spacer spacer spacer
JRSA's IBR Resource Center
Return to Main Page
Background and Status of Incident-Based Reporting and NIBRS
State Profiles
Using Incident-Based and NIBRS Data
References and Related Sites
Available Syntax and Sample Data Files
line divider
IBRRC Fact Sheet
Contribute Information
Site Map
Justice Research and Statistics Association (JRSA) Home
 
Notice of Federal Funding and Federal Disclaimer
 

Standardizing the Display of IBR Data: Relationship Variables

Data Collection and Quality Issues Associated with Relationship Variables

Victim-To-Offender Relationship at a Glance

Segment? Victim
Required? For crimes against person and robbery offenses
Number of Entries? 10
Number of Coding Options? 26

Data Collection

In the Summary reporting system, the victim's relationship is reported only for homicides.  In NIBRS, the victim's relationship to the offender(s) is reported when the victim is the object of a Crime Against Person.  Victim-to-offender relationship data are also reported for Robbery because one of its elements is an assault, which makes it a violent crime.

According to the FBI Data Collection Guidelines, the relationship variable is to be used to report the relationship of the victim to offenders(s) who have perpetrated a crime against person or a robbery against the victim.  Therefore, this data element is to be used only if one or more of the following offenses are reported:

09A Muder and Nonnegligent Manslaughter 11D Forcible Fondling
09B Negligent Homicide 120 Robbery
09C Justifiable Homicide 13A Aggravated Assault
100 Kidnaping/Abduction 13B Simple Assault
11A Forcible Rape 13C Intimidation
11B Forcible Sodomy 36A Incest
11C Sexual Assault with an Object 36B Statutory Rape

When one of these offenses occurs, up to ten entries can be made for each victim.  There are 26 allowable entries:

SE Victim was Spouse FR Victim was Friend
CS Victim was Common-Law Spouse NE Victim was Neighbor
PA Victim was Parent BE Victim was Babysittee
SB Victim was Sibling BG Victim was Boyfriend/Girlfriend
CH Victim was Child CF Victim was Child of Boy/Girlfriend
GP Victim was Grandparent HR Homosexual Relationship
GC Victim was Grandchild XS Victim was Ex-Spouse
IL Victim was In-Law EE Victim was Employee
SP Victim was Stepparent ER Victim was Employer
SC Victim was Stepchild OK Victim was Otherwise Known
SS Victim was Stepsibling RU Relationship Unknown
OF Victim was Other Family Member ST Victim was Stranger
AQ Victim was Acquaintance VO Victim was Offender

The category "Victim was Offender" is to be used in cases where a participant in the incident is a victim and offender in the incident, such as domestic disputes where both husband and wife are charged with assault; double murders; or barroom brawls where many participants are arrested. 

When coding relationships, the FBI relies on the investigating officer to best determine the relationship from the information available during the preliminary investigation.  However, there are a few guidelines offered.  When coding ex-boyfriends/ex-girlfriends, BG (Victim was Boyfriend/Girlfriend) fits better than acquaintance.  Ex-domestic partners would most likely be coded as CS (Victim was Common-Law Spouse), BG (Victim was Boyfriend/Girlfriend), or HR (Homosexual Relationship), depending on the circumstances.  If the victim is a child in a common-law relationship, the most likely code would be CF (Victim was Child of Boyfriend or Girlfriend), where one might assume they could have been common-law.

When reporting these data, keep in mind that the relationship requested is that of the victim to the offender.  Also, the relationship of the victim to each offender is to be reported. 


Data Quality Issues

Relationship codes can be very useful in categorizing domestic violence incidents.   In order to be sure that the offenses truly are domestic in nature, the codes need to be checked against different elements in the data.

bullet.gif (326 bytes)

Check that stepparent is coded correctly
bullet.gif (326 bytes) Check that incest involves only family members

bullet.gif (326 bytes)

Check that relationship codes made sense based on reported ages

bullet.gif (326 bytes)

Check that stepparent is coded correctly

In the summary system, spouse was coded as SP; in NIBRS, however, spouse is SE and stepparent is SP.  As a result of the change, some spouse relationships may be miscoded as stepparents.  This can be checked by looking at the ages of the victims and offenders where stepparents are reported as the victim.

In the 1999 NIBRS file, 2,971 (.1%) of victims are reported as stepparents who have been victimized by their stepchildren. 

There are a couple of ways we can look at this.  First of all, it is unlikely that a stepparent would be younger than the offending stepchild.  When we compare ages, we see that 40% of the victims are younger than the offenders.  Although it is conceivable that stepchildren could be older than their stepparents, it is possible that the relationship is miscoded.  All of these cases should be reviewed.

younger_victim.jpg (17479 bytes)

Since the miscode could be for stepchildren rather than for spouses, we can also look at the age differences between the victim and offender.  An arbitrary age difference can be selected; in this example, victims and offenders within 8 years of each other's age were chosen.

After selecting cases in which both ages are known, we see that 28% of the victims are close (within 8 years) in age to the offenders.  Since it is possible that a stepparent could be within the range of 8 years of his or her stepchildren, we would need to select these cases and review each of them by hand.

We can then look and see how many of the reported stepparent victims who are younger than the offenders compare in age.

stepparent2.jpg (20400 bytes)

This table suggests that 70% of the younger victims may be miscoded stepchildren.   The remaining 30% may be miscoded spouses. 

In matching these segments, it is unlikely that there will be multiple victims in the same incident that are both stepparents.  If this occurs, there was probably a coding error.  Please see Check that relationship codes make sense based on reported ages, below.

Download the SPSS syntax for this data quality check.
Note: Please check that the variable names used in this syntax match the variable names in your data file.  If you need assistance, contact JRSA.


bullet.gif (326 bytes)

Check that incest involves only family members

The offense of incest should only include family members.  In the 1999 NIBRS file, there are 401 victims of incest. The following table lists the relationship of the victim to the offender.

incest_rel.jpg (94588 bytes)

When looking at the relationship of the incest victim to the first offender, 23% of the relationships are outside of the family and 19% are unknown.  Since it is possible that there are more than one offense or offender listed, each of the victims with other than a familial relationship with the offender should be checked to ensure that the correct offense and offender are being compared.  For example, if a victim engages in sexual behavior with her brother and his friend, then two offenders would be listed, and one would be coded as a sibling and the other as otherwise known.  Since when running these tables we cannot know which offender relationship is being displayed, each record should be checked and common sense used.

Download the SPSS syntax for this data quality check.
Note: Please check that the variable names used in this syntax match the variable names in your data file.  If you need assistance, contact JRSA.


bullet.gif (326 bytes)

Check that relationship codes made sense based on reported ages

Although the relationship code should be the victim's relationship to the offender, it is possible that the relationship of the offender to the victim was recorded.  In order to check that relationships are coded correctly, we can look at the ages of parents and children, stepparents and stepchildren.  In general, common sense would tell us that parents are older than their children and that stepparents are older than their stepchildren.  Although it is possible that stepchildren could be older then their stepparents, it is unlikely.  In this example, the age difference between victims and offenders is displayed.  For this example, only single victims with one offender are compared.  In actuality, cases with multiple victims and/or offenders should also be reviewed.  Age differences between parents and children will be discussed first.

parent_age.jpg (30008 bytes)

As the table shows, .5% of the "parents" are younger than their "children".  These cases should be reviewed.

Now we'll look at stepparents and stepchildren.

stepparent_age.jpg (28298 bytes)

Here we see that 38% of the "stepparents" are younger than their "stepchildren".  These cases need to be reviewed to determine whether these are correctly coded or whether they are miscoded stepchildren or spouses.

Download the SPSS syntax for this data quality check.
Note: Please check that the variable names used in this syntax match the variable names in your data file.  If you need assistance, contact JRSA.