Note: those rates span a lifetime. It doesn't mean that every current relationship has a high rate of abuse.

Whatever the temporal basis of comparaison, it kind of means that any relationship involving a cop is more likely to result in abuse. no?

Yes. But it doesn't mean that, given a random sampling of 100 cops in relationships, you could expect ~40 of those to be currently abusive of their partners.

This is also what I initially understood. But still, it means that if I get 100 cops in a room, about 40 have been or will be in an abusive relationship. That is still very telling/surprising.

Does this also mean that of 100 randomly selected people (of whom an ordinary proportion are cops), that about 20 of them have been or will be in an abusive relationship?

I don't mean to downplay the 40 number in the police case (which is very high and kind of terrifying), but, 20% is still a lot! (and so, 40% is even more).

That's, for every 5, 1, on average! That is shockingly high.

wow?

Yelling is domestic violence. I think 20% is a bit low.

I mean at any point in time it's not likely any given person, even an abusive person, is actively doing something. It doesn't mean it's less of a problem.

"Domestic violence occurs in about 40% of households with police officers"... I think that is what the above claim means.

difference between "it occurs in 40% of said households" and "40% of officers are the sole abusers".

Abusers in relationships tend to be abusers in all of their relationships.

Do you have a citation for this statistic?