Just another ComMetrics – social media monitoring, best metrics, marketing metrics weblog

research that matters – identity theft

October 30th, 2007 · No Comments ·

Identity theft has been a real problem for some time. Damages caused by such criminal activity are rising and the pain this causes victims cannot be understimated.

Here we discuss an important report on identity theft by summarising the findings, point out some interesting stuff regarding the methodology and, finally, raise some research issues that need to be addressed if we want to combat identity fraud more successfully than so far.

identity fraud – trends and patterns in the USA
goals descriptions as provided on p. 9 of the study report
1 To explore and identify, from a national (i.e. USA) perspective, key identity theft offense, offender, and case characteristics.
2 To collect and analyze criminal case data for the purpose of establishing an empirically-based profile of identity theft offense, offender, and case characteristics.
3 To isolate those empirically obtained offense, offender and case factors that accurately represent the challenges to effective identity theft control and prevention.
4 To convert the aggregation and analysis of identity theft crime case data into a substantive and formative guide to aid the successful control and prevention of identity theft.

The study, was based upon archival data. For a summary of the cases used by the researchers we copied their first Table and placed it here:

50% of the cases were classified as Fraudulent Use of Account Numbers, Fraudulent Access Device Applications, Stolen Bank Issued Cards, Financial Institution Fraud (FIF) Involving Check Fraud, Counterfeit Bank Issued Credit Cards, Counterfeit Commercial Checks, and Counterfeit State Driver’s licenses. if you cannot see the above Table please click

Figure 1 displays the most frequent primary case types represented by the 517 cases

The Findings were based on the 517 cases as listed in the above Table. These cases were closed between January 2000 and March 2007.

Researchers reported their findings 2007-10-22 at a conference held in Virginia. Amongst other findings the following was reported:

– found that 54% of the offenders were black, 38% were white

– nearly one-third of the offenders that got caught were female,

– about 71% had no previous criminal record, and

– many cases involved sloppy data-security practices by businesses like retail stores.

press release – CIMIP Study Reveals New Findings About ID Theft Cases Closed Secret Service Cases Reveal Methods, Victims, Perpetrators

Data Limitations are spelled out on p. 20 of the report (we quote the paragraph below):

    The data used in this study was collected from Secret Service cases related to identity theft that were opened and closed between January 2000 and March 2007 and made available to the research team. These cases were referred to and accepted by the Secret Service during that time period. This data does not represent all of the identity theft cases that were investigated and prosecuted during this time period by the Secret Service and other law enforcement agencies. The characteristics of cases that were not referred to and/or accepted by the Secret Service, but investigated by local or state law enforcement or another federal entity (e.g. USPIS, FBI), may differ, as may conclusions drawn from them concerning trends and patterns. However, the differences may not be great and the findings of this study should be applied to state and local law enforcement efforts. The researchers recognize that there is an unknown figure of identity theft crimes.

We do not quite understand how the authors could state that the differences may not be great regarding investigations and findings by local police authorities. As well, based on the above information one can, of course, say that the Secret Service may have those cases where pre-liminary data suggested that the case could result in a court verdict for the accused. Thus, some complex cases, or quite evil ones where the tracks left behind by the culprit may not be included in this database because the Secret Service may have decided that it ain’t worth the effort.

Nonetheless, check out the report here:

Gary R. Gordon, Donald J. Rebovich, Kyung-Seok Choo, Judith B. Gordon (October 2007). Identity fraud trends and patterns: Building a data-based foundation for proactive enforcement. Utica, NY: Center for Identity Management and Information Protection
Utica College (available online, see this link) 74 pages


This is an important study also because it sheds some light on demographics of the offerenders including gender, race, date of birth, place of birth, and criminal history (see p. 19). At a next step it would be interesting to understand better if such demographic variables can be used to control for particular effects. For instance, after having controlled for gender and age, does educational level of the offender or restitution that had to be paid still have a significant effect upon the dependent variable, such as type of offense as outlined in the Figure 1 above taken from the study.

On pp. 55-56 the report also discusses the offender methods and provides numerous categories that were used. For instance, for Internet e-mail, phishing, hacking, malware/viruses and for technology categories are such as computers to: scan documents, product documents, computer printers to produce documents, photo copier and typewritier. These are too many categories to control for the level of technical savvy or complexity that was used by the culprit for succeeding with identity theft.

There is much to do to clarify many more issues but this study is a nice trail blazer that everone should browse through to understand the challenge. Nice job and well done.

A word of caution maybe, the Law Enforcement Training section on p. 70 seems to be a bit a bit too fast by making links between the findings and investiative work. For instance, suggesting that the demographics identified (e.g., large percentage of offenders were black and about a third were women) could contribute to continued and improved investigation and prosecution may be true but more research is needed using additional control variables to make sure what could be behind these spurious demographic effects.

To make it more convenient for you to take advantage of CyTRAP Labs’ offerings, just provide us with your e-mail address below. You can personalize your subscription to make it suit your needs.


→ No CommentsTags: empirically · female · goal · identity · offender · offenders · offense · theft