You might have misread that. It says 2% of global adults (“across all countries”) are level 5. It also says that level 5 does not officially exist, and level 4 is the highest level on the PIAAC scale.

According to that article, 12% of Americans are in the highest level measured. Over 13% of Americans have post-graduate degrees. And while on the scale you linked to, the U.S. ranks lower than a lot of countries, don’t ignore the fact that the spread of absolute scores above the U.S. is in the single digit percentages. Framing these particular scores as representing some kind of glaring education problem here seems to be over-emphasizing minor global differences and ignoring some context.

I hope this gives you reasons to feel more optimistic. We have plenty of room to improve, but it is clearly not as bad as you thought.

If you go to https://nces.ed.gov/pubs2014/2014008.pdf for a more detailed look at the data, you'll find that the fraction of adults world-wide who score at level 4/5 is the same as the USA. In age breakdowns, more US 55-65 year olds are at that top literacy bracket, and in every other age cohort is worse.

Therefore there is no reason to believe that the US at the level 5 is any different than the global average. And there is reason to suspect that the USA is getting worse over time.

Incidentally I was merely stating it as data and didn't blame this on a glaring education problem here. That would be a much longer discussion.

It is clear that the earlier claim "under 2% of Americans read beyond a grade 12 level" is not true, and not what your supporting links actually say. It's absolutely clear that your speculation "Odds are than under 1% could read that [Prezi's] terms of use and actually understand it." is both not true and dramatically pessimistic.

Yes, legalese is dense. Hardly anyone reads it because it takes a lot of time, not because nobody is capable. How many people read below a 12th grade level (whatever that means) are using Prezi in the first place?

I'm just pointing out that your initial interpretation of the data was off by an order of magnitude.

There absolutely is a level 5. It was defined and measured in the survey. Both links confirm that.

But then they did not break it down in the reporting because few enough people fell into it for their survey to produce reasonable statistics on a per country level.

Now you are correct that I should not have said "under 2%". But given that the portion of the USA at level 4/5 matches the international portion at the same level (both cases 12%), it seems likely that the portion of the USA at level 5 (which the survey was not large enough to estimate) is about the same as the international average (which the survey measured at 2%).

Therefore even though the survey did not report a figure for the USA, it is likely that the real figure is around 2%.

I stand by my belief that this terms of service is sufficiently complex that even fewer can understand it.

You are right that the average Prezi user is more likely to be literate than the average American.

This figure is meaningless, which is part of why they didn’t report it. You’re fixating on a number that doesn’t inform you.

Whatever their level 5 means, it excludes at least 90% of the people who wrote large technical and impenetrable documents just to graduate from school, not including all the practicing lawyers, doctors, engineers, scientists, and writers in the country.

There’s a disconnect between using the 2% number to represent any kind of literacy threshold, and reality. It may have measured something, but this “2%” does not represent a limit on number of people who have the capability to read Prezi’s terms of service.

> I stand by my belief that this terms of service is sufficiently complex that even fewer [than 2%] can understand it.

That’s unfortunate, I’ve failed to make a convincing argument. I was hoping to help you see more clearly that’s not what your data actually says, and is also contradicted by other data as well. The data isn’t wrong, but your interpretation is.