Computer technology in law enforcement crime solving: a preliminary survey

¶ … Computers in Solving Non-computer-Based Crime

This proposal for research involves a survey of law enforcement officials to determine how much they use computers as a tool to help solve crimes that are not committed by computer, such as murder and robberies. Using a questionnaire that utilizes a numerical scale for responses with opportunities for written comments as well, it will quantify the results and indicate areas for further research. Since little research has been done in this area, it should be considered a preliminary study.

Law enforcement has traditionally struggled to keep up, technologically, with the criminals they are charged with catching. During Prohibition, gangsters had machine guns and government agents did not. Now, in the year 2002, law enforcement may have a powerful and relatively new tool available to them in the war against crime: computers.

However, an exploration of the literature reveals little if any systematic study about the uses of technology in law enforcement. Isolated articles appear describing what various agencies do, but no one has surveyed government agencies to see if they use computers as a crime-solving tool, and if so, how they use it. No studies have been done regarding whether the use of computer technology is more efficient or less, and under what circumstances they are likely to be an advantage or a liability. There are no formal studies on what agencies use, how they have chosen what they use, whether they think they made a good choice, whether the software supports or hinders their efforts, and whether computers are viewed overall as a useful tool by those charged with solving crime.

This project proposes quantifying how law enforcement agencies use computer technology in the process of solving crimes. It would be a preliminary, or pilot study, and would raise more questions than it answers. In addition, as computer technology evolved, both the questions and the answers will change over time, but only by carefully examining the best role for computers is in crime investigation can law enforcement agencies make informed decisions that improve their efficiency rather than interfering with it.

Significant questions regarding this expensive technology have not been asked. What crimes most benefit from computer-assisted crime solving techniques? What are those techniques, and how do they apply to solving various types of crimes? How expensive is the technology, and is it cost-effective? Do computers make it easier for agencies to share information, or do the formats used by different agencies make it hard to share information? If so, how could those problems be solved?

Since technology is only as good as the people who use it, how do those asked to incorporate computers into their crime-fighting techniques feel about it? Are they allowed to be innovative in their use, setting up data bases as they think will serve them best, or are there departmental formats that must be followed? If an individual investigator can set up his or her own organizational system for him or herself, is that time-efficient, or could the time be better used following more traditional investigative patterns?

Once information regarding a crime has been computerized in some way, does it become rigid, or do investigators ever go back to the original data to look at it in its original form? How does the department protect against the problem of important information being left out, or inaccurate information being entered?

There is no established protocol for any investigative agency to add technology to its crime-solving techniques. It is suspected that individual states and local agencies may be gradually adding technology to their arsenal tools without any specific plan. Systematic research regarding how various agencies have implemented technology as a crime-solving tool could be valuable information for other organizations beginning that process themselves. Systematic study of how to use the potentially powerful computer as efficiently as possible, rather in a random way, would greatly enhance law enforcement efforts to solve serious crimes as quickly and accurately as possible.

SEARCH OF THE LITERATURE

One area that has received attention is in the use of computers to commit crimes. When the computer is the weapon used by criminals to commit a theft (monetary or information), it makes sense to use computers to solve that crime. Many countries have formed task forces to fight computer crime. For instance, Great Britain has formed a National Hi-Tech Crime Unit (ANA, 2001) to combat computer-related crime. These task forces focus on crime committed on the Internet. While important, it represents a small amount of serious crime committed today.

More broadly, white-collar crime has become a serious problem in the United States, with one in every three households a victim (Johnston, 2002). Since white-collar crime is committed by businesses, including investment firms, insurance companies, telecommunications and credit card fraud, computers are often a significant part of those crimes, and often play a significant part in solving and prosecuting them. Another area that has benefited from computer investigation is child pornography. (Garber, 2001) Again, however, the crime is often computer-based as well as solved using computer technology. In all three of these examples, the computer is used as a tool in investigation because computers were used in the commission of the crime. This demonstrates that law enforcement has managed to match technology with the criminals involved but does not demonstrate any expansion of computer use in the solving of non-technologically-based crimes, such as murders.

The FBI has explored the possibility of using computers as a possible crime-solving tool since the mid-1980's (Garber, 2001), but once again, the focus was on computer-based crime. The authors note that they have developed sophisticated techniques, including specialized software, to improve their investigation of computer crime. However, development of similar software for solving non-computer-based crimes lags far behind (Mountain, 1999) In addition, the International Association of Computer Investigative Specialists (IACIS) reports that computer-driven investigations have not completely supplanted more traditional detective work, and that they frequently have to examine the data manually (Garber, 2001) So it seems that computers remain a too, but not a complete investigative solution.

A slightly broader use of computers is achieved by confiscating the computers of suspects. The IACIS reports that one murder suspect left incriminating information on his computer about how he poisoned his wife, (Garber, 2001) leading to his arrest and conviction. However that is more serendipity than use of computers to solve crimes.

The National Crime Information Center (NCIC) reports the first use of computer-aided detective work in 1968, when the FBI used a new fingerprint database to identify a fingerprint found on the gun used to kill Martin Luther King. This early computer use was cumbersome and inefficient, and came up with 1,200 possible matches (Hitt, 2000). As computers faster and more powerful computers have been developed, their potential as a crime-solving tool has increased significant. In the 1996, it only took computers one hour for a computer to turn up the fact that an Oklahoma state trooper had stopped Timothy McVeigh 88 miles from Oklahoma City, one hour after the bombing of the federal building there (Hitt, 2000). The NCIC system is one use of computer where specific facts exist: 99% of its activity comes from agencies other than the FBI accessing the information compiled in their database on crimes and criminals (Hitt, 2000). In addition to fingerprints, this data base tracks guns either reported stolen or recovered in some other way. It includes incidences of parole violation, escape, and failure to appear in court. Since 1995 it also tracks the theft of hazardous materials.

The NCIC database also can compare crimes as well as criminals, and is cross-matched so that vehicles or guns connected to the incident will be pulled up out of the database. Some recent experiences with the NCIC database demonstrate how technology can create new problems instead of solving them. When a feature was added to the data base holding information requests for five days, to see if additional new connections could be made, the system was quickly overloaded, and those using the data base were back to the situation they had in 1968 when they had 1,200 possible fingerprint matches: the inquiring agency got more possible connections than they could process or evaluate (Hitt, 2000) This is a clear example demonstrating that caution has to accompany the implementation of technology.

The FBI has also used computer technology to investigate crime occurring in the United States with international connections, such as drug activities and organized crime. Computers are used in these cases to organize information gathered (McFeely, 1999) In addition, the FBI has developed new techniques for investigating organized crime. They call this model the "Enterprise Theory of Investigation." This theory proposes that no matter what crimes organized criminals may commit, their goal is to make money (McFeely, 1999). Using this information, they can use computers as well as other methods to demonstrate organized intent to commit crime as a group.

Finally, software developers are beginning to look at the issue of how computers can best be used as a crime-solving tool. (Mountain, 1999) reports on new CAD software intended to recreate crime scenes in digital format for use in trials. The programs involve a library of symbols. While such programs may add to graphic presentation in a trial, questions may remain about whether it would be useful to use such a program during investigations.

The general public may also have a strong interest in seeing computers and the Internet used as a crime-solving tool. "All Points Bulletin" is a new website devoted to information on current crime investigations (ANA, 1998). However, the literature search did not turn up a similar site dedicated to those actually solving the crimes. Security might be a significant issue for such sites, raising the question of how agencies might easily share information across city, county and state lines as needed. For instance, a web site dedicated to the murders ultimately charged to Ted Bundy might have been useful for the many states involved: Washington, Oregon, Colorado, Utah and Florida.

The lack of concrete information regarding computer use outside computer crime is remarkable. It is expected that this research will demonstrate that law enforcement officials are some distance away from such efficient sharing of information. It is hoped the research will demonstrate the range of uses currently used for crime investigation, and whether those using them judge them effective or not. Its results will have to preliminary, a starting point for gathering valuable information about how best to use the power of the computer and the Internet in crime investigation.

Whether the respondent considers the computer a valuable tool for this purpose

Whether investigators use computers for this purpose as they see fit, or whether the department has an organized approach to using computers as a crime-solving tool

Whether they think their equipment is suitable for their investigative needs

Whether they think budget allocations are adequate to support computer use as needed in the solving of crimes

How many years their department has used computers in investigations, and to what extent

What improvements they have noted in the time they have used computers for this purpose (ex: decreased time required for a response; whether the responses are too broad, including too much information, or too narrow, eliminating important information

How their department interweaves the use of computer with more traditional ways of organizing and analyzing information

When they think computer use is an advantage, and when it is a liability

If they ever use outside private investigators for this purpose, and if so, whether those outside investigators use computers to gather information on a crime or criminal

These questionnaires will be sent out to 10 state departments of investigation and ninety local police departments. They will be addressed to specific individuals. Follow up letters will be sent to non-respondents. It is assumed that 75% of respondents will respond, and recipients will be sent copies of the results for their trouble. Respondents will be anonymous and referred to only by a number so they can express their thoughts freely.

Numerical responses will be analyzed using a chi-Square analysis to generate statistically analyzed results. Verbal responses will be evaluated for positive or negative remarks. Once the information has been gathered, it will be entered into a spreadsheet.

A scale will be devised to rate those responses so they can be analyzed using a Chi-square as well. Internal reliability will be determined by comparing statistical results of the numerical responses to the quantified results of written comments. Internal reliability will be achieved if those responses agree at.75. If written responses indicate opinions different than those obtained from numerical responses, the results will be judged unreliable.

Because this research represents a pilot study, generating as much information as possible from the questionnaire will counterbalance the relatively low number of respondents. However, to encourage respondents to participate, as many questions as possible will be answered with a numerical rating of 1-5. 1-3 was considered to encourage participation, but the judgment has been made that using a scale of 1-5 will generate more precise information.

There will be periodic opportunities for the respondents to provide additional information in written form if they so choose, but it won't be required to participate.

This research design does contain some flaws. The problem of referral bias is a very real one. That is, it seems reasonable that those most interested in using computers will be the most likely to respond. The pattern of those who do and do not respond may correlate with respondents who are and are not comfortable with computers, and the researcher will have no way of knowing this.