Studies show that teachers are more effective after a few years of classroom experience, so this new development would be quite disturbing… if it were remotely true.

MORE NEWS: Know These Before Moving From Cyprus To The UK

According to the most recent data from the National Center for Education Statistics, only 9% of government school teachers had less than three years of classroom experience in 2011-12. Even charitably assuming that by “most” Weingarten meant only 50.1%, there would have had to have been massive layoffs and unprecedented hiring in the last two years. Since the number of teachers has not changed significantly in that time, Weingarten’s claim assumes that about 1.4 million experienced teachers were replaced by new recruits since 2012. The latest NCES data showed only 8% of government school teachers leaving the profession after the 2008-09 school year, which is fewer than 275,000.

In other words, Weingarten would like us to believe that the number of teachers leaving the profession has increased five-fold in five years. Even half that number would have resulted in screaming headlines across the nation. It simply did not happen.

These figures are especially hard to believe when government school teacher “accountability” systems routinely rate nearly all teachers as “effective”—even when those same systems categorize schools as low-performing. Here’s an example from Michigan reported this morning, where school performance is somehow declining as teacher “effectiveness” supposedly improves:

For the last two years, every high school student in the Lansing School District received a letter from the district stating that all three high schools are on the state’s low-performing watch list.

The letters are mandated after a school is designated as one of the worst performing in the state by finishing in the bottom 5 percent academically.

Yet, according to the district, the effectiveness of its teachers is increasing significantly while it has had more schools put on the state’s low-performing watch list.

In 2009-10, the district had one school on the persistently lowest achieving list. That number increased to two in 2010-11 and then eight in 2011-12 and six in 2012-13. The 2013-14 list of low performing schools will be released later this month by the Michigan Department of Education.

In 2011-12, the Lansing School District rated all 887 of its teachers as “effective” — the second highest of four ratings available. In 2012-13, 337 teachers received “highly effective,” 456 received “effective,” 20 received “minimally effective” and 1 received “ineffective.”

In 2013-14, 363 teachers received “highly effective,” 301 received “effective,” 16 received “minimally effective” and 1 received “ineffective.”

In three years, only two out of 2,382 teacher evaluations (or 0.08%) rated teachers as “ineffective” in the failing schools. Would they have us believe that there was just an influx of rotten kids?

This problem is not only rampant across Michigan, but across the nation. Nearly all of New York’s teachers were rated “effective” while two-third of students were failing reading and math tests. In Indiana, fewer than 0.5% of teachers were rated “ineffective” last year, though 320 of 2,114 schools received a “D” or “F” grade. In 2013, only 4% of Louisiana teachers were rated “ineffective” though about 28% of schools received a “D” or “F” grade.

Government teacher evaluation systems too often fail to identify ineffective teachers and union rules often make it almost impossible to fire incompetent teachers, yet the AFT’s president would have us believe that there has been an unprecedented number of government school teachers replaced in the last two years. If you believe that, she also has a magic new diet pill to sell you.

MORE NEWS: How to prepare for face-to-face classes

Of course, this isn’t the first time the AFT has muddied national discourse about education policy with dubious “data.”

[Hat tip to Bob Bowdon of Choice Media and Dave Dorsey of the Kansas Policy Institute.]

UPDATE: A Tweeter brought to my attention a 2012 study showing that the modal (most common) years of experience for teachers was under two years in 2007-08. However, by 2011-12, the most common years of experience for teachers was five years  (see page 12). So even assuming that Weingarten meant “modal” instead of “most,” her data was several years out of date. Moreover, citing the modal years of experience is still somewhat misleading given that the average viewer is likely to interpret her statistic as referring to “most” though the mode accounts for less than 9% of total teachers.

Authored by Jason Bedrick