May 2003 — Special Feature
Print this article | Email this articleClick here to receive your FREE subscription to T.H.E. Journal
Data-Driven Decision-Making
The importance of this phenomenon cannot be overemphasized: Only when data becomes genuinely useful and commonplace in the classroom will teachers and administrators welcome it. And only when it is useful will data quality improve. This lack of 'clean' data - i.e., data that is timely and accurate - is the bane of researchers and analysts. Up until this point, no one cared too deeply if a student name on a school record didn't exactly match the student name on a test record, because no one was really doing anything with the test data.
However, if we really want to follow how a particular child performs on tests over time, we need all of his or her data to align. The benefits to districts are palpable as well. For instance, districts in Texas improved their graduation rates dramatically when students who were listed as dropouts were matched with the other names they were enrolled under and taken off the dropout list.
It also bears mentioning that the disuse of data for decision-making is not unique to schools. Only recently have private for-profit firms begun to appreciate the power and utility of the large data sets that they generate. For example, Wal-Mart shouldered aside Sears and Montgomery Ward in large measure because it is one of the nation's premiere users of on-time, online data - leveraging it to provide precise and infinitely detailed inventory control, shelf-stocking and ordering, as well as just-on-time delivery. Among other advantages, this permits the generation of substantial economies that are passed onto customers in the form of lower prices.
In the mysterious realm of data-driven decision-making, the following three dimensions are essential in order to make it work:
- Data warehousing with decision-support tools. By this I mean a strategic, relational database that can be queried to answer any question for which there is a quantitative answer and for which digital data is available. How many kids are taking Advanced Placement courses in terms of race, sex and ethnicity? How many are taking algebra? Where are absences concentrated? Is there a pattern to 'hard' course taking?
- Standards-based, curricular alignment. This is essential to successful instruction and assessment. What is on the agenda to teach, what is actually taught, what is actually learned? These are the questions that must be answered to fashion instructional programs that work.
- Community engagement. Getting everyone involved in the enterprise as a virtual partner means management of a dynamic Web portal by the institution for its entire school community. E-mail for all is a good place to start, as are activity calendars and posting school report cards. Indeed, the demographic and test-score data contained in the data warehouse should be easily accessible to all Web site visitors, so long as student privacy rights are respected. And the sky is the limit, because in today's busy world a virtual school is just as important as a real school.
Data for Entrepreneurs
The cement that holds this together is entrepreneurship, a much overworked and misunderstood term. The great French economist Georges Says once said that an entrepreneur is a person who develops a product or a process that no one else knows they need. I believe that we have entrepreneurial-minded educators nationwide who lack the essential data they need to dream up new products and processes that will help all our students realize their truest potentials.
