPRLS`�[V�W\YZ\L�H�OPNOLY�LK\JH[PVU��^P[O�H�Z[H[PZ[PJHSS`�ZPNUPÄJHU[�
negative correlation (p<.001). 

;HISL���PSS\Z[YH[LZ�[OH[�ZJOVVSZ�^P[O�WVSPJL�VɉJLY�WYLZLUJL�LHYULK���� �
�������OPNOLY�VU�Z[HUKHYKPaLK�[LZ[�ZJVYLZ��VU�H]LYHNL��HM[LY�JVU[YVSSPUN�
for crime rates at the school’s location and the school setting’s 
WVW\SH[PVU�KLUZP[ �̀�;OL�]HYPHUJL�VM��������PU�Z[HUKHYKPaLK�[LZ[�ZJVYLZ�
^HZ�L_WSHPULK�I`�WVSPJL�VɉJLY�WYLZLUJL�PU�ZJOVVS�ZL[[PUNZ��YH[L�VM�JYPTL�
PU�[OL�HYLH��HUK�ZJOVVS�ULPNOIVYOVVK�WVW\SH[PVU�KLUZP[ �̀�:[HUKHYKPaLK�
[LZ[�ZJVYLZ�HUK�VɉJLY�WYLZLUJL�PU�ZJOVVS�OHK�H�Z[H[PZ[PJHSS`�UVU�
ZPNUPÄJHU[�JVYYLSH[PVU��/PNOLY�[LZ[�ZJVYLZ�OHK�H�Z[YVUN�Z[H[PZ[PJHSS`�
ZPNUPÄJHU[�ULNH[P]L�JVYYLSH[PVU�^P[O�SV^�HUK�TVKLYH[L�JYPTL�SL]LSZ�PU�[OL�
school neighborhood (p<.001) when high crime rates were the reference 
group. With metropolitan cities as the reference group, suburbs and 
Y\YHS�JVTT\UP[PLZ�OHK�H�Z[H[PZ[PJHSS`�ZPNUPÄJHU[�ULNH[P]L�JVYYLSH[PVU�^P[O�
[OL�SPRLSPOVVK�VM�OH]PUN�OPNOLY�Z[HUKHYKPaLK�[LZ[�ZJVYLZ��W#�������

After controlling for school crime rates and population density, there 
^LYL�������������TVYL�Z\ZWLUZPVUZ�^OLU�H�WVSPJL�VɉJLY�^HZ�WYLZLU[�
PU�ZJOVVS��7VSPJL�VɉJLY�WYLZLUJL�PU�[OL�ZJOVVS��JYPTL�YH[L�PU�[OL�HYLH��
HUK�WVW\SH[PVU�KLUZP[`�VM�[OL�ZJOVVS�ULPNOIVYOVVK�L_WSHPULK� ���VM�
the variance in suspension rates. As shown in Table 2, the number of 
Z\ZWLUZPVUZ�HUK�[OL�WYLZLUJL�VM�VɉJLYZ�H[�ZJOVVS�YL]LHSLK�H�Z[YVUN��
Z[H[PZ[PJHSS`�ZPNUPÄJHU[�WVZP[P]L�JVYYLSH[PVU���W#������>OLU�TL[YVWVSP[HU�
cities were the reference group, suspension rates were positively 
correlated with towns and rural communities, with a strong statistical 
ZPNUPÄJHUJL��W#������

RESTORATIVE CIRCLES

9LNHYKPUN�YLZ[VYH[P]L�JPYJSLZ��[LHJOLY�[YHPUPUN��Z[\KLU[�:,3��HUK�VɉJLY�

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   13   

involvement in discipline, school attendance, could be determined with 
H�������]HYPHUJL��0U�;HISL����YLZ[VYH[P]L�Q\Z[PJL�\ZL�OHK�H�Z[H[PZ[PJHSS`�
UVU�ZPNUPÄJHU[�JVYYLSH[PVU�VM��������������^P[O�ZJOVVS�H[[LUKHUJL��
9LZ[VYH[P]L�Q\Z[PJL�\ZL�OHK�H�Z[H[PZ[PJHSS`�PUZPNUPÄJHU[�WVZP[P]L�JVYYLSH[PVU�
with higher education attainment, as seen in Table 3. Teacher training 
PU�JYPZPZ�PU[LY]LU[PVU�HUK�WYL]LU[PVU�OHK�H�Z[H[PZ[PJHSS`�ZPNUPÄJHU[�
WVZP[P]L�YLSH[PVUZOPW�VM�������������^P[O�OPNOLY�LK\JH[PVU�H[[LUKHUJL�
�W#������:[\KLU[�:,3�HSZV�OHK�H�Z[YVUN�Z[H[PZ[PJHSS`�ZPNUPÄJHU[�WVZP[P]L�
YLSH[PVUZOPW�VM���������� ��^P[O�OPNOLY�LK\JH[PVU�H[[HPUTLU[�HJYVZZ�[OL�
ZJOVVS��W#�������7VSPJL�VɉJLYZ�PU]VS]LK�PU�Z[\KLU[�KPZJPWSPUL�OHK�H�
Z[H[PZ[PJHSS`�ZPNUPÄJHU[�ULNH[P]L�JVYYLSH[PVU�VM��������������^P[O�OPNOLY�
education rates (p<.05). 

When examining interventions of restorative justice use, teacher 
[YHPUPUN��Z[\KLU[�:,3��HUK�VɉJLY�PU]VS]LTLU[�PU�KPZJPWSPUL�[V�KL[LYTPUL�
Z[HUKHYKPaLK�[LZ[�ZJVYLZ��[OLYL�^HZ�H�������]HYPHUJL��9LZ[VYH[P]L�
Q\Z[PJL�\ZL�OHK�H�Z[YVUN��Z[H[PZ[PJHSS`�ZPNUPÄJHU[�WVZP[P]L�JVYYLSH[PVU�VM�
������������^P[O�OPNOLY�Z[HUKHYKPaLK�[LZ[�ZJVYLZ��W#�������9LJLU[�[LHJOLY�
[YHPUPUN�PU�JYPZPZ�PU[LY]LU[PVU�HUK�WYL]LU[PVU�OHK�H�Z[H[PZ[PJHSS`�ZPNUPÄJHU[�
ULNH[P]L�YLSH[PVUZOPW�VM���������� ���^P[O�OPNOLY�Z[HUKHYKPaLK�[LZ[�ZJVYLZ�
(p<.05). 

When changing the predictor and controls to restorative justice use, 
[LHJOLY�[YHPUPUN��Z[\KLU[�:,3��HUK�VɉJLY�PU]VS]LTLU[�PU�KPZJPWSPUL��
[OL�Z\ZWLUZPVU�YH[LZ»�]HYPHUJL�^HZ��������(Z�ZOV^U�PU�;HISL����^OLU�
JVU[YVSSPUN�MVY�[LHJOLY�[YHPUPUN��Z[\KLU[�:,3��HUK�VɉJLY�PU]VS]LTLU[��
YLZ[VYH[P]L�JPYJSLZ�PU�[OL�JSHZZYVVT�OHK�H�^LHR��Z[H[PZ[PJHSS`�ZPNUPÄJHU[�
negative relationship of -0.05 (0.02) with high suspension rates (p<.05). 
:[\KLU[�:,3�HUK�VɉJLY�PU]VS]LTLU[�PU�KPZJPWSPUL�HSZV�OHK�Z[H[PZ[PJHSS`�
ZPNUPÄJHU[�YLSH[PVUZOPWZ�^P[O�Z\ZWLUZPVU�YH[LZ��W#������4VYL�Z[\KLU[�
:,3�TLHU[��������������ML^LY�Z\ZWLUZPVUZ��>OLU�VɉJLYZ�^LYL�PU]VS]LK�
in discipline, suspension rates increased by 0.05 (0.02). 

DISCUSSION

;OL�WYLZLUJL�VM�WVSPJL�PU�2����W\ISPJ�ZJOVVSZ�ÄYZ[�ILNHU�PU�[OL�� ��Z�
HUK�OHZ�YPZLU�Z[LHKPS`�ZPUJL�[OL�:HML�:JOVVS�(J[�VM��  ���)YV^U������"�
Coon & Travis, 2012). The authors are not aware of existing literature 

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE



14  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

L_HTPUPUN�[OL�SPRLSPOVVK�VM�WHY[PJ\SHY�ZJOVVSZ�OPYPUN�WVSPJL�VɉJLYZ"�
however, information on the demographics, prior disciplinary practices, 
and socio-economic statuses of schools most likely to incorporate law 
enforcement would be useful for future research. 

Police presence in school is predictive of greater odds of suspension, 
which can push students through the school-to-prison pipeline 
(McKenna & White, 2017). Alternatives to traditional discipline, such as 
YLZ[VYH[P]L�Q\Z[PJL�JPYJSLZ��JHU�TP[PNH[L�[OL�LɈLJ[Z�[OH[�SLHK�Z[\KLU[Z�[V�
entering this pipeline. This paper’s analysis demonstrates that the use 
of restorative justice, teacher training, and student SEL had a positive 
correlation with higher education attainment and attendance and 
decreased suspension rates. 

-\Y[OLYTVYL��WVSPJ`THRLYZ��WVSPJL�KLWHY[TLU[Z��HUK�ZJOVVS�VɉJPHSZ�
JVU[PU\L�[V�L_WHUK�WVSPJPUN�WYVNYHTZ�Z\JO�HZ�[OL�+61��*VTT\UP[`�
6YPLU[LK�7VSPJPUN�:LY]PJLZ�WYVNYHT��*67:���HUK�TVYL��4J2LUUH�
�
7L[YVZPUV���������*67:��^OPJO�PZ�WHY[�VM�[OL�=PVSLU[�*YPTL�*VU[YVS�HUK�
3H^�,UMVYJLTLU[�(J[�VM��  ���OPYLZ�HUK�[YHPUZ�JVTT\UP[`�WVSPJPUN�
professionals through government funded grants (McKenna & Petrosino, 
2022). This money could potentially have been used instead to fund 
YLZLHYJO�Z\WWVY[LK�WYVNYHTZ�SPRL�:,3��6]LYHSS��VU�H�Z`Z[LTPJ�SL]LS��
[OL�<�:��MLKLYHS�NV]LYUTLU[�PZ�MHPSPUN�[V�WYV]PKL�HKLX\H[L�YLZV\YJLZ�[V�
schools to properly educate the growing population of students with 
KPɈLYPUN�ILOH]PVYHS�ULLKZ��+HYSPUN�/HTTVUK��������

Investing millions of dollars in law enforcement programs may contribute 
to increasing student chances of criminal legal system involvement 
�.V[[MYLKZVU�L[�HS����������;OL�ZHTL�M\UKPUN�JV\SK�IL�\[PSPaLK�[V�WYVTV[L�
restorative justice and alternative means of discipline that keep students 
PU�ZJOVVS��0[�PZ�JY\JPHS�[V�YLJVNUPaL�TVYL�WLKHNVNPJHSS`�ZV\UK�WYHJ[PJLZ�
[V�HKKYLZZ�ZJOVVS�]PVSLUJL�HUK�OLSW�Z[\KLU[Z�TH_PTPaL�[OLPY�LK\JH[PVUHS�
opportunities, such as restorative circles, which were shown to decrease 
Z\ZWLUZPVU�YH[LZ�HUK�PUJYLHZL�Z[HUKHYKPaLK�[LZ[�ZJVYLZ��7VSPJ`THRLYZ�
should enact legislation to provide students with counseling services, 
mental health services, and SEL programs that promote anger-
THUHNLTLU[�Z[YH[LNPLZ�HUK�JVUÅPJ[�YLZVS\[PVU��0TWYV]PUN�[OL�X\HSP[`�
and strength of teaching approaches and classroom management skills 

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   15   

OH]L�WYV]LU�[V�IL�HTVUN�[OL�TVZ[�LɈLJ[P]L�^H`Z�[V�LUOHUJL�ZJOVVS�
safety (Nickerson et al., 2021). 

This paper’s analysis demonstrates that suspensions decrease when 
schools implement crisis prevention and intervention training, restorative 
Q\Z[PJL�WYHJ[PJLZ��HUK�:,3�WYVNYHTZ��(KKP[PVUHSS �̀�[OL�ÄUKPUNZ�PUKPJH[L�
[OH[�YLZ[VYH[P]L�Q\Z[PJL�WYHJ[PJLZ�OH]L�H�WVZP[P]L�PTWHJ[�VU�Z[HUKHYKPaLK�
test scores. This reinforces themes presented in previous literature, such 
HZ�[OL�JVUULJ[PVUZ�IL[^LLU�PUJYLHZLK�Z\ZWLUZPVU��ZJOVVS�WVSPJL�VɉJLY�
presence, and decreased access to higher education (Petrosino et al., 
2012; Weisburst, 2019). This paper contributes to existing literature by 
TLHZ\YPUN�[OL�PTWHJ[�VM�ZJOVVS�WVSPJL�VɉJLY�WYLZLUJL�HUK�YLZ[VYH[P]L�
Q\Z[PJL�JPYJSLZ�VU�H[[LUKHUJL�HUK�Z[HUKHYKPaLK�[LZ[�ZJVYLZ��

LIMITATIONS

Police presence and restorative circles in schools were examined to 
build on existing literature addressing the school-to-prison pipeline 
and its disproportionately negative impact on students of color. 
Unfortunately, student demographics were not available within the data 
set used, making it impossible to determine potential relationships 
outside of the area’s crime level and population density. The dataset 
KPK�UV[�PUKPJH[L�[OL�Z[\KLU[�[V�ZJOVVS�WVSPJL�VɉJLY�YH[PV�VY�HU`�V[OLY�
PUKPJH[PVU�VM�ZJOVVS�JOHYHJ[LYPZ[PJZ�[OH[�OPYLK�ZJOVVS�WVSPJL�VɉJLYZ�PU�
percentages. The dataset only provided the number of public schools 
tabulated into school characteristics such as neighborhood crime level, 
ZJOVVS�ZPaL��KP]LYZP[ �̀�HUK�Z[\KLU[�HWWYV_PTH[L�OV\ZLOVSK�PUJVTL�
SL]LS��(KKP[PVUHS�SPTP[H[PVUZ�PUJS\KL�[OL�]HY`PUN�X\HSP[`�VM�YLZ[VYH[P]L�
justice circles across schools that participated in the survey, potentially 
HɈLJ[PUN�[OL�YLNYLZZPVU�V\[JVTL�ZPUJL�MHJPSP[H[VY�JVTWL[LUJ`�JV\SK�
shift the outcome of the restorative justice circles.

CONCLUSION 

;OL�YVSL�WVSPJL�VɉJLYZ�WSH`�PU�ZJOVVSZ�ULNH[P]LS`�PTWHJ[Z�Z[\KLU[�
SLHYUPUN��[OLYLI`�HɈLJ[PUN�[OLPY�M\[\YL�LK\JH[PVUHS��LTWSV`TLU[��HUK�
criminal legal involvement. It is crucial to understand better ways to hold 
students accountable for their decisions, address school violence, and 

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE



16  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

meet student needs. In order to engage students without perpetuating 
pathways to the juvenile legal system, schools need to provide SEL 
WYVNYHTZ�HUK�YLN\SHY�[YHPUPUN�[V�[LHJOLYZ��WYPUJPWHSZ��HUK�Z[HɈ�[V�
support positive behavior. This analysis has shown that restorative 
Q\Z[PJL�\ZL�^HZ�[PLK�[V�OPNOLY�Z[HUKHYKPaLK�[LZ[PUN�ZJVYLZ�HUK�OHZ�[OL�
ability to mitigate the impacts of suspension rates. Considering the 
impact that the school-to-prison pipeline can have on youth, changes 
must be made to transition to pedagogically sound practices that are 
proven to improve academic outcomes.

REFERENCES

)YV^U��)����������<UKLYZ[HUKPUN�HUK�HZZLZZPUN�ZJOVVS�WVSPJL�VɉJLYZ!�(�JVUJLW[\HS�HUK�
methodological comment. Journal of Criminal Justice���������� �������O[[WZ!��KVP�VYN��
10.1016/j.jcrimjus.2006.09.013

*VSVYHKV�:[H[L�<UP]LYZP[`��U�K���>O`�3H[PU_�L&�O[[WZ!��LSJLU[YV�JVSVZ[H[L�LK\�HIV\[�^O`�
latinx

Coon, J. K., & Travis III, L. F. (2012). The role of police in public schools: A comparison of 
principal and police reports of activities in schools. Police Practice and Research, 13(1), 
15-30.

+HYSPUN�/HTTVUK��3����������9HJL��PULX\HSP[`�HUK�LK\JH[PVUHS�HJJV\U[HIPSP[`!�;OL�PYVU`�
of ‘No Child Left Behind.’ Race Ethnicity and Education������������������O[[WZ!��KVP�
org/10.1080/13613320701503207

Gase, L. N., DeFosset, A., Perry, R., & Kuo, T. (2016). Youths' perspectives on 
the reasons underlying school truancy and opportunities to improve school 
attendance. The Qualitative Report����������  ��O[[WZ!��ZLHYJO�WYVX\LZ[�JVT�
VWLU]PL^��K�L�����H�HK�K���K H��K�J�JLɈ����&WX�VYPNZP[L$NZJOVSHY
JIS$�����

Gottfredson, D. C., Crosse, S., Tang, Z., Bauer, E. L., Harmon, M. A., Hagen, C. A., 

�.YLLUL��(��+����������,ɈLJ[Z�VM�ZJOVVS�YLZV\YJL�VɉJLYZ�VU�ZJOVVS�JYPTL�HUK�
responses to school crime. Criminology & Public Policy��� ����� ��� ����O[[WZ!��KVP�
VYN�������������� ����������

Gump, S. (2005). The cost of cutting class: Attendance as a predictor of student success. 
College Teaching, 53(1), 21-26. http://www.jstor.org/stable/27559212

Hulvershorn, K., & Mulholland, S. (2018). Restorative practices and the integration of 
social emotional learning as a path to positive school climates. Journal of Research in 
Innovative Teaching & Learning.

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   17   

Johnson, S. L., Bottiani, J. H., Waasdorp, T. E., & Bradshaw, C. (2018). Surveillance or 
ZHMLRLLWPUN&�/V^�ZJOVVS�ZLJ\YP[`�VɉJLY�HUK�JHTLYH�WYLZLUJL�PUÅ\LUJL�Z[\KLU[Z��
WLYJLW[PVUZ�VM�ZHML[ �̀�LX\P[ �̀�HUK�Z\WWVY[��;OL�1V\YUHS�VM�(KVSLZJLU[�/LHS[O!�6ɉJPHS�
Publication of the Society for Adolescent Medicine, 63(6), 732-738. https://www.science 
KPYLJ[�JVT�ZJPLUJL�HY[PJSL�WPP�:������ ?��������

McKenna, J. M., & Petrosino, A. (2022). School policing programs: Where we have been 
and where we need to go next. National Institute of Justice. https://www.ojp.gov/
WKɉSLZ��UPQ����� ��WKM

McKenna, J. M., & White, S. R. (2017). Examining the use of police in schools: How roles 
may impact responses to student misconduct. American Journal of Criminal Justice, 
����������¶�����O[[WZ!��KVP�VYN���������Z���������� �����

5HUJL��1��7����������:[\KLU[�Z\Y]LPSSHUJL��YHJPHS�PULX\HSP[PLZ��HUK�PTWSPJP[�YHJPHS�IPHZ��Emory 
Law Journal��������������������O[[WZ!��ZZYU�JVT�HIZ[YHJ[$�������

Nickerson, A. B., Randa, R., Jimerson, S., & Guerra, N. G. (2021). Safe places to learn: 
Advances in school safety research and practice. School Psychology Review, 50(2-3), 
���������O[[WZ!��KVP�VYN������������� ��?���������� ��

6Y[LNH��3���3`\IHUZR �̀�4���5L[[SLZ��:���
�,ZWLSHNL��+��3����������6\[JVTLZ�VM�H�YLZ[VYH[P]L�
circles program in a high school setting. Psychology of Violence���������� ¶�����O[[WZ!��
KVP�VYN���������]PV�������

Padgett, Z., Jackson, M., Correa, S., Kemp, J., Gilary, A., Meier, A., Gbondo-Tugbawa, K., 
and McClure, T. (2020). School Survey on Crime and Safety: 2017–18 Public-Use Data 
[Data set]. National Center for Education Statistics, Institute of Education Sciences, U.S. 
Department of Education. Washington, DC. http://nces.ed.gov/pubsearch

Petrosino, A., Guckenberg, S., & Fronius, T. (2012). ‘Policing schools’ strategies: A review 
of the evaluation evidence. Journal of MultiDisciplinary Evaluation, 8, 80–101.

9LUKSLTHU��*��4����������+V�H[[LUKHUJL�WVSPJPLZ�PTWYV]L�Z[\KLU[�WLYMVYTHUJL&�;OL�
relationship among attendance, class policies, and grades. NACTA Journal,������������
�� ��O[[WZ!��^^ �̂QZ[VY�VYN�Z[HISL� �������

Tanner-Smith, E. E., & Fisher, B.W. (2017). Visible school security measures and student 
academic performance, attendance, and postsecondary aspirations. Journal of Youth 
and Adolescence���������� �������O[[W!��KVP�VYN���������Z�� �������������

Tucker, J. W., & Vance, A. (2016). School surveillance: The consequences for equity and 
privacy��5H[PVUHS�(ZZVJPH[PVU�VM�:[H[L�)VHYKZ�VM�,K\JH[PVU��O[[WZ!��ÄSLZ�LYPJ�LK�NV]��
fulltext/ED582102.pdf

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE



18  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

<�:��+LWHY[TLU[�VM�,K\JH[PVU��+6,�����������Data snapshot: School discipline. https://
ocrdata. ed.gov/assets/downloads/CRDC-School-Discipline-Snapshot.pdf

Wachtel, T. (2016). +LÄUPUN�YLZ[VYH[P]L!�/PZ[VY �̀ International Institute for Restorative 
7YHJ[PJLZ��O[[WZ!��^^ �̂PPYW�LK\�KLÄUPUN�YLZ[VYH[P]L�OPZ[VY`

Weiler, S. C., & Cray, M. (2011). Police at school: A brief history and current status of 
ZJOVVS�YLZV\YJL�VɉJLYZ��The Clearing House: A Journal of Educational Strategies, 
Issues and Ideas������������������O[[WZ!��KVP�VYN������������ ������������� ��

Weisburst, E. K. (2019). Patrolling public schools: The impact of funding for school police 
on student discipline and long-term education outcomes. Journal of Policy Analysis and 
Management, 38(2), 338-365. https://doi.org/10.1002/pam.22116

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   19   

APPENDIX A

Table 1: Descriptive Statistics

VARIABLES Total School  
Resource 
6ɉJLYZ�
Participate in 
Discipline

No School  
Resource 
6ɉJLYZ�
Participate in 
Discipline

Student  
Involvement 
in Restorative 
Circles

No Student 
Involvement 
in Restorative 
Circles

Attendance 93.156 93.123 93.225 93.110 93.196

Higher  
Education

62.693 62.276 63.550 63.900 61.922

Standardized 
Tests

18.263 ������ 18.137 20.318 ��� � 

Suspension 
Rates

���� 0.508 ����� ����� 0.522

Observations 2,762 1,859 903 1,077 1,685

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE



20  |  COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  

APPENDIX B

Table 2: Law Enforcement and Academic Achievement

VARIABLES Attendance Higher  
Education

Standardized  
Tests

Suspension  
Rates

Sworn law enforcement  
participate in discipline

-0.16 -1.69 0.79 0.06**

(0.31) (0.93) (0.67) (0.02)

Moderate crime level 1.35+ 9.56*** -12.29*** 0.00

(0.69) (2.07) ���� � (0.05)

Low crime level 2.51*** 21.70*** -20.85*** 0.01

(0.65) (1.96) ������ ������

Suburb locale ���� 1.51 �������� 0.00

(0.38) (1.15) (0.83) (0.03)

Town locale -0.01 -12.95*** -2.10+ 0.10**

���� � ���� � (1.07) (0.03)

Rural locale 0.19 ��������� �������� 0.08**

������ (1.31) ��� �� (0.03)

Constant 90.95*** � ������ 38.22*** �������

������ (1.93) (1.39) ������

Observations 2,762 2,762 2,762 2,762

R-squared 0.01 0.12 0.13 0.01

Standard errors in parentheses 
*** p<0.001, ** p<0.01, * p<0.05, + p<0.10

LAW ENFORCEMENT AND RESTORATIVE CIRCLES IN SCHOOLS



COLUMBIA SOCIAL WORK REVIEW, VOL. XIX  |   21   

APPENDIX C

Table 3: Restorative Circles and Academic Achievement

VARIABLES Attendance Higher  
Education

Standardized  
Tests

Suspension  
Rates

Restorative Circles ����� 1.38 ������� -0.05*

(0.33) (1.16) (0.85) (0.02)

Teacher Training -0.38 ����� -2.12* -0.05+

(0.38) (1.32) (0.97) (0.03)

Student SEL -0.32 ������ 0.81 -0.08*

������ (1.59) (1.16) (0.03)

Sworn Law Enforcement ����� ������ 0.75 0.05*

(0.32) (1.11) (0.81) (0.02)

Constant  �� ���� 57.19*** �������� 0.66***

���� � (1.70) ������ ������

Observations 1,859 1,859 1,859 1,859

R-squared 0.00 0.01 0.02 0.01

Standard errors in parentheses 
*** p<0.001, ** p<0.01, * p<0.05, + p<0.10

GABRIEL BAUTISTA, CAITLIN E. MELLO, JENNIFER SONG, AND RICHARD A. UNITE


