Making Sense of Statistics
Statistics are meaningful numbers that reveal impor- tant information. Statistics are of crucial importance to social scientists, policy analysts, and decision makers because they replace vague adjectives such as “many,” “most,” and “few” with precise numbers.
Both victimologists and criminologists gather their own data to make their own calculations, or they scrutinize official statistics, which are com- piled and published by government agencies. Why do they pore over these numbers? By collecting, computing, and analyzing statistics, researchers can answer intriguing questions. Accurate and credible statistics about crimes and victims are vital because they can shed light on these important matters:
Statistics can be calculated to estimate victimization rates, which are realistic assessments of threat levels that criminal activ- ities pose to particular individuals and groups. What are the odds various categories of people face of getting robbed or even murdered dur- ing a certain time period? Counts (such as death tolls) answer the question “How many?” Better yet, rates (number of persons who get robbed out of every 100,000 people in a year)
can provide relative estimates about these disturbing questions.
Statistics can expose patterns of criminal activity. Patterns reflect predictable relation- ships or regular occurrences that show up during an analysis of the data year after year. For instance, a search for patterns could answer these questions: Is it true that murders generally occur at a higher rate in urban neighborhoods than in suburban and rural areas? Are robberies committed more often against men than women, or vice versa?
Statistical trends can demonstrate how situa- tions have changed over the years. Is the bur- den of victimization intensifying or subsiding as time goes by? Are the dangers of getting killed by robbers increasing or decreasing?
Statistics can provide estimates of the costs and losses imposed by illegal behavior. Estimates based on accurate records can be important for commercial purposes. For example, insurance companies can determine what premiums to charge their customers this year based on actuarial calculations of the typical financial expenses suffered by policyholders who were hospitalized last year after being wounded by robbers.
Statistics can be used for planning purposes to project a rough or “ballpark figure” of next year’s workload. Law enforcement agencies, service providers, and insurance companies can anticipate the approximate size of their case- loads for the following year if they know how many people were harmed the previous year.
Statistics also can be computed to evaluate the effectiveness of criminal justice policies and to assess the impact of prevention strategies. Are battered women likely to lead safer lives after their violent mates are arrested, or will they be in greater danger? Do gun buyback programs actually save lives or is their impact on the local murder rate negligible?
Finally, statistical profiles can be assembled to yield an impression of what is usual or typical
68 CH APT ER 3
Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. WCN 02-200-203
about victims in terms of their characteristics such as sex, age, and race/ethnicity. For example, is the widely held belief accurate that most of the people who die violently are young men from troubled families living in poverty- stricken, big-city neighborhoods? Portraits based on data can also provide a reality check to help ground theories that purport to explain why some groups experience higher rates of predation than others. For example, if it turns out that the frail elderly are robbed far less often than teenagers, then any theory that emphasizes only the physical vulnerability of robbers’ targets will be totally off-base or at best incomplete as an explanation of which groups suffer the most, and why.
However, statistics might not only be used, they can also be abused. Statistics never speak for themselves. Sometimes, statistics can be circulated to mislead or deceive. The same numbers can be interpreted quite differently, depending on what spin commentators give them—what is stressed and what is downplayed. Cynics joke that statistics can be used by a special-interest group just like a lamppost is used by a drunkard—for support rather than for illumination.
Officials, agencies, and organizations with their own particular agendas may selectively release statistics to influence decision makers or shape public opinion. For example, law enforce- ment agencies might circulate alarming figures showing a rise in murders and robberies at budget hearings to support their arguments that they need more money for personnel and equipment to bet- ter protect and serve the community. Or these same agencies might cite data showing a declining number of victimizations in order to take credit for improving public safety. Their argument could be that those in charge are doing their jobs so well, such as hunting down murderers or preventing robberies, that they should be given even more funding next year to further drive down the rate of violent crime. Statistics also might serve as evidence to argue that existing laws and policies are having the intended desirable effects or,
conversely, to persuade people that the old meth- ods are not working and new approaches are necessary.
Interpretations of mathematical findings can be given a spin that may be questionable or debatable—for example, emphasizing that a shelter for battered women is “half empty” rather than “half full,” or stressing how much public safety has improved, as opposed to how much more prog- ress is needed before street crime can be considered under control. As useful and necessary as statistics are, they should always be viewed with a healthy dose of scientific skepticism.
Although some mistakes are honest and unavoidable, it is easy to “lie” with statistics by using impressive and scientific-sounding numbers to manipulate or mislead. Whenever statistics are presented to underscore or clinch some point in an argument, their origin and interpretation must be questioned, and certain methodological issues must be raised. What was the origin of the data, and does this source have a vested interest in shap- ing public opinion? Are different estimates available from other sources? What kinds of biases and inac- curacies could have crept into the collection and analysis of the data? How valid and precise were the measurements? How were key variables operationalized—defined and measured? What was counted and what was excluded, and why?
Victimologists committed to objectivity must point out the shortcomings and limitations of data collection systems run by the government. They try to interpret statistics without injecting any particular spin into their conclusions because (it is hoped) they have no “axe to grind” other than enlightening people about the myths and realities surrounding the crime problem.