I am a member, and I think the AHA does good work, and they certainly have scientific affiliates, but I wouldn't say their data goes much more beyond polling these days either. More so on the Brewers Association side. I have seen just as many posts from people not happy with them that won't be renewing their memberships. I would argue membership data from free resources, particular those associated with brewing supply websites such as Northern Brewer would be more telling of the state of the homebrewer industry. Data Mining web search data is most certainly a valid scientific research method. However you would need more than simple keyword search tools to get accurate data. Plus I would say it depends on what you are looking to get out of the data as well. In this case, all we are really concerned with is an interest trendline. For that purpose, and with an appropriate selection of keywords, the Google tool gets us close enough. For sales data, I would beg and bribe manufacturers and retailers for their sales data. That would be the primary data I would rely on for justifying bringing a new product to market. If the sales are down, forget it. But that doesn't seem to be the case recently.