You are here

Effective Data Mining Protocols to Enhance Student’s Learning

My presentation will start from addressing the needs and importance of data mining and data-driven decision-making (DDDM) in education.  Next, I will present current data systems available for education use in United States (U.S.) and the possible development of further data systems.  After that, two types of data mining protocols will be discussed and compared.  Lastly, I will synthesize what a local classroom teacher finds as useful data, what they use with this data, and how it affects student’s academic and behavioral improvement.

Abstract
Data-driven decision-making (DDDM), a systematic process to gather, store, analyze, and apply data to make decisions to increase performance, has been proved to be successful in business, yet not in education. DDDM, as it applies to education, still reminds to be a challenge due to the confusing to read and not easily accessible data for teachers to use in their daily instruction.  Despite all the challenges, educators believe that data has the capacity to improve student learning. Thus, it is essential to know how to effectively dig the data deeper to enhance teacher’s teaching.

In US, current educational data systems include Aeries, Destiny, OARS, Cruncher, etc.  The data collected from these systems range from grades to suspensions and state mandated testing scores. The quantitative data has very limited instructional use because these numerical data only tells how well student did on a test, rather than what and why students did well on.

The purpose of this presentation is to provide effective data mining protocols to enhance teacher’s teaching. To effectively apply DDDM, a review of literature suggests two protocols. One is Datnow and Hubbard’s  (2015) 11-step protocol for teachers to use that involves obtaining performance feedback, diagnosing the cause for underperformance, and drawing up plans that specify how each student currently performs. Another practical protocol proposed by Kennedy, Mimmack, and Flannery (2012) is for staff to complete a data gallery walk. In the gallery walk, staff is assigned into small groups and each group is given multi-year data on a particular topic.
If schools are to utilize DDDM effectively, they must have easy to read and up-to-date data that is readily accessible. Then, they must assign time to analyze the data, to set goals, to plan strategies, to implement the plan, and to reevaluate the goal and plan.