That's not an accurate summary of the linked article's conclusions. It makes no conclusion about the percentage of accusations that are false. Rather, it takes as a seemingly arbitrary premise that 8% of accusations are false. From the article:
> Lisak lists seven studies he considers credible, which find false accusation rates of 2.1%, 2.5%, 3.0%, 5.9%, 6.8%, 8.3%, 10.3%, 10.9%....We will stick with the article’s figure of 84,000 reported rapes per year and 8% false accusation rate, for a total of 6,750 falsely accused.
The article's conclusion is actually about the chance that any particular man will be falsely accused in his lifetime. I can't say whether that particular conclusion is accurate. I can only observe that it is meaningless unless one accepts the 8% assumption. Which is rather suspect, considering it appears to have been cherry-picked from amongst the seven studies the author referenced.
> Which is rather suspect, considering it appears to have been cherry-picked from amongst the seven studies the author referenced.
It's near the middle of the range, though (a bit on the high end), so it seems like a reasonable assumption.
TLDR: About 20% (with a wide margin of error) of rape accusations are false.
> "1 in 4 women is raped" was a fabrication by Mary Koss.
Not entirely. According to a CDC report (http://www.cdc.gov/violenceprevention/pdf/nisvs_report2010-a...), 1 in 3 women have experienced rape, physical violence, or stalking by an intimate partner. That said, some have criticized this report for taking too broad a definition, e.g. (http://www.washingtonpost.com/opinions/cdc-study-on-sexual-v...).