Is there a statistically significant difference in height between different parts of the U.S.?

Each of these 5,000 dots represents one person who reported their height in a national survey. Your job is to split the country into two parts any way you like. Then we'll run a statistical significance test to compare heights between the two sides.

First, adjust this slider to spin the line and split the country into two parts

Each of these 5,000 dots represents one person who reported their height in a national survey. We're sampling people one at a time and updating our estimate of the average height on each side of the line as we go, shown in the bottom plot. The error bars represent uncertainty in estimating these averages. As we sample more and more people, our estimates settle down and our uncertainty decreases.

Believe it or not, after a few hundred samples we should have enough data to conclude that there's a statistically significant difference between heights on both sides of the line, no matter where we draw the line!

So what exactly does this mean?

Does it mean that there's a huge difference in heights on each side of the line?