FINDINGS

Seventy-eight studies met our criteria. The statistical techniques used across these studies included basic difference-of-means tests, chi-square tests, bivariate correlations, and multiple types of multivariate analysis. Fifty-four of these studies included measures of consumption and/or availability of alcohol; 19 included measures of marijuana; 11 cocaine; 4 crack cocaine; 12 heroin/opiates; 7 other types of drugs (e.g., barbiturates, amphetamines, PCP, etc.); 29 undifferentiated measures (e.g., “drug use”); and 8 combined measures of alcohol and illegal drug consumption.

Alcohol

Alcohol Consumption

Table 11.1 summarizes the comparison-level vote count. Turning first to the separate tests of alcohol and violent or nonviolent crime, we see that measures of various levels of alcohol consumption were positively related to violence in

Independent

Total

X

%

О

%

Variable

Comparisons

Comparisons

Comparisons

Alcohol

Consumption

Violent Nonviolent Violent vs. Nonviolent

  • 78
  • 105
  • 39
  • 53
  • 66
  • 9
  • 68
  • 62.9
  • 23.1
  • 0
  • 0
  • 4
  • 0
  • 0
  • 10.3

Alcohol

Intoxication

Violent Nonviolent Violent vs. Nonviolent

  • 9
  • 10
  • 37
  • 9
  • 9
  • 27
  • 100
  • 90
  • 73
  • 0
  • 0
  • 1
  • 0
  • 0
  • 2.8

Alcohol

Availability

Violent Nonviolent Violent vs. Nonviolent

  • 42
  • 23

n/a

  • 29
  • 20
  • 69.1
  • 87
  • 0
  • 0
  • 0
  • 0

Undifferentiated1 Illegal Drug Consumption

Violent Nonviolent Violent vs. Nonviolent

  • 64
  • 85
  • 32
  • 21
  • 45
  • 11
  • 32.8
  • 52.9 34.4
  • 4
  • 1
  • 16
  • 6.3
  • 1.2
  • 50

Marijuana

Consumption

Violent Nonviolent Violent vs. Nonviolent

  • 39
  • 65
  • 11
  • 21
  • 42
  • 5
  • 53.8
  • 64.6
  • 45.5
  • 1
  • 2
  • 3
  • 2.6
  • 3.1
  • 27.3

Cocaine

Consumption

Violent Nonviolent Violent vs. Nonviolent2

  • 16
  • 26
  • 6
  • 7
  • 7
  • 2
  • 43.8
  • 26.9 33.3
  • 2
  • 1
  • 4
  • 12.5
  • 3.8
  • 66.7

Crack Cocaine Consumption

Violent

Nonviolent

  • 6
  • 14
  • 4
  • 10
  • 66.7
  • 71.4
  • 0
  • 1
  • 0
  • 7.1

Heroin, Opiates, “Narcotics”

Violent Nonviolent Violent vs. Nonviolent

  • 19
  • 40
  • 3
  • 8
  • 16
  • 2
  • 42.1
  • 40
  • 4
  • 1
  • 0
  • 21.1
  • 2.5
  • 0

Other Illegal Drug3

Consumption

Violent Nonviolent Violent vs. Nonviolent

  • 17
  • 57
  • 9
  • 10
  • 30
  • 6
  • 58.8
  • 60
  • 2
  • 1
  • 0
  • 11.8
  • 1.8
  • 0

“Substance Use” (Combined Alcohol and Drug)

Violent Nonviolent Violent vs. Nonviolent

  • 19
  • 18
  • 2
  • 11
  • 15
  • 1
  • 57.9
  • 83.3
  • 50
  • 2
  • 0
  • 0

10.5

  • 0
  • 0

Ж Coefficient in the predicted direction (greater neighborhood instability, disorder, more crime or greater stability, collective efficacy, less crime)

  • 0 Coefficient is in the opposite direction of that predicted
  • 1 Authors did not distinguish between drugs
  • 2 Includes both cocaine and “crack” cocaine
  • 3 Includes other drug categories not listed here such as amphetamines and benzodiazepines

53 out of 78 statistical tests (68%), with no comparisons in the negative direction. Considering that “statistical significance” allows for about 5% of the comparisons to be significant due to chance, it is clear that the findings favor the hypothesis that alcohol and violence are correlated. This is no surprise. However, alcohol consumption was similarly positively related to nonviolent crime in 66 out of 105 models (63%), which was unexpected.

Study level vote counts (Table 11.2) show that a larger percentage of studies had a preponderance of comparisons (PoC3) in the predicted direction for violent offending (57%) than nonviolent offending (46%). Both exceed the proportion we might expect by chance, both are substantial. The difference between these proportions causes us to tentatively conclude that there is evidence for a differential, more consistent, impact of alcohol consumption on violent versus nonviolent offending.

 
Source
< Prev   CONTENTS   Source   Next >