I propose to address educational reform with a crowdsource approach. This proposal has implications from the local level all the way up to the federal level, because it is a nationwide program that includes some federal funding but is under local control.

This proposal is part of a nationwide approach that we should share with other state parties. I will describe how we can do this at a state and local level, and also how it would work nationally. In addition we can try this out as an internal educational policy development tool, even before it is written into law.

Here's how it would work.

1. The government would create a website where anyone can propose an idea for a pilot program to be tried out at a single American school. We'd get millions of overlapping ideas, some idiotic and others brilliant.

We can put this site into service right now, to gather educational reform ideas and to prove that the idea can work. And also to smooth out the inevitable bumps.

1. Any citizen could vote thumbs up or thumbs down for every idea on the site. In this way, the most popular ideas would rise to the top and the dumb ones would sink into obscurity.

An idea would pass through three stages, in which each stage provides more information than the last. This is clarified near the bottom of this explanation.

We should develop a way to ensure that people are unique when they vote on an idea. We need to ensure fairness while making sure that privacy concerns are fully addressed.

1. Any local school district could sign up for any of the programs, providing the name of a school where they would try out that program if given the option. Conservative school districts would pick different ones than those in more liberal areas, and that's okay.

2. Decision makers would use objective (non-ideological) criteria to decide which pilot programs to fund. For example, it might fund all programs where five or more school districts had signed up, and there was at least a 70% level of popular support. The cost of the program might be factored into the equation, so that programs with the "most bang for the buck" would be more likely to succeed.

3. A fund would be created to cover setup costs of the pilot programs. For each chosen program, a specified amount of dollars would be allocated to cover startup costs. The local district would be responsible to continue funding the normal operations of the school in the same way that they are already funding it.

Local districts will sign up only for ideas they believe will work. They'll get funds to help, but they won't want to be left with a non-functioning school if the idea fails disastrously.

This fund could be created at the state level, but the ultimate idea is to have a national fund if this is a nationwide program.

1. A set of objective criteria will be used to measure the performance of a pilot program. It will include academic performance, but also college acceptance rates, graduation rates, incarceration rates and other numbers. We're not just testing math and spelling, we also want to know if students are successful in life, after they leave. We have to do this using objective criteria that can be measured and applied without being controversial. Everyone wants a school that causes incarceration rates to go down and college acceptance rates to go up.

School reforms are often "gamed" by teachers and administrators eager for funding. For example, using a standardized academic test, teachers may "teach to the test" to improve their numbers with no actual benefit to students. If we measure graduation rates, administrators could simply soften graduation requirements. For these reasons, it is important to include external factors such as college acceptance rates, mean and median income after graduating, etc.

1. Schools are measured against themselves, not against other schools. So the numbers before the pilot program at School X are compared to the numbers once the program is in effect, in the same School X. If you compare one school to another, then you're comparing apples to oranges. Compare a school to itself, and you're definitely comparing apples to apples.
   

This approach is designed to correct for the problem that socio-economic factors may skew performance measures. However, a problem could occur when the school population changes in the middle of the reform process. In addition to comparing schools to themselves, other objective criteria may need to be added to the mix, in order to control for this issue.

1. If the numbers get worse after several years, the pilot program is decommissioned. If the numbers improve, the program is expanded to cover additional schools. In this way, the most successful programs grow in size while the ones that "look good on paper" but don't really work, are simply phased out.
   

This program is designed to be both popular and effective. It's crowdsource, so everyone who has ideas can participate, and most of them will thus support the program. Only the ideas that people love, and that school districts want to try, will be chosen. The ones that work will grow, and the ones that fail will vanish. Liberal districts can try different ideas than conservatives: nobody will feel that they are having an ideology imposed on them, unless politicians tamper with the program. Districts that don't like this program are under no obligationto participate.

Clarification related to how the crowdsource website would work, and how people could specify information about a proposed plan: I envision a three-stage progression for an idea. The first stage is a simple text description and is completely free-form, to lower the entry barrier so that anyone can participate in the process. The site should allow for loosely-moderated discussions where anyone can comment on a proposal and interested parties can work together.

Ideas that generate a lot of interest and support can progress to the second stage, where there would be a more formal template to fill out that would answer a specific set of questions designed to provide comprehensive information about the proposal. (Because this is crowdsource, there could be multiple stage 2 documents associated with a single stage 1 proposal. A stage 2 proposal can be a collaboration of interested parties, and can draw from one or more stage 1 proposals.) Stage 2 would need to answer every question a school district needs to know in order to express their interest in the proposal. For that reason, school districts would have the ability to ask any questions they want, as part of stage 2.

Stage 3 would be the most formal. In that stage, a comprehensive plan would clearly specify every aspect of the plan including estimated costs, a detailed implementation plan, a teardown plan in the event the program is decomissioned, etc. Stage 3 would contain all the information that the government needs to make a funding decision, and that the local school district needs to run the plan, and information required to determine compliance, etc.