Watch Yer Graphs

It's pretty easy to get up to mischief with excel. Here's a wonderful post from the hyperlipid blog that recharts one of the results from Framingham in a more honest way. This seems like such a fundamental task of any science reporting: when presented with a graph, check the axes! Are they clear? Do you know what they mean? What's the scale? Good lord, what the hell is going on with the X axis? I guess the saving grace for mainstream reporting is that even the first graph would likely never make it to print (just the ostensible conclusion).