First, people are generally going to be less committed to a relationship with aggression. The differential effects of intimate terrorism and situational couple violence: Findings from the National Violence Against Women Survey.
Second, a body of studies shows that commitment (think of dedication, here) inhibits negative behaviors, including aggression, which would partly explain why people who are more dedicated to their partners will report less of it.
Even though many break up, it raises the question of why so many of these couples stay together—a subject addressed by the focus on commitment in both papers we describe here.
There are many different published theories of commitment in romantic relationships.
Based on the idea of inertia, you might expect that controlling for some aspects of constraint would lower the degree to which cohabitation was associated with aggression; but living together remained associated with aggression with aggressive relationships continuing.
Selection, Inertia, and Asymmetrical Commitment Do the increased constraints of cohabiting make it more likely that people in aggressive relationships will remain in those relationships, or is something else in the mix?
Similarly, in our Relationship Development Study, a national longitudinal sample of 1,278 emerging adults in unmarried relationships (aged 18 to 34), we found that 48 percent of the unmarried adults reported some sort of physical aggression in the history of their relationship.
Among those with aggression, the odds were five times greater that they would remain together over the next two years if they were cohabiting versus dating (even when controlling for a number of other important variables).
Commitment Dynamics Hitting is especially common in relationships during earlier stages of life.
The percentages in our sample are likely higher due to the wider age range and other differences.
While the percentages in either study may seem high to you, they are consistent with many other studies of those in these earlier stages of life.