Lecture 5: Victimization

Victimization: Lecture Overview

Good luck in your exam this week! This week we consider the patterns evident in observable statistics regarding victimization, but we also consider what patterns are evident in our perception of victimization. The patterns in our perceptions are consistently different from the patterns reported by official statistics-gathering agencies.

Our lecture subjects this week are:

Get Ready to Celebrate!

2015 Crime in the United States -- Getting Ready to Go

Cue that Kool and the Gang song and get ready to spin your disco ball:  the FBI is about to release its annual Crime in the United States report for 2015 (why not 2016? The year hasn’t finished). This report will compiles all the Uniform Crime Reports across the nation for the year 2015.  You can tell that the report is about to be released when clicking this link sends you to a login page (for reporters and other insiders with early access).  Sometime in the next week, that link will be open to everybody.  Once it opens, you’ll find me referencing these statistics from now on where possible in the course.

If you’re like me and consider a fresh data release to be some kind of holiday, then you should seriously consider graduate school.  My friend, you can get paid for maintaining your information habit.

Reports of Victimization Risk and Actual Victimization Risk

Prediction of Victimization Risk by Crime Type
In our anonymous course survey at the beginning of the semester, I posed the following challenge:

“Make your best guess by completing this sentence: in one year’s time, 1 out of ____ people in the United States will become a victim of motor vehicle theft.”

I issued identical challenges for the crimes of murder, rape and robbery. How well did you predict the risk of these types of crime? It turns out that most of you strongly overestimated the risk of criminal victimization.

For motor vehicle theft, the median guess was that 1 out of 10 people in the United States would become a victim of motor vehicle theft in one year’s time. By “median,” I mean that half of students guessed a number lower than 1 out of 10 and half of students guessed a number higher than 1 out of 10. The actual data (based on the 2014 Crime in the United States report of UCR data) indicates that 1 in 463 people in the United States was actually a victim of motor vehicle theft last year. 81.4% of students guessed the motor vehicle theft rate in America to be higher than Uniform Crime Reports data suggests.

The median student guess was that 1 out of 150 people in the United States would become a victim of murder in one year’s time. Available data (based on Uniform Crime Reports) indicates that 1 in 22,222 people in the United States are victims of motor vehicle theft in a year’s time. 77.8% of students guessed the murder rate in America to be higher than Uniform Crime Reports data suggests..

The median student guess was that 1 out of 14 people in the United States would become a victim of rape in one year’s time. Available data (based on Uniform Crime Reports) indicates that 1 in 2.597 people in the United States are reported as victims of rape in a year’s time. 96.3% of students guessed the rape rate in America to be higher than Uniform Crime Reports report data suggests.

Finally, the median student guess was that 1 out of 50 people in the United States would become a victim of robbery in one year’s time. Available data (based on Uniform Crime Reports) indicates that 1 in 960 people in the United States are reported as victims of robbery in a year’s time. 77.8% of students guessed the robbery rate in America to be higher than Uniform Crime Reports data suggests.

What does this mean? Does it mean that students in the criminology course overestimate the occurrence of crime? I might be tempted to say “yes,” but there’s another possibility. It could be that as students of crime, you are correctly reporting the victimization rate for these crimes while the Uniform Crime Reports data is consistently underestimating the victimization rate. Underreporting of some crimes to police and by police, particularly the crimes of rape and robbery, is a well-known phenomenon. On the other hand, it’s hard to argue that the murder rate is wildly underreported, since in the United States the disposition of the population is well-tracked. At least in that instance, students appear to very strongly overestimate the occurrence of murder in the United States. Our country is much safer, from murder at least, than we think it is.

Prediction of Victimization Risk by City in Maine
In our anonymous course survey at the beginning of the semester, I asked you to guess (without looking at crime statistics) which of the following cities in Maine had the highest violent crime rate:

South Portland

The following were the top guesses for the Maine city with the highest violent crime rate:

#1: Lewiston (42.9% of responses)
#2: Portland (25.0% of responses)
#3: Augusta (7.1% of responses)
#4: Bangor (7.1% of responses)
All Others: (17.9% of responses)

From that list of cities, the following are the actual four Maine cities with the highest violent crime rate in 2014, as reported in the 2014 Crime in the United States report of Uniform Crime Reports statistics:

#1: Biddeford (488.2 violent crime reports per 100,000 population)
#2: Augusta (416.7 violent crime reports per 100,000 population)
#3: Waterville (330.8 violent crime reports per 100,000 population)
#4: Sanford (369.3 violent crime reports per 100,000 population)

The only community that appears in both lists is Augusta, and not many students chose Augusta to sit atop the rankings. The other top three communities that students guessed would be highest in crime are not found on the actual top 4 list. Guesses and reality regarding the city of Lewiston and Portland are particularly discordant; students by far guessed that Lewiston’s crime rate would be the highest in the state. In reality, the crime rate for Lewiston and Portland are ranked right in the middle of cities listed above, neither strikingly low nor strikingly high. It is a typical city; why did these two communities attract so much attention as a high-crime city?

I invite you to discuss the topic in the comments section appearing at the end of this lecture.  What’s your theory for the difference?

Whatever the cause, the stark difference between “common sense” and crime in Maine’s larger cities shows us why it’s always better to collect data than to go with our gut feelings about what we’ve heard about the communities we live in: sometimes the stories we hear have truth to them, but sometimes the stories we tell turn out to be inaccurate. Careful measurements by social scientists have a role to play in finding out the difference.

Gender, Actual Risk of Victimization and Fear of Victimization

There is a strong gender gap in fear of crime; in a Gallup poll, 50% of women report being afraid to walk alone at night, while only 22% of men report that fear. This gender gap has replicated repeatedly in social research (Hale 1996).

Other inaccuracies in fear of crime exist; for instance, residents of rural areas appear to be more likely to fear they will be the victim of a terrorist attack although urban areas are actually attacked more often (May et al 2011). But the gender gap in fear of crime has perhaps attracted the most attention of researchers. What accounts for this large difference? Lane and Fox (2013) identify four explanations in scholarly research:

Explanation 1. Actual differences in vulnerability to crime. The idea here is that people accurately receive risks and opportunities in their environment. According to this account, because adult women on average tend to be smaller than adult men, they might be actually more likely to be victims of crime, and that an actual risk of crime leads to a greater fear of crime. This idea is not supported by available data. According to the National Crime Victimization Survey:

  • The rate of robbery victimization is 2.7 men per 10,000 persons and 1.7 women per 10,000 persons
  • The rate of assault victimization is 18.3 men per 10,000 persons and 14.3 women per 10,000 persons
  • The rate of aggravated assault victimization is 3.9 men per 10,000 persons and 2.8 women per 10,000 persons

The NCVS-reported rate of rape is the only category of violent crime for which the rate of victimization is higher for women (1.3 women per 10,000 persons) than for men (0.3 men per 10,000 persons). The gender gap in murder (data about which cannot be collected in a crime victims’ survey) is evident from Uniform Crime Reports. In 2011 (the last year for which data is available), 9829 murder victims were male and 2813 murder victims were female.

In short, with the significant exception of rape, men are actually significantly more likely to be victims of crime. Differences in fear do not appear to be a simple expression of actual differences in victimization risk. Interestingly, however, Hilinksi et al (2011) found that women in a college setting often referred to their size as a reason why they perceived themselves to be at a greater risk for crime. Although men are at a greater risk, this account has cultural power.

Explanation 2. Patriarchy, or male domination of women. According to this theory, fear is not simply a consequence of the physical trend that men on average are larger and stronger than women. It would also be important to note that men tend to hold positions of higher power, resource control and legitimate authority than women do. In a patriarchal system, men as a class of people would be able to control women as a class of people in part through actual threat of criminal violence against women, and in part through the cultural definition of free movement in public as “dangerous” for women and therefore to be prohibited.

The former avenue of patriarchal control is evident in stalking statistics, in which, according to data from the National Crime Victimization Survey the stalking of females appears to be an activity strongly dominated by male offenders, while the stalking of males appears to be carried out equally by males and females (Catalano et al 2009: 6). The latter avenue of patriarchal control is described in a Manisha Roy’s ethnography of Bengali Women, a culture in which to protect women’s “modesty” and “safety,” Hindu women of certain classes in the mid-twentieth-century were not permitted to travel outside a home compound without a male escort (Roy 1992). Are there similar if perhaps less noticeably extreme constraints based on “modesty” and “safety” for women in the United States?

Explanation 3. Socialization. As Lane and Fox (2013) put it, it is possible “that women are socialized to be fragile and submissive whereas men are socialized to be tough and dominant.” Think of Walter Miller’s theory of focal concerns from Walsh textbook Chapter 4: if men are brought up to value assertiveness, resilience and strength, then they may not be inclined to admit feelings of fear, even to themselves.

Explanation 4. The Shadow of Sexual Assault. This account stems from the work of sociologist Kenneth Ferraro (1996), and asserts that women interpret their risk of victimization in terms of their risk of being victimized by rape; the “master account” of crime for women in our culture involves the either perception that non-rape criminal victimization may lead to rape of women, or that being raped may lead to additional criminal victimizations including murder. According to the Shadow of Sexual Assault Thesis, women will only perceive a greater risk of violent or property crime to the extent that they perceive a high risk of being raped. Women who do not perceive a high risk of raped, according to the theory, will not perceive otherwise a high risk of other criminal victimizations. The Shadow of Sexual Assault Thesis has received empirical support not only in Ferraro’s original (1996) study, but also in later studies of fear of gang violence (Lane and Fox 2013; Lee et al 2011), on-campus victimization (Hilinski et al 2011; Lee et al 2011), property crime (Lane and Fox 2013; Lee et al 2011) and violent crime (Lane and Fox 2013).

Finding Homophily in Crime

In a study of crime in San Antonio Texas, Cancino, Martinez and Stowell (2009) find that the most common pattern of robbery is within ethnic groups, not between them. Why? It turns out that people rob those with whom they rub elbows, those who live in their own neighborhoods — and many neighborhoods in San Antonio are strongly segregated by ethnic group.

People don’t just live in neighborhoods; they occupy a more virtual space as well, a space defined by activities of daily life — family, school, work, sports, dining, entertainment, and worship. All of these social environments tend to draw very similar people along lines of age, race, income, educational status — and just as individuals tend to rob people in their own physical neighborhood, so they tend to victimize people in their own social environment. The consequence is that people victimize people who are like them. The pattern that people are especially likely to associate with others like themselves is called “homophily,” and it occurs for all sorts of kinds of interaction, of which a victimizing interaction is one (McPherson, Smith-Lovin and Cook 2001). We’ve discussed this pattern briefly before, but it’s worth a review.

Homophily occurs by age in violent crime. In this chart using 2008 National Crime Victimization Survey data (sadly not updated since the important study by Baum et al 2009), the perceived age of the offender in stalking cases is compared to the age of a stalking victim:

Table 1. Perceived age of the stalking offender, by age of the victim
Age of the victim
18-20 21-29 30-39 40-49 50 or older
 Offender Under 18 10.9 0.7 1.8 2.1 2
 age 18-20 41.6 5.7 2.3 2.9 1
21-29 23.3 48.2 13.8 8.8 3.8
30-39 5.1 23 37.6 16.7 16.3
40-49 6.7 7.7 20.8 34.2 18.7
50 or older 2.4 5.9 9.9 21.6 34.6
Age of offender unknown 10 8.8 13.9 13.7 23.6
 Total % 100  100  100 100 100
Number of victims 349,490 929,080 752,690 722,890 663,660

The numbers are percentages, adding up in each column to 100%.  In every column, the offender age with the highest percentage of cases is the category that is the same age category as the victim. Older people are disproportionately stalked by older people; younger people are disproportionately stalked by younger people.

The last row of the above table contains the number of victims of stalking in each age category. As we know from an earlier lecture, most criminal activity like stalking is carried out by the young — people in their late teens and early 20s. Because people disproportionately stalk those who are similar to them in age, victims of stalking are disproportionately young. Victims of stalking in a range of just three years of age — 18 through 20 — are nearly half the size of the pool of all stalking victims aged 50 and up (a much larger age range).

Well, according to 2014 National Crime Victimization Survey data (remember, you’ve read about this survey), this age-crime trend also bears out for victims of violent crime in general, beyond stalking:

Percent in Certain Age Ranges in the United States who have been the victim of at least one violent crime in the last year (source: National Crime Victimization Survey)


Cancino, Jeffrey, Ramiro Martinez Jr. and Jacob I. Stowell. 2009. “The Impact of Neighborhood Context on Intragroup and Intergroup Robbery: The San Antonio Experience.” Annals of the American Academy of Political and Social Science 623: 12-24.

Catalano, Shannan, Erica Smith, Howard Snyder and Michael Rand. 2009. “Female Victims of Violence.” Bureau of Justice Statistics: Accessed 4/1/2013 at http://bjs.gov/content/pub/pdf/fvv.pdf/.

Cullen, Andrew. 2011. “Struggle and progess: 10 years of Somalis in Lewiston.” Kennebec Sun Journal, December 18.

Ferraro, Kenneth F. 1996. “Women’s Fear of Victimization: Shadow of Sexual Assault?” Social Forces 75: 667-690.

Hale, Chris. 1996. “Fear of Crime: A Review of the Literature.” International Journal of Victimology 4: 79-150.

Hilinski, Carly M, Kindsey E. Pentecost Neeson and Heather Andrews. 2011. “Explaining the Fear of Crime Among College Women, in their own Words.” The Southwest Journal of Criminal Justice 8(1): 112-127.

Lane, Jodi and Kathleen A. Fox. 2013. “Fear of Property, Violent, and Gang Crime Examining the Shadow of Sexual Assault Thesis Among Male and Female Offenders.” Criminal Justice and Behavior 40(5): 472-496.

Lane, Jodi, Angela R. Gover and Sara Dahod. 2009. “Fear of Violent Crime Among Men and Women on Campus: The Impact of Perceived Risk and Fear of Sexual Assault.” Violence and Victims 24(2): 172-192.

Lee, Daniel R. and Carly M. Hilinski-Rosick. 2012. “The Role of Lifestyle and Personal Characteristics on Fear of Victimization among University Students.” American Journal of Criminal Justice 37: 647-668.

McPherson, Miller, Lynn Smith-Lovin, and James M. Cook. 2001. “Birds of a Feather: Homophily in Social Networks.” Annual Review of Sociology 2001: 415-444.

Roy, Manisha. 1992. Bengali Women. Chicago: University of Chicago Press.

3 thoughts on “Lecture 5: Victimization”

  1. My theory on the difference of which city has the highest crime rates could be due to people believing that the cities with the highest populations would have the highest crime rates as well as the cities with the highest diversity would also cause that.
    Media attention to crimes could also contribute to the belief that a certain city has a higher crime rate. If a particularly horrible crime happens in a city, such as that body burning case in Bangor, people could automatically assume that city is a very crime ridden place.

    1. Karla,

      Thanks for writing in. Almost as important as what your predictions are is why you make them. Why would you say that size and diversity would increase crime rates?

      Best, Prof. Cook

  2. I wasn’t actually meaning that size and diversity INCREASES crime rates, just people’s perception of that crime rate.
    Crime is normally more prevalent in more dense populations. The more rural, the less crime. Or at least that is what I have been led to believe before taking this class. As far as the diversity goes, it would be linked to stereotyping. Blacks are thought to be more crime driven by not only the general public but also the police and justice system. This belief would prompt people to assume that the crime levels would be higher in a black neighborhood as well as the crimes are being committed more by blacks. The media’s twist on also portraying blacks usually in a negative light contributes to that overall thought process.

Leave a Reply

Your email address will not be published. Required fields are marked *