> anecdotal evidence, ie, "what worked for them", is still valid data, and should be valued as such It's useful information, but it's also limited information.

> anecdotal evidence, ie, "what worked for them", is still valid data, and should be valued as such It's useful information, but it's also limited information. If Atkins worked for your buddy two cubes down, that might indicate that it's worth trying yourself, but it's less useful in terms of making, say, national level nutritional policy. Though I suppose in this case I'm more worried about lobbying than anecdotal data.