Someone pointed me to a letter from State Superintendent Jack O'Connell to school districts regarding problems with the CalPADS data system. CalPADS is the system that has been in progress since 2002. The state has spent millions of dollars on the project. They just rolled it out this year. Now, they're admitting that things aren't working quite as smoothly as they'd hoped. The State Superintendent wrote:

Due to unacceptable system performance issues that occurred during the rollout of CALPADS, I have directed IBM and CDE staff to focus all resources over the next two months on stabilizing the system.

The first step was to halt any additional changes to CALPADS while the system is thoroughly reviewed, top to bottom, to correct deficiencies and ensure that the system performs efficiently. Between now and late March, the CALPADS project team is performing a comprehensive system review and testing process on all CALPADS functions and components. This process will culminate in the release of a properly functioning system targeted for release on March 29, 2010. In the meantime, all previously scheduled CALPADS submittals from local educational agencies (LEAs) have been put on hold. Obviously, that will change our data collection processes in the near term.

The deadline for submitting Fall 1 data (including enrollment counts, graduate counts, and dropout counts) has been extended indefinitely and the Fall 1 information will be the only data required to be submitted to CALPADS in the 2009–10 school year. Other data submissions (Fall 2, Spring 1, and End-of-Year) will be accomplished through traditional means, the specifics of which will be communicated to your staff in a separate letter.

So, not only are they delaying implementation until at least the end of March and doing a top-to-bottom review to figure out why the system doesn't work as advertised, but they're also abandoning the system for any other data uploads this year. Districts will be using the same old systems to upload all the other data for the year.

I think this is a good example of what happens when you design a data system by committee. Feature creep kills the developers and it doesn't end up doing its most basic function very well either. Hopefully, they'll work out the issues eventually. I think it is really important for California to have a good data system for student and teacher data. If we actually had access to this data, there are lots of incredible research studies that could be done to find out what works and what doesn't work in public education.