I think its a good start. it will provide preliminary data to refine the next steps.
KJJ that could happen but if the participant reads the instructions about who is considered gifted and how its designated for the study that will head off alot of overachieveing parents.
I dont think there are all that many ubermoms of the mundane trolling the web to participate in polls they arent really the target population for. there might be a few that find it and want to be funny by putting in weird activities, ( jonlaw? you did that didnt you?)
but these possible occurances dont make the overall data not valuable.
I for one am very interested in learning the width and breadth of activities self-identified parents of gifted kids report of the activities of their children vs the reports of self identified kids.
just because a study is limited does not mean it has no value.

Wren, I also dont understand the difficulty with picking one activity. social behavior models would predict that most parents would pick the activity the kid spends the most time at, the one the parent thinks is most important or is most focused on, the newest activity or the one the kid enjoys the most. For the purpose of this preliminary work i dont think its an issue. whatever the reason for picking, the answers are still valid and of interest. and its not a flaw. its just a different question than the first question. would you really want to answer those questions for every activity you wrote down in question one. most people wouldnt bother because the survey would have become too tedious and they would just quit.