Know your sample!

Almost in the final stages of my exploratory data collection, I realize that some of my starting points weren't the most adequate. I had only a vague idea about who my sample would be and a much vaguer idea still how to get hold of it.

I defined one of my samples as "successful farmers" without making clear that I mean by "success" someone who is able to live exclusively from the income derived from farming. Therefore I haven't systematized strategies beforehand how to access successful farmers and how to distinguish them from the crowd digging the backyards.

My sampling idea was "snowballing", something really simple as decribed for example in Wikipedia. But: how to find the right entry points to start snowballing?! If you ask an antisocial farmer for further contacts you land on a dead-end of your pretty snowballing strategy. I followed my supervisors advice and tried everything, in order to find something that works. Just: I haven't found anything that works properly. If I phone the farmers listed in the yellow pages, half of them are already dead or out of farming. If I go to farming associations and wait behind the door to catch a farmer, only old housewives turn up asking for subsidies... If I ask someone if he can indicate me a farmer I am told "Oh, real farmers? There is a man in village YZ who still sells some grapes..." I follow up the indications and find an old retired emigrant who in fact sells some grapes to improve his cash. Can I consider him to be a "successful farmer"?

Well... in the end, what I really need and do is to gather a profile: "Who is still in farming?"; "What are their liveilhood strategies?". But, as it happens always, sitting at my far away desk, it took me weeks to formulate the question "How do you explain the success of your farm?". And the answers I'm getting are:"I might skip next week."; "Success? I'm not gonna reply that one."