59.3% United States  United States
8.8% United Kingdom  United Kingdom
4.9% Canada  Canada
4.1% Australia  Australia
3.5% Philippines  Philippines
2.6% Netherlands  Netherlands
2.4% India  India
1.6% Germany  Germany
1% France  France
0.7% Poland  Poland

Today: 57
Yesterday: 370
This Week: 1121
Last Week: 2221
This Month: 5709
Last Month: 6822
Total: 130308
PDF Print Email
User Rating: / 0
Prison & probation
Written by Avelardo Valdez   
Saturday, 22 April 1995 00:00

Illegal Drug Use, Alcohol and Agressive Crime Among Mexican-American and White Male Arrestees in San Antonio
Avelardo Valdez, Ph.D.*; Charles D. Kaplan, Ph.D.**; Russell L. Curds, Jr., Ph.D. *** & Zenong Yin, Ph.D.****
This research was supported by National Institute on Drug Abuse grant DA07234-02 to the Hispanic Research Center, University of Texas at San Antonio.
*Hispanic Research Center, The University of Texas at San Antonio.
**Department of Psychiatry, The University of Limburg, Maastricht, The Netherlands.
***Department of Sociology, The University of Houston.
****Division of Education, The University of Texas at San Antonio.
Please address reprint requests to Avelardo Valdez, Ph.D.,
Hispanic Research Center, University of Texas at San Antonio, 6900 North Loop West, San Antonio, Texas 78249-0603.
Journal of Psychoactive Drugs vol27(2) Apr-Jun 1995

The relationship between substance abuse and violence is more complex than earlier surmised. Research has indicated that being intoxicated on alcohol or other drugs increases aggressive behavior, such as homicide (Goldstein 1989; McBride 1981; Wolfgang & Strohm 1956), domestic violence (Frieze & Knoble 1980; Orford et al. 1975), and adolescent violence (Altschuler & Brownstone 1991; Hartstone & Hansen 1984; Tinklenberg et al. 1981). Other findings indicate that intoxication does not consistently lead to aggressive and violent behavior (Levinson 1983) and that the level of aggression varies from one drug to another (Jung 1994). Nonetheless, the process by which substance use is related to violent behavior is relatively unknown, particularly as it is differentiated among distinct populations.

There are comparatively few studies that focus on the direct relationship between drugs and violence. One of the studies that focuses on this subject argues that there are three possible linkages (Goldstein 1985). The first is the "psychopharmacological" type, in which the chemical effects of the drug lead to violence; the second is "economic compulsive," in which the need to support a drug habit leads to violence committed in the course of engaging in crimes to obtain money to buy drugs; and the third is "systemic," in which violence is committed in the course of drug dealing.

Others have suggested that there are social and cultural influences that mediate the relationship between violence and drugs among groups (Fagan 1990). These influences include distinct ethnic patterns and class and subcultural determinants. These researchers also indicated that the interaction of personality, social network, and situation or setting provides a powerful influence on violent behavior. Differences in drug use patterns between males and females, for instance, are often explained by differences in normative expectations and socialization.

Various studies have found that types of violent incidents tend to vary by ethnic group and type of drug used. Goldstein (1989) found that heroin, for instance, was five times more associated with economic-compulsive violence among Hispanics than for Blacks or Whites. Alcohol was the predominant drug involved in violent events for all groups; however, more so for Hispanics (28%) than Whites (17%) or Blacks (16%). In other findings, crack cocaine (which is associated with episodic violent acts) has been found to be more popular among Blacks than Mexican Americans. 1

Many studies cite the importance of alcohol as a catalyst for violent responses (Jung 1994) and as an important factor in physical aggression, such as rape (Abbey 1991; Klassen & Wilsnack 1986) and homicide (Spunt et al. 1990; Welte & Abel 1989). In several studies, alcohol was found to be present more often than any other drug (or other drugs combined) in homicide victims (Mugford 1992). Dembo and colleagues (1987) found alcohol to be more closely associated with the commission of homicides among over 400 persons incarcerated for homicide. Alcohol is also associated with domestic violence, particularly among blue-collar working-class men (Straus & Gelles 1986; Gerson 1978). However, many of these studies caution that violence might have occurred without the use of alcohol.

Research on Mexican-Americans suggests that there is a relationship between drug use and violence, particularly for men. Several studies have found that violent delinquency is related to early experiences within the immediate family, transferred to peer activities, and linked to the use of licit and illicit drugs (Carpenter et al. 1988; Newcomb & Bender 1988; Dembo et al. 1985; Horowitz 1983). Specifically looking at Texas Mexican-American youths, Watts and Wright (1990) found the strongest association (among several variables) between use of illicit drugs and violent delinquency. However, their interpretation "that drug use and participation in drug-using peer groups contributes greatly to delinquency and violence" requires more study.

Violence within the Mexican-American population is often focused on gang-related violence. Compared to other types of violence, gang violence seems to be more episodic, more likely to take place in public places, and more random, with less prior contact between young victims and suspects (Lopez & Mirande 1992; Moore 1990; Vigil 1988).

Violence associated with Mexican-American gangs is sensationalized through the media, and gang violence is perceived to be associated with illicit drugs. As a result of this attention, particularly in the Southwest, the public often perceives gangs and violence as predominantly a Mexican-American phenomenon (Lopez & Mirande 1992). This type of publicity, as well as that which focuses on Mexican-American prison gangs and on homicide and domestic violence by this group, reinforces the stereotype that Mexican-Americans are inherently more violent than the general population. As for other groups, however, the connection between drugs and violence is not as clear as one might expect.

The present study focuses on the relationship between aggressive crime (of which violent crime is a subtype) and alcohol and other drug use among Mexican-American and White male arrestees in San Antonio. It also explores the differential effect of alcohol and other drug use among this particular sector of Mexican-Americans and Whites, and how ethnicity, class, and culture mediate these differences.


Data for analysis were obtained from the Drug Use Forecasting (DUF) program in San Antonio, Texas. The DUF program is a national monitoring system of the Department of Justice that both administers urine tests and interviews adult arrestees in local jails on a quarterly basis in 24 cities throughout the United States. Both large cities (e.g., Houston and Miami) and smaller cities (e.g., Ft. Lauderdale and San Antonio) are represented. The purpose of the DUF program is to provide each city with estimates of drug use among arrestees and information for detecting changes in drug use trends. The DUF interview also focuses on drug use and criminal history, demographic background, and other variables (U.S. Department of Justice 1990; Wish & Gropper 1990). The validity of drug test data of arrestees has been demonstrated in numerous studies (Smith & Polsenberg 1992; Balenko, Fagan & Chin 1991; Goldkamp, Gottfredson & Weiland 1990). The limitations of the DUF methodology have also been recognized and include problems of nonrepresentativeness of the criminal population and data adequacy and completeness (Goldstein et al. 1992).

Enrollment in the DUF sample occurs during booking, when arrestees are approached about participating in the study. Data are collected in central booking stations in the 24 cities for approximately seven days each quarter. Local staff obtain voluntary and anonymous urine specimens and interviews from the arrestees. Approximately 225 males and 100 females are sampled in each site. DUF inclusion criteria limit the number of male arrestees charged with the sale or possession of drugs, since these persons are more likely to have used drugs, and would exaggerate the estimate. DUF samples are also selected so as to limit the number of persons charged with petty offenses or other drug offenses. Persons charged with traffic offenses or vagrancy are excluded. Where there is a choice of arrestees, persons are selected by arrest charge in the following order: nondrug felony, nondrug misdemeanor, drug felony, and drug misdemeanor. Because of their small numbers in most jurisdictions, all female arrestees are approached, regardless of charge. In practice, most persons in the DUF samples are charged with nondrug felony offenses.

The response rate in DUF samples is high, with over 90% of those arrestees approached agreeing to be interviewed and over 80% of these providing urine samples. Urine specimens are analyzed for 10 drugs: cocaine, opioids, marijuana, PCP, methadone, benzodiazepines, methaqualone, propoxyphene, barbiturates, and amphetamines. The urine test can detect use of most drugs up to three days before testing, and marijuana and PCP use can be detected up to several weeks before testing.

The San Antonio DUF sample used in this study was collected in 1992. The sample consisted of 1,014 arrestees drawn from over 25,000 arrestees incarcerated in San Antonio in 1992. The characteristics of the 1992 San Antonio sample are presented in Table I. Nearly 60% of the sample were of Mexican origin, 21% were White, and 16% were B lack. 2 The Mexican-American subsample is roughly equal to the proportion of this group in the larger San Antonio metropolitan area. Compared to the other 23 DUF cities, San Antonio has the largest Hispanic arrestee population sample. Female arrestees comprised nearly one-third of the sample. The sample is relatively young and uneducated, with 46% under 25 years and 40% not having completed high school. Around one-third of the sample is employed full time. These socioeconomic characteristics correspond to the general population of San Antonio, which is one of the poorest large metropolitan areas in the United States (Partnership for Hope 1991).

The San Antonio sample has comparatively lower drug use compared to other DUF cities, with slightly over 50% testing positive for drug use. Thirty percent (30%) of the sample tested positive for cocaine, 24% for marijuana, and 14% for opioids. Only a small percentage of the sample tested positive for the other seven drugs. A major difference from other DUF cities is that in San Antonio the percentage of women testing positive was lower than for men. This difference may be related to the high proportion of Mexican-American women among the San Antonio arrestees. Mexican-American women have been found to subscribe to more traditional gender roles, which may discourage the use of illicit drugs (Mirande 1979).



Analysis employed descriptive and inferential procedures for multiple contingency tables. The first aim of the analysis was to describe the patterns of variation that existed in contingency tables in terms of percentage differences. The second aim of the analysis was to abstract and summarize the results of the descriptive procedure by comparing various models representing the interaction of crime, alcohol and other drug use, and ethnic variables.

Using a four-way asymmetrical analysis, logitmodels were tested to examine the relationships between the response variable, the types of crimes charged (nonaggressive versus aggressive) and a set of exploratory variables: ethnicity (White versus Hispanic), drug test results (positive versus negative), and alcohol use (infrequent versus frequent). The logic-analysis allows the specification of a subset of relevant models to be tested for their adequacy of fit (Kennedy 1992). For the current study, the

. logic-models were built by selecting only the interaction models in which the types of crimes charged was present. A Baysean approach to log-linear analysis was applied whereby the statistical testing was oriented not to determine whether various models simply fit the data, but rather to determine the relative value of a given model compared to the other candidate models (Raftery 1986). Component G 2 tests as well as Baysean Information Coefficients (BIC) were calculated to evaluate the log-linear models using the SPSS PC program (Kennedy 1992).

Variables for analysis were recoded into dichotomous or trichotomous categorical variables. The crime (aggressive-nonaggressive) variable was based on the charge for which the arrestee was booked. Aggressive crimes included extortion/threat, homicide, kidnapping. robbery, sex offenses, assault, family offenses, obstruction of police, public peace/disturbing, and sexual offense (rape). Nonaggressive crimes included burglary, prostitution, drug sale, weapons, fight/escape/bench, forgery, fraud, larceny/ theft, probation/parole violation, stolen property, stolen vehicle, under the influence, drug possession, fare beating, liquor, obscenity, driving while intoxicated (DWI), and driving violations (not DWI). Given the nature of these aggressive crimes, they represent the major behaviors associated with violence. 3

The DUF interview only records self-reported frequency of alcohol use, making it impossible to measure heavy drinking in the manner corrunon in alcohol epidemiological surveys, which usually employ some sort of quantity-frequency index. Thus, alcohol use was categorized as a dichotomy of frequent and occasional use based on an item that measured how many times per month the arrestee drank. An arrestee was coded as "frequent" if drinking ten or more days per month and "occasional" if less than that threshold. This cutoff point was based on previous studies that define different categories of drinking (Clark & Midanik 1982; Cahalan, Cisin & Crossley 1969). A respondent who drank ten times or more could not be classified as a "heavy" drinker according to this previous work, but would fall into an intermediate level this study categorizes as "frequent."

Significant differences between male and female arrestees on aggressive crimes as well as the females' low levels of drug use resulted in the exclusion of female arrestees from further analysis (see Table 1). Similar to other crime data, the San Antonio data indicate that female arrestees comprised only a small percentage of aggressive crime offenders (Blount et al. 1991). In this sample, 10 (3%) of the aggressive crimes were committed by women. In order to focus the analysis, only the two largest groups in the sample, Mexican-Americans and Whites, were included for further analysis. Blacks were excluded from further analysis because of socioeconomic characteristics distinct from both Whites and Mexican-Americans that might confound the differences in the ethnicity-nonethnicity variable (National Institute on Drug Abuse 1991; Partnership for Hope 1991). Thus, the ethnicity variable consisted of a dichotomy between Whites and Mexican-Americans.

Bivariate and trivariate contingency tables were crosstabulated and descriptively evaluated in terms of percentage differences. On the basis of this exploratory analysis, loglinear models were constructed using a forward elimination procedure in order to minimize the suppression of subtle interaction effects that hypothetically were of interest. Alternative models were tested and compared with independent procedures, BIC and Component Chi-Square values, in order to increase the reliability of the model fitting



The relationship between drug use (testing positive or negative for any drug) and aggressive crimes among Mexican-American and White males is presented in Table 11. In comparing Mexican-American and White men who tested negative, Whites were much more likely to be arrested for aggressive crimes (Chi-square (1)=6.93,p=.009). However, for those who tested positive, there was a trend for more Mexican-American men to be arrested for aggressive crimes, although this was not statistically significant 4

The relationship between testing positive for cocaine, marijuana, and opioids (heroin) and aggressive crimes is presented in Table III. Although none of these relationships proved to be statistically significant, the data indicate that testing negative for any of the three drugs was more associated with aggressive crimes than was testing positive. The relationship holds for both Mexican-Americans and Whites, although it was slightly higher for Whites. That is, persons who were under the influence of drugs when arrested were less likely to be involved in aggressive crimes than those who were not under the influence of drugs. This was especially the case for opioid users, followed by marijuana users and cocaine users. In fact, arrestees testing negative were more than twice as likely to have committed an aggressive crime. In comparing Mexican-Americans and Whites, Mexican-Americans were more likely to test positive for any of these drugs than were White arrestees, particularly for marijuana. Overall, Mexican-Americans who tested positive were more likely to engage in aggressive behavior.

The relationship between alcohol use and aggressive crime shows that arrestees, as a whole, committed fewer aggressive than nonaggressive crimes. Although not presented here, levels of alcohol use were not related to levels of aggressive crime, with 42% of both occasional and frequent users committing aggressive crimes, while 58% committed nonaggressive crimes. This was the case for both Mexican-American and White arrestees (see Table II), although there is a trend for Whites to more likely engage in aggressive behavior regardless of level of alcohol use.

The interaction of alcohol and other drug use on aggressive crime is presented in Table II. In observing the subtotals for each category, combining Mexican-Americans and Whites, frequent alcohol use and testing negative for drugs has the highest association with aggressive crime (60%). The lowest association with aggressive crimes were those that were frequent drinkers and tested positive for drugs (27°!0), this was even lower than the occasional drinkers (34%'0) testing positive for drugs. In other words, arrestees were less likely to commit an aggressive crime if they were under the influence of drugs than if they were not under the influence of drugs, even when one controls for alcohol. This difference was statistically significant for both occasional (Chi-square (1)=9.91, p=.002) and frequent drinkers (Chi-square (1)=16.25, p=.000).

In comparing Mexican-Americans and Whites, one of the major differences is that Whites who are frequent drinkers, but who test negative for other drugs, are more associated with aggressive crimes than Mexican-Americans (see Table IV). Among arrestees who are frequent alcohol users and test negative for drugs, 52% of MexicanAmericans were arrested for aggressive crime, compared to 83°!0 of Whites. The percentage for Whites is the highest rate of involvement in aggressive crime appearing in Table IV. On the other hand, one of the striking differences between Mexican-American and White arrestees is for frequent drinkers who test positive for drugs. MexicanAmericans in this category are twice (31%) as likely to be arrested for aggressive crimes as Whites (14%). These data indicate that combining drug use and alcohol is much more associated with aggressive crimes for Mexican-Americans than for Whites.

The results of the log-linear analysis conducted to abstract and summarize the complex relationships found in the contingency tables are presented in Table V. Seven alternative models are specified representing the associations and interactions of crime with drug use, alcohol, and ethnic categorical variables. Model 2, representing the association of drug use and crime, fits the data best (G 2=25.10, df--l, p<000). However, the other models do provide insights into the relationships among the variables. Model 1, representing the crime by alcohol association, is not significant. Alcohol does become significant in Model 7, the fully saturated model in which all interactions are included (G 2=4.52, df=1, p<05) and appears as a trend (G 2=3.36, df=l, p<10) only in Model 4 in association with drug use. Ethnicity becomes a relevant variable in interaction with drug use in Model 5 (G 2=4.82, df=1, p<05), as well as in the fully saturated model. It should be noted that alcohol and ethnicity become relevant to aggression only as mediators of the strong drug use and crime association.

The results of the BIC test generally confirm the results obtained from the component G 2 test. Model 2, with a BIC value of -18.60, is clearly the best model. The one exception is Model 3, which is comparatively a strong model by the BIC test, but is nonsignificant by the component G 2 test. This discrepancy is probably an artifact of the procedure having to do with the relatively small change in the Residual G 2 effected by Model 3 after the large change effected between Model 1 and Model 2. Thus, by the BIC test, Model 3 must be seen as indeed important. The crimeethnicity interaction indicates that ethnicity is almost as important an influence on aggressive crime as the (non)influence of drug use. White arrestees who do not use drugs are the subgroup most likely to be involved in aggressive crimes, while Whites who frequently drink and use drugs will be more likely not to commit an aggressive crime.

Together with the mediating variable of alcohol, ethnicity greatly modulates the generally strong association of (non)drug use on aggression. Mexican-Americans who both frequently drink and also take other drugs will be more likely to commit aggressive crime than Whites in that subgroup.




The results of this analysis indicate a complex but interpretable pattern between alcohol and other drug use variations and aggressive crime. Consistent with the research literature, alcohol use in San Antonio was found to be associated with aggression and, furthermore, to be more associated with aggression than drug use (see De La Rosa, Lambert & Gropper 1990). However, the conditioning variable of ethnicity was found to interact with the drug and alcohol variables to produce the novel result that in San Antonio not using drugs is generally more associated with aggressive crimes than using drugs. The explanation for this pattern hits at the very core of the frequent pleas that the drug and violence nexus is not a simple and straightforward causal relationship. Numerous intervening and conditional variables come into play to modulate the drug use and violence relationship.

The methodological importance of including interaction terms in future research is apparent given the results of this study. Alcohol is more positively associated with aggressive crimes than are other drugs, but it becomes a clearly significant pattern for Whites who drink frequently. However, when frequent alcohol use is combined with other drug use there is a closer association with aggressive crime for Mexican-Americans than for Whites. These findings are supported by other researchers (see Fagan 1990) who have concluded that alcohol does not consistently lead to aggressive and violent behavior in a simple and straightforward way. This suggests that there are many social and cultural influences that mediate the relationship of violence and drugs. These influences, subtle in effect, are nonetheless critical.

An important New York research program focused on drugs and violence has been especially sensitive to these subtle interactions (Goldstein et al. 1991; Johnson et al. 1985). In one study, the New York group specifically focused on race/ethnicity and gender differences in the drugs-violence relationship (Spunt et al. 1990). They found that alcohol was the predominate drug involved in violent events for all groups. In contrast to findings of the present study, however, this result is stronger for Hispanics than for Whites or Blacks.

A plausible explanation for this difference is in the sampling variations in the two cities. In New York, a snowball sample was drawn from drug users and distributors who lived on the Lower East Side of Manhattan. Many of the initial contacts came from a local methadone maintenance treatment program. In contrast, the San Antonio sample was more broad based and included respondents who were not involved in the hard-core drug scene. In particular, this was the case for the New York Hispanic sample, which represented only a narrow segment of the Hispanic population. The same sampling differences are evident in regard to comparing the White samples in each city. What may be concluded is that in hard-core drug user groups in a context of a "culture of violence," alcohol is more connected to violence than other drugs for Hispanics than it is for Whites. This interaction was also seen in San Antonio, where frequent-drinking Hispanics who tested positive for other drugs were more involved in aggressive crime than any other group, other than frequent drinking nondrug tested Whites. In a more general population, such as represented by the San Antonio sample, violence seems to be more associated with alcohol for Whites than for Hispanics.

Another factor that may explain the city differences is that the Hispanic population in San Antonio is predominantly Mexican-American, while in New York it is Puerto Rican, Cuban, and Dominican. Different drinking subcultures may occur among Hispanic groups, accounting for the difference in results found in the two cities. Cultural differences between Whites in New York and San Antonio also come into play. The strong association with alcohol and aggressive crimes among Whites in Texas may be due in part to larger cultural values in the American South where alcohol-related aggression forms a persistent and uniform pattern that has been described as "belligerent drinking." 5

Apart from these differences, the New York results parallel the situation found in San Antonio in several important ways. In general, alcohol, not the use of other drugs, is most related to aggression; and, as in San Antonio, there are differences between Whites and Hispanics. In San Antonio, the highest percentage of aggressive crimes was committed by Whites who drank frequently and tested negative for drugs (83%), a full 31 points higher than their Mexican-American counterparts who also drank frequently and tested negative. The same pattern, though weaker, could be observed for those who drank only occasionally and also tested positive and also tested negative.

The San Antonio research shows that the use of drugs can have an effect on aggressive crime, and subsequently violence, for specific subgroups defined by interactions of alcohol and ethnicity. While Whites overall engage in slightly more aggressive crime than Mexican-Americans, one of the exceptions is Mexican-Americans who are frequent drinkers and use other drugs. However, even within this most violent Mexican-American subgroup, the percentage is small (31 %) compared to most of the other subgroups of Mexican-Americans and Whites, which average about 50%. The drug/alcohol nexus is noteworthy because of the difference between Mexican-Americans (31 %) and Whites (14%) in this category. For Whites the drug effect is most extreme. Whites who both test positive and drink frequently are, by far, the lowest subgroup for aggression. For Whites, at least, the drugs seem to be functioning to "soften" the hard frequent drinking pattern and, in contrast to the Mexican-Americans, to mitigate greatly the belligerent drinking culture.

The results of this research suggest that public perceptions linking violence with Mexican-Americans, particularly when compared to Whites, obscure the true picture. In this study, a select portion of the Mexican-American population, those arrested, is associated with less violent type crimes than their White counterparts. In fact, the White frequent drinker who does not use other drugs is the most violent group. Thus, prevention efforts should not only be targeted to drug use in the Mexican-American community (although this is also indicated in the results), but also toward Whites' belligerent drinking. Furthermore, more comparative research involving other DUF cities should be conducted to explore the ethnic and alcohol interaction effects on drug use and violence under different conditions.




1. These data are from a series of focus groups of heroin and cocaine users in San Antonio conducted by this research team as part of a larger project on intravenous drug users.

2. The DUF questionnaire uses the generic term Hispanic to identify all persons of Latin American and Caribbean backgrounds. In the San Antonio DUF sample,

it is assumed that the majority of those identified as Hispanic are of Mexican origin since approximately 90% of the Hispanics in this urban area are of this group. White refers to all non-Hispanic Whites.

3. A distinct violent crime variable was constructed using the Uniform Crime Report (UCR) schema. UCR violent crimes include homicide, robbery, sex offenses, and assault. Because of the relatively low percentage (9%) of violent crimes in the sample, and the high percentage of aggressive-nonaggressive crimes (31%), the analysis focuses on the latter.

4. The same relationship was found when using the Uniform Crime Report (UCR) constructed violence variable, although it was slightly more pronounced. Twice as many Whites were likely to be engaged in violent crime than Mexican-Americans when testing negative for drugs, although there was no difference between these groups when testing positive. Although not presented here, the UCR violence crime variable generally paralleled the findings associated with aggressive crime.

5. Novelist Cormac McCarthy (1985) used the term "culture of redemption" to describe Texas' violent frontier tradition among early White settlers. These values may help explain why the state of Texas has the most liberal homicide defense laws in the nation, such as the "fighting words" law. These factors should be considered when addressing issues of violence among Whites in Texas as compared to other regions of the United States.



Abbey, A. 1991. Acquaintance rape and alcohol consumption on college campuses: How are they linked? Journal of American College Health 39:165-69.

Altschuler, D.M. & Brownstone, P.J. 1991. Patterns of drug use, drug trafficking, and other delinquency among inner-city adolescent males in Washington D.C. Criminology 29:589-622

Balenko, S.; Fagan, J. & Chin, K. 1991. Criminal justice responses to crack. Journal of Research in Crime and Delinquency 28:55-74.

Blount, WR.; Danner, TA.; Vega, M. & Silverman, I.J. 1991. The influence of substance use among adult female inmates. In: R. Dembo. & L.E. Williams (Eds.) Drugs and Crimes. Lanham, Maryland: University Press of America.

Cahalan, D.; Cisin, I.H. & Crossley, H.M. 1969. American Drinking Practices: A National Study of Drinking Behavior. New Brunswick, New Jersey: Rutgers Center for Alcohol Studies.

Carpenter, C.; Glassner, B.; Johnson, D.D. & Loughlin, J. 1988. Kids, Drugs, and Crime. Lexington, Massachusetts: Lexington Books.

Clark, W .B. & Midanik, L. 1982. Alcohol use and alcohol problems among U.S. adults: Results of the 1979 national survey. In: Alcohol Consumption and Related Problems. Monograph No. 1. Rockville, Maryland: National Institute on Alcohol Abuse and Alcoholism.

De La Rosa, M.; Lambert, E.Y. & Gropper, B. 1990. Introduction: Exploring the substance abuse-violence connection. In: M. De La Rosa, E.Y. Lambent & B. Gropper. Drugs and Violence: Causes, Correlates, and Consequences. NIDA Research Monograph 103. Rockville, Maryland: National Institute on Drug Abuse.

Dembo, R.; Grandon, G.; LaVoie, L.; Schmeidler, J. & Burgos, W. 1985. Parents and drugs revisited: Some further evidence in support of social learning theory. Criminology 24 (1): 85-104.

Dembo, R.; Williams, L.; Wish, E.D.; Dertke, M.; Berry, E.; Getreu, A.; Washburn, M. & Schmeidler, J. 1987. The relationship between physical and sexual abuse and illicit drug use: A replication among a new sample of youths entering a juvenile detention center. International Journal of the Addictions 23 (11): 1101-123.

Fagan, J. 1990. Intoxication and aggression. In: M. Tonry & J.Q. Wilson (Eds.) Drugs and Crime. Chicago: University of Chicago Press.

Frieze, I.H. & Knoble, J. 1980. The effect of alcohol on marital violence. Paper presented at the annual convention of the American Psychological Association, Montreal.

Gerson, L.W. 1978. Alcohol-related acts of violence: Who was drinking and where the acts occurred. Journal of Studies on Alcohol 39:1294296.

Goldkamp, J.S.; Gottfredson, M.R. & Weiland, D. 1990. Pretrial drug testing and defendant risk. Journal of Criminal Law and Criminology 81 (3): 585-652.

Goldstein, P.J. 1989. Drugs and violent crime. In: N.A. Weiner & M.E. Wolfgang (Eds.) Pathways to Criminal Violence. Beverly Hills, California: Sage.

Goldstein, P.J. 1985. The drugs/violence nexus: A tripartite conceptual framework. Journal of Drug Issues 15:493-506.

Goldstein, PJ.; Bellucci, PA.; Spunt, B.A. & Miller, T. 1991. Volume of cocaine use and violence: A comparison between men and women. Journal of Drug Issues 21 (2): 345-67.

Goldstein, PJ.; Henry, J.; Brownstein, H.B. & Patrick, J. 1992. Drug related homicide in New York: 1984 and 1988. Crime and Delinquency 38 (4): 459-76.

Hartstone, E.C. & Hansen, K.V. 1984. The violent juvenile offender: An empirical portrait. In: R.A. Mathias, P DeMuro & R.A. Allinson (Eds.) Violent Juvenile Offenders: An Anthology. San Francisco: National Council on Crime and Delinquency.

Horowitz, R. 1983. Honor and the American Dream. New Brunswick, New Jersey: Rutgers University Press.

Johnson, B.D.; Goldstein, PJ.; Preble, E.; Schmeidler, J.; Lipton, D.S.; Spunt, B. & Miller, T. 1985. Taking Care of Business: The Economics of Crime by Heroin Abusers. Lexington, Massachusetts: Lexington Books.

Jung, J. 1994. Underthe Influence: Alcohol and Human Behavior. Pacific Grove, California: Brooks/Cole.

Kennedy, J.J. 1992. Analyzing Qualitative Data: Log-Linear Analysis for Behavior Research. New York: Praeger.

Klassen, A.D. & Wilsnack, S.C. 1986. Sexual experiences and drinking among women in a U.S. national survey. Archives of Sexual Behavior 15:363-92.

Levinson, D. 1983. Alcohol use and aggression in American subcultures. In: R. Room & G. Collins (Eds.) Alcohol and Distribution: Nature and Meaning of the Link. Washington, D.C.: U.S. Department of Health and Human Service, U.S. Health Service.

Lopez, J. & Mirande, A. 1992. The gangs of Orange County: A critique and synthesis of social policy. Atzlan 19 (1): 125-46.

McBride, D.C. 1981. Drugs and violence. In: J.A. Inciardi (Ed.) The Drug-Crime Connection. Beverly Hills, California: Sage.

McCarthy, C. 1985. Blood Meridian. New York: Random House.

Mirande, A. 1979. La Chicana: The Mexican American Woman. Chicago: •   University of Chicago Press.

Moore, J. 1990. Gangs, drugs, and violence. In: M. De La Rosa, E.Y. Lambert & B. Gropper. Drugs and Violence: Causes, Correlates, and Consequences. NIDA Research Monograph 103. Rockville, Maryland: National Institute on Drug Abuse.

Mugford, S. 1992. Licit and illicit drug use, health costs and the crime connection in Australia: Public views and policy implications. Contemporary Drug Problems 19:351-85.

National Institute on Drug Abuse. 1991. National Household Survey on Drug Abuse: Population Estimates. DHHS Publ. No. (ADM) 921887. Washington, D.C.: U.S. GPO.

Newcomb, M.D. & Bender, P .M. 1988. Consequences of Adolescent Drug Use: Impact on the Lives of Young Adults. Beverly Hills, California: Sage.

Orford, J.; Guthrie, S.; Nicholls, P; Oppenheimer, E.; Egert, S. & Hensman, C. 1975. Self-reported coping behavior of wives of alcoholics and its association with drinking outcome. Journal of Studies on Alcohol 36:1254-267.

Partnership for Hope. 1991. Pride and Poverty: A Report on San Antonio. San Antonio, Texas: Partnership for Hope.

Raftery, A.R. 1986. Choosing models for cross-classifications. American Sociological Review 51:145-46.

Smith, D.A. & Polsenberg, C. 1992. Specifying the relationship between arrestee drug test results and recidivism. Journal of Criminal Law and Criminology 83 (3): 364-77.

Spunt, B.J.; Goldstein, PJ.; Bellucci, PA. & Miller, T 1990. Race/ ethnicity and gender differences in the drugs-violence relationship. Journal of Psychoactive Drugs 22 (3): 293-303.

Straus, M. & Gelles, R.J. 1986. Societal change and change in family violence from 1975 to 1985 as revealed by two national surveys. Journal of Marriage and the Family 48:465-79.

Tinklenberg, J.R.; Murphy, P.; Murphy, PL. & Pfefferbaum, A. 1981. Drugs and criminal assaults by adolescents: A replication study. Journal of Psychoactive Drugs 13 (3): 277-87.

U.S. Department of Justice. 1990. Uniform Crime Reports 1989. Washington, D.C.: U.S. GPO.

Vigil, J.D. 1988. Group process and street identity: Adolescent Chicano gang members. Ethos 16 (4): 421-45.

Watts, W.D. & Wright, L.S. 1990. The drug use-violent delinquency link among Mexican-Americans. In: M. De La Rosa, E.Y. Lambent & B. Gropper. Drugs and Violence: Causes, Correlates, and Consequences. NIDA Research Monograph 103. Rockville, Maryland: National Institute on Drug Abuse.

Welte, J.W. & Abel, E.L. 1989. Homicide: Drinking by the victim. Journal of Studies on Alcohol 50:197-201.

Wish, E.D. & Gropper, B.A. 1990. Drug testing by the criminal justice system: Methods, research, and application. In: M. Tonry & J.Q. Wilson (Eds.) Drugs and Crime. Chicago: University of Chicago Press.

Wolfgang, M.E. & Strohm, R. 1956. The relationship between alcohol and criminal homicide. Quarterly Journal of Studies on Alcohol 17A11-25.