In the below plot: Clicking a data point will open that Wayback archive page.

The colors are coded by the department website.

For each scrape, we calculate how far it is from the 2011 scrape (we divide the scrape count by the census count, and say a scrape is "suspicious" if that proportion is greater than 2 or less than 0.5). We divide the number of suspicious scrapes by the total number of scrapes: that is the ratio displayed next to each university name.