Yes, it can be altered, or important and salient points can be left out (more on that in a moment), but it doesn't even HAVE to be ALTERED. A great deal is in the INTERPRETATION and PRESENTATION of the data.
Prime example of this is an anti-gun article I was reading on the bang list which claimed, tragically, that for every one woman who killed an attacker (I'm quoting from memory here but believe I'm real close), 51 women were killed by guns.
The point was, GUNS ARE INHERENTLY BAD.
And naturally, many in this forum said, "Well, if those 51 women had had handguns to defend themselves with, it probably would've been 52-0." That may have been stretching reality, since it might've been more like 26-26, but the point is: the statistic could BETTER be used to justify handgun defense!
Now here's what I mean by omitting relevant information:
Twenty-nine percent of high-school boys have at least one firearm; most are intended for hunting and sporting purposes.
Six percent say they carry a gun outside the home.
The National Institute of Justice, 1998
If I'm not mistaken, this is from a study which ALSO reported (as an incidental result) that CHILDREN WHOSE PARENTS BUY THEM LEGAL GUNS ARE MUCH LESS LIKELY TO COMMIT CRIMES THAN THOSE WHO OBTAIN GUNS ILLEGALLY, AND, INTERESTINGLY, ARE ALSO SOMEWHAT LESS LIKELY TO COMMIT A CRIME THAT CHILDREN WITHOUT A GUN IN THE HOME.
But we aren't told that part in the cited compilation of statistics...