Scoring the JVQ-R2

The JVQ-R2 can be scored in a variety of ways.  Below, we provide guidelines on how to reproduce many of the scores that we use in the National Survey of Children’s Exposure to Violence (NatSCEV) and other work.

Example
Let’s say you are interested in physical abuse by a caregiver.  What scoring options can be applied?

Item level scores.  First, there is just the answer to the “Physical Abuse by a Caregiver” screener.  Anyone who answers “yes” to this question can be classified as a victim of physical abuse (for individual assessment and diagnostic purposes, we strongly recommend following up to get more details). 

Re-scored item.  You can enhance the information in the single item on “Physical Abuse by a Caregiver” by counting caregiver-perpetrated assaults reported to other items.  For example, a child might say yes to the screener on “Assault with a Weapon,” and then identify the perpetrator as a parent.  Note:  You can only do this if you ask the follow-up question on perpetrator identity (in the Full Interview or Abbreviated Interview). 

Module score.  Physical abuse by a caregiver can be combined with screeners on other types of maltreatment, such as neglect and psychological abuse, to determine the total percentage of maltreated youth in a sample. 

Physical assault aggregate.  Physical abuse by a caregiver is one of several different items on physical assault.  Another way we use JVQ responses is to aggregate across all the items on a particular type of victimization (physical assault, sexual assault, witnessing violence, etc).  Physical abuse by a caregiver is included in the aggregate for physical assault.  This score tells you the percentage of youth who had been physically assaulted by any perpetrator, including caregivers as well as peers, siblings, strangers, etc.  

Is it OK to create our own scores? 

Yes.  We get asked this question frequently, and we recognize that new scores might be useful, due to the specific goals of a project or the items selected for inclusion in a project.  For example, although we separate child maltreatment and exposure to family violence, some users may prefer to combine these into a single “family violence” aggregate. 

Categories of Scoring Options
 

Item-level Scores

Can be used with:  Full Interview, Abbreviated Interview, Screener Sum Version, Reduced Item Version
Many users of the JVQ-R2 use the results at the item level, because each item measures a specific form of youth victimization, and many of these, such as teen dating violence, have clinical or research findings that are unique to that form. 

Re-scores

Can be used with:  Full Interview and Abbreviated Interview
Each screener question has follow-up questions that can be used to collect details on the incident, such as perpetrator identity and whether injury occurred.  The follow-up items from the JVQ can be used to greatly increase the precision of the scoring and improve the correspondence of answers with official crime and child protection categories. 
Improved estimates can be obtained by incorporating follow-up data into rates.  For example, some assault screeners are general items that do not specify a class of perpetrators, while others are perpetrator-specific.  This is because past research shows that generic questions about assault are important for identifying stranger and some acquaintance assailants, but often miss assaults by intimates and peers.  To obtain the most complete rates for perpetrator-specific items, follow-up data should be used.  For example, the rate for Teen Dating Violence can be enhanced by assaults reported to other, more general assault screeners that were described as committed by boyfriends or girlfriends (in the follow-up). 
We also use responses to follow-up questions about weapon-use, injury, and sexual penetration to create some special scores for the most severe assaults (see details below). 

Module Scores

Can be used with: Full Interview, Abbreviated Interview, Screener Sum Version
Each module can be scored to produce a one-year or lifetime rate for that module (depending on the form used).  Because of the potential overlap among items (for instance, peer assault and bullying could occur in the same incident), it is NOT recommended that the frequencies be used at the module level.  Rather, module scores should be used as dichotomous scores.  Thus, a “yes” or 1 for a module indicates that at least one form of victimization on that module was reported, whereas a “no” or zero indicates that no forms of victimization on that module were reported. 

Possible module scores include:  Any Conventional Crime (Module A), Any Child Maltreatment (Module B), Any Peer or Sibling Victimization (Module C), Any Sexual Victimization (Module D), and Any Witnessing or Indirect Victimization (Module E).

Aggregate Scores

Can be used with: Full Interview, Abbreviated Interview, Screener Sum Version, Reduced Item Version
Some forms of victimization are asked about in more than one module.  For example, although the basic assault questions are in the Conventional Crime module, a total rate for all assaults should include caregiver and peer assaults as well, which are asked about in the Child Maltreatment and Peer and Sibling Victimization modules, respectively.  For this reason, we have developed several Aggregate Scores.  As with Module Scores, these are scored dichotomously as “yes” if any victimization in that aggregate is reported and “no” if no form of victimization included in that aggregate was identified. 

Detailed Scoring Instructions for the JVQ-R2

The following are some of the most common re-scores and aggregates that we routinely use in publications from the National Survey of Children’s Exposure to Violence dataset (for example, Finkelhor et al., 2009). 

Commonly Used Aggregate Scores (combining answers to multiple screeners)

 

  1. Property Crime Aggregate = YES if any “yes” to:  Robbery (C1) or Personal Theft (C2) or Vandalism (C3).
  2. Physical Assault Aggregate = YES if any “yes” to: Assault with Weapon (C4) or Assault without Weapon (C5) or Attempted Assault (C6) or Kidnapping (C8) or Bias Attack (C9) or Physical Abuse by Caregiver (M1) or Gang or Group Assault (P1) or Peer or Sibling Assault (P2) or Nonsexual Genital Assault (P3) or Dating Violence (P6).
  3. Sexual Assault Aggregate = YES if any “yes” to:  Sexual Assault by Known Adult (S1) or Nonspecific Sexual Assault (S2) or Sexual Assault by Peer (S3) or Forced Sex (Including attempts) (S4).  Note: This aggregate differs from the Sexual Victimizations module because it only includes incidents that involve nonconsensual physical contact. 
  4. Any Maltreatment  -  Child experienced any maltreatment victimization.  Includes all of the items in Any Child Maltreatment (Module B) score, plus any Sexual Assault by Known Adult (rescored).
  5. Witness to Intimate Partner Violence - Child saw one parent get hit (for example, slapped, hit, punched, or beat up) by another parent, or parent’s boyfriend or girlfriend.  Includes item level scores from W1, EF1, EF2, EF3, EF4, and EF5, plus rescores from W3 or W4 if both victim and perpetrator were parents. 
  6. Any Witnessed Violence - Child witnessed any violence.  Includes W1, W2, W3, W4, ECV1, ECV2, ECV3, EF1, EF2, EF3, EF4, EF5, or EF6 (if child saw or heard the violence).
  7. Any Indirect Exposure to Violence - Child was told about or saw evidence of a violence event in child’s household or community (any of screener items W5, SC1, SC2, ECV1, ECV2, ECV3, EF1, EF2, EF3, EF4, EF5, or EF6 if told about or saw evidence of violence).

Common Re-scores

  1. Assault with Weapon  (rescored)  - Someone hit or attacked child on purpose with something that would hurt (like a stick, rock, gun, knife or other thing).  Includes item level scores from C4, plus rescores from M1, P1, P2, P3 and P6 if a weapon was involved.  (Items C5, C8 and C9 have no weapon data)
  2. Assault with Injury  (rescored) - Someone hit or attacked child, and child was physically hurt when this happened. (Hurt means child felt pain the next day, had a bruise, a cut that bled, or a broken bone.)  Includes rescores from C5, C8, C9, M1, P1, P2, P3 and P6 if an injury occurred.
  3. Assault without Injury  (rescored) - Someone hit or attacked child, and child was not physically hurt when this happened. No weapon was used.  Includes rescores from C5, C8, C9, M1, P1, P2, P3 and P6 if no injury occurred.
  4. Assault by Sibling  (rescored)  - Child was attacked by a sibling.  Includes rescores from C4, C5, C8, C9, M1, P1, P2, P3 and P6 if any perpetrator was a sibling.
  5. Assault by Non-sibling Peer  (rescored) -  Child was attacked by a peer, not including any sibling.  Includes rescores from C4, C5, C8, C9, M1, P1, P2, P3 and P6 if any perpetrator was a peer (juvenile), but no sibling was involved.
  6. Dating Violence  (rescored) - A boyfriend or girlfriend of child, or someone child went on a date with, slapped or hit child.  Includes item level scores from P6, plus rescores from C4, C5, C8, C9, P2 and P3 if any boy- or girlfriend perpetrator was involved. (P6 only asked of youth 12 years or older.)
  7. Completed Rape  (rescored) -  Someone forced child to have sexual intercourse and put any part of their body inside child.  Includes rescores from S1, S2, S3 and S4 if penetration occurred.
  8. Forced Sex (including attempts)  (rescored) -  Someone forced, or attempted to force, child to have sexual intercourse.  Includes item level scores from S4, plus rescores from S1, S2 and S3 if penetration occurred. (Note:  completed rape can be identified from S1, S2 and S3 rescores, but attempted forced sex cannot.)
  9. Sexual Assault by Known Adult (rescored) - An adult the child knows touched child’s private parts, made child touch their private parts, or forced child to have sex.  Includes item level scores from S1, plus rescores from S4 if any known adult perpetrator was involved.
  10. Sexual Assault by Adult Stranger (rescored) - An adult the child does not know touched child’s parts, made child touch their private parts, or forced child to have sex.  Includes item level scores from S2, plus rescores from S4 if any adult stranger perpetrator was involved.
  11. Sexual Assault by Peer  (rescored) - A peer made child do sexual things.  Includes item level scores from S3, plus rescores from S4 if any peer (juvenile) perpetrator was involved.
  12. Statutory Rape/Sexual Misconduct (rescored) - Our rates are based on reports to S7 of children aged 12 to 15 years (the item is only asked starting at age 12) who describe a partner who is at least 18 and more than 4 years older than they youth, and with whom they willingly engaged in sexual behavior.  Note:  Statutory rape laws vary considerably by state, and with the full interview it is possible to define this differently to match the laws of a particular state or locality.  This item is typically asked of all youth aged 12 to 17 years. 
  13. Any Sexual Victimization (rescored) - Child experienced any sexual victimization.  Includes any victimization from S1, S2, S3, S4, S5, S6 or statutory rape/sexual misconduct, as rescored.  (Note:  this is similar to Any Sexual Victimization (Module D) score, but item S7 is not counted unless it meets statutory sexual offense rescore criteria.)
  14. Physical Abuse by Caregiver (rescored) - An adult in child’s life hit, beat, kicked, or physically abused child in any way.  Includes item level scores from M1, plus rescores from C4, C5, P2 and P3 if any parent perpetrator was involved.  (Note:  other assault items can also be used in rescores if parent perpetrators are present.)
  15. Custodial Interference/Family Abduction (rescored) - A parent took child, kept child, or hid child to prevent child from being with another parent.  Includes item level scores from M4, plus rescores from C8 (Kidnapping) if any parent or live-out relative was involved.
  16. Witness to Parent Assault of Sibling  (rescored) - Child saw a parent hit, beat, kick, or physically abuse a sibling.  Includes item level scores from W2, plus rescores from W3 or W4 if a parent was perpetrator and any victim was sibling.