Juvenile Transfer and Deterrence Reexamining The E

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/238432120
Juvenile Transfer and Deterrence: Reexamining the

Effectiveness of a “Get-Tough” Policy
Article in Crime & Delinquency · March 2011

DOI: 10.1177/0011128708319111
CITATIONS READS
32 1,185
2 authors:
Kareem Jordan David L Myers

American University Washington D.C. University of New Haven
29 PUBLICATIONS 420 CITATIONS 29 PUBLICATIONS 496 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Kareem Jordan on 28 February 2014.
The user has requested enhancement of the downloaded file.

Crimehttp://cad.sagepub.com/
& Delinquency
Juvenile Transfer and Deterrence: Reexamining the Effectiveness of

a ''Get-Tough'' Policy
Kareem L. Jordan and David L. Myers
Crime & Delinquency 2011 57: 247 originally published online 11 July 2008
DOI: 10.1177/0011128708319111
The online version of this article can be found at:

http://cad.sagepub.com/content/57/2/247
Published by:
http://www.sagepublications.com
Additional services and information for Crime & Delinquency can be found at:
Email Alerts: http://cad.sagepub.com/cgi/alerts
Subscriptions: http://cad.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
Citations: http://cad.sagepub.com/content/57/2/247.refs.html
>> Version of Record - Mar 9, 2011
Proof - Jul 11, 2008
What is This?
Downloaded from cad.sagepub.com at University of Central Florida Libraries on November 29, 2011
Article
Crime & Delinquency
57(2) 247–270
Juvenile Transfer © 2011 SAGE Publications
DOI: 10.1177/0011128708319111
and Deterrence: http://cad.sagepub.com
Reexamining the
Effectiveness of a
“Get-Tough” Policy
Kareem L. Jordan1 and David L. Myers2
Abstract
Although research has examined the effectiveness of juvenile transfer on
recidivism, there has been a lack of research done in assessing how well
juvenile waiver to adult court meets the criteria necessary for deterrence to
occur (i.e., certainty, severity, and swiftness of punishment). The purpose of
this study is to assess how well juvenile transfer meets these criteria, using
data on 345 youths legislatively waived to adult court in Pennsylvania. The
findings indicate that there is greater punishment severity in adult court, but
there is no difference in punishment certainty between the two court systems.
In addition, court processing occurred more quickly in juvenile court. In other
words, only one element of deterrence theory is achieved with juvenile transfer.
Implications for subsequent research and policy are discussed.
Keywords
decertification, juvenile transfer, legislative waiver, offense exclusion, Act 33
From the late 1980s through the mid-1990s, American youth violence became
a major public concern (Zimring, 1998). Aggregate juvenile violent crime
arrest rates increased by more than 60% from 1988 to 1994 (Snyder, 1998),
and the total number of juvenile arrests for murder rose by more than 100%
during this same time period (Cook & Laub, 1998; Zimring, 1998). In addition
1
University of North Florida
2
Indiana University of Pennsylvania
Corresponding Author:
Kareem L. Jordan, University of North Florida, Atlanta, GA 30302
Email: [email protected]
248 Crime & Delinquency 57(2)
to surges in violent youthful offending, violent victimization of adolescents

also increased substantially. It not only appeared that juveniles were becom-
ing more involved in violent acts, but they also were suffering at the hands
of their peers. In this context, serious and violent delinquents started to be
branded as “super-predators” (Dilulio, 1995), who were thought to be not
only more dangerous than previous generations of youth but also younger
when they started exhibiting their violent behavior.1
As a result of growing public concern about youth violence, legislatures
and juvenile courts responded with a number of “get-tough” policies and
strategies (Taylor, Fritsch, & Caeti, 2002). For example, more juvenile
offenders were sent to detention and correctional facilities and for lengthier
periods of time. The most popular approach, though, was for states to make
it easier to “transfer” or “waive” certain youths to adult criminal court (Torbet
et al., 1996).
In an examination of the use of juvenile waivers, there are several intended
purposes, including increased punishment, transfer of the most serious offend-
ers so the juvenile court can concentrate on those more amenable to treatment,
and deterrence (Mears, 2003). In Mears’s (2003) critique of the empirical
research on juvenile waiver, he indicates the important gaps in the transfer
literature, strongly encouraging the need for more studies that assess how
effective transferring youth to adult court is in achieving these purposes.
Although there are several goals of transfer, the focus of this study is to
examine the effectiveness of adult court punishment for violent youthful
offenders in achieving punishment certainty, severity, and swiftness of pun-
ishment, which are the three components necessary to achieve specific deter-
rence from subsequent recidivism (Becarria, 1764/1963).
Several researchers have examined the accountability of juvenile transfer
in terms of deterrence theory. Some research has been done on the effect of
juvenile transfer on both general deterrence (see, e.g., Jensen & Metsger,
1994; Risler, Sweatman, & Nakerud, 1998; Singer & McDowall, 1988) and
specific deterrence (see, e.g., Bishop, Frazier, Lanza-Kaduce, & Winner, 1996;
Fagan, 1995, 1996; Myers, 2001, 2003b; Podkopacz & Feld, 1996; Winner,
Lanza-Kaduce, Bishop, & Frazier, 1997). This research generally shows that
transfer has either no effect on recidivism or that transferred youths actually
exhibit higher recidivism than nontransferred youths. Regardless of these
findings, it is important that when discussing deterrence, researchers also
should assess how consistent juvenile transfer is with the three elements of
deterrence (Beccaria, 1764/1963). In other words, is there a higher likelihood
of punishment certainty, severity, and swiftness of punishment in adult court
for these youths?
Jordan and Myers 249
Past Empirical Research

Punishment Certainty
In examining the first element of deterrence theory, Beccaria (1764/1963)
stated that punishment should be certain. He stated that the strong impression
made by the high likelihood of punishment would result in people’s refrain-
ing from committing crime. If consistent with deterrence associated with
juvenile transfer, certainty of punishment should be higher among transferred
offenders as compared to nontransferred offenders. The early descriptive res
earch on juvenile transfer and punishment certainty (i.e., conviction) reports
high conviction rates (between 76% and 96%) among transferred offenders
(Bishop & Frazier, 1991; Bishop, Frazier, & Henretta, 1989; Champion, 1989;
Gillespie & Norman, 1984; Houghtalin & Mays, 1991; Thomas & Bilchik,
1985). Although these could be viewed as high conviction rates, it does not
necessarily follow that transfer increases the likelihood of conviction. Res
earchers were better able to assess the effect of transfer itself on conviction
when they used comparative methods to examine the issue.
Early comparative research suggests that transferred offenders experience
higher conviction rates than nontransferred offenders (Eigen, 1981a, 1981b;
Rudman, Hartstone, Fagan, & Moore, 1986). However, these studies suffer
from a lack of multivariate analyses, meaning other explanatory factors (e.g.,
offense seriousness and prior record) were not statistically controlled for in
the analyses. More recent research using multivariate techniques generally
do indicate that violent transferred offenders experience a higher conviction
rate than similar youths retained by the juvenile court (Fagan, 1995; Myers,
2001, 2003a). Fagan’s (1995) research does, however, suggest that the likeli-
hood of conviction may be offense based, with less of a difference in likeli-
hood of conviction existing among nonviolent offenders. Myers (2001,
2003a) also found a higher probability of target conviction (i.e., conviction
on the offense that triggered the initial transfer to adult court) among trans-
ferred offenders. This is an indication of a lower likelihood of reduced charges
in adult criminal court. In other words, transferred youths appear less likely
to benefit from plea bargaining and more likely to be convicted of the crime
that initiated waiver to adult court.
Punishment Severity
The second element of deterrence theory is punishment severity. Beccaria
(1764/1963) stated that for deterrence to take place, punishment should be
severe enough to outweigh the benefits derived from the illegal act. Beccaria
did stress, however, that punishment should not be overly severe. In other
words, the punishment should be proportionate to the crime. When examin-
ing deterrence theory, research generally assesses the harshness of the sentences
imposed on offenders. Punishment severity can be assessed in juvenile trans-
fer research by examining the likelihood of incarceration as well as incarceration
length among transferred youths. If juvenile transfer is to be effective in
deterring youths, it would be expected that there would be increased punish-
ment severity among transferred offenders.
Early descriptive research found that between 64% and 67% of the trans-
ferred and convicted offenders were incarcerated (Bishop et al., 1989; Bishop
& Frazier, 1991; Houghtalin & Mays, 1981; Thomas & Bilchik, 1985). Again,
because comparison groups were not used, sound conclusions cannot be
made. Other research that used comparison groups suggests that transferred
youths experience a greater likelihood of incarceration than nontransferred
youths (Eigen, 1981a, 1981b; Fagan, 1995; Podkopacz & Feld, 1995; Rudman
et al., 1986) and were sentenced to longer incarceration time than nontrans-
ferred offenders (Eigen, 1981a, 1981b; Rudman et al., 1986). However,
explanatory factors were not controlled for in those analyses, meaning the
results could have been caused by other unmeasured factors (e.g., offense
seriousness). Fagan’s (1995) research found very little difference in incar-
ceration time between transferred and nontransferred offenders, and Fritsch,
Caeti, and Hemmens (1996) found longer confinement time among youths in
adult court but also indicated the need to examine actual time served (instead
of the official sentence imposed by the court), as transferred youths served
only a small portion of their sentences. More recent multivariate research does
indicate that serious and violent transferred offenders are more likely to be
incarcerated (see, e.g., Kupchik, 2006; Myers, 2001) and more likely to be
sentenced to significantly longer incarceration times than similar retained
offenders (Lemmon, Austin, Verrecchia, & Fetzer, 2005; Myers, 2001, 2003a).
This suggests that transfer is effective in generating longer sentences (i.e., puni
shment severity) for these offenders, as compared to those in juvenile court.
Punishment Swiftness
The final element of deterrence theory is swiftness of punishment. According
to Beccaria (1764/1963), if punishment quickly follows the illegal act, offend-
ers will be able to associate the punishment with that act. An increased separation
between the punishment and the act will result in a disassociation of the two,
and a decreased likelihood of deterrence among offenders. A common mea-
sure of punishment swiftness is the length of time it takes to process offenders’
cases (i.e., time between arrest and final disposition). This does assume,
however, that arrests occur soon after the alleged illegal act took place. The
date of final disposition will be when the final “punishment” is imposed.
Youths who are transferred to adult court, therefore, should experience shorter
case-processing time (i.e., punishment swiftness) than similar youths in
juvenile court.
The research, however, is consistent in showing that that transferred offend-
ers experience significantly longer case-processing time than those who are
retained in juvenile court (Fagan, 1995; Myers, 2001, 2003a; Rudman et al.,
1986). Though the U.S. Constitution and state constitutions provide defen-
dants in adult criminal court the right to speedy trials, this time frame is typi-
cally shorter than state processing time requirements regarding offenders
tried in juvenile court. Also, because of court docket backlogs, along with the
exercise of due process rights (e.g., bail and right to a jury trial), many cases
in the adult system require lengthy processing times.
Limitations of Prior Research

As mentioned earlier, prior research has examined the effectiveness of juve-
nile transfer on specific deterrence (see, e.g., Fagan, 1995; Myers, 2003b;
Winner et al., 1997). However, there is little research that examines the puni
shment outcomes in terms of certainty, severity, and swiftness of punishment
(see, e.g., Fagan, 1995; Rudman et al., 1986), which then affects the level of
offender recidivism (Becarria, 1764/1963). Even less research has been done
examining all three elements of deterrence theory using multivariate analyses
(see, e.g., Myers, 2001, 2003a).
The most important limitation of prior research is in regard to the statisti-
cal analyses. All the research that compares youths in juvenile and adult
court uses a quasi-experimental design because of the unethical and impracti-
cal nature of random assignment. When using quasi-experimental designs,
sample elements in one group (e.g., youths in juvenile court) already differ
from sample elements in the other group (e.g., youths in adult court) (Shadish,
Cook, & Campbell, 2002), meaning the groups are statistically nonequiva-
lent on factors other than just the court in which their cases are processed
(i.e., selection bias). One way of minimizing the validity threat of selection
bias is through statistically controlling for factors that distinguish the groups.
Past research on juvenile waiver that used multivariate analyses tends to sta-
tistically control for many factors (e.g., offense seriousness, prior record, age,
race, etc.), but unmeasured factors likely still exist between the two groups.
Put differently, even after controlling for other explanatory factors, the type
of court still is not the only difference between the two groups, meaning the
results of those studies need to be interpreted with caution.
The current study advances this research area, as many studies, to date,
do not statistically account for selection bias. A limited number of studies
do, though, control for selection bias through various methods (see, e.g.,
Kupchik, 2006). We, therefore, use the Heckman two-step approach, a statis-
tical technique that allows us to control for this validity threat (Bushway,
Johnson, & Slocum, 2007; Heckman, 1976). This study examines the certainty,
severity, and swiftness of punishment for youths processed in adult court (as
compared to youths processed in juvenile court), after controlling for selection
bias. Through the use of this technique, the results will be more meaningful,
allowing firmer conclusions to be drawn and a more concrete understanding
of the effectiveness of juvenile waiver in achieving this particular goal.
Methodology
The data include all youths who had preliminary hearings in three Pennsyl-
vania counties (i.e., Philadelphia, Allegheny, and Dauphin) between March
1996 (the implementation date of Pennsylvania’s legislative waiver law, Act 33;
Pennsylvania Juvenile Court Judges’ Commission, 1996) and December 1996
and who were determined to meet the Act 33 criteria for juvenile court exclu-
sion.2 First, except in the case of murder, the juvenile must be at least age 15
at the time of the offense and not older than 17. Second, the offender must be
charged with a violent crime (i.e., rape, involuntary deviant sexual inter-
course, aggravated assault or aggravated indecent assault, robbery, robbery
of a motor vehicle, voluntary manslaughter, kidnapping, or an attempt, con-
spiracy, or solicitation to commit any of these crimes). Third, the youth either
must have used a deadly weapon in the offense or must have been previously
adjudicated for an excluded offense (except if the current offense is aggra-
vated assault, which is not included in the repeat offender clause).
There is, however, a provision in place for offenders who initially are
waived under Act 33 to be decertified to juvenile court. Juveniles can request
a hearing to have their respective cases considered for decertification. The
burden is on the juvenile to establish, by a preponderance of the evidence,
that decertification will serve the “public interest” (Pennsylvania Juvenile
Court Judges’ Commission, 1996). In considering decertification, criminal
court judges are required to use the same criteria that juvenile court judges
use in making a judicial waiver decision (e.g., child’s culpability, age, matu-
rity, prior record, and amenability to treatment). If the case is decertified, it is
then processed in juvenile court. If the case is not decertified, it remains in
adult court. Numerous visits were made to each research site to examine these
Table 1. Descriptive Statistics for All Continuous Variables N = 345
Variable M SD Min. Max.

Age 16.71 0.80 15.01 18.00
Prior referrals 1.74 2.27 0.00 13.00
Prior violent referrals 0.62 1.01 0.00 5.00
Case-processing time (natural log) 5.21 1.10 0.69 7.27
youths’ case files (e.g., probation, public defender, and district attorney files),
after which the final sample was 345, of which 144 were decertified to juvenile
court and 201 were not decertified (i.e., those retained in adult court).3
Because random assignment to groups is not possible with this type of
research, the two groups of offenders are likely to differ in regard to factors
other than type of court. As discussed earlier, criminal court judges can decer-
tify youths where it serves the “public interest,” meaning that additional fac-
tors do likely distinguish the two groups of offenders. Therefore, we employ
the Heckman two-step approach to statistically control for the differences
between the groups (see, e.g., Bushway et al., 2007; Heckman, 1976). In
employing the first step of the Heckman approach, we first modeled the
probability of being included in the decertified group. The estimates were
then used to calculate a lambda term (i.e., inverse Mill’s ratio). The lambda
term represents the unmeasured factors that has an effect on the decertification
decision.4 The second step in this approach involves using the lambda term,
as a variable, in subsequent multivariate analyses when comparing decerti-
fied and nondecertified youths. The Heckman two-step approach allows for
unbiased estimates in the analyses, because the selection bias factor is statis-
tically controlled for with the lambda term.
Measures
Tables 1 and 2 present descriptive statistics of all the variables employed in
the current study for the entire group of violent youths.
Dependent Variables
There are five dependent variables examined in this study, which are meant
to measure the elements of deterrence theory. Conviction and target convic-
tion both pertain to punishment certainty. With regard to the current research,
conviction refers to whether any charges were substantiated in either juvenile
or adult court. Target conviction is whether a youth was convicted on the
statutory excluded offense that triggered the initial waiver to criminal court.
Table 2. Descriptive Statistics for all Nominal and Ordinal Variables
Variable % n
Decertification
Decertified 42 144
Nondecertified 58 201
Gender
Male 88 304
Female 12 41
Race
Non-White 87 301
White 13 44
Offense
Aggravated assault 41 142
Robbery 53 182
Other offense 6 21
Role
Primary role 71 244
Nonprimary role 4 15
Unknown role 25 86
Weapon
Firearm 55 189
Nonfirearm 27 95
Unknown weapon 18 61
Attorney
Public defender 46 158
Nonpublic defender 54 187
County
Allegheny 16 55
Dauphin 6 21
Philadelphia 78 269
Conviction 68 234
Target conviction (of convicted youth) 83 194
Incarceration (of convicted youth) 77 181
Type of sentence
Probation 47 162
Short sentence 29 100
Long sentence 24 83
The next two dependent variables are indications of punishment severity and
apply to youths who had charges substantiated against them in either juvenile
or adult court. Incarceration refers to whether convicted offenders (n = 235)
were sentenced to secure confinement. This is a dichotomous variable.
The second measure of sentence severity is type of sentence, which is mea-

sured at the ordinal level. This variable explores the type of sentence convicted
offenders received. There were three indicators of sentence type: probation,
short sentence (i.e., less than 2 years in secured confinement), and long sen-
tence (i.e., more than 2 years in secured confinement).5 For decertified youths
in juvenile court, actual time served in secured confinement was used as a
basis for determining the sentence type for those not receiving probation. For
nondecertified youths in adult court, the minimum sentence imposed in adult
court was used for those who did not receive probation.6
Case-processing time pertains to the punishment swiftness for the offend-
ers. For all offenders, this variable represents the length of time from arrest
to final disposition.7 Because this variable was positively skewed in its distri-
bution of values, it was necessary to transform it by taking its natural log
(Mertler & Vannatta, 2005). Through transforming this variable, we can meet
the assumption of normality when employing multivariate analyses (Lewis-
Beck, 1980).
Independent Variables
The key independent variable in the current study was decertification. This vari-
able does not represent a true “treatment,” for it was not possible to randomly
assign youths to treatment and control groups for this research. As discussed
earlier, to be decertified to juvenile court, the adult criminal court judge must
find by a preponderance of the evidence that it serves the public interest for this
youth to be processed in juvenile court, taking into consideration the child’s
culpability, age, maturity, prior record, amenability to treatment, and so on.
Several control variables were also used in the subsequent analyses. Because
this research was a quasi-experimental design, a strong attempt was made to
statistically consider factors that have been shown to have an effect on the
transfer decision, court outcomes, or both: age at referral (see, e.g., Fagan &
Deschenes, 1990; Kinder, Veneziano, Fichter, & Azuma, 1995; Myers, 2003b;
Podkopacz & Feld, 1996), gender (see, e.g., Leonard & Sontheimer, 1995;
Wordes, Bynum, & Corley, 1994), race (see, e.g., Fagan & Deschenes, 1990;
Feld, 1995; Leonard & Sontheimer, 1995; Minor, Hartmann, & Terry, 1997),8
the number of prior referrals to juvenile court (see, e.g., Barnes & Franz,
1989; Fagan & Deschenes, 1990; Fagan, Forst, & Vivona, 1987; Myers,
2003b), and the number of times youths previously were referred to the juve-
nile court for a violent offense (Myers, 2001).
It was not possible to examine a direct measure of offense severity using
only the offense for which the youths were charged. In Pennsylvania during
1996, most Act 33 youths (94%) were charged with either aggravated assault
or robbery. Therefore, offense was measured as three dichotomous variables:
aggravated assault, robbery, and other offenses.
The type of weapon used during the commission of the crime also has
been shown to affect the decision to transfer a youth to adult court and other
court outcomes (see, e.g., Myers, 2003a; Sridharan, Greenfield, & Blakley,
2004). The use of a firearm also can be used as an indicator of offense sever-
ity, as compared to other weapons used in a crime. It was not possible to
determine the type of weapon used for approximately 18% of the offenders
(n = 61). Therefore, the unknown category was used as its own separate group,
resulting in three dichotomous weapons variables being developed: firearm,
nonfirearm, and unknown weapon.
A neglected area of research when examining court outcome data has
been the potential effect of the role a youth played in the offense. It is possi-
ble that judges in both juvenile and adult courts may hold offenders more
accountable if they played a central role in the offense, as compared to those
offenders who did not play as big a part (see Spohn, 2002, for a fuller discus-
sion). In the current research, this potential relationship was explored. Limi-
tations in the data resulted in 25% (n = 86) of the values for this variable to
be missing. Therefore, this variable was treated similarly to the firearm vari-
able, resulting in three dichotomous “role” variables: primary role, nonprimary
role, and role unknown.
Another variable that tends to be neglected in past juvenile transfer res
earch is the potential effect of attorney type. Although attorney type largely
has been ignored in juvenile transfer research, the effect of type of attorney
on juvenile court outcomes has nevertheless been examined in several stud-
ies (see, e.g., Burruss & Kempf-Leonard, 2002; Langley, 1972). For the cur-
rent study, attorney type was measured as whether the offender’s attorney was
a public defender.9
Finally, the county of the offender also was used as a control variable in
the analyses. Prior research has found that the type of county (i.e., rural, sub-
urban, or urban) has an impact on certain court outcomes (see, e.g., Lemmon
et al., 2005; Myers, 2001). Mears (2003) also indicated that there may be
county differences in outcomes related to juvenile waiver.
Results
Statistical Analyses
Three of the dependent variables (conviction, target conviction, and incarcera-

tion) were measured as a dichotomy, meaning multivariate logistic regression
Table 3. Logistic Regression Estimates for the Impact of Decertification on

Conviction (N = 345)
Variable B SE Wald Exp(B)
Decertification –0.133 0.272 0.241 0.875
Age 0.174 0.285 0.372 1.190
Gender (male = 1) 0.817 0.432 3.572 2.264
Race (non-White = 1) –1.331 0.535 6.192* 0.264
Prior referrals 0.516 0.306 2.838 1.675
Prior violent referrals –0.410 0.216 3.618 0.663
Robbery –0.146 0.759 0.037 0.864
Other offense –1.196 1.007 1.410 0.303
Nonprimary role –0.713 1.430 0.249 0.490
Unknown role 0.348 0.869 0.161 1.417
Nonfirearm –1.353 1.559 0.753 0.259
Unknown weapon –0.215 0.687 0.099 0.806
Public defender –0.369 0.795 0.215 0.692
Constant 0.913 2.753 0.110 2.491
–2 log likelihood 392.291
Model chi–square 41.149**
Cox and Snell R2 .112
Nagelkerke R2 .157
*p < .05. **p < .01.
was the appropriate method of estimation (Menard, 2002). Sentence type was
measured as three ordinal level outcomes: probation, short sentence, and
long sentence. Therefore, ordinal regression (polytomous universal model)
was chosen as the appropriate statistical technique to use (Menard, 2002).
The final dependent variable, the natural log of case-processing time, was
measured continuously, so ordinary least squares multiple regression was used
(Lewis-Beck, 1980).
Multivariate Results
The multivariate analyses are presented in Tables 3 through 7. Because some
variables had extremely high standard errors in some statistical models, they
were excluded from some of the analysis. The high standard errors were
because of some categories having zero cells. According to Menard (2002),
When [the odds] are 0 or 1 for an entire group of cases, as defined by

the value of a categorical independent variable, the result will be a very
high estimated standard error for the coefficient associated with that
category. (p. 78)
Table 4. Logistic Regression Estimates for the Impact of Decertification on Target

Conviction (n = 235)
Age 0.142 0.403 0.124 1.152
Gender (male = 1) 0.830 0.727 1.301 2.293
Race (non-White = 1) 0.474 0.529 0.805 1.607
Prior referrals 0.161 0.287 0.316 1.175
Prior violent referrals –0.087 0.280 0.097 0.917
Robbery 0.098 0.796 0.015 1.102
Other offense –0.968 0.776 1.554 0.380
Nonprimary role –0.836 0.848 0.972 0.433
Unknown role –0.197 0.872 0.051 0.821
Nonfirearm –0.996 1.384 0.518 0.369
Unknown weapon 0.014 0.925 0.000 1.014
Public defender –0.742 0.866 0.735 0.476
Dauphin 1.443 0.860 2.814 4.233
Philadelphia 1.144 0.651 3.089 3.140
Constant –0.937 5.342 0.031 0.392
Model chi-square 24.939
Nagelkerke R2 .167
*p < .05. **p < .01.
For this reason, several variables were excluded from some of the subsequent
multivariate analyses, meaning the findings should be interpreted with extreme
caution.
Punishment certainty. In examining punishment certainty in Table 3, the
findings for this statistical model indicate that there is no significant differ-
ence in likelihood of conviction between youths processed in juvenile and
adult court. Put differently, the likelihood of punishment in one court was
similar to that of the other court. One variable, however, was significant in
the model. Non-Whites were significantly less likely to be convicted than
Whites (b = –1.33, p < .05). The race finding was interesting, given that
some literature suggests that minorities (either generally or specific non-
White races) have less favorable outcomes in juvenile court than Whites (see,
e.g., Frazier & Bishop, 1995; Leiber, Johnson, Fox, & Lacks, 2007).
To examine the potential effect of decertification on target conviction,
only convicted offenders were included in this model. Table 4 provides the
Table 5. Logistic Regression Estimates for the Impact of Decertification on

Incarceration (n = 235)
Age 0.032 0.443 0.005 1.033
Gender (male = 1) 0.444 0.796 0.311 1.559
Race (non-White = 1) 0.922 0.542 2.897 2.514
Prior referrals 0.477 0.338 1.992 1.611
Prior violent referrals 0.001 0.334 0.000 1.001
Robbery 0.313 0.867 0.130 1.367
Other offense 1.006 1.164 0.747 2.735
Nonprimary role 1.576 1.032 2.329 4.834
Unknown role 0.286 0.919 0.097 1.332
Nonfirearm –1.465 1.511 0.939 0.231
Unknown weapon 0.387 0.985 0.155 1.473
Public defender –0.665 0.974 0.466 0.514
Dauphin –0.564 0.762 0.547 0.569
Philadelphia –0.090 0.687 0.017 0.914
Constant 0.410 5.755 0.005 1.507
Model chi-square 57.194**
Nagelkerke R2 .327
*p < .05. **p < .01.
logistic regression estimates for the determinants of target conviction. When

statistically controlling for the effect of other explanatory factors, decertifi-
cation was negative and highly insignificant (b = –.34, p = .46). In other
words, there was no difference in target conviction for youths processed in
the two court systems. This finding suggests that youths are being convicted
for Act 33 offenses similarly, regardless of the court in which the cases are
processed.
Punishment severity. Table 5 presents the logistic regression estimates for
the determinants of incarceration, which considered only those offenders
who were either adjudicated delinquent in juvenile court or convicted in adult
court (n = 235). After controlling for other factors, the decertification coef-
ficient was close to reaching statistical significance (b = –.69, p = .055),
suggesting that decertified youths have a lower likelihood of incarceration
than nondecertified youths. This difference suggests a somewhat greater chance
for incarceration in the adult system, but again the difference was statistically
insignificant.
Table 6. Ordinal Regression Estimates for the Impact of Decertification on

Incarceration Time (n = 235)
Variable B SE Wald
Decertification –1.921 0.340 31.855**
Age –0.121 0.359 0.114
Gender (male = 1) 0.400 0.653 0.376
Race (non-White = 1) 1.298 0.442 8.608**
Prior referrals 0.234 0.249 0.883
Prior violent referrals –0.069 0.198 0.123
Robbery 0.069 0.685 0.010
Other offense 0.696 0.706 0.972
Nonprimary role 1.247 0.768 2.625
Unknown role 0.542 0.696 0.606
Nonfirearm –1.160 1.218 0.906
Unknown weapon –0.166 0.696 0.057
Public defender –0.996 0.783 1.617
Dauphin 0.071 0.597 0.014
Philadelphia 0.209 0.534 0.153
Model chi-square 103.620**
Nagelkerke R2 .405
McFadden .207
*p < .05. **p < .01.
Table 6 shows the ordinal logistic regression model, which examines the
effect of decertification on sentence type, and it includes all of those offend-
ers who were convicted (n = 235). The coefficient for decertification was
positive and significant (b = –1.921, p < .01), indicating that decertified
offenders were sentenced to less confinement time than nondecertified offend-
ers. In addition, race was statistically significant in the model: Non-Whites
were more likely to get increased confinement time, as compared to Whites.
Punishment swiftness. The final model, which is presented in Table 7, exam-
ines the effect of decertification on the natural log of case-processing time.
The results indicate that decertified youths experience significantly shorter
case-processing time than nondecertified youths (b = –0.37, p < .01). In other
words, case processing in juvenile court is shorter than its adult counterpart.
The model also revealed other significant factors in explaining the natural
log of case-processing time. Males experienced significantly longer case-
processing time than females (b = .55, p < .01). Youths processed in Dauphin
Table 7. Ordinary Least Squares Regression Estimates for the Impact of

Decertification on the Natural Log of Case Processing Time (N = 345)
Variable B SE Beta T
Decertification –0.372 0.124 .167 2.988**
Age 0.011 0.126 .008 0.090
Gender (male = 1) 0.545 0.200 .160 2.733**
Race (nonwhite = 1) –0.009 0.203 .003 0.046
Prior referrals –0.038 0.133 .079 0.288
Prior violent referrals 0.084 0.093 .077 0.906
Robbery 0.019 0.342 .008 0.054
Other offense 0.365 0.460 .079 0.794
Nonprimary role –0.086 0.650 .016 0.132
Unknown role 0.466 0.370 .183 1.261
Nonfirearm 0.007 0.696 .003 0.011
Unknown weapon –0.039 0.284 .013 0.136
Public defender –0.335 0.350 .152 0.957
Dauphin 0.677 0.280 .147 2.419*
Philadelphia 0.135 0.190 .051 0.709
Constant 4.543 1.253 3.626**
F 4.125**
R2 .168
*p < .05. **p < .01.
County experienced longer case-processing time than youths in Allegheny

County (b = .68, p < .05).
Discussion and Conclusion

When looking at comparison studies focused on violent youthful offenders,
the findings from the current research are contrary to prior studies that exam-
ined the effect of transfer on conviction (Eigen, 1981a, 1981b; Fagan, 1995;
Myers, 2001, 2003a), which found that youths in adult court have a higher
likelihood of conviction (or punishment certainty) than similar youths in
juvenile court. One study did suggest that violent offenders in both court sys-
tems have virtually identical conviction rates (Rudman et al., 1986).
In contrast to the findings from previous studies (e.g., Myers, 2001), the
findings from this study point to similarities between the two courts in both
conviction on any offense and conviction on a target offense (among those
convicted). Also, the results from these multivariate analyses indicate that
Act 33 is not achieving a higher certainty of punishment in adult court in
terms of conviction likelihood. Some criminal justice professionals (i.e., judges,

prosecutors, and public defenders) suggest that there is little or no difference
in likelihood of conviction between juvenile and adult court for serious and
violent offenders (Jordan, 2006). Because serious offenses were committed,
prosecutors are less likely to file charges (in either juvenile or adult court)
where there is not enough evidence to substantiate that the offense was com-
mitted. Therefore, both courts have an increased probability of having simi-
lar rates of conviction.
An explanation for the similarity between the two courts on target convic-
tion involves the role of plea bargaining in the process. Criminal justice pro-
fessionals indicate that there is an increased pressure to plead cases in adult
court because of the uncertainty of juries and the backlog of the court docket,
which will lower the likelihood of conviction on target offenses (Jordan,
2006). Because the goal of the juvenile court is to provide treatment to youth,
there is less pressure to adjudicate youth on the original Act 33 offense,
because offenders likely will receive the same disposition regardless of the
charge. Therefore, between plea bargaining in the adult system and limited pres-
sure to convict on a target offense in the juvenile system, there ends up being
little difference between the court systems on target conviction likelihood.
Whether looking at conviction on any offense or on a target offense, it seems
clear that youths legislatively waived and processed in adult court have a
similar chance of conviction than youths decertified to juvenile court. If deter-
rence is one of the goals of Act 33, it appears that the element of increased
punishment certainty is not present. According to Beccaria (1764/1963), cer-
tainty of punishment is the most important element in achieving deterrence,
as those who believe they will be punished will be more likely to refrain from
committing criminal acts.
Although not the focus of our study, the finding regarding race was that
minorities were less likely to be convicted than Whites. The research is gen-
erally mixed when it comes to the role race plays in juvenile court outcomes.
A review of existing research found that race does affect juvenile court deci-
sion making but more so in the early stages of the process (Engen, Steen, &
Bridges, 2002). In addition, methodological issues (e.g., measurement of race)
greatly affect results regarding race, resulting in mixed findings.
The quantitative findings also indicate that among those convicted, there
was no significant difference between decertified and nondecertified offend-
ers in the likelihood of incarceration (i.e., punishment severity). This finding
is not consistent with the prior research using comparison groups, which found
that transferred offenders in adult court had a higher likelihood of incarcera-
tion than comparable nontransferred offenders in juvenile court (Eigen, 1981a,
1981b; Fagan, 1995; Kupchik, 2006; Myers, 2001, 2003a; Podkopacz & Feld,
1996; Rudman et al., 1986).
On the other hand, when examining sentence type, a different result emerged,
as adult court offenders were sentenced to longer confinement time than sim-
ilar juvenile court offenders. This finding is consistent with most research
using comparison groups and assessing the incarceration length of transferred
and nontransferred offenders (Eigen, 1981a, 1981b; Fritsch et al., 1996;
Myers, 2001, 2003a; Podkopacz & Feld, 1996; Rudman et al., 1986). How-
ever, exceptions do exist (Fagan, 1995). In general, it appears that in terms of
length of confinement among incarcerated offenders, Act 33 is achieving its
goal of greater punishment severity. The main explanation for this finding is
that in adult court, lengthier sentences are enabled and sometimes mandated,
especially if a gun was involved. In Pennsylvania, presumptive sentencing is
used in adult criminal court, where offenders must serve their minimum
confinement before being considered for parole. However, in juvenile court,
indeterminate sentencing is used, and Pennsylvania juvenile courts have
jurisdiction over offenders only until age 21.
Some may appreciate more than others that youths who are processed and
incarcerated by adult courts are given harsh sentences. Act 33 and the result-
ing confinement time produce lengthier incapacitation, whereby youths can-
not recidivate during the time they are in secure custody. Therefore, from a
public-safety point of view, many would perceive society to be safer because
some of these youths are incarcerated for several years or more.
Finally, in examining case-processing time (or punishment swiftness), it
was found that it takes significantly longer for youths processed in adult court
to have their cases processed than similar youths processed in juvenile court.
This finding is consistent with prior research (Fagan, 1995; Myers, 2001,
2003a; Rudman et al., 1986). As mentioned earlier, this finding likely is
because of the slow nature of the criminal justice system because of increased
due process rights (e.g., bail, jury, etc.) and court backlogs. However, in the
juvenile court system, state law requires youths to be processed within a pre-
scribed period of time, which necessarily shortens the amount of time until a
case is disposed in the system. Although the U.S. and state constitutions allow
for defendants in adult court to have a speedy trial, they sometimes waive
that right to have more time to prepare for trial. However, we were not able
to include a measure on whether youths processed in adult court waived this
right and to determine its impact on case-processing time.
The results of the current study do not lend support for one of the objec-
tives of juvenile transfer, creating a greater deterrent to youthful offenders, at
least in terms of the three criteria specified in deterrence theory. There is no
increased punishment certainty or punishment swiftness for youths processed

in adult court. The study’s findings support only one element of deterrence
theory, punishment severity, which Beccaria (1764/1963) clearly states is the
least important of the three elements in providing a deterrent effect. There-
fore, if Act 33 is meant to deter subsequent offending, increasing punishment
certainty and swiftness should be a priority instead of increasing the sanc-
tions (i.e., punishment severity) for these youthful offenders.
Limitations and Directions for Future Research

One limitation of this research is the sample. We examined only one state for
this study. Although we examined multiple counties, we examined only youths
who were transferred under one Pennsylvania law. Given this, it is difficult
to draw general conclusions about juvenile transfer. Subsequent research
should examine transferred youths in multiple states, allowing firmer conclu-
sions to be made regarding the certainty, severity, and swiftness of punishment
in waiving youths to adult court.
Future research should explore the effect of type of court on court out-
comes while controlling for selection bias. As mentioned earlier, because this
research can only be quasi-experimental, the “treatment” can never be iso-
lated in the analyses. However, a body of knowledge accumulated using data
of an adequate sample size, data from multiple jurisdictions, relevant control
variables, and multivariate analyses can only build the knowledge of both
academics and practitioners on juvenile transfer and decertification, which
then could be used as one source for guiding criminal justice (including juve-
nile justice) policy.
Finally, researchers should examine juvenile transfer among different effec-
tiveness outcomes. Deterrence is only one measure of effectiveness, but it
has been noted that it is not the exclusive goal of waiving youths to adult
court. To obtain a more complete understanding on the effectiveness of this
practice, we must examine its impact among several outcomes.
Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interests with respect to the authorship
and/or publication of this article.
Funding
The author(s) received no financial support for the research and/or authorship of this
article.
Notes
1. Although youth violence did increase during this time period, it did not have the
projected increase that many believed. See Zimring (1998) for a fuller discussion.
2. All cases used for this research were Act 33 cases. This was verified at each res
earch site through examination of case files, court records, probation files, and so
on. The court of original jurisdiction in these cases was the adult system. None of
the cases was waived to adult court by a Pennsylvania juvenile court judge. After
youths have been formally charged with an Act 33 offense (along with the other
criteria mentioned earlier), they are automatically in the adult system. The juvenile
court does not have the legal authority to transfer Act 33 “juveniles,” because the
youths completely bypass the juvenile justice system and go to adult court.
3. This total sample was reduced from the original sample of 447. First, several
youths in the original data set did not meet Act 33 criteria. For instance, some
youths in the original data set were charged with murder. Although this charge is
an “excludable” offense, it is not an Act 33 offense, as youths charged with murder
were automatically waived to adult court prior to Act 33. Second, some youths
were in the data set multiple times because of committing more than one Act 33
offense during 1996. Third, some youths were either younger than 15 or older than
17, meaning they were not Act 33 offenders. Finally, some cases were dismissed
at the preliminary hearing, meaning they had no court outcomes that were relevant
to this particular research.
4. Multiple lambda terms were computed for the analyses depending on whether we
examined the entire sample of youthful offenders or a subsample, such as when we
included only convicted offenders for particular models. The appropriate lambda
term (i.e., selection bias correction) was entered as an independent variable in each
of the multivariate models.
5. The original coding for incarceration time was the amount of time imposed by the
court (i.e., for nondecertified youths) or the amount of time served (i.e., for decer-
tified youths), all of which was measured in months. However, the original coding
would pertain only to youths who were sentenced to some form of incarceration
(n = 181). This would have drastically reduced the sample, which then would have
severely limited the independent variables available to be entered in the statistical
model. Therefore, sentence type was used, allowing all convicted offenders to
be included in the analysis (n = 235). Based on Pennsylvania’s criminal justice
system, where 2 years or less in secure confinement is considered county jail time
and anything more than 2 years is considered state prison time, the measure of sen-
tence length (for both decertified and nondecertified offenders) reflected this. If
offenders in either court received 2 years or less, it was coded as a short sentence;
if offenders received more than 2 years, it was coded as a long sentence. The rest
of the youths were placed on probation.
6. For nondecertified offenders serving a county sentence (i.e., short sentence),

county judges can order parole prior to the minimum sentence being served
(Jordan, 2006; Myers, 2001). However, for offenders with a state-level sentence,
presumptive sentencing is used where at least the minimum term of incarceration
imposed by the judge must be served prior to consideration of parole. Therefore,
when examining incarceration length for nondecertified offenders, most of the
offenders would have served at least their minimum sentence, making it a conser-
vative estimate of incarceration length.
7. An argument could be made that only convicted offenders should be included
in the analysis of case-processing time, because deterrence theory may be con-
cerned only with those who actually are found guilty of committing offenses (i.e.,
swiftness of punishment). In other words, there should be some substantiation of
an illegal act occurring for youths to associate with punishment (i.e., the purpose
of punishment swiftness). This scenario was explored among the group of con-
victed offenders (n = 235), and the results were similar to that of the model that
used all the offenders. At both the bivariate and multivariate level, decertified
convicted youths had significantly shorter case-processing times than convicted
nondecertified youths.
8. Although 87% of the offenders in the data set were non-White, they also were
predominantly African American (more than 90% of the non-Whites were African
American).
9. Two separate issues were considered at this stage. First, the original coding of
the variable included public defenders, private attorneys, and court-appointed
attorneys. However, statistical tests revealed no significant differences between
private and court-appointed attorneys in decertification. In addition, court-
appointed attorneys are private attorneys who are hired by the court to take cases.
Therefore, private and court-appointed attorneys were combined into a non–pub-
lic defender group.
References
Barnes, C. W., & Franz, R. S. (1989). Questionably adult: Determinants and effects of
the juvenile waiver decision. Justice Quarterly, 6(1), 117-135.
Beccaria, C. (1963). On crimes and punishments. Indianapolis, IN: Bobbs-Merrill.
(Original work published 1764)
Bishop, D. M., & Frazier, C. E. (1991). Transfer of juveniles to criminal court: A case
study and analysis of prosecutorial waiver. Notre Dame Journal of Law, Ethics
and Public Policy, 5, 281-302.
Bishop, D. M., Frazier, C. E., & Henretta, J. C. (1989). Prosecutor waiver: Case study
of a questionable reform. Crime & Delinquency, 35, 179-201.
Bishop, D. M., Frazier, C. E., Lanza-Kaduce, L., & Winner, L. (1996). The transfer
of juveniles to criminal court: Does it make a difference? Crime & Delinquency,
42(2), 171-191.
Burruss, G., & Kempf-Leonard, K. (2002). The questionable advantage of defense
counsel in juvenile court. Justice Quarterly, 19, 37-67.
Bushway, S., Johnson, B. D., & Slocum, L. A. (2007). Is the magic still there? The
use of the Heckman two-step correction for selection bias in criminology. Journal
of Quantitative Criminology, 23, 151-178.
Champion, D. J. (1989). Teenage felons and waiver hearings: Some recent trends,
1980-1988. Crime & Delinquency, 35(4), 577-585.
Cook, P. J., & Laub, J. H. (1998). The unprecedented epidemic in youth violence. In
M. Tonry & M. H. Moore (Eds.), Crime and justice: A review of research (Vol. 24,
pp. 27-64). Chicago: University of Chicago Press.
Dilulio, J. J. (1995, November 27). The coming of the super-predators. Weekly Stan-
dard, 1, pp. 23-28.
Eigen, J. P. (1981a). The determinants and impact of jurisdictional transfer in Phila-
delphia. In J. C. Hall, D. M. Hamparian, J. M. Pettibone, & J. L. White (Eds.),
Major issues in the juvenile justice information and training: Readings in public
policy (pp. 333-350). Columbus, OH: Academy for Contemporary Problems.
Eigen, J. P. (1981b). Punishing youth homicide offenders in Philadelphia. Journal of
Criminal Law and Criminology, 72(3), 1072-1093.
Engen, R. L., Steen, S., & Bridges, G. S. (2002). Racial disparities in the punishment
of youth. A theoretical and empirical assessment of the literature. Social Problem,
49(2), 194-220.
Fagan, J. (1995). Separating the men from the boys: The comparative advantage of
juvenile versus criminal court sanctions on recidivism among adolescent felony
offenders. In J. C. Howell, B. Krisberg, J. D. Hawkings, & J. J. Wilson (Eds.), A
sourcebook: Serious, violent, and chronic juvenile offenders (pp. 238-260). Thou-
sand Oaks, CA: Sage.
Fagan, J. (1996). The comparative advantage of juvenile versus criminal court sanc-
tions on recidivism among adolescent felony offenders. Law and Policy, 18, 79-113.
Fagan, J., & Deschenes, E. P. (1990). Determinants of judicial waiver decisions for vio-
lent juvenile offenders. Journal of Criminal Law and Criminology, 81(2), 314-347.
Fagan, J., Forst, M., & Vivona, T. S. (1987). Racial determinants of the judicial trans-
fer decision: Prosecuting violent youth in criminal court. Crime & Delinquency,
33(2), 259-286.
Feld, B. C. (1995). The social context of juvenile justice administration: Racial dispari-
ties in an urban juvenile court. In K. K. Leonard, C. E. Pope, & W. H. Feyerherm
(Eds.), Minorities in juvenile justice (pp. 66-97). Thousand Oaks, CA: Sage.
Frazier, C. E., & Bishop, D. M. (1995). Reflections on race effects in juvenile justice.
In K. K. Leonard, C. E. Pope, & W. H. Feyerherm (Eds.), Minorities in juvenile
justice (pp. 16-46). Thousand Oaks, CA: Sage.
Fritsch, E. J., Caeti, T. J., & Hemmens, C. (1996). Spare the needle but not the punish-
ment: The incarceration of waived youth in Texas prisons. Crime & Delinquency,
42(4), 593-609.
Gillespie, L. K., & Norman, M. D. (1984). Does certification mean prison: Some pre-
liminary findings from Utah. Juvenile and Family Court Journal, 35(13), 23-34.
Heckman, J. J. (1976). The common structure of statistical models of truncation,
sample selection and limited dependent variables. Annals of Economic and Social
Measurement, 5, 475-492.
Houghtalin, M., & Mays, G. L. (1991). Criminal dispositions of New Mexico juve-
niles transferred to adult court. Crime & Delinquency, 37(3), 393-407.
Jensen, E. J., & Metsger, L. K. (1994). A test of the deterrent effect of legislative
waiver on violent juvenile crime. Crime & Delinquency, 40(1), 96-104.
Jordan, K. J. (2006). Violent youth in adult court: The decertification of transferred
offenders. New York: LFB Scholarly.
Kinder, K., Veneziano, C., Fichter, M., & Azuma, H. (1995). A comparison of the
dispositions of juvenile offenders certified as adults with juvenile offenders not
certified. Juvenile and Family Court Journal, 46, 37-42.
Kupchik, A. (2006). The decision to incarcerate in juvenile and criminal courts. Crim-
inal Justice Review, 31(4), 309-336.
Langley, M. (1972). The juvenile court: The making of a delinquent. Law and Society
Review, 7, 273-298.
Leiber, M. J., Johnson, J., Fox, K., & Lacks, R. (2007). Differentiating among racial/
ethnic groups and its implications for understanding juvenile justice decision mak-
ing. Journal of Criminal Justice, 35, 471-484.
Lemmon, J. H., Austin, T. L., Verrecchia, P. J., & Fetzer, M. (2005). The effect of
legal and extralegal factors on statutory exclusion of juvenile offenders. Youth
Violence and Juvenile Justice, 3(3), 214-234.
Leonard, K. K., & Sontheimer, H. (1995). The role of race in juvenile justice in Pennsylvania.
In K. K. Leonard, C. E. Pope, & W. H. Feyerherm (Eds.), Minorities in juvenile justice
(pp. 98-127). Thousand Oaks, CA: Sage.
Lewis-Beck, M. S. (1980). Applied regression: An introduction. Thousand Oaks, CA:
Sage.
Mears, D. (2003). A critique of waiver research. Critical next steps in assessing the
impacts of laws for transferring juveniles to the criminal justice system. Youth
Violence and Juvenile Justice, 1(2), 156-172.
Menard, S. (2002). Applied logistic regression analysis (2nd ed.). Thousand Oaks,
CA: Sage.
Mertler, C. A., & Vannatta, R. A. (2005). Advanced and multivariate statistical meth-
ods: Practical application and interpretation (3rd ed.). Los Angeles: Pyrczak.
Minor, K. I., Hartmann, D. J., & Terry, S. (1997). Predictors of juvenile court actions
and recidivism. Crime & Delinquency, 43(3), 328-334.
Myers, D. L. (2001). Excluding violent youths from juvenile court: The effectiveness
of legislative waiver. New York: LFB Scholarly.
Myers, D. L. (2003a). Adult crime, adult time: Punishing violent youth in the adult
criminal justice system. Youth Violence and Juvenile Justice: An Interdisciplinary
Journal, 1(2), 173-197.
Myers, D. L. (2003b). The recidivism of violent youths in juvenile and adult court: A
consideration of selection bias. Youth Violence and Juvenile Justice, 1(1), 79-101.
Pennsylvania Juvenile Court Judges’ Commission. (1996). The Juvenile Act. 42
Pa.C.S. Sec. 6301 et seq. Harrisburg, Author.
Podkopacz, M. R., & Feld, B. C. (1996). The end of the line: An empirical study of
judicial waiver. Journal of Criminal Law and Criminology, 86(2), 449-492.
Risler, E. A., Sweatman, T., & Nakerud, L. (1998). Evaluating the Georgia legisla-
tive waiver’s effectiveness in deterring juvenile crime. Research on Social Work
Practice, 8(6), 657-667.
Rudman, C., Hartstone, E., Fagan, J., & Moore, M. (1986). Violent youth in adult
court: Process and punishment. Crime & Delinquency, 32(1), 75-96.
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-
experimental designs for generalized causal inference. Boston: Houghton Mifflin.
Singer, S. I., & McDowall, D. (1988). Criminalizing delinquency: The deterrent effects
of the New York Juvenile Offender Law. Law and Society Review, 22(3), 521-535.
Snyder, H. N. (1998). Juvenile arrests 1997 (Report No. NCJ 173938). Washington,
DC: Office of Juvenile Justice and Delinquency Prevention.
Spohn, C. C. (2002). How do judges decide? The search for fairness and justice in
punishment. Thousand Oaks, CA: Sage.
Sridharan, S., Greenfield, L., & Blakley, B. (2004). A study of prosecutorial certifica-
tion practice in Virginia. Criminology and Public Policy, 3(4), 605-632.
Taylor, R. W., Fritsch, E. J., & Caeti, T. J. (2002). Juvenile justice: Policies, pro-
grams, and practices (2nd ed.). Woodland Hills, CA: Glencoe/McGraw-Hill.
Thomas, C. W., & Bilchik, S. (1985). Prosecuting juveniles in criminal courts: A legal
and empirical analysis. Journal of Criminal Law and Criminology, 76(2), 439-479.
Torbet, P., Gable, R., Hurst, H. I., Montgomery, I., Szymanski, L., & Thomas, D.
(1996). State responses to serious and violent juvenile crime (Report No. NCJ-
161565). Washington, DC: Office of Juvenile Justice and Delinquency Prevention.
Winner, L., Lanza-Kaduce, L., Bishop, D. M., & Frazier, C. E. (1997). The transfer
of juveniles to criminal court: Reexamining recidivism over the long term. Crime
& Delinquency, 43(4), 548-563.
Wordes, M., Bynum, T. S., & Corley, C. J. (1994). Locking up youth: The impact
of race on detention decisions. Journal of Research in Crime and Delinquency,
31(2), 149-165.
Zimring, F. E. (1998). American youth violence. New York: Oxford University Press.
Bios
Kareem L. Jordan is an assistant professor at University of North Florida. He
received his PhD from Indiana University of Pennsylvania in 2005. His recent publi-
cations have appeared in Criminal Justice Studies, American Journal of Criminal
Justice, and Youth Violence and Juvenile Justice. His research and teaching interests
include juvenile delinquency, criminological theory, race and crime, and quantitative
methods and statistics.
David L. Myers is a professor of criminology and Interim Vice Provost for Research
and Dean of Graduate Studies at Indiana University of Pennsylvania. His recent pub-
lications have appeared in Criminal Justice and Behavior, Criminal Justice Review,
and Crime Prevention and Community Safety. He is the editor of Criminal Justice
Policy Review and author of Boys Among Men: Trying and Sentencing Juveniles as
Adults (Praeger, 2005).
Erratum
Crime & Delinquency
57(3) 489
Erratum © 2011 SAGE Publications
Reprints and permission: http://www.
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0011128711407760
http://cad.sagepub.com
Jordan, K. L., & Myers, D. L. (2011). Juvenile transfer and deterrence:

Reexamining the effectiveness of a “Get-Tough” policy. Crime & Delin-
quency, 57(2), 247–270. Original DOI: 10.1177/0011128708319111
In the above article, at the bottom of page 247, Kareem L. Jordan’s corre-
spondence address was erroneously mentioned as—University of North Flor-
ida, Atlanta, GA 30302.
The correct correspondence address is—University of North Florida,
Department of Criminology and Criminal Justice, 1 UNF Drive, Jackson-
ville, FL 32224.
View publication stats

Juvenile Transfer and Deterrence Reexamining The E

Uploaded by

Copyright:

Available Formats

Juvenile Transfer and Deterrence Reexamining The E

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Juvenile Transfer and Deterrence Reexamining The E

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Juvenile Transfer and Deterrence: Reexamining the

Article in Crime & Delinquency · March 2011

Kareem Jordan David L Myers

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Juvenile Transfer and Deterrence: Reexamining the Effectiveness of

The online version of this article can be found at:

Email Alerts: http://cad.sagepub.com/cgi/alerts

>> Version of Record - Mar 9, 2011

Proof - Jul 11, 2008

Kareem L. Jordan1 and David L. Myers2

to surges in violent youthful offending, violent victimization of adolescents

Past Empirical Research

Limitations of Prior Research

Table 1. Descriptive Statistics for All Continuous Variables N = 345

Variable M SD Min. Max.

Table 2. Descriptive Statistics for all Nominal and Ordinal Variables

The second measure of sentence severity is type of sentence, which is mea-

Three of the dependent variables (conviction, target conviction, and incarcera-

Table 3. Logistic Regression Estimates for the Impact of Decertification on

When [the odds] are 0 or 1 for an entire group of cases, as defined by

Table 4. Logistic Regression Estimates for the Impact of Decertification on Target

Table 5. Logistic Regression Estimates for the Impact of Decertification on

logistic regression estimates for the determinants of target conviction. When

Table 6. Ordinal Regression Estimates for the Impact of Decertification on

Table 7. Ordinary Least Squares Regression Estimates for the Impact of

County experienced longer case-processing time than youths in Allegheny

Discussion and Conclusion

terms of conviction likelihood. Some criminal justice professionals (i.e., judges,

increased punishment certainty or punishment swiftness for youths processed

Limitations and Directions for Future Research

Declaration of Conflicting Interests

6. For nondecertified offenders serving a county sentence (i.e., short sentence),

Jordan, K. L., & Myers, D. L. (2011). Juvenile transfer and deterrence:

You might also like