Wednesday, February 20, 2008

Gangs and Crime - or is it Crime and Gangs?

Tita, G., & Ridgeway, G. (2007). The Impact of Gang Formation on Local Patterns of Crime Journal of Research in Crime and Delinquency, 44 (2), 208-237 DOI: 10.1177/0022427806298356

Blogging on Peer-Reviewed ResearchBefore we head away from gangs completely, I wanted to bring up this paper looking at the connection between gangs and crime. Obviously, most gangs are heavily associated with crime in the areas in which they operate. The 10% of youth who join gangs are responsible for 80-90% of the crime that occurs in a given community. So which comes first? Basically, there are three theories for how the two are related:
  • social facilitation - once a large group of juveniles congregate, they become more likely to commit crimes, and it is easier for them to do so (more lookouts, getaways, etc.).
  • selection - the juveniles who are already committing crimes are likely to band together out of common interests, or for utilitarian reasons.
  • enhancement - this combines facilitation and selection. Essentially, once the juveniles who are already committing crimes band together, they become much more effective and efficient.
The social facilitation and enhancement theories both predict that crime rates will go up after the establishment of a gang, while selection theory predicts that it will be relatively stable. Furthermore, facilitation can be separated from enhancement in that facilitation predicts there will be more people committing the crimes, whereas enhancement predicts that the same people will just be committing more crimes.

To test which one of these theories holds, the authors went to a site of emerging gang activity: Pittsburgh, PA. In addition, they wanted to test the theory that gangs inhibit certain types of crime, like property violence against local shopkeepers. (Personally, this sounds more like Organized Crime and "protection" to me, though I know gangs in LA run similar rackets with the local street vendors.) The investigators were able to draw causal conclusions from their dataset because prior to 1992, there was no gang activity in Pittsburgh. Looking at the two years before this time (90-91), the two years during "establishment" (92-93), and the two years postgang (94-95) it's possible to discern trends in crime rates that correspond with the introduction and maintenance of gang activity.

The authors obtain their crime data from 911 call records (problematic, yes, but they attempt to address these fears in the paper). They also try to disaggregate their data by defining, with the help of locals, what constitutes gang turf, and denoting areas as "carriers" of gang activity, or "set spaces," neighboring areas adjacent to gang turf, and non-neighbors. Comparisons of different types of crimes in these different areas is shown in Figure 2, and in Figure 3 they weight the crime rates to eliminate confounding factors, like racial makeup, income level, etc. There's an assumption that crime rates in non-neighboring areas are generally unaffected by gang activity, so comparing rates in these two areas can give an estimation of the effects of gangs on crime.

Looking at Figure 3, it's clear that after gangs gained a foothold in the 92-93 time period there were spikes in reportings of drug crimes, robberies, and shots fired in gang neighborhoods. All of these tapered off in the 94-95 period, returning to pre-gang levels or below. Looking at assaults and burglaries, there was no difference between gang and non-adjacent neighborhoods, and they both showed general downward trends over time, apparently unaffected by the introduction of gangs.

So what conclusions can be drawn from this data? It would appear that gangs aren't involved in assaults or burglaries, though they also don't result in the protective effects that were hypothesized. Drug crimes, robberies, and gunshots all increased, but only briefly. What could be causing this? Is it a need for the gang to establish itself as a force in the community with a concentrated burst of crime? Or is it a result of increased police focus and refinement of tactics having a real effect on crime? Alternately, it could possibly just be citizen fatigue, with residents getting tired of constantly calling 911, or having little to show for it when they do. Other possibilities are that residents are being threatened, or that they may actually benefit from the increased economic activity of the drug trade. Ultimately, it looks like the selection theory is not supported, and that gang activity actually results in new crimes and criminals, and doesn't just round up the old ones.

Monday, February 4, 2008

A Little Clarification on Psychopathy

HARE, R. (2006). Psychopathy: A Clinical and Forensic Overview Psychiatric Clinics of North America, 29 (3), 709-724 DOI: 10.1016/j.psc.2006.04.007

Guay, J., Ruscio, J., Knight, R., & Hare, R. (2007). A taxometric analysis of the latent structure of psychopathy: Evidence for dimensionality. Journal of Abnormal Psychology, 116 (4), 701-716 DOI: 10.1037/0021-843X.116.4.701

Blogging on Peer-Reviewed Research So there's some confusion about the differences between the diagnoses of psychopathy and Antisocial Personality Disorder (APD). Well, to make matters worse, the Diagnostic and Statistical Manual of Mental Disorders (DSM) goes and equates the two. Dr. Geiselman has said in his lectures that the difference, in his experience, is that psychopathy is characterized by delusions, hallucinations, and other sensory phenomena. Okay, for the test you're going to want to go with this definition; however, I tend to prefer the definition offered by one of the primary researchers on psychopathy, Robert Hare. Did I ever mention how much great forensic research comes out of Canada? Hare's based in the University of British Columbia, and he developed the Psychopathy Checklist thirty years ago, which has become the instrument to assess psychopathy.

As summed up in a recent overview (Hare 2006), psychopathy can be subdivided into four main factors:
  1. interpersonal (glibness, narcissism, conning behaviors, etc.)
  2. affective (lack of remorse, shallow affect, etc.)
  3. lifestyle (need for stimulation, parasitism, impulsivity, etc.)
  4. antisocial (criminal versatility, poor control, etc.)
Notice there's no mention of hallucinations or disordered thinking among these factors. So how does this compare with APD? According to the DSM, they're synonymous. However, taking a real look at them, APD is diagnosed entirely by manifest behaviors, while psychopathy is an attempt to define a personality type by the underlying cognitive and relational patterns. As Lykken (quoted in Hare 2006) puts it: "Identifying someone as ‘having’ APD is about as nonspecific and scientifically unhelpful as diagnosing a sick patient as having a fever or an infectious or a neurological disorder." APD is the symptom; psychopathy is the disease. There's significant overlap in diagnoses of APD and psychopathy, but there are also significant numbers of people who fail to meet both definitions.

There have been numerous attempts to put psychopaths into a taxon, a group used for classification - either you're psychopathic or you're not. However, these attempts have consistently failed to find evidence supportive of a taxonomic system. Hare (2007), in a large and diverse sample, again tried to perform a taxometric analysis, but could only find evidence for dimensionality. What does this mean? Well, there are several reasons why a legitimate taxon could be covered up in an analysis, but if psychopathy is truly dimensional, it means that there are no clear, defining lines between those with psychopathy and those without. It may just be a scale on which a given person is either high, moderate, or low.