GANGS AND GUNS:
A TASK FORCE REPORT FROM THE
National Gang Crime Research Center
CO-PRINCIPAL INVESTIGATORS:
George W. Knox, Ph.D., Director,
National Gang Crime Research Center
James G. Houston, Ph.D., Department
of Criminal Justice, St. Ambrose University, Davenport, Iowa.
John A. Laskey, M.A., professor of
Criminal Justice, Morton College.
Thomas F. McCurrie, M.S.,
Managing Editor of Journal of Gang
Research; adjunct faculty, Dept. of
Criminal Justice, Chicago State
University.
Edward D. Tromanhauser, Ph.D.,
Chairman, Dept. of Criminal Justice,
Chicago State University.
COPYRIGHT 2001: National Gang Crime Research Center.
EXECUTIVE SUMMARY
of the Gangs and Guns Task Force Research Report
Six gang researchers at three universities collaborated to carry out an extensive study of gangs and guns in the midwest involving 1,206 survey respondents which included 505 gang members. Four social contexts were used for the survey: eight county jails from the farmland to the urban central area (891 inmates), matched pair design samples from a Chicago public high school and an inner city program, and a sample of gang members in a private suburban probation program.
I. The jail study showed:
*** Gang problems in the midwest show a large "ripple effect" where many variables show the consistent effect of increasing in magnitude the closer one is to the urban central area, and decreasing in severity through the outlying areas and heartland to its lowest level in the farmland areas.
*** Gang membership is a variable that significantly differentiates many variables about firearms, violence, behaviors and beliefs.
*** Gang membership can be predicted with 81 percent accuracy using discriminant analysis.
*** Gang density, that is the percentage of inmates who are gang members, using a more restrictive definition of gangs, showed percentages two to ten times higher than the parameter estimated in a recent federal national assessment of gangs in corrections.
II. The high school study showed:
*** Some 87 gang members were matched with 87 demographically identical non-gang members.
*** These are essentially the same gangs represented in other social contexts studied.
*** The gang member profile is similar to that found in other social contexts.
III. The probation program study showed:
*** Some 69 gang members in a suburban-based probation had much the same "gang profile" as elsewhere, but with some new important twists.
*** Fewer suburban gang members originally joined the gang primarily "to make money", but did so mostly for essentially social rather than economic reasons. They are atypical in this regard compared to the jail inmate gang member.
IV. The inner city program study showed:
*** Some 36 gang members were matched with 36 demographically identical non-gang members.
*** The gang member profile is comparable to that found in other social contexts.
V. A Combined Analysis of Gang Members in All Contexts showed:
*** There is no difference on most variables when comparing "Peoples" and "Folks" gang members. That is, on most human traits they are they same thing differing only symbolically, not objectively.
*** There is little difference in the basic gang member profile about guns and violence across social contexts. Gang members show the higher risk profile regardless of social context.
*** A gang risk continuum exists showing a consistent violence escalation effect from the lowest level of risk (non-gang member with no gang friends) to the highest level (active gang member).
*** An analysis of factors associated with gang members who attempt to leave the gang was made. This showed, generally, a hardening effect; where the gang member who has never attempted to quit the gang appears to have a higher commitment to the gang, and is consistently more 'hard core' in regard to findings about gun crimes and gun violence.
VI. Conclusions include the following:
*** It is possible to profile individual gangs by threat analysis. For example, weapons access, acquisition, and usage patterns, level of violence threat (i.e., police shoot-outs), and related variables of interest. But research on a larger scale would have to be done to ensure specific gang identity sample development. The ideal design would ensure at least 100 members of each gang organization of interest.
*** The research findings reported here about gang density demand a serious national assessment of gangs in corrections be undertaken, because a recent federal report is believed to grossly underestimate the problem to the point of mathematical fantasy.
Acknowledgements
We are indebted to a large number of individuals and agencies that helped this research project come to fruition.
We gratefully acknowledge the cooperation of eight county jails for purposes of data collection. We thank the eight sheriff's of these respective counties, and the correctional officers and other staff at these eight jails who assisted us. We did promise full anonymity to the jails if they wanted it, so at this time all we are permitted to say is they exist in Illinois and Iowa.
A probation supervision program allowed us access to a sample of non-incarcerated gang members, and for this we are exceedingly thankful for the assistance of Charla Waxman.
We thank, as well, a high school principal who allowed us to collect a sample from the general student population for the research reported here. The identity of this school must remain anonymous because it is the site of an ongoing research about the impact of gang problems in public schools. Suffice it to say we are very grateful to the administrators and teachers involved.
We are also grateful for the access to three community based social service programs for a sample provided by Reginald King, a graduate student in the Department of Criminal Justice at Chicago State University. We especially thank Beverly Hicks at Chicago State University's Department of Criminal Justice for laboring on this project.
We gratefully acknowledge the help of several research assistants who are Chicago State University criminal justice students for their help in the Chicago area: Dawn Collins, Mary Chambers, Kelly Gandy, and Kendra Evans. We are also thankful for the assistance from the clerical staff of the National Gang Crime Research Center. Finally, we want to thank over a thousand respondents that participated in this survey research. We thank the 505 gang members who completed the survey, and a larger number of non-gang members, for their time and input to this analysis.
GANGS AND GUNS:
A Task Force Report
from the
NATIONAL GANG CRIME RESEARCH CENTER
INTRODUCTION
The issue of "gangs and guns" is one of the more interesting criminological topics and is one of the least researched issues today. This is true in spite of, (1) the enormous social policy implications surrounding the use of firearms by gang members and the devastating consequences in terms of human violence, (2) in spite of the good intentions of federal agencies whose duty it is to promote useful knowledge in this area, and (3) in spite of the fact that this is an enduring issue that is gaining increased salience and public attention.
Empirical research on the relationship between gangs and guns is woefully lacking. Some rather simple questions beg the attention of criminologists and policy makers alike: do gang members represent a significantly higher threat when it comes to armed violence in America when compared to other offenders or the general public? Do gang members really have better access to firearms, including military weapons and explosive devices, and do they really use such weapons in committing crimes? Do gang members pose a more serious threat of violence in terms of shoot-outs with police? Where is the evidence --- we must ask --- for those most inclined by either academic theory or practitioner insight to claim that gang members seek out weapons more than others, use weapons more than others, and use more deadlier weapons than non-gang members in the United States? Further, we ask for empirical evidence from serious research on the issue, not simply casual answers from a cursory understanding of the complexity of the modern day gang problem.
In spite of millions of dollars spent by government agencies on gang intervention and prevention service programs, no such program that works has yet surfaced in the literature. In addition, little useful knowledge on the issue of gangs and guns has been added to the literature from this source.
The research reported here seeks empirical answers to some of these questions about gangs and guns.
REVIEW OF THE PREVIOUS LITERATURE
The first known document having relevance to gangs and guns is a military report on the long campaign by the British Army to suppress the Budhuk gangs in India (Sleeman, 1849). The robber gangs of India operated much like a guerilla army. These gangs were effectively outfitted through the open and underground market with nearly the same quality of weapons used by the British Army itself. The only weapon that proved effective against the Budhuk gangs that the gangs did not possess was artillery. The strategy that proved most effective in gang suppression according to Sleeman (1849) was intelligence and housing gang inmates in terms of their level of threat.
In the American literature, clearly the parallel to Sleeman is found in Green (1847), which showed that firearms played a significant role in the criminal operations of early American gang. We call any group whose members knowingly benefit from crime a gang. Hate groups are also considered gangs here, because it does not matter if the offense is an income producing crime, it can be a political or ideological crime. Some commentators on the gang problem in America and elsewhere claim much confusion about what a "gang" means. From our point of view, it is sufficient to know that crime is involved in a group or organizational context, be it informal or formal in structure and function. Some would include deviance as a condition defining a gang. Some would be so generalized to define any youth group that is not adult supervised by formal social control agencies as a gang.
The controversy over "what is a gang" is not ours, as we are content with the definition already advanced. Further it is a moot issue, as the kinds of gangs we are examining are those who pose a danger genuine enough to evoke strong penal sanctions from the systems of formal social controls in our society. Many are gangs who do not want to be called "gangs", they want to be called organizations and nations. They have proliferated throughout the United States and their identities are no secret to criminal justice agencies or the general public: Gangster Disciples, Vice Lords, Latin Kings, Aryan Brotherhood, Aryan Nation, etc. Further, these are for the most part highly structured gangs, most of which have written constitutions and a formalized infrastructure (Knox, 1993).
The issue of weapon access is given little attention in the recent NIJ document on gangs and simply cites Spergel for issues of gang homicides. It concludes, without any data, but rather on the basis of impressions, that gangs have more weapons and may have a higher propensity to use deadly weapons. It does cite a school study to the effect that gang members were nearly three times as likely than non-gang members to say firearm access was "easy" (Callahan and Rivara, 1992). Its conclusion is that they are uncertain that gangs today have greater access to guns, it only appears they are more willing to use them. This is based on comments from one prosecutor in one major urban jurisdiction.
The participant observation study by Hagedorn and former gang member Perry Macon gave the estimate that about one half of African-American gang members in Milwaukee reported having one or more guns, never purchased legally, more often than not from the underground or stolen (Hagedorn and Macon, 1988: p.144).
The small quantitative study of gang members in Detroit revealed similarly that most gang members have a gun in their home (Taylor, 1990: p. 130), regardless of the level of gang sophistication. Oral history research on imprisoned gang leaders has recently revealed the interesting and believable story of how in Illinois one such leader routinely carried a fully loaded .25 caliber semi-automatic as an inmate in the Illinois Department of Corrections. Such anecdotal evidence can also be easily found through newspaper coverage of gangs and gang members.
Jankowski (1991: p. 121) provides a paragraph on how the theft and resale of guns is a component of gang business. No statistics are provided, but it is claimed that gangs do have access to fully automatic weapons and routinely buy and sell other guns including sawed off shotguns.
Short and Strodtbeck (1965: p. 257) commented to the effect that firearms are status symbols in the gang subculture. We know from a small survey of Illinois National Guardsmen that 9.2 percent report having been at one time approached by a gang member who wanted to acquire military weapons or ordnance (e.g., grenades, etc).
Student survey research on gang problems in schools reveals that gang members are significantly more likely than non-gang members in the public high school population to report carrying a concealed gun for purposes of protection (Knox, Laske, and Tromanhauser, 1992: p. 56).
Some of the most direct empirical evidence on the relationship between gangs and guns to date comes from the research by Sheley and Wright (1993, p. 9). Generally, it found greater rates of possessing and carrying firearms among gang members than among non-gang members using confined juvenile and high school juvenile samples. Sanders analyzed police reports in San Diego, concluding that gun use in gang incidents had increased over time in the 1980s (1994: pp. 56-57).
This brief look at the literature summarizes the overlap between the topics of gangs and guns. Obviously there is a paucity of research material on the subject. We come to a rather sad realization regarding the issue of gangs and guns. It is the notion that so much attention and discussion is given to this relationship, and it is felt by all to be a problem, but few researchers have advanced much generalizable knowledge about the issue. It is in this sense that about a year ago the plan was made to create a small task force to carry out a research project on gangs and guns and present the findings to the criminological community and others interested in the relationship between gangs and guns. This report, providing data collected from a number of different social contexts (jails, high school, juvenile probation) is a result of that plan.
THE TASK FORCE APPROACH
Developing their own unique research agendas and hypotheses to be tested the researchers associated with this project collectively agreed that similar data would to be collected on gang members found in different social contexts. Sharing in costs, and labor, the researchers helped to identify sites for data collection.
At a very early point all researchers submitted their contributions to the item pool that would be used to develop the common survey instrument. There was some overlap and redundancy, but overall the researchers had entirely different interests. Therefore the hypotheses they would test with the data were diverse in nature.
This Task Force approach was planned, coordinated, and implemented for the purpose of developing further empirical knowledge about the relationship between gangs and guns. All researchers are considered equal co-principal investigators. As another element in the Task Force approach it was agreed that a common report would be issued that reflects the basic empirical findings from this collective research project.
While individual researchers were free to develop and extend their individual analyses using the data, the overall report would provide the basis for some consensus between the researchers. All researchers examined and helped edit almost weekly updates of the report over a period of several months of analysis time. Thus, everyone reviewed, edited, and contributed to the thinking in the common report to be issued.
There are many different ways to examine and account for the relationship between gangs and guns, and it was for this reason that the researchers also agreed to present their individual findings at a professional forum --- the Annual Meeting of the American Society of Criminology. This was to be the outlet for presenting individual scholarly analyses of the problems associated with armed gang violence. Thus, individual papers presented at this criminology meeting may extend an analysis far beyond anything that is within this Task Force report. This report simply provides the basic results obtained when different social contexts were used to survey gang members and non-gang members on the issue of guns and gang violence.
THE DIFFERENT SOCIAL CONTEXTS USED FOR THIS STUDY
The plan was to examine the "gangs and guns" issue in a variety of different social contexts. The choice of sites for the research was based on satisfying the need for such variation. The reasoning was that a diverse cross-section of gang members could be obtained if a variety of such different social contexts were used.
One site turned out to be a complete mistake. It was a recreational program serving a large public housing complex on Chicago's southside. This program had recently given a $30,000 contract for "gang consulting" to a major gang activist who is also a Gangster Disciple advisor. It is a site of "Midnight Basketball" games between rival gangs. The program director at the site asked the gang leaders for their cooperation. The gang leaders agreed to help with the project. It meant they could get some "goodies" if they would fill out the surveys. The gang leaders got enough goodies for 300 surveys and returned 108. Many of these survey instruments were fraudulently completed (usually in the same ink color and same handwriting). We were, simply, the victims of the "hired hand" phenomenon in offender research. All data from this social context were discarded.
However, other sites did prove effective for broadening the social contexts in which to study gangs and guns, because we were personally in charge and one or more of the researchers were on the scene at all times for data collection. The other social contexts include eight jails, a probation supervision program, a regular public high school, and a sample from a social service agency serving inner city youths on Chicago's southside. The way the Task Force worked was to identify the work requirements and find the least costly approach to completing that work. Travel to other cities, overnight stays, etc, were assumed to be personal expenses. Any direct hard costs such as photoduplication and purchases of "goodies" for use as honorariums were shared equally by prior agreement between the researchers.
Overall, the research reported here is therefore based on using the same survey in the following social contexts: (1) Eight county jails (N = 891 inmates), (2) a matched-pair analysis from a large urban public high school (N = 87 non-gang members, N = 87 gang members), (3) a matched-pair analysis from a social service/recreation program (N = 36 non gang members, N = 36 gang members), and (4) a suburban probation program (N = 58 gang members). Thus, combining these social contexts, a total of N = 1,206 respondents are represented in this study, of which 505 are self-reported gang members. We can further break down gang membership categories into the People/Folks alliance as follows in our analysis: N = 161 Peoples, N = 183 Folks.
ORGANIZATION OF THE REPORT
The first section (I.) examines the findings from an analysis of survey results at eight county jails in the midwest.
The second section (II.) examines results from the same survey where the respondents were high school students in a large urban public high school setting.
The third section (III.) examines the results from the same survey using a suburban probation program sample.
The fourth section (IV.) examines an inner-city service and recreation program where we were able to use the same survey.
The fifth section (V.) combines gang members surveyed from these four social contexts for a closer analysis.
The last section (VI.) presents our overall summary of findings, conclusions, and recommendations for future research.
I. THE SUMMER 1994 STUDY OF EIGHT JAILS
One megajail, four larger jails and three smaller jails were chosen from the midwestern United States. While the sample includes one of the megajails, the sampling strategy intentionally sought out smaller and rural jails as well. The rationale for selecting these settings outside of the context of the large urban setting stems from recent research on gang proliferation and gang migration which showed the gang problem extends far into the "heartland" or geographical areas not previously considered as having a "gang problem".
Another reason has to do with how this type of geographical setting provides an ideal way to test certain assumptions about gang density --- that is, the proportion of gang members within the correctional population. The matter of gang density is addressed in a later section of the report as the "six percenter issue" which stems from the recent conclusion (ACA, 1993) that only six percent of the American prison population are gang members.
The research goal was to achieve this census by having each and every inmate cooperate, not just a sample of some inmates from the overall jail population. This research sought a saturation sample of inmates in seven county jails, a goal that was substantially met. The research procedure required sampling from the megajail, where the most secure unit was chosen. Figure 1 shows the types of jails used in this research.
Figure 1
The Eight Jails Surveyed for the Gangs/Guns Study
The Various Types of Jails Used in This Study
Farmland Areas Heartland Areas Urban Outlier Urban Central
*********** ************* ************ *******
3 in Iowa 2 in Iowa; 1 in Illinois Illinois
1 in Illinois
THE RESEARCHERS VISIT THE JAILS IN THE MIDWEST U.S.A.
The jail inmate component of this study began with surveying smaller jails in the Heartland and culminated with a survey of inmates in the maximum security unit of the megajail located in the urban central area. It is possible then to examine the extent to which a classical "ripple effect" occurs, comparable to the concentric zone hypothesis of early Chicago researchers, but expanded in geographical scope to the midwestern United States. Our urban central geographical area is the site of the megajail, holding over 5,000 inmates, and it is a thirty minute drive from an outlying county where we also surveyed the entire jail. Three larger sized and three small county jails round out the analysis. The three
larger jails constitute what we call the "heartland" area of the midwest. and the three rural smaller jails are viewed as farmland areas.
One of the larger jails in the Heartland is an old classic lock-up matching the century old architecture of its city. Just minutes away separated by a river one of the other larger jails is a modern new building exemplifying what is meant by the modular style. It is a difference between old and new. In the older Iowa jail, the correctional officers knew the names of most all inmates and open communication existed. In the newer Illinois jail, we watched as an inmate threw a hard plastic cup at correctional officer and we wondered if the social climates of these two jails were really equivalent. In the older jail the inmates could smoke, in the newer jail inmates could not smoke. The newer jail also had more personnel. The newer jail had truly adapted to the times as well in another respect. At the door where the visitors entered the jail, in large conspicuous letters read the warning "NO GANG COLORS OR GANG CLOTHING ALLOWED IN THIS BUILDING". This warning in the newer jail notifies all entrants of its zero-tolerance policy on gangs. These two larger jails, in close proximity to each other, provided the basis for assessing validity and reliability issues.
Overall, however, one megajail, four larger county jails, and three smaller county jails were used for the jail inmate component of this research. Two of the larger county jails are located in Illinois, the other five jails are all located in Iowa; and of course the megajail is that in a large metropolitan area. In none of the jails did a policy exist where all known members of any particular gang were concentrated in a single jail unit. "They are very mixed up" was the general response, implying a concern about any single gang being able to exert too much control in a given area. Staff at the jails generally knew who most of the gang members were, because of the nature of their gang-related offenses that brought them to the jail.
METHODOLOGY
As a collective unfunded pro bono research project all researchers early in the research offered the types of items they wanted to include in the survey instrument. All researchers reviewed the item pool and changes, modifications, and improvements in style, syntax, and construction were made. Further review was given in terms of item order and validity testing contingencies were established. Thus, all researchers had input into item content, item order, phraseology, structure, and length as well as written instructions to be read by the inmates as a normal part of instrumentation. A complete copy of the survey instrument is provided for the reader in Appendix A of this report.
Access to the jail population was typically obtained through first working with the sheriff's office in each county. Dates were set for data collection and three or four of the researchers were on site during all data collection at the larger jails. The team member based in Iowa was able to survey the three smaller rural county jails alone. The Iowa based researcher had career experience in correctional counseling in the federal bureau of prisons and was remarkably adaptive to the psychological states of inmates encountered in the two jails. This researcher became the first speaker during all social encounters with inmates, although all researchers at one point or another contributed points of explanation on our purpose and intent and plans for use of the data, explaining over and over again this was a completely anonymous survey and "no names were needed".
By prior arrangement the plan for data collection was that of a complete saturation, that is a census of every jail inmate confined on the Saturdays that we visited the county jails for purposes of data collection. Generally, with one jail facility as the exception, almost all inmates participated in the survey. And, of course, at the megajail where we faced a population of nearly ten thousand, a subsystem census approach was used --- surveying all inmates in certain types of divisions within the larger jail complex. In almost all jails, therefore, nearly 95% plus of the inmates who were in the jail participated, and those inmates who were asked to participate who did not participate were rare. In one jail, to illustrate this rare situation of an inmate not participating in the survey, one inmate in the medical ward still had a large wound in his chest from a recent gun shot involving a highly publicized murder case as well. This was the only inmate at the jail who refused to cooperate with us and we suspect it was based on the physical discomfort this particular inmate was facing. Other than the one exception discussed below, the data collection plan for this research was that of a census of the jail facilities, by seeking to have a saturation sample of the inmates. Once in the jails almost all inmates cooperated.
One reason for the high cooperation we received from jail inmates was the honorarium we promised as instant gratification. This was possible in all but two of the jails. In the inmate economy, cigarettes or "squares" are the leading mode of exchange. Thus where it was permissible we offered a free package of name brand cigarettes to anyone who completed a survey. We did tell all inmates as a part of the routine "presentation" that we would physically inspect the surveys when collecting them to make sure they were completed. In point of fact, when we collected the surveys we made only a cursory examination and generally collected them "as is". In the non-smoking jails we provided small commercially packaged bags of cookies and snacks as the honorarium. In two of the larger jails we could not distribute cigarettes or food items or hygiene items (i.e., combs), so we simply spent more time trying to motivate the inmates by appealing to them to help us with this multi-university research project. In the two jails where we could not give out honorariums we were able to get only about half of the inmates to participate in the survey, making this situation the exception to the otherwise general result of being able to generate a complete census of each jail facility.
A usable sample size of N = 74 inmates was obtained from three smaller county jails in the "farmland" of the midwest. Additionally, a sample size of N = 370 inmates was obtained from three larger county jails in the "heartland" of the midwest. A larger jail in the urban outlying area of the midwest generated a another sample of N = 211 inmates. And finally, some N = 236 inmates were sampled from the high security division we chose for analysis at the megajail, which we call the urban central area jail. This generates a combined jail inmate sample of N = 891 for purposes of analysis here. Details on the validity of this data are discussed next.
VALIDITY
The validity of the survey research on jail inmates was established through a number of different techniques described here. Seven of these validity issues are discussed in greater detail below. Overall, when all seven issues are combined, we reach the conclusion that little validity threat exists with our data. In fact, for seven different reasons we conclude that high validity exists with the findings reported on our survey research involving jail inmates.
1. Face Validity. The survey instrument itself is easily structured, contains no double-barrelled questions, the items use unambiguous language, and in many cases actually replicates previous well known prior research (e.g., Wright and Rossi, 1986). Face validity here is high.
1. Covert Observation. Opportunity for covert observation of the inmates completing the questionnaires allowed for several other types of validity control. After handing out pencils and surveys the researchers often returned to the control room of the jail where the video monitors allowed watching the inmate behavior in the cell houses. There was no evidence from this type of covert observation that any threat to validity occurred such as one inmate completing another inmate's survey form, nor was there any inmate behavior visible to us that implied any type of collective fraud. Almost all inmates were remarkably cooperative.
2. Overt Observation. Gender is a specific forced-choice item on the questionnaire. However, when collecting the completed surveys from all female inmates their instruments were immediately marked "female" for later comparison with self-report data. We did this for both jails. In no instance did any respondent lie about their gender. If "random responses" been given by the inmates some cases would have emerged which misidentified this gender variable in comparison with the overt observation and marking procedure.
3. Zero Tolerance for Data Entry or Transcription Errors. All survey data stored electronically for purposes of computerized statistical analysis were cross-checked against source documents (i.e., the survey instruments). The data was checked and re-checked and contains no validity threat from transcription errors.
4. Few Unusable Survey Instruments Detected. To illustrate the value of notifying the inmates we would casually check the surveys before giving out the honorariums, at the two larger county jails that border each other, only one random response pattern was found where the jail inmate simply marked the first choice for almost all of the 96 plus questions. This one questionnaire was excluded and not used in the analysis. All other surveys contained well differentiated response patterns implying they were completed by cooperative individuals giving their best level of effort. The inmates were motivated to complete the surveys in one sense because this provided an interesting distraction from the boredom of routine jail activity. No "tainted" data is therefore included in our sample. Generally, we felt the inmates made a good faith effort to complete the survey instruments.
5. An acceptable level of trust was established. Four of the researchers were typically on hand in most jails surveyed where they approached each cell-house area and explained in detail the purpose of the survey research. Much small chit chat and friendly discussion often ensued when we introduced ourselves as professors and three of the four professors were from Chicago. Jokes about legal representation and lawyers generally were common in these informal social exchanges. The survey instrument asks for no name and we told inmates verbally that no names were needed. We told them they could drop the survey's in a box as we would collect them, which is the procedure that we used. In spite of this, however, a goodly number of the inmates put their full name and cell-house unit number on the front, inside, or rear of the survey instrument along with personal memoranda to the effect they would like to "have personal interviews" and provide further information for us. We feel on the basis of the above procedures and observations therefore that a sufficient level of trust was established with the inmates to get relatively honest answers. We recognize generally however as is the criminological axiom: offenders have the tendency to over-report their positive attributes and under-report their stigmatizable attributes. However this axiom applies to all criminological research involving real law offenders.
6. High Cognition of the Meaning of the Survey Items Implies clearly we are measuring what we purport to measure. A large number of the inmates had the tendency to write notes and comments in the margins of the survey instrument on a large and wide variety of issues. These are highly emotive comments implying clear cognition of the true meaning of the survey items. For example one respondent in response to the question (CASE # 178) "Did you vote in the last presidential election?" in the survey instrument answered "NO" and commented in his own hand writing in the margin the reason for his non-voting behavior to the effect that "they are all crooks". More typical was a common pattern of emotive accentuation, where for example (CASE # 190) in response to item # 91 ("Do the gangs in prison basically exist in the same form on the streets?") answered "YES" but marked in the column in his own handwriting "Definitely!". Another respondent (CASE # 164) in response to item # 91 answered "YES" and wrote in "I know from my past 16 years in prison". These types of frequent unsolicited explanations for their answers in the survey instrument provide further evidence that these inmates clearly understood the meaning of the questions. The higher the cognition, the higher the validity. There was nothing in this survey an inmate did not understand. We never asked to narrate their opinion of "deterrence theory" or "strain theory" or any abstract concept. The survey contains very concrete survey items. No survey was received with any comment to the effect "I do not understand what you mean here", and there was not a single inmate in either jail who needed help to understand any question from us.
7. High Internal Consistency of Response Patterns Emerged Upon Data Analysis Implying Low Levels of Deceit. In the Minnesota Multiphasic Personality Inventory (MMPI) full form of 566 true or false questions, a "lie scale" exists by being able to compare response to a question early in the form with a similar question latter in the form. The present researchers wanted to anticipate and head off any potential criticism along these lines and developed items rather similar in nature and placed these non-sequentially in the instrument in order to provide this basis for a "validity check" of internal consistency at time of data analysis. Here we explain the nature of this internal consistency validity control technique and how the results support the high validity in the present research.
The first test is one where we would capture an offender who was so paranoid he or she would lie about present age and age of first arrest. Item # 13 on the survey instrument asks for "At what age were you first arrested for any offense?" and provides a blank space for the actual two digits elicited (i.e., "_____Years old"). Item #74 on the survey instrument asks "What is your age in years? I am ____ years old today". An inmate who would engage in early intentional deceit in responses to the survey could therefore be detected by comparing these two items, and deceit cases would obviously exist where the value of Item # 13 exceeds that of Item # 74. A simple computer check allows for directly testing this validity hypothesis. Do we see a lot of cases where an inmate claims his/her first arrest was "24" and then provides several pages later a current age of say "19"?
Using a simple compute statement a variable was made called LIETEST1 and was calculated for all cases in the sample as follows:
COMPUTE ITEM1374 = missing
IF ITEM13 > ITEM74 THEN LIETEST1 = 1
/***COMMENT "1" means deceit
IF ITEM13 <= ITEM74 THEN LIETEST1 = 0
/****COMMENT "0" means consistency, it is logically possible
IF ITEM13 = missing THEN LIETEST1 = missing
IF ITEM74 = missing THEN LIETEST1 = missing
Upon creating this new internal validity check variable called "LIETEST1" it was simply analyzed for a full frequency distribution. The results showed that 11.9 percent qualified as missing, having missing data on one or more of the two variables, but that 88.1 percent were "0". Thus no cases emerged where this type of deceit could be detected.
Several other types of internal deceit were tested for. One was where a respondent indicated he/she had served in the U.S. military or armed forces, if they indicated "yes", this group was examined for its frequency distribution on education. The military does not take high school drop outs. None appeared in our sample as well among those self-reporting military service. Another test of suspect validity we reasoned would be if many cases emerged where someone first declares they did vote in the last presidential election and then later in the survey indicates they are not a currently registered voter. The fact is, only about a fourth (24.4%) of the jail inmates claimed to vote in the last presidential election, a figure far below the national average and a believable figure. Further, when asked if they are currently a legally registered voter, some 39 percent claimed this status. But there was not one case in the sample where someone claimed to have voted in the last presidential election and who presently reported they were not a legally registered voter. Thus, high internal consistency emerges in our data.
Finally, the issue of validity threats from "macho" behavior where inmates who claim or brag about their criminal exploits particularly against police was not a pattern found in this research. In point of fact, when asked in a series about firearms use only a small percent of the entire sample claimed this high level of intensity (11.3%). A bragger would have claimed more shooting with police than with other contexts reported below in our findings.
RELIABILITY
Reliability in this kind of survey research on inmates has two major meanings: (1) whether the same questions yield the same results at different points in time, and (2) whether the same questions yield the same results on relatively equal populations in different locations? The reliability issue here, however, relates only to the second of these reliability issues (e.g., inter-jail reliability). The reason is that this research was an anonymous survey and therefore was not designed for longitudinal follow-up on the same respondent.
The issue of inter-jail reliability is still an important one for several reasons. The same survey instrument used in the two larger jails in close proximity to each other should yield rather similar background profiles of the inmate population. Both of the two larger county jails are about the same size population and are geographically located only minutes apart by automobile travel. There are more reasons why the populations should be similar on a number of background risks because of the homogenizing process. That is there is more commonality in the general offender population than differences, especially in two jails located in the same area, separated by only a state line and a river, but otherwise only about two miles apart.
The hypothesis tested here, therefore, ascertains if the same questions generate relatively equal background and risk profiles for the inmate populations in the two larger county jails that are separated only by a short distance and a river. If major differences emerge in basic background factors and on other variables where no reason exists to hypothesize such differences, then a problem of inter-jail reliability would exist, and it could be said that the same questions may yield different results in similar jails. The item analysis results should be relatively equal and this is what an intensive comparative analysis of both jail populations revealed. The item analysis undertaken here involved over 100 variables comparing the distributions between both jails using the Chi-square distribution. There was not one single variable for which a significant difference existed between the two jails which also held up within the gang member subgroup population comparing both jails.
Missing data was not a major problem in this analysis. There was a small amount of missing data which is typical of most survey research of this kind. However, it was random and not systematic in nature. In most cases it meant a respondent skipped over an entire page while folding back the seven page questionnaire.
GENERALIZATIONS BASED ON THE JAIL DATA
While both validity and reliability are at acceptable levels for survey research of this kind, it is important to point out that this research design was not intended to be a national probability sample. Thus, the generalizations based on this jail data cannot be applied to the entire United States local detention population. Further, by intentionally seeking out smaller rural jails in the "Farmland" and in the "Heartland" where the gang problem was not expected to be as large of a problem as it is in large urban areas such as Chicago and Los Angeles, our research strategy captures the parameters at both the lower end of the gang threat level and the higher end of the gang density problem (i.e., Chicago). This differs then from the correctional research strategy employed by Sheley and Wright (1993) who sought out geographical areas known to have serious gang problems. The correctional research strategy employed in this report was therefore designed to capture not only the large urban area known for its gang problems, but also geographical areas not traditionally known for having a large gang problem.
WHERE THE RIPPLE HAS NO EFFECT
The comparison of the four types of jails, which range from the farmland to the central urban location, were expected to vary by gang density. Gang density is the proportion of inmates who are gang members. So any comparison by the geographical area of the jail is also a comparison by gang density. The purpose of this section is to describe those findings from the survey where there are no significant differences in comparing the results for the four types of jails.
The findings in this section are those where no statistically significant difference exists when comparing the four types of jails. The concept "statistically significant difference" is used here in its common social science meaning as the .05 probability level. Thus, any probability level less than .05 (p < .05) is generally considered statistically significant, meaning it could occur by less than one out of twenty times by chance alone. Conversely, a difference is not significant if the probability level is greater than .05 (i.e., p > .05).
The basic four types of jails compared in this analysis are: Farmland, Heartland, Urban Outlying area, and Urban Central area. This type of classification system is not a typology commonly used in jail research. This is a construct that allows us to analyze the jails in terms of what many regard as the geographical origin of most of these gangs (i.e., Chicago and Cook County) and how their effects spread outwardly, like the ripple pattern from a rock hitting the surface of water. Unlike the concentric zone approach which looked at the city area as the total "zone", we are examining a large geographical section of the midwestern United States, where Chicago is assumed to be the starting point of many of the gangs analyzed here.
Attitudes and Beliefs
There exist some attitudes and beliefs for which no significant difference emerges in comparing inmates across the four types of jails. For example, the attitude about whether bail is excessive is not substantially different from one jail to the next. Overall, some 80.4 percent of the inmates felt that bail was excessive, but no significant difference exists in comparing the attitudes of inmates in the four types of jails.
Similarly, the belief that poor people are disproportionately represented in American jails today is another common denominator. No significant difference exists comparing results for this variable across the four types of jails. Overall, some 88.2 percent of the inmates did feel that the poor are overly represented in jails today.
No Difference In Whether They Would Take an $8.00/Hour Job
When given the scenario that if they had the training and the job were available, paying $8.00 per hour, most inmates say they would in fact take the job (94.9%). No significant difference emerges for this variable among inmates across the four types of jails. Actually, there is little variation to explain here, as almost all inmates indicated they would take the job. In the subsequent section of this report where we discuss significant differences, it is fair to alert the reader now that other differences appear to exist which have implications regarding employment for the gang member or inmate population.
No Difference In School Extracurricular Involvement
The survey asked the inmates whether they participated in any adult supervised extracurricular activities such as sports, band, service clubs, etc, while they were in school. The results show that 72.2 percent of all inmates report they did in fact have the benefit of such socialization services while in school. No significant difference exists comparing the inmates on this factor across the four types of jails (farmland, heartland, urban outlying, urban central).
No Difference In School Suspensions/Expulsions
The survey asked the inmates whether they have ever been suspended or expelled from a school for disciplinary problems. Overall, some 60.9 percent of all inmates reported they had in fact experienced this type of school disciplinary sanction. No significant difference emerged in comparing the results from the four different types of jails.
No Difference in Availability of Handguns
The inmates were asked in the survey "how difficult do you think it would be for you to get a handgun when you are released from jail". Some 14.2 percent of the inmates indicated it would be "very difficult". Some 11.7 percent indicated it would be "somewhat difficult". But most of the inmates (74.1%) indicated that it would not be difficult at all to acquire a handgun after their release from jail. No significant difference exists in comparing the level of difficulty for the availability of handguns to these inmates across the four types of jails.
No Difference in Engaging in Civil Law Suits
The survey asked the inmates "have you sued anyone in civil court for a wrong that was done to you". Some 15.4 percent of all inmates reported that they have sued someone in civil court. However, no significant difference emerged in comparing this variable across the four different types of jails.
Beliefs About The Deterrent Value of Strict Laws
The survey asked the inmates whether they felt we would have less crime if our laws were more strict. Only 27.5 percent of the inmates overall actually agreed that there was any deterrent value in stricter laws. Most of the inmates (72.5%) disagreed with the idea that less crime would ensue if our laws were more strict. Thus, no significant difference exists in this belief comparing the four different types of jails.
Believing The Legal System is Best Way To Settle Conflicts
The survey asked the inmates whether they agreed or disagreed with the idea that "the legal system is the best way to handle disputes and settle conflicts". Overall, 41 percent of the inmates agreed that the legal system is the best way to settle conflicts, and 59 percent disagreed. No significant difference emerged in comparing this factor across the four different types of jails.
Beliefs Supporting Lex Talionis: An Eye for An Eye
The survey asked the inmates whether they agreed or disagreed with the idea that "the best form of justice is simply an eye for an eye". This is the classical lex talionis concept of justice. Some 39 percent of the inmates agreed that the best form of justice is an eye for an eye. Most (61%) disagreed with this idea. No significant difference emerged in comparing this belief across the four different types of jails.
Beliefs About Whether it is Better to Live by the Law of the Jungle Than the Law of the Land.
The survey asked the inmates whether they agreed or disagreed with the belief that "it is better to live by the law of the jungle than the law of the land". Some 43 percent of the inmates, overall, agreed that it is better to live by the law of the jungle than the law of the land. On the other hand, some 57 percent of the inmates overall disagreed with this belief. However, no statistically significant difference emerged when comparing the four different types of jails (farmland, heartland, urban outlying, urban central) regarding this belief.
No Difference in Whether Inmates Have Used a Silencer
The survey asked the inmates whether they have ever used a gun equipped with a silencer or sound suppressor. Some 16.8 percent of the inmates overall reported that they used a gun equipped with a silencer. However, no significant difference emerged in comparing this variable across the four different types of jails.
No Difference in Availability of Explosive Devices
The scenario presented to the inmates in the survey was as follows: "after you are released from jail, and if you really needed to get your hands on some illegal explosives (dynamite, military explosives, hand grenades, etc), how hard would it be for you to do so". This vignette style question yielded the result that about half (47.8%) felt it would be "very hard" to acquire such explosive devices. Just over a fourth (28%) felt it would be "somewhat hard" to acquire explosive devices. But a fourth (24.2%) felt it would not be hard at all to acquire such illegal explosive devices. Further, no statistically significant difference emerged in comparing the availability of explosive devices across the four different types of jails. The implication here is that explosive devices are as hard or as easy to acquire in Chicago as they are in the rural areas; that is, geographical area makes no difference in this study which is limited to the midwestern United States.
No Difference in Preferences for Assault Rifles Vs. Handguns
The survey asked the inmates whether they would "prefer an assault rifle or a handgun for criminal activity". Some 22.7 percent of the inmates preferred the assault rifle for criminal activity. Most of the inmates (77.3%) preferred the handgun for criminal activity. Thus, the handgun not the assault rifle appears to be the weapon of choice for criminal activity among the majority of inmates surveyed here. Further, no statistically significant difference emerges in comparing this preference for assault rifles or handguns across the four different types of jails (farmland, heartland, urban outlying, and urban central areas).
Beliefs About Whether Most Criminals Are Morally Impaired
The survey asked the inmates whether they agreed or disagreed with the idea that "most criminals have not benefited from a fully developed set of moral beliefs". Most inmates (62%) agreed with the idea that most criminals are morally challenged individuals. Some 38 percent disagreed with this idea. However, no statistically significant difference emerged in comparing this variable across the four different types of jails. It should be pointed out, and will be detailed in a later section about significant differences, that a related question about their own moral development was significant and appears to further have some relationship to gang membership and type of jail surveyed as well.
Believing That There Are Many White People Who Are Not Prejudiced
The survey included two different questions about race relations, the one discussed was not significant comparing the four types of jail inmate populations. The survey item asked the inmates whether they agreed or disagreed with the idea that "there are many white people who are not prejudiced". Most of the inmates (79%) agreed that there are many white people who are not prejudiced. Still, 21 percent disagreed with this assumption about race relations. The fact is, however, that no significant difference emerged in comparing inmate beliefs about this factor across the four different types of jails studied (farmland, heartland, urban outlying, and urban central areas).
Believing That Physical Punishment is More Effective Than A Monetary Fine
The survey asked the inmates whether they agreed or disagreed with the idea that "physical punishment is more effective than a monetary fine". Some 36.5 percent of the inmates agreed that physical punishment is more effective than a monetary fine. Most of the inmates (63.5%), however, disagreed with the idea that physical punishment is more effective than a monetary fine. What we do know for sure is that no significant difference emerged in comparing inmate beliefs about this factor across the four different types of jails.
Whether There Are Many Effective Ways Available to Inmates to Settle Grievances Other Than Through Violence Or the Threat of Violence
The survey included a number of questions about conflict and dispute resolution and the predisposition of respondents along these lines, for possible analysis regarding gang membership. The specific language of the survey item asked whether the inmates agreed or disagreed with the idea that "there are many effective ways available to me to settle grievances with other persons other than violence or the threat of violence". Our research indicates that most inmates are certainly cognizant of such alternatives to violence, such that the problem may lay within the area of the extent to which impulsivity overrides the rational assessment of viable alternatives to settling grievances by means of violence. The data shows that 86.3 percent of the inmates agreed that there are such effective non-violent solutions to resolving grievances. Still, some 13.7 percent felt that there were not many effective recourses other than violence. The fact remains, however, that no statistically significant difference emerges on this variable comparing the four different types of jail inmate populations.
Whether They Have Ever Committed A Crime In Or On A Public Housing Project
It is important to point out here that two separate questions were included in the survey about public housing. The one about whether the inmate had ever lived in a public housing project was a variable significantly differentiated by type of jail, and is discussed in a subsequent section of this report. Another question about public housing revealed no such significant differences comparing types of jail inmate populations. This was the question about whether the inmate had ever committed a crime on the property of or inside an apartment of a public housing project. Some 21.8 percent of the inmates reported that they had in fact previously committed a crime in or on the property of a public housing project. However, no significant difference emerged in comparing the four types of jails regarding this aspect of committing crimes in public housing projects. It will be made further clear in examining other samples in this larger study about gangs and guns that there are apparently a number of "outsiders" who commit crimes in public housing projects, that is persons who have never actually lived there who commit crimes there.
Whether They Have Ever Assaulted A School Teacher
The survey asked the inmates whether they had ever assaulted a school teacher. Some 14.3 percent of the inmates, overall, reported that they had previously assaulted a school teacher. No significant difference emerged here comparing the four different types of jail inmate populations on this variable. This is, though, an important variable that contributes to the gang member profile discussed elsewhere in this report.
Whether They Feel Vulnerable to the "Risky Shift" Effect
The survey asked the inmates whether they agreed or disagreed with the self-description that "I am more likely to be violent in the situation of being with my group than I am as a lone individual". In the risky shift phenomenon a person may do things in a group context that they might be less prone to do if they were alone. Some 31.8 percent of the inmates felt they were vulnerable to this "risky shift" phenomenon. There was no significant difference for this variable in comparing the four types of jails studied.
Whether Natural Life Sentences Would Prevent Them From Using A Gun In A Crime
The survey asked the inmates "do you think that if judges had to give an automatic natural life prison sentence for using a gun in any crime that this would really prevent you from ever using a gun in a crime". Overall, 62.4 percent of the inmates expressed the belief that this would in fact prevent them from using a gun in a crime. Yet some 37.6 percent of the inmates indicated that this type of harsh anti-gun measure would not deter them from using a gun in a crime. No significant difference emerged in comparing the four types of jails on this variable.
Whether There Is Anything Police Can Do To Prevent A Gang From Getting Started In Any City in America
The survey asked the inmates whether they felt there was anything that police can to actually prevent a gang from getting started in any city in America. Most of the inmates (72.2%) felt there was nothing that the police could do to actually prevent a gang from getting started in any city in the USA. There was no significant difference comparing the four types of jails regarding this variable.
Whether The Gangs in Prison Basically Exist In The Same Form On the Streets
The survey asked the inmates whether the gangs in prison basically existed in the same form on the streets. Most of the jail inmates (70.9%) felt this was true. No significant difference existed comparing this variable across the four different types of jails. Obviously, groups like the Aryan Brotherhood were once thought to be only prison-based gangs. Even in Illinois prisons and jails the "Northsiders" were felt to be a prison-based only gang. The fact remains that both of these two types of gangs do exist outside of the social context of prison life. Most serious gangs in correctional institutions today have their counterparts, and usually their origins, in the community.
Whether The Same Internal Gang Codes of Behavior That Apply To Prison Gang Members Also Apply To The Members Of The Same Gang On The Streets
Two-thirds of the inmates (68.3%) reported that the gang codes apply in or out of a correctional facility. Examples of such gang codes of behavior behind bars have been reported in the literature (see American Jails, The Magazine of the American Jail Association, 1993, January-February, pp. 45-48). This research found no significant difference regarding this particular variable (whether gang codes or its internal written laws and rules) comparing the four different types of jails.
Whether The Inmates Ever Thought They Might Be Caught
So much traditional criminological thinking about law offenders tends to assume that human beings all benefit from rationally calculating the risks and benefits of current behavior in relationship to a potential future reward or punishment. The survey specifically asked the inmates "prior to committing a crime did it ever occur to you that you might be arrested". Two-thirds of the inmates (67.6%) did in fact acknowledge the possibility of being caught for their crime. Still, some 32.4 percent of the inmates state that it did not occur to them they might be arrested prior to committing a crime. No significant difference exists on this variable comparing the four different types of jails.
The Cultural Universal: The Natural Law of Opposition Group Formation Applied To Gangs
There is a social phenomenon applied to gangs (Knox, 1993) that is called the law of opposition group formation. It basically means that in any social system, where one group threat arises that its natural counterpart (Crips, then Bloods; Folks, then Peoples; etc, etc) arises as well. This cultural universal is very consistent with our knowledge about race relations, racial conflicts, and the rise of gangs along ethnic or racial dimensions. For example, a Chinese gang arising in response to threats or conflicts with a non-Chinese gang in a New York City environment; or a Middle Eastern gang like The Arabian Posse (aka: TAP Boyz, or TAPN for The Arabian Posse Nation) in a city like Chicago during the enmity towards Arab-Americans which arose during the Gulf War, which was when the TAP Boys first arose.
In the present research two techniques were used to ascertain the type of gang alliance that is typical of gangs in the midwestern United States: the difference between "Peoples" and "Folks". The present research cannot speak directly to the issue of the etiology of opposition gang group formation, that is when it occurs over time, but it can speak to the related issue of parity in gang alliances. That is to say, if the name of an inmate's gang is obtained and this is cross-checked with known alliances along the People-Folks dimension, and additionally if the inmate self-identifies as being aligned with either the People or Folks, then the present study is in fact able to assess whether one or the other type of alliance dominates or whether parity exists between these two alliances.
First it is important to note for the non-researcher, that the distinction called the "Peoples/Folks" dimension is not a classification system that covers all gangs in corrections, nor is it a typology that covers all gangs in any city in the midwestern portion of the United States nor anywhere else. There will always be gangs that exist that will be an exception to such larger "Gang Nation" identifications. Some white extremist gangs, for example, would not be included in any group identifying itself as "People" or "Brothers"; nor would white extremist gangs be likely to identify with "Folks" or the "Brothers of the Struggle" (B.O.S.). However, like the west coast distinction between B-Boys (Bloods) and C-Boys (Crips) there is such a large amount of data to work with that it is still worthwhile to examine differences if any in terms of their own density represented within the jail inmate population.
Figure 2 below shows the various type of individual gang names represented in the present sample of inmates from eight different jails in the midwest. These are classified outside of the correctional environment as either People or Folks in the exact same way they are on the streets. These are obviously the more structured, or more organized criminal gangs.
FIGURE 2
THE TYPES OF GANGS IN THE PRESENT JAIL INMATE STUDY
CLASSIFIED BY THE PEOPLE-FOLKS DIMENSION
OF GANG ALLIANCES
Peoples Gangs Folks Gangs
Four Corner Hustlers Almighty Ambrose
Black P. Stone Nation Brothers of the Struggle
Conservative Vice Lords Black Disciples
Familia Stones Black Gangster Nation
Honky Heads Black Souls
Insane Unknowns Gangster Disciples
Insane Vice Lords F.B.I.
Latin Counts Harrison Gents
Latin Kings Insane Spanish Cobras
Loco Boys Insane Two-Twos
Northsiders Simon City Royals
Mickey Cobras Maniac Disciples
Traveling Vice Lords Satans Disciples
Twin City Boys Two Six Nation
Unknown Vice Lords Young Latin Organization
Vice Lords
Figure 2 does not include the following gangs in this particular type of gang alliance: D.A.S.H. (Des Moines Area Skin Heads), Aryan Brotherhood, Aryan Nation, Aryan Resistance, Black Guerilla Family, Crips, Grim Reaper M.C. gang, Mexican Mafia, Nation of Islam, and the New Nation. The inmates of these gangs are not a part of the type of alliance known as the difference between "Peoples" and "Folks" gang nations.
Finally, we can now examine the data where we know the type of gang alliance (People or Folks) in relationship to the presence of such gang members within the gang member population of the four different types of jails. These results are shown in Table 1 below.
TABLE 1
DISTRIBUTION OF INMATES FROM PEOPLE OR FOLKS GANGS
BY THE FOUR DIFFERENT TYPES OF JAILS STUDIED
IN THE MIDWESTERN UNITED STATES
The Four Different Types of Jails Studied
Urban Urban
Alliance: Farmland Heartland Outlying Central
People 2 13 31 73
Folks 0 11 40 83
Chi-square = 3.07, p = .38
As seen in Table 1, no significant difference emerges in comparing the relative proportions of people or folks among inmate gangs represented in the jail inmate populations of the four different types of jails studied in the midwest. Relative parity exists along the dimension of this type of gang alliance in the jails studied here.
All this means is that inmate gang members do have such alliances and that because an individual gang member may be the only representative of his or her particular gang inside the facility, that the convenience of gang alliances allows the inmate to "ride with" other inmates along potential lines of conflict and violence. Some individual types of gangs do dominate, for example the GD's or the Gangster Disciples are the single largest individual "Folks" gang represented in this sample. Where an inmate may be the only one represented from his gang in the jail, there is obviously more than some benefit to having a larger alliance however tenuous it may be.
Clearly, once classified along the dimensions of the People and Folks alliance, inmate gang members are about equal in their proportions in the four types of jails studied here. This is consistent with the law of natural group opposition formation.
Within The Inmate Gang Population
Among the jail inmates who were self-reported gang members as measured by having ever joined a gang, there was no significant difference comparing the four types of jails by whether or not their gang was racially mixed, nor by whether the members are really treated equal. Nor was there any difference by whether or not they had attempted to leave the gang across the four types of jails. Nor was there any significant difference in terms of those who said they would rather die than give up their allegiance to the gang. Nor was there any significant difference in terms of whether the respondents agreed that gang membership helps in obtaining firearms. Nor was there any difference in whether or not the respondent indicated he had been "violated" (i.e., received a beating) by their own gang for a "violation" (one third of the gang members reported being violated). Nor was there any difference in whether or not they had been beaten up by a member of a gang they did not belong to (over half, 56.7%, of the gang members report such a beating).
THE RIPPLE EFFECT IN MIDWEST JAILS
Eight county jails in the midwestern United States were used to create the jail inmate sample. This was further divided into the geographical area for the jails to create four different types of jails: (1) farmland jails, (2) heartland jails, (3) urban outlying area, and (4) urban central area. Most of the gangs in Figure 2 discussed earlier --- the types of gangs represented among the inmates studied here --- have their origins in Chicago. It is well known from other research that gang proliferation in recent years basically means that today many Chicago-based gangs, just as many Los Angeles based gangs, can be found in many different parts of the United States. In fact, such gangs have been reported as being a problem throughout almost all states including Alaska, Hawaii, and the commonwealths of Guam, Puerto Rico, and the Virgin Islands.
Here we examine where the "ripple effect" has a significant impact. This means that statistically significant differences exist among the inmates the further out we get from the urban central area jail. If the stone hits the water, the issue is how far out does the ripple extend geographically, where the stone in this analogy is the fact that most of these gangs have their origins in Chicago, and also where it will be demonstrated that significant variation in gang density exists when comparing farmland, heartland, urban outlying, and urban central types of jails.
Thus, this section of the our larger study of gangs and guns deals specifically with those factors for which significant differences exist comparing the four different types of jails studied in the midwest.
Whether Inmates Feel More Betrayed By Friends and Family
The survey asked the inmates whether they felt that some persons who may have been friends or even family members find it easier to break promises to persons like themselves who are in jail. Table 2 provides the results of this inmate feeling of betrayal by the four different types of jails studied in the midwest.
TABLE 2 THE DISTRIBUTION OF FEELINGS OF BETRAYAL AMONG INMATES IN FOUR DIFFERENT TYPES OF JAILS IN THE MIDWEST The Four Different Types of Jails Urban Urban Farmland Heartland Outlying Central
Do you think that some
persons who may have
been friends or even
family find it easier
to break promises to
persons like yourself
who are in jail?
False 14 47 40 25
True 58 312 167 201
Percent True 80.5% 86.9% 80.6% 88.9%
Chi-square = 7.99, p = .04
As seen in Table 2, urban central area jail inmates feel the most betrayal, but this is not a straight linear difference in terms of the four types of jails.
Difference in Type of Family Structure
The inmates reported a simple majority of being from intact families (i.e., composed of both father and mother and siblings) outside of the large urban central area jail. However, a significant difference does exist here in comparing the four different types of jails. In the farmland jails 17.5 percent of the inmates indicated they come from a mother-only family structure, compared with 31.8 percent for inmates in the heartland, and 35.8 percent for inmates in the urban outlying area, and 49.1 percent for inmates in the urban central area. This is a very significant difference (p < .001) that is completely consistent with the expected "ripple effect". The further one goes from the central urban area jail, the less likely we are to find inmates who come from a mother-only family structure.
Difference in Percent Not Completing High School
The inmates also reported their last grade completed in the survey. From this data it was possible to calculate the percentage of inmates who were not high school graduates nor had they completed the GED. In the farmland jails 17.5 percent had not completed high school, compared with 31.8 percent for inmates in the heartland jails, compared with 35.8 percent of the inmates in the urban outlying jail, and finally 49.1 percent of the inmates in the urban central jail. This is a very significant difference (p < .001) and it is consistent with expected ripple effect: the further away from the central urban area jail one goes, the more likely the inmate is to have completed high school; or conversely, the closer one is to the central urban jail area the more likely the jail inmate is to have not completed high school or the GED.
Difference in Percent Who Have Never Had a Full-Time Job
The inmates were asked about their employment status at the time they were arrested and had to be detained in jail. One of the possible response modes was "never had a full time job". Thus, the inmate who never had a full-time job in his or her life is a measure of socialization and economic opportunity. Our data shows that there is a significant difference along this dimension comparing the four different types of jails. For example, in the farmland jails only 5.4 percent of the inmates reported that they had never held a full-time job in their life. Some 8.1 percent of the inmates in the heartland jails, and 9.3 percent of the inmates in the urban outlying jail, reported that they had never held a full-time job. In the urban central jail, however, some 27 percent of the inmates reported never having held a full-time job. This is a significant difference by employment experiences (p < .001) that is consistent with the ripple effect. The further one goes from the central urban area, the more likely the inmate is to have had such full-time job experience. The lack of any such previous full-time work experience is most profoundly felt in the urban central area.
Difference in Percent of the Inmates Who Have Previously Served Time in a Juvenile Correctional Institution
A significant difference emerges here, but is it is not strictly consistent with the ripple effect. Some 27 percent of the inmates in the farmland jails, 29.3 percent of the inmates in the heartland jails, 19.7 percent of the inmates in the urban outlying jail, and 30.7 percent of the inmates in the urban central jail all acknowledged they had in fact been incarcerated as juveniles. The only difference that does emerge here is a much lower percentage for juvenile incarceration among inmates in the urban outlying jail. This may be a statistical artifact. Thus, no consistent pattern or ripple effect appears here regarding whether the inmates have been previously incarcerated as a juvenile.
Difference in Whether the Inmates Have Permanent Tattoos
A significant difference emerges here in comparing the four types of jails by whether the inmates report having permanent tattoos, however it is not a pattern that is consistent with the ripple effect. In the farmland jails some 55.4 percent of the inmates had permanent tattoos. In the heartland jails, this was 48.3 percent and in the urban outlying jail this was 47.1 percent. In the urban central jail some 62.5 percent of the inmates had permanent tattoos. Thus, again, the difference that emerges here is the much higher proportion for this variable among inmates in the urban central area.
Difference in Gang Density: The Percentage of Inmates Who Report Having Ever Joined A Gang
Table 3 below shows the ripple effect for gang density among inmates in the four different types of jails studied in the midwest. Here we see a relatively consistent pattern in terms of gang density. Gang density is the percentage of inmates who are gang members in any correctional institution or system. Gang density is measured here by whether the inmates report having ever joined a gang. As previously discussed, for the most part the types of gangs included in this study are of the more organized variety where it not easy to just "walk away" from the gang. These are rather "blood in, blood out" types of gangs and are more formalized in their organization.
TABLE 3 THE DISTRIBUTION OF HAVING EVER JOINED A GANG AMONG INMATES IN FOUR DIFFERENT TYPES OF JAILS IN THE MIDWEST
The Four Different Types of Jails Urban Urban Farmland Heartland Outlying Central
Have you ever joined a gang?
NO 61 299 124 66
YES 13 63 83 165
Percentage Yes 17.5% 17.4% 40% 71.4%
Chi-square = 189.7, p < .001
The trend in Table 3 shows there is really no major difference in gang density among the farmland and heartland jails, but rather that a consistent ripple effect exists outward to the heartland area. These findings have enormous implications for recent federally funded research, where a national assessment of the gang problem in corrections basically concluded that only six percent of the American prison inmate population were gang members. This matter will be discussed in a later section of this report as the "six percenter issue".
Table 3 shows that gang density dissipates the further one moves from the urban central area. Geographical variation therefore exists in the gang density problem facing American jails today. All of the eight jails surveyed in the present research yielded inmates who were self-reported gang members. Obviously, this differs dramatically in terms of the ripple effect.
Gang Density: Using Current Gang Membership In Testing the Ripple Effect
Further evidence for this comes from a completely separate question in the survey which asked the inmate whether they were currently a member of a gang or gang organization. The results for this test are provided in Table 4.
TABLE 4 THE DISTRIBUTION OF CURRENTLY BEING A MEMBER OF A GANG AMONG INMATES IN FOUR DIFFERENT TYPES OF JAILS IN THE MIDWEST The Four Different Types of Jails Urban Urban Farmland Heartland Outlying Central
Are you currently a member
or associate of any gang or
gang organization? NO 60 311 149 81
YES 11 48 49 133
Percentage Yes 15.4% 13.3% 24.7% 37.8%
Chi-square = 166.0, p < .001
Thus, the same linear effect is noticed here in Table 4 for current gang membership as that found in Table 3 for having ever joined a gang. The effect extends consistently away from the urban central type of jail up to the point of the heartland. The farmland and heartland jails are relatively equal in gang density. The larger increase occurs the closer one gets to the urban central jail inmate population.
Difference In Having Ever Owned an Assault Rifle
The survey asked the inmates whether they have ever owned what could be considered an "assault rifle". While a significant difference emerges here comparing the four types of jails, the real effect is that higher effect from the urban central jail compared with all outlying and rural jail inmates. A third (33.3%) of the inmates in the urban central jail reported having owned an assault rifle. This compares with a fourth of the inmates in all other jails having ever owned an assault rifle (22.5% farmland; 21.6% heartland, and 23.7% urban outlying jail).
The Real Difference About Assault Rifles: Having Ever Used An Assault Rifle in Committing A Crime
A separate survey question asked whether they had ever used an assault in committing a crime. The results of this test are provided in Table 5.
TABLE 5 THE DISTRIBUTION OF USING AN ASSAULT RIFLE
IN COMMITTING A CRIME AMONG INMATES
IN FOUR DIFFERENT TYPES OF JAILS IN THE MIDWEST The Four Different Types of Jails Urban Urban Farmland Heartland Outlying Central
Have you ever used an assault
rifle in committing a crime?
NO 68 322 175 183
YES 3 35 29 45
Percentage Yes 4.2% 9.8% 14.2% 19.7%
Chi-square = 17.4, p = .001
Table 5 shows the consistent ripple effect for this variable about using an assault rifle in committing a crime as reported by inmates in the four different types of jails. The closer one is to the urban central area, the more the inmate is likely to report having used an assault rifle in committing a crime.
The Difference In Carrying Concealed Firearms Behavior
The replication of a survey item used in a well known previous study of firearms use among offenders (i.e., Rossi and Wright), asked the inmates about their concealed gun carrying behavior. What emerges here is that again a consistent ripple effect is seen. Some 77.9 percent of the farmland jail inmates report having never carried a concealed gun. This reduces to 63.7 percent for heartland jail inmates, and 57.5 percent for urban outlying jail inmates. Finally, in the urban central jail some 40.6 percent of the inmates report having never carried a concealed gun. Thus, the significance here is that the closer one is to the urban central area, the more likely the inmate is to have engaged in some form of this behavior of carrying a concealed firearm.
The Inconsistent Difference In Stealing Firearms
When the inmates were asked about whether they had ever stolen a pistol or rifle, the real difference that emerged here is the low rate of this behavior in the farmland area (12.5%). Some 25.4 percent of the heartland jail inmates admitted to having stolen a firearm, compared with 30.3 percent of the inmates in the urban outlying area jail. But 24.8 percent of the urban central jail inmates also admitted to stealing firearms. Thus the effect here is not consistent with the ripple effect.
The Ripple Effect in Offenders Using Sawed-Off Shotguns to Commit a Crime
The survey asked the inmates whether they had ever used a sawed-off shotgun to commit a crime. Here, as seen in Table 6, a consistent ripple effect emerges comparing the results for the four different types of jails.
TABLE 6 THE DISTRIBUTION OF USING A SAWED-OFF SHOTGUN
TO COMMIT A CRIME AMONG INMATES IN FOUR
DIFFERENT TYPES OF JAILS IN THE MIDWEST The Four Different Types of Jails Urban Urban Farmland Heartland Outlying Central
Have you ever used a
sawed-off shotgun to
commit a crime? NO 70 331 179 189
YES 2 28 23 38
Percentage yes 2.7% 7.7% 11.3% 16.7%
Chi-square = 16.8, p = .001
As seen in Table 6, the closer one gets to the urban central jail area, the more likely the inmate is to report having ever used a sawed-off shotgun to commit a crime. This ranges from a low 2.7 percent among jail inmates in the farmland jails, to 16.7 percent of the inmates who have used this type of firearm in committing a crime in the urban central area.
The Ripple Effect in Offender Firearm Ownership
The survey asked the inmates whether, since 1968, they have ever legally owned any firearm (this means legally purchasing it yourself, and registering it in your name). The data from the jail inmates studied here shows a consistent ripple effect where at the urban central area the inmates report the lowest extent of legally owning firearms, which increases incrementally the further one moves from the urban central area. These results are shown in Table 7.
TABLE 7 THE DISTRIBUTION OF HAVING LEGALLY OWNED A GUN SINCE 1968 AMONG INMATES IN FOUR DIFFERENT
TYPES OF JAILS IN THE MIDWEST The Four Different Types of Jails Urban Urban Farmland Heartland Outlying Central
Since 1968 have you ever
legally owned any firearm