Nonreciprocal and reciprocal dating violence and injury occurrence among urban youth.

To the Editor:

Within public health research, “reciprocal” or “mutual” violence is defined as relationship violence perpetrated by both partners in the same relationship.1–2 Michael Johnson3 coined the phrase “common couple violence,” which he defined as the perpetration of violence by both partners in a romantic relationship during a specific interaction (e.g., disagreement). The study of reciprocity, however, has not been without controversy. The terms “reciprocity,” “mutual violence,” and “sex symmetry” are used interchangeably in the literature to suggest that males and females are both violent in dating or intimate relationships. This has often been reduced to “women are just as violent as men,” resulting in a very polarized field. 4–5

The Youth Violence Survey: Linkages among Different Forms of Violence (Linkages), described by Swahn, Alemdar, and Whitaker,6 asked participants who indicated they had been on a date in the last 12 months about their dating violence experiences, using ten behaviorally specific items assessing a large spectrum of increasingly violent behaviors (e.g. scratched, hit/slapped to punched or hit with something that could hurt to threatened with a weapon and hurt badly enough to need medical care). For victimization, the item was “Thinking about the last 12 months, how often has someone you have been on a date with done the following things to you?” For perpetration, it was “Thinking about the last 12 months, how often have you done any of the following things to someone that you have been on a date with?” These victimization and perpetration items, however, do NOT specify if the violence occurred in the same relationship. Given research indicating that one third of adolescent relationships last less than one month and another third last less than five months, 7 the reported violence likely did not occur within the same relationship and is likely not reciprocal. Although Swahn et al.6 note this possibility in their limitations by saying “...findings may pertain across dating relationships and as well as to multiple partners” (sic)(p. 267), they fail to acknowledge that their data do not assess reciprocity.

The use of definitions consistent with the literature, particularly in the study of reciprocity, is critical to appropriately interpret and use research findings. Research in this area must strive to use valid methods of data collection (e.g., collecting victimization and perpetration data from one member of a relationship about the violence experiences of both members) in order to make any claims about reciprocity. Swahn et al.’s6 paper measures the associations between physical dating victimization and perpetration, some of which may have been reciprocal, and demographic variables. However, the meaningful interpretation of sex differences in the experience of reciprocal physical dating violence, reported by Swahn et al.6, is severely limited based on their analyses.

In sum, Swahn et al.6 do not measure reciprocity as it has been defined in the literature on intimate partner and dating violence. A reader who is not intimately familiar with the Linkages data, however, may not understand this fact given the title of the paper and limited information presented. While they acknowledge that their findings may apply across dating relationships and to multiple partners, adequate information is not provided to allow readers to have a full understanding of how their operational definition of reciprocity affects their ability to measure this construct. We contend that the operational definition of dating violence reciprocity used by Swahn et al.6 is fundamentally flawed and the paper cannot reach its intended goal “...to determine the scope and prevalence of dating violence reciprocity among teens...” (p. 265).