Data-driven instruction for improving student performance

¶ … Driven Instruction

Definition and Meaning

Data-driven instruction is a type of instruction where the teacher uses student performance as the benchmark for planning their instruction. Student performance in this case is defined in terms of their result or scores in various assessments. The teacher uses these results to identify those areas that students were able to grasp effectively and those in which they are still encountering difficulty; and then structures their subsequent instruction in such a way that it focuses more on the latter (Thompson, 2008). Data-driven instruction, therefore, basically has three fundamental elements -- baseline data (where the students are currently), clear goals (where they ought to be in a pre-determined period), and regular assessments (a mode of benchmarking their performance in relation to the set goals over time) (Thompson, 2008). Research has shown data-based instruction to be an effective way of improving student performance (Thompson 2008).

What I personally learnt from the Use of data-supported instruction

I used data-supported instruction multiple times during my practicum and, well, I learnt a variety of things from my use of the same. The most important of these perhaps is how to be effective in the use of data-driven techniques. The process of data-driven instruction is a three-stage process that involves targeting, focusing stage, and finally, planning. In the targeting stage, the teacher administers an assessment on a particular content with the aim of identifying i) the strengths -- areas in which the class performed generally well; ii) the challenges -- areas where the overall performance of the class in relation to the tested content is fair; and iii) critical needs -- areas in which the class put up a poor performance, say a majority of the students failed the tested content (Thompson, 2008). Once this data has been obtained, there is need then to inject meaning into it. This takes place in the focus/analysis stage, where the teacher subjects the 'critical needs' identified earlier to intense scrutiny with the aim of identifying the specific concept(s) that students did not grasp. For instance, the target phase may reveal that students are weak in relation to the use of adjectives. This identifies 'the use of adjectives' as a critical need, but the teacher then needs to conduct an analysis to determine exactly where the problem lies -- are students having problems identifying adjectives, or using the same to improve their own writing?

The results of this evaluation then serve as the basis for the final (planning) stage, where the teacher acts on the information obtained from the analysis by structuring their lesson or instruction plans to address the specific need that has been identified (Thompson, 2008). How effective one is in using data-based instruction to improve student performance will depend, partly, on how effective they are in identifying critical needs, analyzing them, and structuring their programs to internalize them.

I also learnt that besides these, the success of data-driven instruction also depends on the school's organizational culture, which is manifested in the code of conduct and student aspirations. Data-driven instruction would only be effective if the school culture promotes positive learning behaviors and encourages the use of DDI for school-wide improvement (Fenton & Murphy, 2015).

Should all instruction be data-supported?

Well, despite the inherent benefits of data-supported instruction in learner outcomes, I do not believe that it would be a good idea to have all instruction as data-supported; if anything, teachers should not rely entirely on DDI in the teaching process. This is particularly because as the number of assessment tests increases, students are likely to get sick of it, and begin seeing these tests as a joke, at which point they (the tests) may no longer be able to realize their intended objectives. Moreover, the continuous use of tests to assess student performance could culminate in the tendency of teachers 'teaching to the test' and preparing students in a manner that increases their chances of passing tests as opposed to one that expands their reasoning capacity and enhances their ability to think critically. Finally, the large number of tests used in data-supported instruction increases teachers' workload, and this could have a negative effect on their productivity in class. Some schools require teachers to run as many as four pre-tests in a week; for a teacher on multiple subjects, this could be cumbersome and could be detrimental to their ability to give their best in class. For these reasons, the use of data-supported instructions should not just be limited; it ought not to be taken as an everyday practice.

Difference between data-supported instruction and evidence-based methods of instruction; are they of equal importance?

Evidence-based instruction is a type of instruction where a teacher chooses his/her instructional program, strategies or techniques based on their proven success in past research studies. In other words, the teacher's choice of teaching strategy or technique is driven primarily by the fact that there is valid, trustworthy and reliable "evidence to suggest that when the program is used with a particular group of children, the children can be expected to make adequate gains in reading achievement" (The International Reading Association, 2002, n.pag). Evidence-based instruction can, therefore, be used to supplement data-supported instruction -- once the teacher has identified the specific areas in which his/her students are still encountering difficulty, he/she is supposed to re-teach the same using a more appropriate instructional method or strategy. Evidence-based techniques can be crucial in helping them identify the methods that would be most appropriate to that particular group of students. Towards this end, whereas evidence-based instruction relies on secondary data that has been validated through research, data-driven instruction relies on primary (first-hand) data that has not undergone any kind of validation. Although both techniques are crucial to the teaching and the learning process, I would not rank them as equal because whereas evidence-based instruction is recommended for use in all instruction, data-supported instruction is only effective if used to a very minimal extent.