What's shocking is that similarity in values don't seem to be a predictor of happiness in a relationship. Could it be that for those in relationships with diff

What's shocking is that similarity in values don't seem to be a predictor of happiness in a relationship. Could it be that for those in relationships with differences, if the difference were significant enough, they merely wouldn't be in a relationship at all, and wouldn't be in the sample data?