A. 3.72.:
, a»: .

.1215 t
3.: z
o «t.

 

 

 

THE. .9 3:3

lllllllllllllllllllllllllllllll'Ill.llllllllllllllllillill

129301771054

 

LIBRARY I

Michigan State
University

 

This is to certify that the

thesis entitled
THE EFFECTS OF PEERS, FAMILY, AND

THE ENVIRONMENT ON JUVENILE HANDGUN OWNERSHIP

presented by

Regan Marie Suhay

has been accepted towards fulfillment
of the requirements for

hmwsJegee in W\
6 OS“ u

flwfiWflfl

Major profess

Date / (/Y/Zfi

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE
ALICE ii 299%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1!“ COMM“

THE EFFECTS OF PEERS, FAMILY, AND
THE ENVIRONMENT ON JUVENILE HANDGUN OWNERSHIP

By

Regan Marie Suhay

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Criminal Justice

1 999

ABSTRACT

THE EFFECTS OF PEERS, FAMILY, AND
THE ENVIRONMENT ON JUVENILE HANDGUN OWNERSHIP

By

Regan Marie Suhay

A strong gun subculture exists within the United States of America, and in
recent years the number of juveniles participating in this phenomenon has been quite
high. Social learning theory contends that juveniles who are surrounded by significant
others who support the gun subculture will likely acquire these same beliefs. Thus,
youths' involvement with firearms is encouraged by influential groups, such as peers,
the family, and their environment. Results indicate that juveniles are significantly
affected by these three socializing forces. In short, juveniles who come from families
that display pro-gun values, who have peers who exhibit pro-gun behavior, and who
live in environments characterized by a high gun prevalence are much more likely to

own handguns than juveniles who are not exposed to these influences.

ACKNOWLEDGMENTS

My name may appear on this thesis, but many contributed to its completion.
The time I have spent in East Lansing has been made bearable by my friends, namely
Dana Pasquali, Debbie Schaming, Kelly Begley, Alon Neches, and Scott Stoneking.

I would like to thank those on my thesis committee, Charles Corley, Sheila
Maxwell, and Christina Dejong. In all my years of schooling l have had the pleasure of
learning from a few, select, truly gifted teachers; Christina Dejong is one of them. I
have learned much from her classes and benefitted greatly from the guidance she has
given me throughout this experince. I would not ever have finished if it wasn‘t for her
dedication, help, and support. Thanks Christina.

Finally, I would like to stress the role my family has played in not only the
completion of this degree, but also in my life. I come from a wonderful home with two
loving parents and two very special siblings, my sister Paige and my brother Ryan. My
family has made me everything I am, I am so grateful to them for everything they have
given me. This year was especially hard due to the death of my little brother, there are
times when I just wanted to stay at home and not return to Michigan. But, following
Ryan's footsteps l dealt with what was presented and l forged on. This paper is

dedicated to Ryan Thomas Suhay.

iii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... vi

INTRODUCTION ............................................................................................................. 1
CHAPTER 1

DIFFERENTIAL ASSOCIATION THEORY ...................................................................... 4
CHAPTER 2

JUVENILE DELINOUENCY MEASURES ...................................................................... 11
CHAPTER 3

DATA 8. METHODS ....................................................................................................... 15

Data ................................................................................................................... 15

Independent Variables ....................................................................................... 16

Family ..................................................................................................... 19

Peers ...................................................................................................... 20

Environment ........................................................................................... 21

Attitude ................................................................................................... 23

Dependent Variable ........................................................................................... 24

Gun Ownership ...................................................................................... 24

Control Variables ............................................................................................... 25

Age ......................................................................................................... 26

Race ....................................................................................................... 26

Gang Measures ...................................................................................... 27

Status ..................................................................................................... 28

Methods ............................................................................................................. 29
CHAPTER 4

RESULTS ...................................................................................................................... 30

Bivariate Analysis .............................................................................................. 30

Multivariate Analysis .......................................................................................... 35
CHAPTER 5

CONCLUSION .............................................................................................................. 43

Data Findings ................................................................ . ................................... 44

Data Limitations ................................................................................................. 47

Policy Implications ............................................................................................. 50

APPENDIX A ................................................................................................................. 52

iv

BIBLIOGRAPHY ........... . ................................................................................................ 54

LIST OF TABLES

Table 1 - Respondents' demographic characteristics .................................................... 15

Table 2 - Reliability analysis for independent and control variables .............................. 17

Table 3 - Bivariate analysis (ANOVA).
The effects of the independent and control variables on

juvenile handgun ownership ........................................................................... 30
Table 4 — Logistic Regression.

The effects of the control and independent variables on juvenile
handgun ownsership ...................................................................................... 36

vi

mm

The United States of America is a nation with a pronounced gun subculture.
The Bureau of Alcohol, Tobacco, and Firearms estimates in 1993 there were 223
million firearms in America (Bureau of Justice Statistics, 1995). This figure includes 79
million rifles, 77 million handguns, and 66 million shotguns (BJS, 1995). The report
also notes that the number of seized, destroyed, lost, or inoperable firearms is
unknown. In addition, some 211,000 handguns and 382,000 long guns were stolen in
noncommercial thefts in 1994 alone (National Institute of Justice, 1997). The FBl's
National Crime Information Center stolen gun file contains over two million reports, sixty
percent of which are reports of stolen handguns (BJS, 1995).

In recent years, America has witnessed an increasing number of juveniles
owning, possessing, and canying firearms (Ash, Kellen'nann, Fuqua-Whitley, &
Johnson, 1996; Zimring, 1996). Along with the rise of firearm contact, there has also
been a marked increase in the number of juveniles committing crimes with guns and
the number of juvenile deaths caused by firearms. In 1996, 33,436 juveniles were
arrested for possessing a weapon (United States Department of Justice, 1997). Gun-
related mortality accounts for almost half of all deaths among African-American teens
(Ash et al., 1996). In addition, more US teenagers die from gunshot wounds than from
all natural causes of disease combined (Ash et al., 1996). Just as gun possession has
increased in recent years, so has the use of guns in homicides committed by juveniles
from 57% in 1978 to 78% in 1992 (Zimring, 1996).

It is evident that juveniles are becoming increasingly involved with firearms. Not
only are youths owning guns at a higher rate than ever before, but they are also
committing more offenses and dying more often as a result of this heightened level of
gun contact (Center to Prevent Handgun Violence, 1993). Juveniles are obtaining guns

from peers, family members, gangs, theft, and street persons (Sheley 8 Wright, 1995).

In order to stop this trend, research must be conducted to determine why youths are
choosing to participate in the gun subculture. Once policy makers are educated about
the matter, proactive changes may be implemented to stop the corruption of American
youth.

Social Ieaming theory contends that individuals Ieam deviant attitudes and
behaviors in the same way as they do conforming attitudes and behaviors (Hansell &
Wiatrowski, 1981; Sutherland, 1939). Furthermore, Ieaming occurs within intimate
groups made up of those who hold significant meaning in one's life (Burgess 8. Akers,
1966). Differential association theory expands upon these statements and offers a
more specific explanation as to how a person becomes delinquent. The theory states
that a person becomes delinquent because of an excess of definitions favorable to law
violation over definitions unfavorable to law violation (Sutherland, 1947). Significant
others include peers, the family, environment, school, and church. These groups
impact an individual's acquisition of attitudes and, ultimately, his or her decision to act
lawfully or unlawfully.

A great deal of research has found relationships between these groups and
youths' pro-delinquent attitudes and behaviors (Elliot, Huizinga, & Ageton, 1985;
Inciardi, Horowitz, 8. Pottieger, 1993: Winfree, Backstrom, 8. Mays, 1994). Researchers
have repeatedly found that peers influence youths' decisions, especially with regard to
firearm habits (Emler 8. Reicher, 1995; Inciardi et al., 1993; Jensen & Rojek, 1992;
Lizotte, Tesoriero, Thomberry, & Krohn, 1994; Sheley 8. Wright, 1995; Sutherland 8.
Cressey, 1978; Winfree et al., 1994). Additional research has stressed the role of the
family on a child's socialization (Lizotte et al., 1994; Loeber 8. Dishion, 1983;
Patterson, DeBaryshe, & Ramsey, 1989). The family is crucial to youths' Ieaming
process, as it can encourage juveniles to act legally or illegally (Ash et al., 1996). Geo,
Cullen, 8. Link (1997) found that the family actively affects a youth's socialization with

respect to firearms.

A third factor, the environment, serves as an important socializing force as it
introduces youths to the gun subculture. The atmosphere in which a youth is raised
may contain its own set of rules, norms, values, and consequences that affect juvenile
gun ownership (Staples, 1982). As a result, the youth's surroundings require
definitions, attitudes, and behaviors which are radically different from greater society's.
This is the case with urban America, as the inner-city has it's own "code of the streets"
by which youths must Ieam to live (Wilkinson 8. Fagan, 1996).

The purpose of this thesis is to further explore the inter-connectedness of peers,
family, environment, and juveniles' firearm behavior. Specifically, the youths' peers,
families, and neighborhoods are examined to determine whether they are supportive of
the youths' use of firearms. The findings produced will shed light onto the individual
and combined impact each of these socializing factors have on juvenile handgun

ownership.

Hymtheses

Alternative: Juveniles who are exposed to families, peers, and environments which
encourage gun ownership are likely to own handguns.
Null: Juveniles who are not exposed to families, peers, and environments

which encourage gun ownership are not likely to own handguns.

Chapter 1 - Differential Association Theory

Over the years, many different theories have attempted to explain why people.
engage in delinquent behavior. The spectrum of theories ranges from those that claim
criminal acts are freely chosen, to others that state they are a result of outside forces.
In recent years, sociological explanations have emerged as the leading frame of
thought in the field (Jensen, 1981). These sociology-based theories strive to determine
the influences in a deviant's life, namely his/her family, peers, school, church, and
neighborhood. The central thesis of the sociological school is that criminal behavior
results from the same social processes as does conforming behavior (Sutheriand 8.
Cressey, 1978; Vold & Bernard, 1986). In order to explain why a person complies with
or deviates from the status quo, researchers from this school examine the relationships
between the individual and the significant others who influence his/her life. It is these
agents that lead a person to acquire either pro-deviant or anti-deviant attitudes and
behaviors (Akers, Krohn, Lanza-Kaduce, 8. Radosevich, 1979; Burgess 8. Akers, 1966;
Sutheriand, 1947; Winfree et al., 1994).

Social Ieaming theory is an explanation of delinquency focusing on the ways in
which deviance is Ieamed. This frame of thought contends that Ieaming deviant
attitudes and behaviors occurs in the same way that one would Ieam conforming
attitudes and behaviors (Sutheriand, 1939). The social ability model proposes that
delinquents are like nondelinquents, in that they share similar Ieaming processes and
capabilities (Hansell 8. Wiatrowski, 1981). Additional assumptions held by this model
are that delinquents operate similary to non-delinquents in respect to concerns about
social appearance, prestige, and social status (Hansell 8. Wiatrowski, 1981).
Furthermore. the two adolescent groups are alike in that they both can master group
roles and exchange influence and rewards (Hansell 8. Wiatrowski, 1981). In short, both

groups Ieam, internalize, and maintain beliefs and behaviors in a similar fashion.

Learning occurs within intimate groups, through the process of interacting with
other perle (Burgess 8 Akers, 1966; Inciardi et al., 1993; Sutherland, 1947). Much of
the research in this area states that behavior, whether delinquent or non-delinquent, is
learned from those who any high meaning in one's life (Akers et al., 1979; Burgess 8
Akers 1966; Smith 8 Brame, 1994; Sutherland, 1947; Sutherland 8 Cressey, 1978;
Winfree et al., 1994). Sutherland 8 Cressey (1978) suggest that criminal behavior is
learned through the symbolic interaction of intimate groups that have incorporated
criminal values. These groups provide social opportunities to Ieam criminal pattems
and definitions favorable to law violation, in addition to teaching skills necessary to
commit crimes (Sutherland 8 Cressey, 1978). Social Ieaming theory is comprised of
direct conditioning, imitations, definitions, differential associations, and differential
reinforcement (Akers et al., 1979).

Learning occurs in two ways, through direct conditioning and imitation or
modeling (Akers et al., 1979). Direct conditioning is a result of direct observation,
whether through abusive parents, delinquent peers, or violent images witnessed in
society, and leads to pro-delinquent attitudes and behaviors (Goldstein, 1991). Akers
(1985) contends that some Ieaming is nonsocial, but the majority of Ieaming relevant to
delinquent behavior is the result of social interaction or exchanges in which words,
responses, and behavior of other persons directly conditions the youth to acquire
definitions favorable to deviance. Modeling and imitation may occur in the home,
school, or "on the streets" and becomes more formalized in the teenage years (Monti,
1993).

Imitation takes place when a youth is exposed to admired models who act in a
delinquent manner (Akers, 1985). The observer may imitate the modeled behavior if
they "like or respect the model, see the model receive reinforcement, see the model
give off signs of pleasure, or are in an environment where imitating the model's

performance is reinforced" (Baldwin 8 Baldwin, 1981, p.181). As one can see, there

are numerous fashions in which Ieaming occurs, each offering an opportunity for a
youth to develop pro-delinquent beliefs.

It is the acquisition of definitions which shape a person’s opinions and determine
what is and is not acceptable; however, what is acceptable to an individual may not be
acceptable in the eyes of the law. Definitions are "normative meanings which are given
to behavior; that is, they define an action as right or not right" (Akers, 1985, p.49). One
may acquire positive or negative definitions that shape his/her attitudes and,
consequently, his/her behavior.

Differential association has been called the single most important explanatory
concept of social Ieaming theory (Akers et al., 1979; Sellers 8 Winfree, 1990).
Ultimately, the concept resulted in the creation of differential association theory, first
outlined by Edwin H. Sutherland (Vold 8 Bemard, 1986). The theory of differential
association is a variation of social Ieaming theory and contains many of the same
components. The crux of differential association theory is that a person becomes
delinquent because of an excess of definitions favorable to law violation over
definitions unfavorable to law violation (Sutherland, 1947). Differential association
stresses the peer relationship as being the most influential, more so than any of the
other influencers previously listed.

Over the years, a great deal of research has supported the basic premises of
differential association theory (83", 1957; Burgess 8 Akers, 1966; Burkett 8 Jensen,
1975; Winfree et al., 1994). Differential association has also been successfully used to
explain delinquency in terms of adolescent drug and alcohol behavior (Akers et al.,
1979; Krohn, 1974). Upon completion of their study, Akers et al. (1979) stated that the
theory of social Ieaming will be extremely beneficial to research focusing on all types of
delinquent behavior.

Like any theory, differential association has encountered its share of opposition.

Some researchers admit confusion about what comes first, association with delinquent

peers, definition formation, or reinforcement (Gottfredson 8 Hirschi, 1990). Others
found that delinquent peers had little effect on a person's law-abiding and law-violating
definitions (Short 8 Strodbeck, 1965). Another concern of the theory is that it ignores
major influential forces such as the family, school, church, and environment. In this
thesis, I will attempt to include some of these other influential groups as I believe they
also affect the formation of definitions. In addition to focusing on peer relationships
and deviance, indicators focusing on the family, environment, and youth gang
involvement and their effects on attitudes favorable to delinquency and actual
delinquent behaviors will be explored.

Differential association theory proposes that peer relationships have the
greatest impact on a youth's development of pro-delinquency definitions. Research
has found that peers are one of the most important groups in an adolescents life
(Akers et al., 1979; Inciardi et al., 1993; Jensen 8 Rojek, 1992). Inciardi et al. (1993)
state that during the teenage years the primary group shifts from parents to peers as
youths spend more time with their friends, struggle to achieve parental independence,
and strive to be treated as an adult.

Differential association, adopting the basic premise of social Ieaming theory
that Ieaming occurs within intimate groups, contends that peers, whether delinquent or
not, greatly influence youths' definitions, values, and actions (Sutheriand, 1947). Much
of the research in this area has found a relationship between delinquent peers and a
youth's own pro-delinquent attitudes and behaviors (Elliot et al., 1985; Inciardi et al.,
1993; Jensen 8 Rojek, 1992; Winfree et al., 1994). Winfree et al. (1994) have
suggested that knowing about friends' involvement in delinquency is a better predictor
of the youth's own involvement than personal characteristics such as race, sex, or living
situation.

Differential associations depend on the frequency, duration, priority, and

intensity of a relationship (Sutherland, 1947). For this reason, differential association

theory has a natural link to gang research. Gang membership results in expanded
socialization characterized by heightened peer relationships which, almost always,
involves pro-delinquent definitions, attitudes, and behaviors. This statement is
supported by a great deal of research which has found that gangs are characterized by
high levels of delinquent activity (Bjerregaard 8 Lizotte,1995; Monti, 1993; Esbensen 8
Huizinga, 1993; Jensen 8 Rojek, 1992; Sheley, Zhang, Brody, 8 White. 1995). While
Gottfredson and Hirschi (1990) state that a "gang" is merely a gathering of individuals
pursuing their own individual criminality, a greal deal of research supports the claim that
gangs, and the collective atmosphere produced by such a group, strongly influence
their individual members (Esbensen 8 Huizinga, 1993; Howell, 1997; Sheley et al.,
1995). Thus, an individual may very likely acquire and/or maintain pro-delinquent
attitudes and behaviors as a result of joining a gang.

Gang members are subjected to accelerated peer relationships as the gang
often acts as protector, friend, and surrogate family (Monti, 1993). A great deal of time
is spent with the gang; therefore, the frequency of gang contact is quite high (Vigil,
1988). The typical age range of gang members spans a large period, from age 8 to
age 22 (Oehme, 1997). Historically, gang membership was a relatively short lived
period due to member maturation; however, in recent years a new trend has emerged
where youth actively participate in gang activities for a longer period of time (Maxson,
Gordon, 8 Klein, 1985; Spergel, 1983). As a result, the duration of gang relationships
is lengthening.

A look at gang research depicts members as being extremely loyal to their
gang, often referring to it as the most important thing in their life (Padilla, 1992).
Research often highlights the intense bond between members and the organized gang
(Padilla, 1992; Vigil, 1988). Due to these reasons, I believe that gang membership

must be examined when studying differential association and delinquent youth.

There is evidence that differential association can explain gun ownership among
juveniles. Cao et al. (1997) found that people who have been in social situations
supporting pro-gun values often results in the individual acquiring these same beliefs.
These same authors also found that when an individual is trained in firearm handling,
he/she is likely to own a gun. A significant amount of research has indicated that
juveniles are likely to own a gun if their peers own a gun (Blumstein 8 Cork, 1996;
Inciardi et al.,1993; Sheley 8 Wright, 1995; Smith, 1996).

Beyond the peer relationship, several authors have examined the parent-youth
relationship and they have found that family atmospheres that foster definitions and
actions favorable to law violation often produce delinquent children (Santrnan et al.,
1997; Sutherland 8 Cressey, 1978). Adolescent gun ownership is more likely when
parents possess and encourage the use of guns (Caetano, 1979; Lizotte et al., 1994).
Although the concept of "environment" is not typically used when studying social
Ieaming thoery, it is, nonetheless, an important factor. Wilkinson 8 Fagan (1996) refer
to the ”code of the streets" and Osofsky (1995) mention CCV, chronic community
violence, when describing the effect one's surroundings has on his/her socialization.
The inner-city is regarded as a high crime, dangerous, and unstable environment often
characterized by gang presence (Stark, 1987). In addition, Bjerregaard and Lizotte
(1995) report that there is a positive relationship between gang membership and gun
ownership.

As previously stated, differential association is a component of social Ieaming
theory. While differential association focuses on peer relationships and delinquency,
social Ieaming theory includes any group which serves as a significant other to the
individual. This may include, but is not limited to, peers, organizations, family, school,
and church. In this paper, the first three factors (peers, gang membership, and family)
will be examined, but the fourth and fifth factors (school and church) will not be

included due to data limitations which will be outlined in the Data and Methods chapter

of this thesis. In addition to the typical groups being analyzed, the concept of
"environment" is included because a youth's surroundings is an important socializing
factor.

The theory of differential association explains delinquency as a result of an
excess of definitions favorable to law violation over definitions unfavorable to law
violation (Sutherland, 1947). This component of social Ieaming theory focuses on why
a person develops deviant attitudes and why he/she initially becomes delinquent, but
does not explain why delinquent behaviors are maintained. Burgess and Akers (1966)
amended Sutherland's theory of differential association to further the scope of
explanation. These authors are in agreement with Sutherland as to why a person
initially enters the realm of delinquent actions, their contribution to the theory was to
outline why delinquent behaviors are maintained over time. Behavior continues
dependent upon the effects, outcomes, or consequences it has on a person's
environment (Akers, 1985).

Specific to this paper, differential association explains the acquisition of pro-gun
definitions, which, consequently, explains why juveniles choose to own handguns.
Differential reinforcement would have been used to explain juvenile handgun ownership
which exists beyond the initial behavior, however, due to data limitations maintained
handgun ownership cannot be examined. Inciardi et al. (1993) contend that actual
behavior is more important than attitude; thus, peer, parental, gang, and environmental

factors will be measured to gauge their effects on juvenile handgun ownership.

10

Chapter '21— Juvenilejdglinguency measures

The backbone of differential association theory states that a person becomes
delinquent because of an excess of definitions favorable to law violation over
definitions unfavorable to law violation (Suthedand, 1947). While this theory focuses
primarily on youths' peer groups, there are other influencing agents which effect
juveniles' lives. These impressionistic groups include not only peers, but also the
youth's family, church, school, and environment. The data used in this thesis do not
provide measures of the juveniles' church or school histories, but do include items
relevant to peer, family, and environmental influence.

Research has determined that juveniles do, in fact, own firearms and they often
own multiple guns (Ash et al., 1996; Fagan, 1990; Inciardi et al., 1993; Lizotte, Howard,
Krohn, 8 Thombeny, 1997; Wright et al., 1992). This thesis will only include handguns
(revolvers and pistols); rifles and shotguns will be excluded. Juveniles tend to prefer
concealable, large caliber guns (Biskup 8 Cozic, 1992; BJS, 1995). Research has
found that handguns are the gun most commonly used in juvenile crime and that
juveniles are attracted to this type of gun (Center to Prevent Handgun Violence, 1993;
BJS, 1995). DeFronzo (1979) limited his study to handguns because they are more
likely to used for crime, protection. or inflicting injury upon someone than shotguns or
rifles.

Today, the concept of the traditional family is almost extinct. Instead, families
include many varying types including dual parent, single parent, and guardianship.
Whatever the living situation may be, there is a consensus among researchers that the
family is crucial to a child's socialization (Akers et al., 1979; Caetano, 1979; Hirschi,
1969; Lawrence, 1998; Lizotte et al., 1994; Loeber 8 Dishion, 1983; Nye, 1958;
Santman, Myner, Cappelletty, 8 Penmutter, 1997). Just as the family is important to

socialization that results in a youth's conforming behavior, such is the case with deviant

11

behavior (Larzelere 8 Patterson, 1990; Patterson et al., 1989). Researchers have
found that the family does affect youths' pro-gun socialization and that most juveniles
were given their first gun by a peer. older youth, or relative (Ash et al., 1996; Cao et al.,
1997). In addition, a study by Wright et al (1992) found that 79% of inmates and 70%
of students who previously owned guns at one time came from families where male
relatives owned guns. Because of the large amount of research pointing to the
importance of family variables and their affect on youth attitude acquisition and
behaviors, the family will be included as an independent variable.

Youths who feel rejected by their parents often turn to their peers for support
(Lawrence. 1998). At some point in every youth's life, friends begin to have an
influence equal or greater to that of their parents (Lawrence, 1998). A great deal of
social science research asserts that friends significantly influence a youth's attitudes
and behaviors (Akers et al., 1979; Hallinan 8 Williams, 1990; Inciardi et al., 1993;
Jensen 8 Rojek, 1992; Sutherland 8 Cressey, 1978; Winfree et al., 1994). Emler 8
Reicher (1995) interviewed a number of delinquent youth as to why they broke the law,
each youth, in some way, highlighted the importance their peers had in their decision.
Giordano, Cemkovich, 8 Pugh (1986) found that delinquents care more about what
their friends think of them, often influencing their actions as they attempt to win peer
approval. Mason, Cauce, Gonzales, 8 Hiraga (1994) suggest that African-American
youth are more subject to peer influence, which is often negative, as peers become
surrogate parents eariy in life for low-income African-American youth.

Previous research has found a positive relationship between peer gun
ownership and juvenile gun ownership (Inciardi et al., 1993; Lizotte et al., 1994). When
asking juveniles if they were respected by their friends if they owned a gun, 40% of the
respondents who were gang affiliated responded affirmatively while 28% of non-gang
youth responded affirmatively (Decker. Pennell, 8 Caldwell, 1997). In a related study,

Wright et al. (1992) found that 90% of juvenile inmates who owned guns and 57% of

12

students who owned guns had friends who owned guns. An even stronger relationship
has been found betheen juvenile gang membership and gun ownership, as gang
members are more likely to own firearms than non-gang youth (Curry et al., 1994;
Sheley et al., 1995). The impact of gang membership on adolescent gun ownership will
be addressed further in the following chapters.

The concept of environment is defined as the aggregate of social and cultural
conditions that influence the life of an individual or community (Mish et al., 1990). Much
of the research focusing on juvenile gun ownership has found that youths who own
guns come from an environment where firearms are readily available (Decker et al.,
1997; Wright et al., 1992). Another common characteristic found in similar studies is
the high level of violence present in the juveniles' neighborhoods. The inner-city is a
dangerous and explosive environment marked by a high crime rate (Stark, 1987;
Wilkinson 8 Fagan, 1996). In addition, gang problems are common within American
cities, both small and large, and exaggerate the crime rate because of heightened
levels of juvenile violence (Curry, Ball, 8 Fox, 1994).

Many studies have exposed a high rate of victimization in urban America, with
many youth reporting they have been personally threatened, personally know someone
who has been threatened, and have witnessed violence (Ash et al., 1996; Wright et al,
1992). Exposure to high levels of violence becomes a way of life for inner-city
adolescents (Wilkinson 8 Fagan, 1996). Anderson (1994) argues that simply living in
such an environment considerably increases the risk of young people falling victim to
aggressive behavior. This phenomenon is so real that the term chronic community
violence (CCV) has been coined to describe the pronounced risks which accompany
inner-city life (Osofsky, 1995). Chronic community violence is defined as frequent and
continual exposure to guns, knives, drugs, and random violence, conditions which exist

in all urban environments (Osofsky, 1995).

13

The vast majority of research concerning juvenile gun ownership has continually
found that the primary reason listed for owning a firearm is personal protection (Decker
et al., 1997; Hemenway, Prothrow-Stith, Bergenstein, Ander, 8 Kennedy, 1996; Lizotte
et al., 1994; Sheley 8 Wright, 1993: Wright et al., 1992). Juvenile gun ownership may
be a direct result of environmental instability which promotes violence and fear of
victimization. To determine the influence the environment has over juveniles residing
within an environment characterized by CCV, gun accessibility and community violence
will be used as measures of the overall level of violence present in the juveniles'
neighborhoods. These measures will be scaled and examined to determine how they
affect juvenile gun ownership.

Throughout the course of this paper, juveniles' attitudes conceming firearms will
be examined to determine what effect they have, if any, on actual gun ownership.
Attitudes are important to include in analysis because they illustrate youths' pro-gun or
anti-gun beliefs which result from prior socialization. Some researchers contend that it
is more important to study behavior than attitudes, because behavior is what
determines delinquency (Inciardi et al., 1993). Hence, this thesis will study the
relationships among a juvenile's family, peers, and environment to determine what
effects they have on juvenile handgun ownership. In order not to complicate, cloud, or
misconstrue any possible relationships which may emerge, the youths' sex, age, race,
gang membership, and incarceration status will be controlled for so as not to

misrepresent any of the findings.

14

Chapter 3 -- Data 8 Methods

Q§t_a

These data were gathered to study violence committed by and against inner-city
juveniles (see Sheley et al., 1991). The sample population consists of two groups:
inner-city public high school students in grades nine through twelve and incarcerated
youths held in detention facilities near the schools' locations. A total of ten inner-city
high schools and six detention facilities were targeted in four states: California, Illinois,
Louisiana, and New Jersey. Questionnaires were distributed to 835 male inmates and
1,663 male and female students. Data collection took place from January 1991
through April 1991. The two questionnaires differed slightly between the two
populations (students and inmates), but covered the same core subjects for each
sample group. Items questioned the respondents about their demographic
characteristics, school experiences, firearm habits, victimization histories, gang
membership, neighborhood descriptions, delinquent patterns, drug use, and attitudes
concerning guns, general delinquency, and violence. The site selection purposely

targeted cities with high gun-related activities.

 

Table 1. Respondents' demographic characteristics (N = 1641)

Mean age 16.69 years

Percent race distribution 7.7 White
60.3 Black
24.2 Hispanic
8.0 Other

Percent of sample incarcerated 50.9

Percent of sample claiming gang membership 50.3

Percent of sample owning handgun 56.0

 

15

Originally, data were gathered from 835 male inmates and 1,663 male and
female high school students. Because of the absence of comparative female inmate
data, the 904 female students were removed from analysis which resulted in a sample
size of 1594 male respondents. From this sample, 133 cases consisting of persons
who indicated they owned handguns for the purpose of hunting were deleted because
juveniles in some states are legally permitted to own guns for hunting. By excluding
those respondents, the reliability is increased. The final sample size is 1461
respondents. Table 1 describes the respondents' demographic characteristics,
including their age, race, incarceration status, gang membership and handgun

ownership.

Independent Variables

The independent variables used in this thesis will be scales measuring family
gun habits, peer approval of gun ownership, peer gun behavior, gun prevalence in
neighborhoods, and gun attitudes. Scales offer two benefits; they increase the
complexity of the level of measurement of variables, and they avoid reliance on any
single response item. As a result, they provide a more exact measurement of a

concept. Scales are summarized in Table 2.

16

Table 2. Reliabity Analysis for Independent and Control Variables

 

Family gun habits sca_le

 

 

 

 

 

|t_e_rn M211.

1. Do any of your siblings own a gun .3756

2. Have any of your siblings ever given you a gun .1634

3. Do any of your male relatives own a gun .7484

4. Have you ever gone shooting with a relative .3803

Reliability coefficient .5827

Sample size 1065

Peer approval of gun ownership scale

M Mean Percent Agree
1. Friends would look down on me if I didn't own gun 1.5188 9.6

2. No respect from peers it didn't own gun 1.5848 11.5
Reliabilm' coefficient .7713

Sample size 1197

Peer gun behavior scale

[Leg Mpg Percent Agree
1. How many of your friends own guns 2.1860 68.3

2. How often have you gone shooting with friends 3.0387 48.8
Reliability coefficient .7417

Sample size 1188

 

17

 

Gun prevalence in neighborhood scale

 

 

 

lt_e_rp Meg; Percent Agree
1. Have you been threatened with a gun 1.8863 64.0
2. There are lots of guns in my neighborhood 2.4565 79.2
3. How easy is it to get gun 2.3649 75.7
4. Need gun to defend myself 2.4432 75.4
Milimgoefficient .6036

Sample size 1047

Qun attitude scale

Ltm Mill

1. Guns make me feel better .5962

2. Guns are fun .4076

3. It is OK to shoot someone who has hurt you .4867

4. It is OK to shoot someone who doesn't belong .2057

5. It is OK to shoot someone to get what you want .2495

My coeflzient .7381

Sample size 1050.

 

18

 

Gang Influence scale

 

 

 

Item Mean
1. Many gang members cany guns .7057
2. There are many guns around when gang meets .6743
3. Must own a gun to join my gang .0943
4. Must show you can use a gun to join my gang .1743
Reliability coefficient .6068
Sample size 700
Family

A youth's family is an important factor in his decision to participate in or abstain
from delinquent activities (Akers et al., 1979; Nye, 1958). A great deal of research
shows that a child's family is cmcial to his/her socialization (Hirschi, 1969; Loeber 8
Dishion, 1983; Nye, 1958). In accordance with Ieaming theory, it a child is raised in a
family which supports pro-delinquent beliefs and actions, he/she is likely to exhibit pro-
delinquency tendencies. Research has found that delinquency is more probable
among youths whose family members are criminalistic (Santman et al., 1997;
Sutherland 8 Cressey, 1978). In addition, youths who own firearms often come from
families which support pro-gun beliefs and behaviors (Ash et al., 1996; Cao et al.,
1997; Wright et al., 1992). For these reasons, I intend to include the youth's family as
an independent variable to determine if is affects the adolescents' delinquent habits,
namely gun ownership.

There are several items measuring family delinquency and pro-gun attitudes.

The family gun habits scale includes four items, "have any of your siblings ever owned

19

a gun", ”have any of your siblings ever given you a gun", "do any of the males in your
family own guns" and "have you ever gone shooting with a relative". The items in the
scale are dummy coded (0 = no, 1 = yes). From the scale it is apparent that 37.56%
have siblings who own guns, 16.34% have been given a gun by a sibling, 74.84%
report that there is at least one male in their family who owns a gun, and 38.03% have
been shooting with a family member. The .5827 alpha level suggests that the scale
measuring family gun habits is moderately reliable. The sample size of 1065 is
somewhat lower than the 1461 participating in the study. The reason for the
discrepancy is that a number of respondents did not answer the questions (refer to

Table 2).

Peers

A great deal of research exploring all forms of juvenile delinquency has found
that youths are significantly affected by their peers (Akers et al., 1979; Inciardi et al.,
1993; Sutherland 8 Cressey, 1979; Winfree et al, 1994). In the adolescent years,
youths often strengthen peer bonds in an attempt to establish parental independence
(Inciardi et al., 1993). When the friends of juveniles hold attitudes favorable to law
violation, the juveniles often also hold these beliefs (Hallinan 8 Williams, 1990). The
scope of peer influence exceeds mere attitude acquisition as many researchers have
found that delinquent youth often have delinquent peers (Akers et al., 1979). Juveniles
often act in a delinquent manner in an attempt to gain peer approval (Emler 8 Reicher,
1995; Giordano et al, 1986).

Specific to gun violations, it is likely for juveniles to own guns when their peers
are actively involved with firearms (Inciardi et al., 1993; Lizotte et al., 1997). Decker et
al. (1997) questioned youths and found that a significant amount of them receive

respect from friends if they own a gun. These figures are more dramatic in the case of

20

gang members and inmates as a greater percent of these juveniles report peer
approval of gun ownership (Decker et al., 1997; Wright et al., 1992).

The first peer scale includes the following questionnaire items, "my friends
would look down on me if I did not carry a gun" and "in my crowd, if you don't have a
gun, people don't respect you". These two items are coded in a fashion that is similar
the Likert scale (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree) and
produce a .7713 reliability (alpha) level. Looking at Table 2, the percentage breakdown
for the first item is as follows, 62.7% disagreed strongly, 27.6% disagreed, 4.6%
agreed, and 5.0% agreed strongly. In terms of the second item, 58.7% disagreed
strongly, 29.7% disagreed, 5.8% agreed, and 5.7% agreed strongly. The difference
between the number of cases in the scale and the total number of respondents is due
to some of the juveniles failing to answer the item.

The second peer scale, peer gun behavior, includes "how many of the people
do you hang out with own guns" and "have you ever gone shooting with friends".
These items are scaled from 1 through 3, with 1 representing the lowest value and 3
the highest. Of the respondents, 23.1% don't have any friends who own guns, 26.7%
have some friends who own guns, and 41.6% have a lot of friends who own guns.
When asked if they had ever shot guns with friends, 36.3% said they had never, 6.2%
had gone shooting a few times, and 42.6% had been shooting with friends many times.
The peer gun behavior scale carries a .7417 alpha level. The number of valid cases is

smaller than the total number of respondents due to missing data.

Environment

Wilkinson 8 Fagan (1996) have found that the "code of the streets" has a
strong influence over children, adolescents, and young adults. An environment can
have a subculture of its own, complete with values, norms, and rules which may differ

from the rest of society (Miller, 1958). As a result, the environment can act as an

21

influencing agent, affecting what is Ieamed. Rodney Stark (1987) found that certain
neighborhoods continually have high rates of crime and deviance despite complete
residential turnover, suggesting that the environment actively influences its population.
Staples (1982) contends that in the ghetto the highest level of esteem and respect is
reserved for the best streetfighter in the neighborhood and that older males encourage
children to develop aggressive tendencies. Environments characterized by gang
presence result in many youths being informally socialized into the gang subculture at a
very eariy age (Shelden, Tracy, 8 Brown, 1997). Youths living in urban America have
been raised in a subculture which is characterized by guns, drugs, crime, and
victimization (Wright et al., 1992). Wilkinson 8 Fagan (1996) contend that youths must
adapt to this lifestyle and Ieam to act in accordance with the norms, values, and
realities present in the inner-city. The prevalence of firearms and the ease of
accessibility also affect juvenile gun ownership (Blumstein 8 Cork, 1996) Due to these
findings, the concept of environment should be operationalized and examined to
detemine what effect it has on juvenile delinquency.

In an attempt to measure the affect one's environment has on his or her
decision to act in a delinquent manner, several items will be entered into a scale. The
gun prevalence scale (see Table 2) measures the level of perceived and actual risk
present in the youths' neighborhoods. The first item measures the frequency the
respondents have been threatened with a gun in their neighborhoods. The responses
are coded from 1-3; with 1 = never; 2 = a few times, and 3 = many times. The
percentage of juveniles answering never is 36.0%, a few times is 47.0%, and many
times is 17.0%. The second and third scale items measure gun accessibility within the
juveniles' neighborhoods. The second item has three possible values, 1 = very hard to
obtain guns, 2 = moderately easy, 3 = very easy, and asks the youths how easy it is to
get guns in their neighborhoods. The answers are as follows, 20.7% think it's very

hard, 16.7% moderately easy, and 62.5% very easy. The third item, "there are lots of

22

guns in my neighborhood; it's easy to get guns", reports that 24.4% of the juveniles
believe the statement is false and that it is difficult to get guns, 19.3% agree with the
statement and believe it's fairly easy to get guns, and 56.4% of the juveniles feel
strongly about the statement and say it's very easy to get guns. The fourth scale item
is, "the importance of carrying a weapon for self-defense", and is coded from 1-3. An
answer of not important = 1, somewhat important = 2, and very important = 3. Of the
youths who answered the question, 16.5% answered 1, 24.9% answered 2, and 50.5%
answered 3. The missing cases for the four scale items equal 103, 249, 205, 203

respectively; this results in a sample size of 1047 valid cases.

Attitude

Differential association asserts that youth become delinquent when they have
an excess of definitions favorable to law violation over definitions unfavorable to law
violation (Sutherland, 1947). Youths' attitudes are influenced by those who carry
significant meaning in their lives, namely family, peers, and environment. Attitudes, in
turn, affect one's behaviors. A youth who has pro-delinquent attitudes often exhibits
delinquent behavior as a result.

A number of items regarding the juveniles' gun beliefs are entered into an
attitudinal scale to create a reliable measure. These survey items include, "feeling
better is an important reason as to why I own guns", "guns are fun, I like guns", "it is OK
to shoot a person if they have done something to hurt or insult you", "it is OK to shoot
somebody who doesn't belong in your neighborhood", and "it is OK to shoot a person if
that's what it takes to get something you want". Each of these six items are dummy
coded with 0 = not important and 1 = important for the first question (nfelbet) and 0 =
disagree and 1 = agree for the four other items. Looking at Table 2 it is evident that
59.62% of the juveniles feel better when having a gun, 40.76% think that guns are fun,

48.67% believe it's OK to shoot someone who hurts you, 20.57% say it's OK to shoot

23

someone who doesn't belong in your neighborhood. and 24.95% affirm that it's OK to
shoot someone if they have something you want. A reliability analysis shows that the
scale carries a .7381 alpha level which is fairly high. There are 1050 valid cases in this

scale.

Dependent Vam
Gun Ownership

In the United States, it is illegal for minors (persons under the age of 18) to own
firearms. There are several exceptions to this general rule and include provisions such
as hunting, having written parental consent, and military privilege. While the number of
inner-city juveniles who fall under these categories is probably rather small, an effort
will be made to exclude all possible cases. These exceptions will be discussed later.
This thesis will only include handguns (revolvers and pistols); rifles and shotguns will be
excluded. Research has found that handguns are the firearm most commonly used in
crime (BJS, 1995). Additional research has suggested that juveniles are attracted to
handguns, largely because of the ease of obtaining the gun (Center to Prevent
Handgun Violence, 1993). DeFronzo (1979) limited his study to handguns because
they are more likely to used for crime, protection, or harming someone than either
shotguns or rifles.

It is unclear from the data whether any of the gun owning youth have written
parental consent or military privilege; however it is unlikely that either condition is
satisfied. While youths may own firearms with expressed, documented parental
permission, the number of cases of inner-city adolescents who meet this requirement is
(probably minimal. Furthermore, the majority of respondents, 70.1%, are under the age
of 18, and therefore are not permitted to join the military. The juveniles who make up
the other 29.1% are either incarcerated or attending high school, this suggests that

they are not actively involved with the armed forces. Information concerning firearm

24

ownership for hunting purposes is provided, all persons who indicated they owned
handguns for the reason of hunting (n = 133) will be dropped from the study; this
decision is supprted by past research which also chose to eliminate hunters from
analysis (Hill, Howell, 8 Driver, 1985). Because the number of cases dropped is small,
8%, the sample size is not affected as it changes from 1594 male respondents to 1461
male respondents.

The variable "gun ownership" is derived from the responses given to questions
asking if the youth "has ever owned" either a pistol or a revolver. The answers to
these items have been collapsed into one measure (ohandgun) which is dummy coded
where 0 = no, I have never owned a handgun and 1 = yes, I have owned a handgun. A
frequency of the variable reveals an almost equal distribution with 56.0% of the

respondents indicating that they have owned a handgun at some time.

Control Variables

When examining the effect family, peers, and the environment have on youths'
gun attitudes and gun ownership, it is necessary to control for rival influencing
variables. This is done to prevent the emergence of spurious relationships and
inaccurate results. - The following variables will be controlled for so they will not
influence the relationships between the independent variables, gun attitudes, and gun
ownership. In this paper, females will be excluded from analysis largely because of the
absence of female data. Ideally, both sexes would be included so that results may
generalized to all youths and not just males However, this does not pose too large of a
problem as many delinquency researchers elect to focus solely on male subjects.
Excluding females is a common occurrence because males are considerably more
criminalistic than females at all ages and with regard to nearly every type of criminal
offense (Anderson, 1994; Emler 8 Reicher, 1995; Steffensmeir 8 Allan, 1991;

Sutheriand 8 Cressey, 1978). Specific to this particular form of delinquency (juvenile

25

gun ownership) males are more likely to be involved with firearms (Anderson, 1994:
Bjerregaard 8 Lizotte, 1995; Cao et al., 1997; Shelley, Brody, Wright, 8 Williams 1994;
T. Smith 8 R. Smith, 1995).

Age

The age of the juveniles in this study will be controlled to determine which age
group is more likely to hold pro-gun attitudes and exhibit pro-gun behavior. Older
youths have been socialized longer and therefore, may be more likely to be affected by
influencing agents. These data include youths as young as 11 years of age and as old
as 24 years of age. The majority (96.9%) fall within the range of 14 to 19 years of age,

with a mean of 16.73 years and a median and mode of 17 years.

Race

Another variable which will be controlled for is the respondents' race or ethnicity.
The inner-city, which is characterized by a high rate of criminal incidence, is
predominantly made up of minority residents (Cuny et al., 1994). As a result, lower-
income black children are exposed to violence at very eariy ages (Staples, 1982).
Thus, the environment and race of the respondents may interact to affect juveniles'
attitudes and behaviors concerning firearms. Research shows that blacks are more
likely to own guns than whites (Ash et al., 1996; Bjerregaard 8 Lizotte, 1995). It is for
all of these reasons that race must be controlled for when examining the effects of
family, peers, and the environment on gun attitudes and ownership.

Research has suggested that minority youth may experience different family
and peer socializations than non-minority youth (Mason et al., 1994). This may be
related to where the family lives since urban settings are usually characterized by high
crime rates, high victimization rates, and as a result, the family must adapt to inner-city

life in order to survive (Anderson, 1994; Cuny et al., 1994). Minority youth may also be

26

more highly influenced by peers than non-minority youth (Mason et al., 1994; Watts 8
Jagers, 1997). Juvenile gun ownership research has found that black youths are more
likely to own guns than white youths (Ash et al., 1996; Bjerregaard 8 Lizotte, 1995). In
addition, black males are over-represented in the criminal justice system (McGarrell,
1993; Sutherland 8 Cresey, 1978).

In order to account for these variations, race will be controlled so that the
independent variables and their effect on gun ownership may be accurately measured.
Much of the research on this topic examines only black and white gun owning habits,
by controlling for race, we can see how prevalent gun ownership is amongst other
racial groups. The racial breakdown of these data are the following, 7.7% of the

sample is White, 60.3% Black, 24.2% Hispanic, and 8.0% Other.

Gang Measures

A third controlling variable is gang membership. Many varying definitions of a
"gang" have been offered over the years, and still no consensus has been reached
(Knox, 1991; Oehme, 1997). Knox (1991) contends that a group qualifies as a gang
when it is involved in criminal or delinquent activity which benefits the members. This
researcher, and others, note that the benefits may take many forms and are
individualistic to each member (Jankowski, 1991; Knox, 1991; Shelden et al., 1997).
Gangs are often present in urban environments characterized by a low socio-economic
standing, high unemployment, and a large minority population (Bjerregaard 8 Lizotte,
1995; Curry et al., 1994; Santman et al., 1997).

Each of the four states included in this study, California, Illinois, New Jersey,
and Louisiana have cities which report having a gang problem (Curry et al., 1994).
Gangs offer youth acceptance, a sense of belonging, and a place of protection
(Lawrence, 1998). Consequently, the gang has considerable influence over its

members, more so than the influence of other highly delinquent non-gang peers

27

(Howell, 1997). Youths claiming gang membership are more criminal than non-gang
youth (Esbensen 8 Huizinga, 1993; Howell, 1997; Sheley et al., 1995). Also, gang
members are more likely to be involved with guns than are non-gang members
(Bjerregaard 8 Lizotte, 1995; Fagan, 1990; Lizotte et al., 1997; Sheley et al., 1995).

In this thesis, two gang measures will be included as control variables. The first
measure, "are you a gang member", includes 50.3% of the juveniles responding "yes".
The second measure will be used to determine the strength of pro-gun influences
gangs have over their members (see Table 2). The following binary variables comprise
the second measure and have been used to create a scale, "most of the people in my
gang carry guns", "there are always lots of guns around whenever the gang gets
together", "you have to have a gun to join my gang or group", and "you have to show
you can use a gun to be in my gang". The percent of juveniles who answer
affirmatively to these measures are 70.57%, 67.43%. 9.43%, and 17.43% respectively.
The .6086 alpha level suggests that this scale moderately measures the influence
gangs have on juveniles' decisions to own guns. A sample size of 700 may be
problematic; however, it coincides with the number of juveniles in the study claiming to

be gang members.

Status

The final factor which must be controlled for to prevent reporting misleading
results is whether the respondent is currently incarcerated or a student currently
attending high school. There are two reasons why this must be taken into account.
First, inmates are known to have been delinquent; therefore, they have a higher rate of
criminality than non-delinquents (Inciardi et al., 1993). Second, research has shown
that inmates are more likely to own and/or cany firearms than are non-inmates (Ash et

al., 1996; Decker et al., 1997; Smith, 1996; Wright et al., 1992).

28

Much of the research on juvenile delinquency suggests that delinquent youth.
more often than non-delinquent youth, have friends and family members who are
criminal (Sheley et al., 1995). With regard to firearm habits, inmates are more likely to
have a higher degree of gun contact than are non-inmates (Ash et al., 1996; Decker et
al., 1997; Wright et al., 1992). This thesis will control for the status of the respondent
(student verse inmate). Within this sample, 49.1% of the juveniles are inner-city high
school students and 50.9% of the juveniles are inmates being held in detention facilities

relative to the high schools' locations (see Table 1).

Methods

First, bivariate analysis will be conducted to examine the influence each of the
independent variables has on the dependent variable using analysis of variance.
Second, multivariate analyses will be implemented to look at the effect the independent
variables have on pro-gun attitudes and pro-gun behavior (ownership), while controlling
for possible competing factors. Because the dependent variable is dichotomous, it has
only two possible values, the ordinary least squares regression model is inappropriate
(Bachman 8 Paternoster, 1997). The dependent variable, juvenile gun ownership,
must be. dummy coded with 0 = "no" and 1 = "yes" so that the logistic regression model
can be used for multivariate analyses (Bachman 8 Paternoster, 1997). Logistic
regression uses a nonlinear model called the maximum-likelihood method to estimate
the data parameters. From this model, an equation is obtained in order to estimate the
probability of the dependent variable occurring based on the independent variables

(Bachman 8 Patemoster, 1997).

29

Chapter 4 -- Results

Bivariate Analysis

This study uses a dichotomous dependent variable, juvenile gun ownership, and
a number of both categorical and continuous independent and control variables.
Analysis of variance was used to test whether or not the group means of the variables
are equal to zero. If the reported significance level is below .05, then the dependent
variable and the independent variable are significantly related to one another. In the
cases where there are more than two means, a post hoc Bonferroni test was used to
determine which means were significantly different from the others. The findings of

these analyses are presented in Table 3.

Table 3. Bivariate analysis (ANOVA)
The effects of the independent app control variables on juvenile handgun ownership

 

 

 

Variable N Mean Significance
Age .000
1 1-14 74 3243

15-17 934 .5482

18-older 431 .6241

Missing 22

Race .134
White 1 1 1 .6486

Non-white 1 336 5524

Missing 14

Student Status .000
Inmate 743 .7725

Student 718 .3398

Missing 0

G_ang Membethip .000
Gang Member 701 .7660

Non-gang Member 692 .3497

Missing 48

30

 

 

 

 

Variable N Mean Significflce
ang Influence .000
No gang influence 153 .4052

Gang influence 547 .8464

Missing 761

Family Gun Habits .000
0 - low 201 .2090

1 297 .4377

2 310 .6258

3 169 .8166

4 - high 88 .9432

Missing 396

Peter Gun Approvg .001
2 - low 598 .5033

3 218 .5872

4 230 .6522

5 54 .6296

6 38 .6316

7 25 .7200

8 - high 34 .6471

Missing 264

Peer Gun Behavior .000
2 - low 267 .1199

3 189 .2857

4 157 .5287

5 160 .7375

6 - high 415 .9133

Missing 273

Gun Environmgpt .000
1 - low 56 .1250

2 - moderate 245 .4041

3 - high 513 .8207

Missing 647

Gun Attitudes .000
0 - low 256 .2539

1 233 .4335

2 - high 561 .7986

Missing 411

31

Beginning with two of the control variables, age and race, one can see that only
the former is statistically related to handgun ownership. In this analysis, a juvenile's
race does not affect his decision to own a handgun. Age has been coded into three
categories; the first category includes respondents aged 11-14, the second includes
1517 year olds, and the third is made up of individuals over the age of 18 years. As
seen in Table 3, the older the respondent, the more likely he is to own a handgun. The
Bonferroni post hoc test confirms the initial findings and further clarifies the relationship.
The differences in gun ownership between age groups are statistically significant from
one another.

Bivariate analysis of the third control variable, student status, reveals that
inmates are more likely to own handguns than students (77% compared to 34%). In
this sample, being an inmate more than doubles the likelihood of handgun ownership.
Being a member of a youth gang significantly increases the probability of owning a gun,
as 76% of members own a handgun while only 35% of non-members own a handgun.
The relationship between gang membership and handgun ownership is statistically
significant. The influence exerted by a gang over its members to own a handgun is
also important; juveniles who experience gang influence are more than twice as likely
to own handguns than those who are not subjected to gang influence.

Family gun habits is measured using a scale which ranges from zero to four,
where zero represents an absence of gun socialization within the home, and four
represents a high level of pro-gun socialization within the home. There is a definite,
positive relationship between family gun habits and juvenile handgun ownership.
Specifically, the greater the level of pro-gun socialization present in the home (as
defined by sibling and relative gun possession, whether siblings had ever given the
respondent a gun, and frequency of shooting guns with family members), the more
likely juveniles will own handguns. When gun exposure is present in the home,

juveniles tend to follow that example.

32

Items that measure the importance of peer approval in a juvenile's decision to
own a handgun were combined in a scale. Respondents' scores vary from two through
eight, where a value of two indicates that peer approval is not important to the juvenile,
and a value of eight means that peer approval is very important to the juvenile with
regard to handgun ownership. While an ANOVA produces a significance level of .001,
the Bonferroni test indicates that only two groups are significantly different from one
another (group 2 and group 4). The importance of this variable should be questioned,
since there is no obvious pattern to the relationship.

While the effects of peer approval on gun ownership are inconclusive, peer
behavior and the effect it has on juvenile handgun ownership is very clear. This scale
is made up of items describing gun ownership among the juveniles' close friends, and
the frequency with which juveniles and their peers have gone shooting together. There
is a steady positive relationship, as the level of peer gun behavior increases, so does
juvenile handgun ownership. The relationship is statistically significant; juveniles who
have few friends owning guns are less likely themselves to own handguns than
juveniles who have many friends who own guns (12% compared to 91%). The
difference is striking, and the Bonferroni test reveals that the categories are all
significantly different from each other.

The gun prevalence scale was created using measures of the level of violence
and gun accessibility in the juvenile's neighborhood. The possible values of this scale
range from one through three, where one signifies a low level of gun presence in the
immediate environment, 2 represents a moderate level of gun presence, and 3 shows a
high level. As the level of firearm prevalence and accessibility increases, so does the
frequency of juvenile handgun ownership. There is a significant difference in handgun
ownership among youths who live in relatively gun-free neighborhoods versus youths
who live in an environment characterized by a moderate level of gun prevalence and

youths who live in highly volatile areas with a pronounced firearm subculture. Looking

33

at Table 3, the mean level of juvenile handgun ownership for respondents living in a
neighborhood characterized by a low level of firearm presence is .12, compared to .40
in a moderate gun environment, and .82 in a atmospheres containing a high level of
firearm prevalence. In other words, the neighborhood in which the respondent lives
directly impacts the likelihood of him owning a handgun.

Finally, the relationship between juveniles' pro-gun attitudes and gun ownership
was examined. Adolescents who do not harbor many attitudes favoring guns have a
significantly lower rate of handgun ownership than youths who have attitudes
moderately or highly favorable to firearms. Once again, there is a positive relationship
present, and the Bonferroni analysis indicates that each category of pro-gun attitudes

(high, moderate, low) is significantly different from the others.

34

Multivariate Analysis

This thesis uses a dichotomous dependent variable, thus linear regression is not
an option when conducting multivariate analysis. When the dependent variable can
have only two values, the premise supporting ordinary least—squares regression will not
hold true because it is not possible to assume that the distribution of errors is normal.
Instead, logistic regression will be used to estimate the probability that an even occurs.
This model requires fewer assumptions than linear regression and still provides
accurate, reliable data. Because this type of regression analysis is not linear by nature,
the parameters of the model are set using the maximum-likelihood method. The log-
likelihood provides information as to how well this model fits the data; in other words,
do the independent variables adequately measure the dependent variable.

In accordance with logistic regression, the odds ratio coefficient is included to
show the probability of an event occurring. The odds ratio coefficient reveals the
increase or decrease in the odds of the dependent variable occurring with a one unit
change in the independent variable. Thus, if the number in the odds column is greater
than one, the independent variable increases the odds of juvenile handgun ownership.
Conversely, if the number is less than one, the independent variable decreases the
odds of juvenile handgun ownership. Also, if the odds column equals zero, the
independent variable does not affect the dependent variable. The odds ratio was
chosen over the log odds coefficient because of personal preference; both accurately
describe the data. The Nagelkerke R Square statistic signifies how much of the
variation in the dependent variable, juvenile gun ownership, is explained by the
independent variables. Throughout this section the odds coefficient will be used to
describe the relationship between the independent and dependent variables. This

statistic was chosen because it is easier to interpret than the log odds statistic.

35

 

Soyuz .38.

.924

mend mmvd mvvd 3.5 8nd nomd Swd mm oxaméommz

1 Sm; : wand 35m: cam >=Emm

S: t 33 3.555 26.05

. 9.: z 8.... “55:93.8 :30

t on: 2. 8nd 3323 can Loon.

o8... . R: .2653 cam been.

8: i 83 ... Emu : $3 .. :5 t 8..” z 83 E 23:56 260

2. and : vtd 2. N86 2. mmwd : 436 2. mad 2. nmmd E Emuaw

. coed : owed coho 260 856 21.6 30.0 A: QE>>

v 3.? mvoé omoé mead Boé awe... one; oo<
amnuz mmopuz topaz opowuz nmiuz nmSuz owmwuz

Hands Hands “.3qu Hands “mugs «duds flange «aqua»

 

aEmLmEso Sauce: 0.7.93 cc 3339 E03335 new .0550 05 .0 Score of. E23053. 9660.. .v mink

36

In Table 4, a number of models are presented. There are several reasons for
including these models, the most influential being the large number of missing cases
found throughout the data. When looked at individually, the separate independent
variables are missing for a small percentage of cases; however, once all of the relevant
variables are included in the model, there is a large number of missing cases. In order
to safeguard against unrepresentative findings as best as possible, the results are
offered both individually and as a whole so that the reader may be alerted to the
phenomenon of missing data. It is important to note that the independent variable
"gang influence" has not been included in multivariate analysis. The reason behind this
decision is that this variable had a high number of missing cases (761), thus it
interfered with the results produced in logistic regression.

The first model contains only the four control variables, age, race, student
status, and gang membership. Model 1 has the highest possible sample size present
during multivariate analysis, 1360 respondents out of a possible 1461 respondents. In
addition, this model highlights the effect of the control variables on juvenile handgun
ownership and serves as a basis for comparison as other variables are entered into the
equation. In short, Model 1 reveals that age and race are not significantly related to
juvenile handgun ownership. An odds coefficient of 3.486 was produced for gang
membership; thus, being a gang member significantly increases the odds of a juvenile
owning a handgun. In addition, students, as opposed to incarcerated juveniles, are
significantly less likely to own a handgun (.223). Both of the significant variables, gang
membership and student status produce p values less than .001. The Nagelkerke
coefficient is the preferred R Square statistic; in this model is reveals that .241 (or
24.1%) of the variation in juvenile handgun ownership is explained by the youth's age,

race, student status, and gang membership.

37

Models 25 produce similar results and will be discussed below. Each of these
models contain the four control variables mentioned above, and each model contains
one of the independent variables. This was done to examine the effect of that
particular independent variable on the dependent variable, juvenile handgun
ownership. In each of these analyses, the general effects of the various independent
variables were the same on the control variables, with the strength of the odds varying
slightly.

Model 2, which contains peer approval of gun ownership as the independent
variable (N = 1157). The overall findings are similar to Model 1 with age and race not
related to handgun ownership. Once again, gang membership significantly increases
the odds ratio of juvenile handgun ownership (3.180), and students are significantly
less likely than inmates to own handguns (.185). The independent variable, peer
approval of gun ownership, reveals that as gun approval rises, so does the likelihood of
juvenile handgun ownership. The relationship is significant (p < .05) with an odds
coefficient of 1.127. The Nagelkerke R2 for Model 2 (.263) is similar to Model 1 (.241),
thus the inclusion of peer gun approval by itself does not explain much more variance
in juvenile handgun ownership.

The third model includes peer behavior in order to examine the effect peer gun
ownership has on a juvenile's decision to own a handgun. Peer behavior is a measure
of two items: how many of the juveniles' friends own guns, and the frequency that the
juvenile had gone shooting guns with his friends (see Table 4). The original effects the
control variables have on juvenile handgun ownership are the same as those described
in Model 1. The odds coefficients for age, race, student status, and gang membership
are 1.007, .769, .414, 1.711 respectively. The major difference between Model 3 and
Models 1 and 2 is that the strength of the effect of gang membership on juvenile
handgun ownership is reduced. Whereas the first two models produced odds

coefficients of 3.486 and 3.180, the third model only produces an odd coefficient of

38

1.711. Although it remains significant, the strength has decreased noticeably. From
this analysis, one can see that having friends who actively participate in the gun
subculture greatly increases the likelihood of a juvenile exhibiting the same behavior.
Juveniles who have a large number of peers owning and firing guns are significantly
(2.360) more likely to own a handgun. In this model, the Nagelkerke R2 increases as
39.6% of the variation in juvenile handgun ownership is explained by age, race, student
status, gang membership, and peer gun behavior. This figure is notably larger than the
Rz's produced in Models 1 and 2.

Model 4 (N = 1019) examines the influence a juvenile's environment has on his
decision to own a handgun. The environment variable was created using measures of
the prevalence and ease of accessibility of firearms, the need for guns to provide
protection, and the frequency of gun victimization and/or threat within the juveniles'
neighborhoods. This drop in strength is not significant and therefore it may be ignored.
Being a student significantly decreases the odds of juvenile handgun ownership (.229),
while being a gang member significantly increases the odds of juvenile handgun
ownership (2.257). These results are similar to those reported above. The
independent variable, gun prevalence in the juveniles' environment, significantly
increases the odds of juvenile handgun ownership (1.489). Thus, juveniles who live in
areas characterized by high levels of firearm activity are dramatically more likely to own
handguns than juveniles living in relatively gun-free atmospheres. Models 4 and 5
have similar Nagelkerke findings; 44.7% and 44.5% of the variation found in juvenile
handgun ownership is explained by the independent variables found in these
respective models.

The fifth model produces much of the same odds coefficients as the previous
models with age (1.05) and race (.706) not affecting likelihood of gun ownership.
Similariy, being a student significantly decreases the probability of gun ownership

(.232) while gang membership significantly increases the probability of gun ownership

39

(2.261). The attitudes a juvenile has towards guns significantly effects the odds of that
juvenile owning a handgun. Adolescents who feel that guns are fun, who indicate that
having a gun makes them feel better, and who are able to justify the use of guns
against others are significantly more likely to own a handgun (1.659) than juveniles who
express reserve and uncertainty about firearms.

The last independent variable, family gun habits, is included in Model 6. A
sample size of 1036 respondents is slightly higher than the sample sizes of Models 4
and 5 (N = 1019, 1017 respectively). In this analysis, the results for three out of the
four control variables mimic those described above: age is not related to juvenile
handgun ownership, while inmate status and gang membership both increase the
likelihood of ownership. A change in the relationship between race and owing guns
occurs when family gun habits are entered into the regression model. Controlling for
the gun subculture found in the home, white juveniles are significantly less likely to own
handguns than non-white juveniles (.420). Model 6 produces a slightly higher
Nagelkerke R2 statistic as 48.5% of the variation found in juvenile handgun ownership
is explained by the independent variables.

Model 6 is the first model where race becomes a significant factor affecting
juvenile handgun ownership. Further analysis in the form of an independent samples
t-test (which is not presented in this paper) reveals that white youth are significantly
more likely than black youth to come from families which support the use of firearms.
Further explanation of this phenomenon will be offered in Chapter 5. The independent
variable, family gun habits, significantly affects the odds ratio of a juvenile owning a
handgun (2.329). Juveniles who are raised in homes where family members own guns,
have given them a gun, and/or have taken them shooting are socialized to accept guns;
therefore, they are more likely to own a handgun.

Model 7 contains all four of the control variables and each of the five

independent variables. When these nine variables are combined into one logistic

40

regression model, the sample size drops to a mere 799 respondents. Steps were taken
to try and increase this sample size, including or excluding certain variables. altering
scales, and substituting items, but nothing resulted in a higher number of cases. Again,
age is not related to juvenile handgun ownership and students are significantly less
likely than inmates to own handguns (.316); these two relationships have been
constant throughout the models. This model results in several changes with regard to
two of the control variables; race plays a significant role in terms of juvenile handgun
ownership and gang membership loses its significance. In this model. the odds of
white juveniles owning handguns are significantly lower than the odds of non-white
juveniles owning handguns (.460). In the six previous models. gang membership was
highly related to juvenile handgun ownership. In Model 7, this relationship disappears
(1.45). Additional analysis was conducted to examine this phenomenon; an
independent samples t-test revealed that gang membership is significant across each
individual category of the five independent variables. The reason why this relationship
disappears when they are all combined can only be that these five variables (peer gun
approval, peer gun behavior, gun prevalence in the neighborhood, gun attitudes, and
family gun habits) explain why juveniles join a gang. This will be further explored in the
next chapter.

Three independent variables, peer gun behavior, gun prevalence in the
neighborhood, and family gun habits all significantly increase the odds of a juvenile
owning a handgun (1.738. 1.140, 1.557 respectively). Thus, juveniles who have a large
number of friends active in the gun subculture. are surrounded by firearms in their
immediate environment, and who are socialized in families where pro-gun beliefs are
present are much more likely to own a handgun. While pro-gun attitudes increase the
odds of juvenile handgun ownership, this relationship is not significant (1.157). The
only independent variable which does not increase the odds of the dependent variable

occurring is peer gun approval (.940). In Model 7. 56.9% of the variation found in

41

juvenile handgun ownership is explained by the juvenile's age, race, student status,
gang membership, peer gun approval, peer gun behavior, gun environment, pro-gun
attitudes, and family gun habits (as reported by the Nagelkerke R2). This figure is

notable higher than that produced by any of the other six models.

42

Chapter 5 -. Conclpsion

Social learning theory is based on the idea that people Ieam deviant attitudes
and behaviors the same way they Ieam conforming attitudes and behaviors.
Delinquency theories which follow this basic premise stress the similarities found
between delinquents and non-delinquents; namely, both groups share Ieaming
capabilities and internalization patterns (Hansell 8 Wiatrowski, 1981). Learning occurs
within intimate groups made up of significant others (Inciardi et al., 1993; Sutherland,
1947). Thus, delinquent behavior is learned through the individual's interaction with
influential groups that have embraced non-conforming, criminal values. The definitions
learned and the meanings attached to certain acts are the consequence of where
Ieaming occurred. Individuals acquire definitions which label certain attitudes and
behaviors as being acceptable or unacceptable. It is these definitions which will
determine a person's actions.

Differential association has been called the single most important explanatory
concept of social Ieaming theory (Akers et al., 1979; Sellers 8 Winfree, 1990). The
crux of Sutheriand's differential association theory is that a person becomes delinquent
because of an excess of definitions favorable to law violation over definitions
unfavorable to law violation (Sutheriand, 1947). While this extension of social Ieaming
theory preposes that peer relationships have the greatest impact on a youth, it is
important not to ignore the effects other significant groups have on a child's
development. Because differential associations depend on the frequency, duration,
priority, and intensity of a relationship, other influencing agents must be included in
analysis (Sutherland, 1947). Besides peers, the family and the environment both serve

as powerful agents which shape youths' definitions, attitudes, and behaviors.

43

Data Findings

This thesis examines the effects which a juvenile's peers, family, and
environment have on his decision to own a handgun. In an attempt to safeguard
against reporting misleading findings, certain characteristics were controlled for such as
the youths' age, status (whether he was a student or an inmate), gang membership.
and race. Both bivariate and multivariate analyses were conducted. Analysis of
variance was the method of bivariate statistics used; this procedure resulted in every
variable, both control and independent, being significant except the youth's race. In
order to further elaborate upon these preliminary findings, logistic regression was used
to examine the effect one independent variable has on the dependent variable while
controlling for all other competing variables. Table 4 highlights the relationships found
in this analysis. Model 1 (which consists of the four control variables) serves as a base
model which each of the other, more complicated models may be compared to in order
to determine the changes the independent variables produce on the odds of a juvenile
owning a handgun. Here, neither age nor race are significant; however, both student
status and gang membership are significant. Only age and student status will maintain
the original findings produced in Model 1, gang membership and race will undergo
changes once other variables are introduced.

A surprising effect produced in logistic regression was found when examining
the control variable "gang membership". The association of a juvenile with a youth
gang and his subsequent involvement with delinquency has been found to be important
in a great deal of research (Bjerregaard 8 Lizotte, 1995; Monti, 1993; Esbensen 8
Huizinga, 1993; Jensen 8 Rojek, 1992; Sheley et al., 1995). Models 1-6 in Table 4
found similar results; however, Models 7 and 8 found conflicting results. In the first six
regression models, gang membership was consistently found to be significant below
the .001 p-Ievel. In each of these models (beside the first which only looked at the

control variables) only one independent variable was introduced into analysis. Model 7

consists of regression analysis which includes all five independent variables, family gun
habits, peer gun approval. peer gun behavior, gun prevalence in the environment, and
gun attitudes. When these independent variables are all included, gang membership is
no longer significant. Perhaps the reason for this occurrence is that these five
variables explain gang membership; in other words. it is these characteristics which
determine whether or not a juvenile will join a gang. This analysis does not prove this
statement, it only points to it as a possibility. Nevertheless, this finding is important as it
suggests that gang membership should be included in future research.

Another unexpected finding produced in Table 4 is the apparent relationship
between a juvenile's race and his family's gun habits. Throughout Table 4, race is not
significant except when family gun habits is included in analysis. An independent
samples t-test was conducted to further elaborate upon this finding. This analysis
showed that white males are more likely to come from a family which is supportive of a
gun subculture. Although this finding is not significant (X2 = .05), it is important to note.
In this study, including family gun habits in analysis results in race becoming significant
as white males are dramatically less likely to own a handgun than black males. This
finding is surprising, and l have no way to explain this phenomenon.

Several independent variables were used in this thesis, these measures were
constructed for the purpose of measuring the three main significant groups which affect
a juvenile's socialization; family, peers, and the environment. The concept of family
gun habits measures the extent to which a juvenile's family is supportive of the gun
subculture. When controlling for other factors, this variable has a significant effect on
the odds of a juvenile owning a handgun. This finding is supported by research which
has found similar results (Caetano, 1979; Lizotte et al., 1994; Santman et al., 1997).
Thus, juveniles raised in pro-gun households are more likely to own a handgun than

juveniles not raised in this type of home.

45

Two measures of peer influence were used in this paper; peer approval of gun
ownership and actual peer gun ownership. The variable labeled peer approval was
found to have virtually no effect on the odds of a juvenile owning a handgun. This
suggests that juveniles who own handguns are not swayed by their friends' outward
approval of this behavior. Conversely, peer behavior (in the form of owning guns)
significantly affects the likelihood of a juvenile owning a handgun. Youths who have a
large number of friends who own guns are dramatically more likely to own a handgun
than youths who do not have a large number of gun-owning friends. This relationship
is prevalent even when controlling for other rival factors such as family, gang, and
environmental influences. Just as other researchers have noted, juveniles are likely to
own a gun if their peers own a gun (Blumstein 8 Cork, 1996; Sheley 8 Wright, 1665;
Smith, 1996).

Researchers have suggested that a person's environment shapes his life
experiences (Osofsky, 1995: Wilkinson 8 Fagan, 1996). These same authors note that
inner-city environments are often explosive, unstable, and characterized by a high
crime rate. The effects which an environment has on a person may be long-term and
significant, such is the findings of this research. Juveniles who live in neighborhoods
marked by a high prevalence of firearms and victimization are significantly more likely
to own a handgun than juveniles who do not live in this type of neighborhood. A
youth's environment is often overlooked in research, this mistake should not be
continued. The neighborhood which a juvenile lives in may have its own set of rules,
norms, and values. This phenomenon is often referred to as the "code of the streets"
(Wilkinson 8 Fagan, 1996). It is important for researchers to recognize, measure, and
analyze the effect the environment has on a person's criminal behavior.

Also included in this paper was the independent variable measuring the
attitudes which juveniles have towards firearms. Attitudes often precede behaviors,

thus it is assumed that juveniles who harbor pro-gun attitudes are likely to own guns.

46

Although the existence of pro-gun attitudes increases the odds of a juvenile owning a
handgun in this study, this relationship is not significant. This finding is intriguing and
should be further examined in future research efforts.

The overall findings of this thesis are that juveniles who are exposed to
socializing agents which encourage gun ownership are likely to own handguns. In
short, juveniles who are raised by families, with peers, and in environments which favor
firearms are significantly more likely to own a handgun. These results shed light on
why American juveniles are participating in the gun subculture, and may be used to

address this trend.

Data Limitations

This study encountered a number of obstacles, most notable is the issue of
missing cases. However, other limitations included the normal hindrances associated
with secondary data analysis. First, the original researchers neglected to find a sample
of incarcerated female juveniles thus comparisons between female students and
female inmates were impossible. This is often the case with criminal justice research.
Because males are more likely to be involved with illicit activities than females, male
subjects are often the focus of research while females, and their criminal involvement,
are often ignored. Another concern with secondary data is the differing purposes for
which it was collected. The original researchers had different hypotheses in mind, thus
the questionnaires contained items relevant to their research problems. Because of
this, a number of questions which would have complemented this study were not
included due to shortcomings in the data.

Beyond these relatively minor considerations is the issue of missing data.
Because the study was restricted to male juveniles, 904 females cases were
automatically deleted which resulted in a sample size of 1594 male respondents. This

sample was decreased by 133 cases, for reasons which will be explained below.

47

leaving a sample size of 1461 participants. Students and inmates were handed
questionnaires to be completed by themselves. The questionnaires varied in length
from 73 items for students and 84 items for inmates, with many of the items having
multiple parts to answer. For whatever reason, some items were left largely
unanswered. As a result, several of the questions which made up scale items had a
large number of missing cases.

The sample sizes for both the bivariate and multivariate analyses conducted in
this thesis were always recorded (see Table 3 and Table 4) so that the reader may be
aware of the number of cases being included in analysis. The problem is more
apparent in logistic regression and is the reason behind presenting eight models. The
first model, containing only the four control variables, has a strong sample size of 1360
cases. This number declines as more variables are entered into analysis. Typically,
when only one independent variable is included with the control variables, the sample
size averages 1076 cases (Models 26). However, when all of the control and
independent variables are entered into regression, the sample size plunges to 799
cases. Due to the dilemma of missing cases, one of the original independent variables
could not be included in analysis. Gang influence was excluded from regression
because of the enormously high number (761) of missing cases.

The issue of defining juvenile handgun ownership as a delinquent act is another
concern of this study. Persons under the age of 18 are not allowed to own handguns in
the United States without having written parental consent, except for the purposes of
hunting or military training. While it is largely assumed that the majority of inner-city
males participating in this study do not have written parental consent authorizing them
to own handguns, it is not positively known as the original researchers made no attempt
to gather this information. In addition, because 50.9% of the sample is incarcerated, it
is safe to assume that they do not have military privileges which permit gun ownership.

While it is possible that some of the remaining respondents have the ability to legally

48

own a handgun because of military permission. it is not likely that this number is large
enough to affect the research findings.

This study was limited to handgun owners in order to exclude all juveniles who
indicated they owned a gun for hunting purposes (this is why the 133 cases mentioned
above were excluded). It was the intention of this thesis to differentiate between
minors (juveniles below the age 18) and adults; however, missing cases did not allow
this to occur. Of the 1461 respondents, 431 of the males are eighteen years of age or
older. When these cases were deleted from analysis, the dilemma of missing data was
greatly exaggerated. As a result, some of the individuals included in this thesis may
legally own a handgun; however, it is the presumption of this paper that the number of
these persons is minute. In the end, while it is not 100% accurate to label all of these
handguns owners as delinquent or criminal, the great majority of them deserve this
characterization.

The final limitation presented by these data has to do with the inability to
address every aspect of social Ieaming theory. The final component of this theoretical
paradigm includes the concept of differential reinforcement. Whereas differential
association explains why a juvenile chooses to own a handgun, differential
reinforcement explains juvenile gun ownership which exists beyond the initial behavior.
Thus, differential reinforcement deals with behavior maintenance. Because these data
are only a snapshot of the juveniles' lives, there is no way to measure behavior
maintenance or behavior modification. Research efforts which include data over time
are beneficial in that juveniles' behavior may be observed for change, thus specific

efforts to treat youths in different situations may be made.

49

Policy Implications

The findings presented in this thesis may be quickly summarized; a youth's
family, peers. and environment determine whether or not he will own a handgun. This
information carries great importance, and the ramifications of its use may be profound.
In light of this evidence, policymakers have a three tier approach in combating juvenile
gun ownership. First, families need to be educated on what it means to have a firearm
in the house. By simply keeping a gun in the home, families are supporting the gun
subculture by indirectly approving of firearms. Children of these families are Ieaming
by example and, as the research suggests, are likely to mimic this behavior. If a family
chooses to keep a gun in the home, parents/guardians should be encouraged to teach
their children about the dangers associated with firearms. In addition, the gun should
be kept unloaded and hidden from the children. Whether or not a family owns a gun,
they should take steps to address the topic of guns with family members because it is
likely that their sons and daughters will encounter them at some point. The family is a
very strong teaching source and may be used to stress the harm linked to guns.

A second approach to curbing juvenile gun ownership may focus on a youth's
friends. While the family can broach this topic. the school is an ideal forum to tackle
this issue. Schools should teach children about firearms at a very early age because
the majority of inner-city children are faced with the dilemma of gun participation very
early in life. Also, by teaching young children about guns, the schools can include
some youths which otherwise would be absent such as dropouts and incarcerated
juveniles. By educating children about guns in the classroom, society could be sure
that the topic is being addressed, teachers could be available for questions, and
complicating matters such as peer pressure, gang membership, crime, and the law
could all be included when explaining the consequences associated with firearm use.
Adolescents are very susceptible to what their peers are doing, the only way to actively

address this is through institutions which house their friends. In addition to schools.

50

efforts made by authority figures in churchs, community centers, and even detention
facilities may encourage youths to avoid firearms.

The third tier, the environment, may be attacked quite actively. Efforts on this
front would include passing legislation making it increasingly more difficult to obtain
firearms, increasing law enforcement efforts to confiscate guns by‘targeting illegal
suppliers, and "cleaning" up inner-city neighborhoods in an attempt to rid the
environment of illicit gun dealers, drug houses, and gang hangouts. By simply getting
guns off the streets, children are less likely to acquire, carry, and fire guns thereby
making the neighborhood a safer place. Beyond this direct assault on the gun
subculture, another approach may be taken where communities become actively
involved in encouraging definitions, values, and norms contrary to those typically found
in an environment characterized by a high gun prevalence. By taking a community
approach, residents of inner-city neighborhoods can stress the dangers of guns and
encourage youth to stay away from guns and those who use them.

If these three socializing groups were to be increasingly included in society's
efforts to erase juvenile gun ownership, definite changes would likely be seen. It is up
to policymakers, community leaders, children advocates, and the criminal justice
system to empower the family, schools and other institutions which serve children, and
the community so that these influential entities may combat the growing problem

America faces. It is only then that juvenile gun ownership may be stopped.

51

APPENDIX

APPENDIX A

Scaling items taken from respondents' questionnaires

Family gun habits scale
1. Have any of your brothers or sisters ever owned a gun?

Yes = 1, N0 = 2
2. Have any of your brothers or sisters ever given you a gun?
Yes = 1, N0 = 2
3. Do any males in your family own guns?
Yes = 1, No = 2
4. Have you ever shot guns with a relative?
Yes = 1, No = 2

Peer gun approval scale
1. My friends would look down on me if I did not carry a'gun

Agree Strongly = 4, Agree = 3. Disagree = 2, Strongly Disagree = 1
2. In my crowd, if you don't have a gun, people don't respect you
Agree Strongly = 4. Agree = 3, Disagree = 2, Strongly Disagree = 1

Peer gun behavior scale
1. How many of your friends own guns?

None = 1, Some = 2, Most = 3
2. How many times have you shot guns with your friends?
Never = 1, A few times = 2, Many times = 3

Gun prevalence in neighborhood scale

1. Have you ever been threatened with a gun or shot at?
Never = 1, A few times = 2. Many times = 3

2. In my neighborhood there are lots of guns on the streets; they are easy to get
Not true = 1, Somewhat true = 2, Very true = 3

3. Let's suppose you need a gun for some reason and you don't already have one,

how much trouble do you think it would be for you to get the gun you wanted?

A lot of trouble = 1, A little trouble = 2, No trouble = 3

4. How important is it to cany a gun to protect yourself? .
Not important = 1, Somewhat important = 2, Very important = 3

Gun attitude scale

1. How important is it to cany a weapon because it makes you feel better?
Not important = 1, Important = 2

2. Guns are fun; I like guns
Disagree = 1, Agree = 2

3. It is OK to shoot a person if they have done something to hurt or insult you?
Disagree = 1, Agree = 2

4. It is OK to shoot somebody who doesn't belong in the neighborhood?
Disagree = 1, Agree = 2

52

5. It is OK to shoot a person if that's what it takes to get something you want?
Disagree = 1, Agree = 2

Gang influence sca_l_e
. Most of the people in my gang or group carry guns

_8

Yes = 1, N0 = 2
2. There are always lots of guns around whenever the gang or group gets together?
Yes = 1, N0 = 2
3. You have to have a gun to join my gang or group?
Yes = 1, N0 = 2
4. You have to show you can use a gun to be in my gang or group?
Yes = 1, N0 = 2

53

BIBLIOGRAPHY

BIBLIOGRAPHY

Akers, R. L. (1985). Deviant beg/ion A social learning approach (3rd ed.) Belmont:
Wadsworth Publishing Company.

Akers. R. L., Krohn. M. D., Lanza-Kaduce, L., 8 Radosevich, M. (1979). Social
Ieaming and deviant behavior: A specific test of a general theory. American
Sociolocficgrl Review. 44, 636655.

Anderson, E. (1994). The code of the streets. Atlantic Monthly. 273. 81-94.
Ash, P., Kellerrnann. A. L., Fuqua-Whitley, D., 8 Johnson, A. (1996). Gun acquisition

and use by juvenile offenders. The Joumflf the America Medical
Association. 275 1754-1758.

 

Bachman, R. 8 Paternoster. R. (1997). Statistical methods for criminology and
crimmgl justice. New York: The McGraw-Hill Companies, Inc.

Baldwin, J. D. 8 Baldwin, J. I. (1981). Behavior principles in gvemday life. Englewood
Cliffs: Prentice-Hall Publishing.

Ball, J. C. (1957). Delinquent and non-delinquent attitudes towards the prevalence of

stealing. Journal of Criminal Law, Criminology, and Police Science, 48, 259-
274.

Biskup, M. D. 8 Cozic, C. P. (Eds). (1992). Youth Violence. San Diego: Greenhaven
Press, Inc.

Bjerregaard, B. 8 Lizotte, A. J. (1995). Gun ownership and gang membership. The
Journal of Criminal Law 8 Criminology, 86, 37-57.

Blumstein, A. 8 Cork, D. (1996). Linking gun availability to youth gun violence. Law
and Qontemporag Problems, 59, 5-24.

Burgess, R. L. 8 Akers, R. L. (1966). A differential association-reinforcement theory of
criminal behavior. Socigl Prgtyems. 14, 128—147.

Burkett. S. 8 Jensen. E. L. (1975). Conventional ties, peer influence, and the fear of
apprehension: A study of adolescent marijuana use. ciol ical uarterl
_1_6, 522-533.

Bureau of Justice Statistics. Guns used in crime: Firearms, crime; and criminal justice.
[Online]. Available http://www.ojp.usdoj.gov/bjs/publascii/guic.txt, July 1995

Caetano, D. F. (1979). The domestic arms race. Journal of Communication 29 39-
46.

 

54

Cao, L.. Cullen, F. T.. 8 Link. B. G. (1997). The social deterrnlnants of gun ownership:
Self-protection in an urban environment. Criminolqy, 35, 629-647.

Center to Prevent Handgun Violence. (1993. June). Kids car_rying guns: Loopholes in
state and fedeifirggrms laws. Washington. DC: Author.

Curry, G. 0., Ball, R. A., 8 Fox, R. J. (1994). Gang, crime, and lgw enforcgment
recordkeeanQ (NCJ 148345). Washington, DC: National Institute of Justice.

Decker, S. H., Pennell, S., 8 Caldwell, A. (1997). Illegal firearms: Access and use by
arrestees (NCJ 153850). Washington, DC: National Institute of Justice.

DeFronzo, J. (1979). Fear of crime and handgun ownership. Criminology, 17, 331-
339.

Elliot, D. S., Huizinga, D., 8 Ageton, S. S. (1985). Eyplaining delinguency and drug
use. Beverly Hills: Sage Publications.

Emler, N. 8 Reicher, S. (1995). Adolescence and delinguency. Oxford: Blackwell
Publishers Inc.

Esbensen, F. A., 8 Huizinga, D. (1993). Gangs, drugs, and delinquency in a survey of
urban youth. Criminology. 31. 565-587.

Fagan, J. (1990). Social process of delinquency and drug use among urban gangs. In
C. Huff (Ed). Gangs in America. Newbury Park: Sage Publications.

Giordano, P. C., Cemovich, S. A., 8 Pugh, M. D. (1986). Friendships and delinquency.
American Journal of Sociolggy, 91, 1170-1201.

Goldstein, A. (1991). Delinguent gangs: A psychological persgtive. Champaign:
Research Press.

Gottfredson. M. 8 Hirschi, T. (1990). A general theog of crime. California: Stanford
University Press.

Hallinan, M. T. 8 Williams, R. A. (1990). Students' characteristics and the peer-
influence process. Sociology of Education, 63, 122-132.

Hansell, S. 8 Wiatrowski, M. D. (1981). Social disorganization models of deviance. In

G. F. Jensen (Ed.). The sociology of delinguengy: Current issues. Beveriy

Hills: Sage Publications, Inc.

Hemenway, D., Prothrow-Stith, D., Bergstein, J. M., Ander, R., 8 Kennedy, B. P.
(1996). Gun carrying among adolescents. Law and Contemporagy Problems,
5_9, 39-53.

Hill, G. D., Howell, F. M., 8 Driver, E. T. (195). Gender, fear, and protective handgun
ownership. Criminology. 23, 541-553. ‘

55

Hirschi, T. (1969). Causes of delinquency. Berkeley: University of California Press.

Howell, J. C. Youth ggiqs (OJJDP Fact Sheet #72). [Online] Available
http://www.ncjrs.org/txtfiles/fs9772.b<t, December 1997.

Inciardi. J. A. Horowitz, R. ,8 Pottieger, A E. (1993). Street kids street drugs stree__t_

crime. An examination of drug use 8 serious delinguency in Miami. Belmont:
Wadsworth Publishing Company.

Jankowski, M. S. (1991). Islands in the street: Gangs and American urban sociey.
Berkeley: University of California Press.

Jensen, G. F. (1981). Sociology of gelinguencv: Cgrrent issues (Volume 22). Beverly
Hills: Sage Publications.

 

Jensen, G. F. 8 Rojek, D. G. (1992). Delinguencvflig YogtLh Crime (2nd ed).
Prospect Heights: Waveland Press, Inc.

Jessor. GP. (1972). Parents, peers and delinquent action: A test of the differential
association perspective. American Joumgl of Sociolggv. 78‘ 63-72.

Knox, G. W. (1991). An introduction to gangs. Berrien Springs: Vande Vere
Publishing Ltd.

Krohn, M. D. (1974). An investigation of the effect of parental and peer associations
on marijuana use: An empirical test of differential association theory (pp. 75-
89). In M. Reidel 8 T. P. Thombeny (Eds). Crime and delinguency:
Dimensions of deviance. New York: Praeger.

Larzelere, R. E. 8 Patterson, G. R. (1990). Parental management: Mediator of the
effect of socioeconomic status on early delinquency. Criminology, 28. 301-323.

Lawrence, R. (1998). Families, peers, schools. and delinquency. School crime :8;
juvenile justice (pp. 59-88). New York: Oxford University Press.

Lizotte, A. L., Howard, G. J., Krohn, M. D., 8 Thombeny, T. P. (1997). Patterns of
illegal gun carrying among young urban males. Valparaiso University Law
Review 31 375-393.

 

Lizotte, A. J., Tesoriero, J. M., Thomberry, T. P., 8 Krohn, M. D. (1994). Patterns of
adolescent firearms ownership and use. Justice Quarterly, 11. 51-73.

Loeber, R. 8 Dishion, T. (1983). Early predictors of male delinquency: A review.
Psychological Bullentin, 94, 68-99.

Mason. C. A.. Cauce, A. M., Gonzales, N., 8 Hiraga, Y. (1994). Adolescent problem
behavior: The effect of peers and the moderating role of father absence and

56

the mother-child relationship. AmericanJourpaJ of Community Psychology. 22,
723-743.

Maxson. G, Gordon. M., 8 Klein, M. (1985). Differences between gang and non-gang
homicides. Criminology. 23. 209-221.

McGarrell. E.F. (1993). Trends in racial disproportionality in juvenile court processing
1985-1989. Crime 8 Delinggencv. 39. 29-48.

Miller, W. B. (1958). Lower class culture as a generating milieu of gang delinquency.
Journal of Social Issues 14, 5-19.

 

Mish, F. C., et al. (Ed). (1990). Webster's ninth new collegiate dictionagy. Springfield:
Merriam-Webster Inc.

Monti, D. J. (1993). Origins 8 problems of gang research in the US. In S. Cummings 8

D. J. Monti (Eds). Gangs: The origins 8 impact of contemporary youth gangs in
the United States (pp. 3-26). Albany: State University of New York Press.

National Institute of Justice. (1997). Guns in America: National survey on private
ownership and use of firearm. [On-line]. Available:
www.ncjrs.orgltxtfilesl165476.txt

Nye, F. l. (1958). fgmily r_ele_itionships 8 delinguent behavior. New York: John Wiley
Publishing.

Oehme III, C. G. (1997). Gangs, groups, and crime: Perceptions and responses of
commungy‘ organizations. Durham: Carolina Academic Press.

Osofsky, J. D. (1995). The effects of exposure to violence on young children.
American Psycholggist, 50, 782-788.

Padilla, F. (1992). The gang as an American enterprise. New Brunswick: Rutgers
' University Press.

Patterson, G. R., DeBaryshe, B. D., 8 Ramsey, E. (1989). A developmental
perspective on antisocial behavior. American Psycholpgist, 44, 329-335.

Santman, J., Myner, J., Cappelletty, G. C., 8 Pertmutter, B. F. (1997). California
juvenile gang members: An analysis of case records. Journal of Qang
Research 5 45-53.

 

Sellers, C. S. 8 Winfree Jr., L. T. (1990). Differential associations and definitions: A

panel study of youthful drinking behavior. The lntemational Journal of the
Addictions 25 755-771. ‘

 

Shelden, R. G., Tracy, S. K., 8 Brown, W. B. (1997). Youth gangs in American
society. Belmont: Wadsworth Publishing Company.

57

Sheley, J. F., Brody, C. J., 8 Wright, J. D. (1994). Women and handguns:
Evidence from national surveys, 1973-1991. SociaLSIcience Resealrch. 23. 219-
235.

Sheley, J. F. 8 Wright, J. D. (1993). Gun acguisition and possession in selected
juvenile samples (NCJ 145326). Washington, DC: National Institute of
Jusfice.

Sheley, J. F. 8 Wright, J. D. (1995). In the line of fire: Yvogthshquns. and
violence in umgLAmeric; New York: Aldine De Gruyter.

Sheley, Joseph P, Wright, James D., 8 Smith. M. Dwayne (1991). Firearms.
violence, and youth in Califomia, Illinois, Louisiana, and New Jersey. [Computer
file]. ICPSR version. New Orleans, LA: Tulane University, Department of
Sociology [producer], 1993. Ann Arbor, MI: Inter-university Consortium for
Political and Social Research [distributor], 1995.

Sheley, J. F., Zhang, J., Brody, C. J., Wright, J. D., 8 Williams, M. A. (1995). Gang
organization, gang criminal activity, and individual gang members' criminal

behavior. Social Science Quarterly, 76, 53-68.

Short, J. 8 Strodbeck, F. (1965). Group process and gang delinguency. Chicago:
University of Chicago Press.

Smith, M. D. (1996). Sources of firearm acquisiton among a sample of inner-city
youths: Research results and policy implications. Journal of Criminal Justice
24, 361-367.

 

Smith, D. A. 8 Brame, R. (1994). On the initiation and continuation of delinquency.
Criminology, 32, 607-625.

Smith, T. w. 8 Smith, R. J. (1995). Changes in firearms ownership among women,

1980-1994. The Joumal of Criminal Law 8 Criminology, 86, 133-145.

Spergel, I. (1983). Violent gangs in Chicago: Segmentation and integration. Chicago:
University of Chicago.

Staples, R. (1982). Black masculinity: The black male's role in American society. San
Francisco: The Black Scholar Press.

 

Stark, R. (1987). Deviant places: A theory of the ecology of crime. Criminology, 25,
893-909.

Steffensmeier, D. 8 Allan, E. (1991). Gender, age, 8 crime. In J. Sheley (Ed).
Criminology: A chntemporagy handbook. (pp. 161-187). Belmont: Wadsworth
Publishing Company.

Sutherland, E. H. (1939). Principles of criminology. Chicago: JB Lippincott Company.

58

Sutherland, E. H. (1947). Prirgiples of criminology (4th ed). Philadelphia: JB
Lippincott Company.

Sutherland, E. H. 8 Cressey, D. R. (1978). Criminology (10th ed). Philadelphia: J.B.
Lippincott Company

US. Department of Justice, Federal Bureau of Investigation. (1997). Crime in the
Cnited States, 1996. (pp. 241-243). Washington. DC: U. S. Government
Printing Office.

Vigil, J. D. (1988) Barrio gangs. Austin: University of Texas Press.

Vold, G. B. 8 Bernard, T. J. (1986). Theoretical criminology (3rd ed). New York:
Oxford University Press.

Watts, R. J. 8 Jagers, R. J. (Eds). (1997). Manhood development in urban African-
American communities. New York: The Haworth Press, Inc.

Wilkinson, D. L. 8 Fagan, J. (1996). The role of firearms in violence "scripts": The
dynamics of gun events among adolescent males. Law and Contemporagy
Problems 59 55-91. '

 

Winfree Jr., L. T., Backstrom, T. V., 8 Mays, G. L. (1994). Social Ieaming theory, self-
reported delinquency, and youth gangs. Youth 8 Society, 26, 147-177.

Wright, J. D., Sheley, J. F., 8 Smith, M. D. (1992). Kids, guns, and killing fields.
may, Social Science, 8 Modern Society, 30 , 84-89.

Zimring, F. E. (1996). Kids, guns, and homicide: Policy notes on an age-specific

epidemic. Law and Contemporagy Problems, 59, 25-37.

59

":ililillilllliiilr