To prove this point, in 2010, researchers at Oklahoma State University investigated whether nonphysical punishments are also associated with delinquent behavior later in life. They found that psychotherapy, grounding, and sending children to their rooms all make kids more antisocial. The researchers don't actually believe that psychotherapy causes behavioral problems -- they just wanted to show that these kinds of studies identify associations that aren't necessarily causally linked the way you would expect. Discipline is associated with behavioral problems in part because discipline is caused by behavioral problems.
There's also the possibility that parents who spank are different in important ways from parents who don't. A 2010 study published in Pediatrics found that spankers are more likely than non-spankers to be young, depressed, under a lot of stress, and substance abusers -- so it's hard to know whether some of the purported effects of spanking might actually be caused by poor parenting skills in general. And what about genetics? Substance-abusing or depressed parents who spank may have given their kids genetic predispositions to mental health problems, so, again, it's not necessarily the spanking that's the cause of future problems. Some studies try to control for these types of confounding variables, but not all.
Then there are what some researchers call "lumping problems" -- when studies are reported as being about spanking but aren't really about spanking. A 2002 meta-analysis scrutinized data from 88 studies and reported that although physical punishment makes kids more immediately compliant, it also makes them more aggressive in the short and long term and more likely to engage in criminal behavior as adults. This analysis has been cited as some of the strongest, most comprehensive evidence that spanking is dangerous, but it actually looked at far more than just spanking: In 16 percent of the studies, parents were hitting their kids with objects, not with the palms of their hands. It also included punishments such as pinching and slapping. It doesn't surprise me that kids who are hit with objects or pinched or slapped end up more screwed up than kids who aren't. But it's misleading to conclude that these kids were worse off because they were spanked.