I agree regarding influence of parents. I've seen projects that seemed to depend on access to lab equipment from a parent's lab.
Another problem is the competition concept itself, which is myopically focused on the lone genius concept. A lot of real science has to do with noticing others' work and collaborating with them effectively in a complementary way.
Support for Maker fairs may actually make sense for a change.
Collaboration has quickly become the de facto standard in the life sciences. There are examples in every discipline; to give you something specific would only highlight my ignorance of other fields. It happens between members of the same lab group, between groups and between institutions. The single-author publication is becoming a rarity.
For an example from my own experience, researcher A makes a groundbreaking discovery in a disease, and collaborates with mouse expert researcher B at a different institution to develop a model for this. For the most part, any paper in a respected journal with authors from various institutions reflects an aspect of this collaboration.
There are so many. Here is one from a couple of friends. The two groups in this case are the Mars rover scientists, who want to get more usable images of their science targets, and the automated planning and scheduling CS types, who have software that can identify them (if they know what they are looking for), and dynamically change the activity plan in response.
Another problem is the competition concept itself, which is myopically focused on the lone genius concept. A lot of real science has to do with noticing others' work and collaborating with them effectively in a complementary way.
Support for Maker fairs may actually make sense for a change.
Here's a good historical summary which I read on HN: http://www.theatlantic.com/education/archive/2015/03/why-sci...
HN comments: https://news.ycombinator.com/item?id=9207686