When misconduct in scientific research is proved, the punishment is graded according to the severity of the crime. Fabrication gets you fired (Jan Hendrik Schon, Diederik Stapel, Elizabeth Goodwin, Hwang Woo Suk, ...), while falsification (some minor forms, at least) and plagiarism get you some sanctions (you can't apply for a grant for n years, or your students get reassigned to some other group).
I know of only one case where someone was sent to jail, not quite for research miscondut, but for his use of fabricated data in a grant application. In other words, he was jailed for cheating the government using false information.
Here's a case of someone who was sentenced to house arrest for six months for saying certain things in a press release -- even though everything in the press release is factually correct!
Harkonen’s crime, according to the U.S. government, a federal jury and the 9th Circuit Court of Appeals, was willfully overstating in a press release the evidence for benefit of a drug his company made.
The press release described a clinical trial of interferon gamma-1b (sold as Actimmune) in 330 patients with a rapidly fatal lung disease. What’s unusual is that everyone agrees there weren’t any factual errors in the four-page document. The numbers were right; it’s the interpretation of them that was deemed criminal. Harkonen was found guilty of wire fraud in 2009 for disseminating the press release electronically. [...]
“If you applied this rule to scientists, a sizable proportion of them might be in jail today,” said Steven N. Goodman, a pediatrician and biostatistician at Stanford University who submitted a statement supporting Harkonen’s appeal. [Bold emphasis added]