"The association of legalization of abortion and crime rates is a novel and controversial suggestion based on the notion that unwanted children are at greater risk for crime and abortion reduces the number of unwanted children. In other words, unwanted children born after 1973, would have become the criminals who elevated the crime rate in the mid-1990s. Other shave argues that the dramatic drop in crime in the mid-1990s was the result of changing sensibilities. According to Tonry (2004), “Crime rates change slowly, in response to long-term social and normative changes: (p. 112) and when people’s attitude toward a particular criminal behavior changes and becomes less tolerant, citizens are more likely to report crime to the police. Thus, at times when people are less tolerant of drug crime, more drug crimes will be noticed and reported to police, which in turn will inflate the statistical incidence of this type of criminal behavior. Still others have argued that the pea in crime rates in 1980 was the result of the expansion of the crack cocaine market couples with the spread of youth gangs and increased access to firearms. When the crack cocaine market matured and turf battled ended, crime began to decrease. The 1990s crime decrease in all offense categories suggests that “something fundamental was changing in the United States and it affected each of these major crimes in the same ways” (Tonry, 2004, p. 116). For example, historians have found that homicide rates since the 12th century (the earliest time for which quantitative data are available) declined steadily in England, Holland, Belgium, Germany, Scandinavia, and Switzerland from between 30 and 100 per 100,000 in the population to 10 per 100, 000 in the 18th century, to 1 to 2 per 100,000 in the 20th century. Considering more recent periods in Western countries in the mid-1990s, historians have described the violent crime rates as a “U” or “reversed J-shaped” curve (Tonr, 2004). To make sense of criminal behavior at any given time and in a particular society or community, it is important to know whether or not a particular criminal behavior is a statistical anomaly or so common a behavior it can be said to be barely a crime at all, and to pay attention to the differences in the extent of crime types across geographical areas, communities, cultures, and time periods. It is also important to recognize that it is no easy task to compare/contrast and synthesize data across sources, and an even more difficult challenge to compare crime rates across time periods and cultures. For example, UCR crime categories are not the same as NCVS categories and self-report studies such as the NYS and NSDUH measure crimes and crime categories at a level not possible with official police data and victimization surveys (e.g., NSDUH provides data on drug use that are not available through other sources). Official statistics present only a limited picture of the extent of crime in the United States, and making sense of the nature and extent of crime in general or of a particular type of crime requires synthesizing multiple sources of data. Figure 1.2, Table 1.4, and Table 1.5 present UCR data on crime types from 1986 to 2006, the NCVS data for 2005, and a rough look at the difference in the amounts of violent and property crime reported by the UCR and NCVS."