Yes, it's been changed. Mitchell's paper doesn't have a graph but there are plenty that do. On mobile now but if you research this topic (don't use Google) you

Yes, it's been changed. Mitchell's paper doesn't have a graph but there are plenty that do. On mobile now but if you research this topic (don't use Google) you can find graphs that show the per year deltas from older data sources to newer. It's not a secret they do this, they admit it when people find out, but it's never brought up by those who are pushing doom for obvious reasons (+the edits specifically created a simple warming trend). In this case you don't actually need a graph anyway, because the editing in that period is so extreme. Just look at his abstract. He says the world was cooling since 1940 and still is in 1975. Obviously that statement makes no sense if you look at modern databases during the period. To understand why he said that you have to look at older graphs, before climatology became so corrupted. They do indeed show a step and decades long fall in temperatures.