Victim Reporting Behaviors Following Identity Theft Victimization: Results From the National Crime Victimization Survey

Conceptual Frameworks on the Decision to Report

Although there are many theories that speak to the functioning of the criminal justice system (see Bernard & Engel, 2001), two particular criminal justice theories provide insights and possible explanations regarding the influence of situational and victim characteristics in the post-victimization decision-making process of crime victims. First, focal concerns theory, which was developed and utilized by Steffensmeier and colleagues (1998) to account for sentencing outcomes, explains that there are three concerns that guide the discretionary sentencing decisions of judges: the blameworthiness of the offender, protection of the community, and practical constraints and consequences (see also Steffensmeier, 1980; Steffensmeier & Demuth, 2000; Steffensmeier, Kramer, & Streifel, 1993). Theoretically, in the case of the sentencing decision, a more blameworthy offender, who is likely to reoffend, and for whom there is not a compelling reason to do otherwise (e.g., overcrowding, disruption to the lives of children) will be sentenced more harshly than someone who does not fulfill these criteria. Generally, research has supported a focal concerns perspective on sentencing (e.g., Hartley, Maddan, & Spohn, 2007; Steffensmeier & Demuth, 2000; Steffensmeier et al., 1998), parole (Huebner & Bynum, 2006), and police search decisions (Higgins, Vito, & Grossi, 2012), but this perspective has yet to be applied to victim decision making.

The second theoretical framework from the field of criminal justice that is useful in understanding the victim’s decision to initiate the criminal justice process was developed by Gottfredson and Gottfredson (1988) as a general theory of criminal justice decision making. Having surveyed the components of the criminal justice system, Gottfredson and Gottfredson synthesized the existing body of research surrounding actor decision making, and likewise identified three factors that persistently affected decisions across the system: the seriousness of the offense, the offender’s prior criminal record, and the relationship between the victim and the offender. According to Gottfredson and Gottfredson, these three factors heavily affect every decision in the criminal justice process, with more serious offenses, committed by repeat offenders, against strangers, being those most likely to elicit a response from the criminal justice system (e.g., reporting, arrest, charging). So far, this theory has been used to explain victim decision making in only one other study. Reyns and Englebrecht (2010) used the Supplemental Victimization Survey to the NCVS, which assessed the extent and nature of stalking against adults in the United States, to examine the effects of event seriousness, the offender’s prior record, and the victim–offender relationship upon the decision to report stalking victimization to the police. In support of the theory, they found that each of these theoretical concepts was tied to reporting decisions. For example, incidents in which the victim’s property was damaged, the victim was threatened, or in which the victim knew that the offender had a prior criminal record were all more likely to come to the attention of police through victim reporting.

Conceptually, there is some overlap between these two theoretical perspectives. That is, they both underscore the importance of seriousness of the offense (and blameworthiness) in affecting criminal justice decision making (see also Skogan, 1984). Likewise, while the two concepts are not exactly the same, protection of the community and the offender’s criminal history are clearly connected because one of the considerations in determining danger to the community is the offender’s prior record. Where the two perspectives diverge somewhat is in their respective third concepts: practical constraints and consequences and the victim–offender relationship. Although the relationship between the victim and the offender has been identified as a correlate of reporting in numerous studies (e.g., Felson et al., 1999; Felson, Messner, Hoskin, & Deane, 2002; Fisher et al., 2003; Gartner & Macmillan, 1995), the concept of practical constraints and consequences has been more loosely defined and has been operationalized differently across studies (e.g., Johnson, 2006; Spohn & Holleran, 2000; Steffensmeier, 1980). Specifically, in Steffensmeier and colleagues’ (1998) work, practical constraints referred to organizational concerns (e.g., case flow), whereas practical consequences were offender-based (e.g., special needs, health conditions).

Many of these theoretical concepts find reasonably close counterparts when the theories are adapted to the victim’s decision to contact law enforcement. For example, the victim might judge the seriousness of the offense and the blameworthiness of the offender according to how much harm they suffered as a result of the crime (e.g., physically, financially, emotionally). Furthermore, like other actors within the criminal justice system, victims will also likely take into account practical considerations when making the reporting decision, such as whether the issue has already been resolved or whether they know anything about how their personal information was obtained. From a victimization perspective, other theoretical concepts, especially those that are offender-based, are not as straightforward in their application. The offender’s prior record and the danger he or she poses to the community, for example, will not be known to the victim unless the two parties have some sort of relationship. Overall, however, these conceptual frameworks provide a theoretical lens through which the victim’s decision to notify law enforcement of his or her victimization can be better understood and contextualized as a decision within the criminal justice system. Yet, by and large, these theories have yet to be utilized in this way to understand victim decision making.

Data

The data for the present study were collected by the Bureau of the Census in partnership with the Bureau of Justice Statistics in 2012 as part of a supplement to the NCVS. The NCVS is an ongoing self-report rotating panel survey designed to measure the extent and nature of criminal victimization within the United States. Data are collected semi-annually at 6-month intervals from a representative sample of residents aged 12 years and older, excluding persons residing in institutions (e.g., nursing homes, prisons), members of the military living in military barracks, and individuals who are crews of vessels. The response rate to the 2012 NCVS was 87%.

Respondents to the NCVS were selected for inclusion in the present study if they were identified as victims of an identity-related fraud victimization. In doing so, four dimensions of victimization were examined. ¹ First, of the 46,128 persons participating in the NCVS who had a credit card at the time of the survey, 2,109 were identified as victims of credit card fraud. ² Second, 56,202 respondents indicated that they possessed a bank account, and of these, 1,930 experienced bank fraud on this account. Third, we identified victims of fraud for any type of existing account. This measure includes victims of existing credit card and bank frauds, as well as frauds involving other types of existing accounts (e.g., medical insurance, telephone, utilities). Fourth, the reporting behaviors of individuals whose personal information was used to create a new account (e.g., online payment account, credit card, loans) and perpetrate a fraud were also identified. Table 1 provides descriptive statistics for these four dimensions of identity theft victimization as they were reported in the NCVS. From these respective pools of victims, cases with missing data were omitted from the analyses through listwise deletion.

OPEN IN VIEWER

Table 1.

Descriptive Statistics for Types of Identity Theft Victimization.

	n	Victims	M (%)	SD	Minimum	Maximum
Existing credit card accounts	46,128	2,109	5	0.20	0	1
Existing bank accounts	56,202	1,930	3	0.18	0	1
Existing accounts	64,132	443	1	0.08	0	1
Medical insurance	442	14	3	0.17	0	1
Telephone	443	110	25	0.43	0	1
Utilities	442	46	10	0.30	0	1
Online payment	442	169	38	0.48	0	1
Investment	442	5	1	0.10	0	1
Other	442	5	1	0.10	0	1
New accounts	64,132	300	0.5	0.06	0	1
Telephone	291	62	21	0.41	0	1
Credit card	291	110	38	0.48	0	1
Checking or saving	285	32	11	0.31	0	1
Loan or mortgage	289	49	17	0.37	0	1
Medical insurance	285	4	1	0.11	0	1
Auto insurance	285	4	1	0.11	0	1
Online payment (e.g., Paypal)	284	26	9	0.28	0	1
Other	290	2	1	0.08	0	1
Other fraudulent purpose	64132	215	0.3	0.05	0	1
Tax return	212	58	27	0.44	0	1
Medical treatment	211	40	19	0.39	0	1
Apply for job	211	20	9	0.29	0	1
False police report	211	35	16	0.37	0	1
Rent house/apt	210	9	4	0.20	0	1
Apply for government aid	208	17	8	0.27	0	1
Other	212	6	2	0.16	0	1

Measures

Victim reporting to law enforcement

The dependent variable in the current analyses was victim reporting to law enforcement, which was measured with a single survey item asking victims the following: “Did you contact any law enforcement agencies, such as the local police, a sheriff or a federal law enforcement agency, to report [the/the most recent] misuse or attempted misuse of your personal information?” Answers to this question were coded dichotomously (0 = no, 1 = yes). Table 2 indicates that victim reporting varied by type of identity theft, ranging from 6% of victims reporting to 33%.

OPEN IN VIEWER

Table 2.

Scales and Descriptive Statistics.

Variable	Scale/range	Credit card fraud		Bank fraud		Any fraud		New account fraud
		M (%)	SD	M (%)	SD	M (%)	SD	M (%)	SD
Reporting	0 = no, 1 = yes	6%	0.24	12%	0.325	8%	0.28	33%	0.47
Seriousness of the offense
Emotional distress	0 = no, 1 = yes	14%	0.34	26%	0.43	19%	0.39	44%	0.50
Physical problems	0 = no, 1 = yes	5%	0.21	12%	0.32	8%	0.26	27%	0.45
Amount obtained (log)	0.69-9.74	5.49	1.79	5.29	1.66	5.33	1.72	6.22	1.74
Personal loss	0 = no loss, 1 = any personal loss	6%	0.24	16%	0.37	12%	0.32	22%	0.42
The offender
Offender information	0 = no, 1 = yes	4%	0.20	12%	0.32	8%	0.28	23%	0.42
Practical considerations
Problem resolved	0 = no, 1 = yes	92%	0.27	89%	0.31	90%	0.30	69%	0.46
Know how obtained	0 = no, 1 = yes	31%	0.46	41%	0.49	35%	0.48	48%	0.50
How long used	1 = 1 day or less to 7 = 1 year or more	1.99	1.11	1.92	1.10	1.98	1.14	3.39	1.68
Contact institution	0 = no, 1 = yes	7%	0.25	7%	0.26	7%	0.25	37%	0.49
Victim characteristics
Female	0 = male, 1 = female	46%	0.49	51%	0.50	48%	0.50	59%	0.49
Age	Age in years (17-90)	49.92	14.83	44.39	14.46	47.07	14.88	47.46	14.88
Non-White	0 = White, 1 = non-White	11%	0.30	16%	0.36	14%	0.34	19%	0.40
Income	1 ≤ US$5,000 to 14 ≥ US$75,000	12.83	2.42	11.54	3.30	12.16	3.00	11.25	3.66
		n = 1,143		n = 1,140		n = 2,226		n = 113

Credit Card Fraud

Table 3 identifies seven independent variables as important predictors of the victim’s decision to notify law enforcement of credit card fraud victimization. First, emotional distress suffered by the victim, the amount obtained by the offender, and whether the victim suffered a personal financial loss—dimensions of offense seriousness—exhibited positive and significant relationships with the reporting decision. Second, regarding practical considerations and their effects upon reporting decisions, three variables produced statistically significant effects. Those who knew how the offender obtained their credit card information or who contacted their financial institution were at an increased likelihood of deciding to contact law enforcement. Conversely, individuals whose information was used for longer periods of time were less likely to contact law enforcement. Finally, the only victim characteristic linked with the reporting decision was income, suggesting that those with higher incomes were less likely to report their victimization to the police.

OPEN IN VIEWER

Table 3.

Binary Logistic Regression Coefficients, Standard Errors, and Odds Ratios for Reporting Identity Theft Victimization to Law Enforcement.

	Credit card fraud			Bank fraud			Any fraud			New account fraud
	Coefficient	SE	OR	Coefficient	SE	OR	Coefficient	SE	OR	Coefficient	SE	OR
Seriousness of the offense
Emotional distress	0.97**	0.35	2.65	0.50*	0.23	1.64	0.67***	0.21	1.95	1.28*	0.67	3.60
Physical problems	−0.54	0.59	0.58	0.38	0.29	1.47	0.36	0.29	1.43	−0.67	0.65	0.51
Amount obtained	0.53***	0.11	1.70	0.49***	0.08	1.64	0.50***	0.07	1.65	0.56***	0.15	1.75
Personal loss	1.82***	0.44	6.20	0.90**	0.28	2.46	0.98***	0.27	2.66	0.49	0.59	1.63
The offender
Offender information	0.69	0.57	2.00	0.39	0.30	1.47	0.60*	0.27	1.82	0.42	0.60	1.52
Practical considerations
Problem resolved	−0.34	0.43	0.70	0.62	0.34	1.86	0.29	0.28	1.34	0.29	0.56	1.34
Know how obtained	1.68***	0.29	5.36	0.84***	0.22	2.32	0.94***	0.19	2.57	0.52	0.50	1.69
How long used	−0.36*	0.16	0.69	−0.27**	0.11	0.76	−0.25**	0.09	0.78	−0.11	0.16	0.90
Contact institution	2.62*	1.34	13.86	1.66*	0.69	5.26	1.33**	0.48	3.76	0.43	0.86	1.54
Victim characteristics
Female	−0.56	0.30	0.57	0.01	0.21	1.01	−0.17	0.17	0.84	−1.06*	0.52	0.35
Age	−0.01	0.01	0.98	0.01	0.01	1.01	0.00	0.01	1.00	−0.02	0.02	0.98
Non-White	−0.93	0.52	0.39	0.03	0.27	1.03	−0.17	0.25	0.84	0.05	0.55	1.05
Income	−0.20***	0.04	0.81	−0.04	0.04	0.96	−0.08**	0.03	0.92	0.01	0.07	1.01
Constant	−5.43	1.73		−7.32	1.05		−6.30	0.79		−4.15	1.80
McFadden’s R2	.32			.17			.19			.20
n	1,143			1,140			2,226			113

Note. OR = odds ratio.

*

p < .05. **p < .01. ***p < .001.

Bank Fraud

Several of the previously discussed findings were replicated with respect to victims of bank account fraud. In particular, offense seriousness and practical considerations appeared to be influential in the victim’s decision to notify law enforcement of the crime. Specifically, emotional distress, the amount obtained by the offender, and whether the victim personally lost money as a consequence of the crime were positive and significant influences on victim reporting decisions. Furthermore, individuals who knew how the offender obtained their information and those who contacted their financial institution were also at a higher likelihood of contacting law enforcement. Once again, the duration of the incident was negatively related to the reporting decision.

Any Existing Account Fraud

The model including the composite measure of identity theft victimization—indicating any type of existing account fraud—produced very similar findings to the two previously discussed models. Seriousness of the offense, offender information, practical considerations, and victim characteristics all affected the identity theft victim’s decision to contact law enforcement. Again, emotional distress, the amount obtained, and sustaining a personal loss each positively affected the notification decision. Notably, victims who possessed some sort of information about the offender were significantly more likely to contact the police regarding their victimization—a finding unique to this model. With respect to practical considerations, knowing how the information was obtained and communicating with a financial institution were positive predictors of reporting, whereas longer periods of misuse of information were negatively related to notification. Last, victims with higher incomes were at decreased odds of contacting law enforcement following identity theft.

New Account Fraud

The final model in Table 3 includes findings related to the reporting decisions of victims of new account frauds. New account fraud is less common than existing account fraud (see Table 1), but it appears that the factors affecting victim reporting decisions are not entirely disparate. That is, variables measuring offense seriousness were significant predictors of victim reporting. In particular, victims who were emotionally distressed and those who lost larger sums of money to the offender were more likely to contact law enforcement. Furthermore, males were significantly more likely to contact the police in cases of new account fraud in comparison with females. These variable effects, as well as those reported for victims of existing account frauds, are discussed below.

Summary and Discussion of Findings

The purpose of the present study was to address gaps in the existing body of research into victim decision making by identifying determinants of the decision to report among victims of identity-related crimes. In the process, three hypotheses derived from focal concerns theory and Gottfredson and Gottfredson’s theory of criminal justice decision making also were tested and supported. The results provide insights into the factors that influence this important victim decision, while also supporting the expansion of these two theoretical frameworks into the field of victimology as a means of understanding victim decision making. However, these three hypotheses warrant additional discussion, as does the influence of victim characteristics upon reporting decisions.

First, based on Gottfredson and Gottfredson’s theory, it was hypothesized that more serious incidents—characterized by the impact the crime had upon the victim—would be reported to law enforcement. As stated, findings related to seriousness were as expected, with the exception of physical problems, which had no impact upon notification decisions. However, focal concerns theory also emphasizes the importance of the related concept of offender blameworthiness. This concept essentially reflects the likelihood that the offender carried out the crime. Establishing offender blameworthiness from a victimization perspective becomes somewhat tenuous in cases of identity theft, because it is a crime that is often perpetrated remotely by the offender. That is, the victim and offender may never have a face to face encounter. Thus, the victim may not know who the offender is, how the crime was perpetrated, or indeed, that he or she has been victimized until some later time. This issue speaks to the second theoretical concept from Gottfredson and Gottfredson’s theory—the victim–offender relationship.

Second, due to data limitations, a proxy measure of the victim–offender relationship was utilized, measured as whether the victim possessed any information about the offender (or offenders). In cases of existing account fraud, it was twice as likely that victims would contact law enforcement if they had information about the offender than if they had no such information. This suggests that a further exploration into how the victim’s knowledge of the offender affects their decision making is warranted. Appendix Table A1 provides descriptive information related to why victims decided not to report the crime to law enforcement. While it appears that most victims decided not to report for practical reasons, 7% said that they did not contact law enforcement either because they could not identify the offender or because it was a friend or family member. While speculative, this may also account for why the duration of the incident was negatively related to reporting. That is, if someone known to the victim was misusing his or her information, he or she may be hesitant to contact the authorities. However, it may also be that the situation seems less urgent, especially if it has already been resolved. Altogether, a continued focus on the victim–offender relationship is needed to better understand how victims of identity theft make decisions following the crime.

Third, practical considerations affected the reporting decision—notably, whether the victim knew how their information was obtained and being in contact with the relevant financial institution. In the case of the former, there may be a parallel to the sexual victimization literature in that previous research indicates victims of sexual assault are less likely to report their victimization to the police when they are concerned that they lack proof that the incident happened (e.g., Fisher et al., 2003). So too, victims of identity theft may feel essentially that if they have nothing helpful to tell the police that will aid in the apprehension of the offender, then there is no reason to notify them of the crime. Relatedly, persons who have contacted their credit card company, bank, or other organization may see no reason to also contact law enforcement—a position supported by the data in Appendix Table A1. Overall, within the context of focal concerns theory, it appears that victims consider the practical purpose or end result of their reporting behaviors, and if the problem has already been addressed or they have insufficient information to aid the police in responding to the crime, then reporting is viewed as unnecessary.

Finally, while victim characteristics have featured prominently in the victim decision-making literature, gender, age, and race did not influence victim reporting decisions related to identity theft victimization. Instead, income emerged as the most important victim characteristic affecting decision making. Specifically, persons with lower incomes were more likely to contact the police and initiate the criminal justice process as compared with those with higher incomes. Perhaps income is one of the criteria victims include in their perceptual shorthand when evaluating their situation and deciding whether or not to call the police. The potential for additional research into the effects of income is discussed further below.

Limitations and Recommendations for Future Research

Although the present study addressed several questions from the field related to victim reporting behaviors following identity theft victimization, and utilized two notable theories from criminal justice to interpret the results, the findings should nevertheless be considered in light of the study’s limitations. These limitations also highlight possible areas of inquiry for future research. Namely, while the NCVS provides a rich and robust source of secondary data, using such data can create challenges with respect to measuring theoretical concepts. In particular, the effects of concepts relevant to the victim–offender relationship, and the blameworthiness and dangerousness of the offender could not be assessed. To fully test these two theoretical perspectives, original data will need to be collected to serve this purpose. A related issue involves the types of identity theft included in the present analyses. The focus of the current study centered on forms of identity theft involving account frauds, and while these are the most common forms of identity theft according to the NCVS, future research should focus on other types of fraud involving misuse of personal information (e.g., driver’s license, social security) and associated reporting behaviors.

Furthermore, as previously mentioned, income may be a key factor in a victim’s perceptual shorthand. Specifically, income could be one of the metrics by which victims evaluate and/or perceive their situation, interacting with other correlates of decision making to ultimately determine whether victims contact the police. For example, perhaps individuals with lower incomes are more likely to know the perpetrator. It may also be that individuals with higher incomes do not judge their situation to be as serious because they can afford to wait for the credit card company, bank, or other organization to resolve the problem. In addition, given that most individuals do not ultimately incur out of pocket costs as a result of the forms of identity theft included in the current study, it would also be interesting to explore whether and under what circumstances these organizations are in contact with law enforcement. In many cases, it is ultimately these entities that suffer a financial loss following the crime.

Last, contextual factors, such as the jurisdiction in which the offense occurred or would be reported might also represent a practical consideration that guides the victim’s decision to report (e.g., Higgins, Hughes, Ricketts, & Wolfe, 2008; Lane & Sui, 2010). State laws vary, and victims may be uncertain as to the appropriate agency to contact, especially if their information is being misused in another state. Although only a small number of victims in Appendix Table A1 spoke to the jurisdiction of the offense by indicating that the crime occurred in another state or outside the United States, those who revealed that they did not know what agency to contact or that they did not know they could contact the authorities might have been affected by these jurisdictional considerations. This possibility represents one of the ways that future research can elaborate upon the present study, but again, this will require original data.

All told, the present study advances several areas in both criminal justice scholarship and within the field of victimology by exploring situational and victim characteristics affecting reporting of identity theft to law enforcement and expanding existing criminal justice theories into previously unexplored territory to better understand this decision. This initial exploratory effort is imperfect, but continued examination into victim decision making generally, and especially following identity-related crimes, is needed. Such inquiries are important because if researchers can identify the reasons behind reporting and non-reporting behaviors, perhaps policy makers can encourage police notification when appropriate, ⁶ participation in the criminal justice process by victims when necessary, and ultimately, a more efficient and effective means of responding to identity-related criminal threats.