So about those student information systems...
Earlier last week I read the article "Student data systems going south." I was struck by the similarity to my own post from a couple months ago (Scheduling (enough said)), in which I actually called for an open-source solution to these student data systems. For any of you fortunate enough to be unfamiliar with these systems, they are large, enterprise database solutions for tracking a plethora of student information. Everything from student schedules to grades, to demographics, to discipline, all stored in a database accessible via the web. The exisisting systems are exceedingly expensive and still largely immature and, instead of making our lives easier as educators, have made our lives astoundlingly more challenging.
Did you know that real information technology planning still has value, even for us educators? As it turns out, I'm not sure that the folks who designed these systems knew that. More importantly, I have serious doubts that the people who signed on the dotted line to purchase and install these systems had the least sense that an information system should meet the requirements of a majority of its users.
When information systems are driven by policy and politics rather than user requirements, failure is virtually guaranteed. Whether in the public or private sector, IT planning must be requirements-driven. However, in the case of the current crop of student information systems, IT planning is being driven by data reporting dictated by state and federal agencies, in particular, No Child Left Behind. While very little thought was given to the actual business rules and workflow in place in our school districts, massive sets of tables designed to provide "rivers of data" (according to the New York Times article on these systems) are creating headaches across the board. While our colleagues in private industry will surely understand the way in which database design is ideally reflective of the way a company does business, politicians, educators, and school administrators generally haven't taken too many data modeling courses in college.
As an example, a number of staff members in our superintendent's office have commented as to how much easier it is to assemble data for state and federal reports now that we have rolled out our particular flavor of student information system. However, my district (currently on the edge of significant teacher layoffs and substantial budget cuts) is paying a consulting group to create a student transcript, since these are no longer natively available as they were with our old system. Although integration of longitudinal data and built-in transcripts are supposed to be available in future versions of the software, graduating seniors, whose college matriculation is dependent on complete transcripts, tend to be less than sympathetic.
States are currently spending hundreds of millions of taxpayer dollars on these systems. Yet our district (and many others like ours) can't afford the full-time database administrators or the server farms it takes to run these systems effectively. My Introduction to Computing students never understand why I make them memorize and simulate the Systems Development Lifecycle until they are blue in the face. Then I start talking about our student management system and the countless ways it fails to meet the requirements of 95% of its users. Suddenly, it becomes much clearer. What was that IT commandment that has been floating about on ZDNet? Put thy users first above all else. More politicians should read ZDNet.
Anyone with some helpful tips on getting these systems to suit your needs on a shoestring, please talk back below - I'm sure there are lots of us who could use some help.