Sunday, May 31, 2015

Student Articulation Lessons Learned

We are halfway through our student articulation meetings using the new articulator sheet that uses imported data from the STAR RTI screener. Some elements are going well. Being able to change the data visualizations using drop down menus has been amazing. The graphs and charts are hugely useful. The new system is less subjective, more visual, and has cut our articulation meetings down by a third. There are also several things that need to be made better for next year. This post will serve as a reflection that will hopefully inform (remind) my 2016 revision.

Behavior is a subjective measure, but one that is crucial to consider for creating balanced classes. The system currently uses one metric for behavior. As we discuss students it is clear that this isn’t dimensional enough. Next year the system will include a metric for Social Skills and another for Attention to Task. These are both under the umbrella of behavior but mean something different for the chemistry of a classroom.

Link ALL Data to Student ID
Student information (Name, Birthday, gender, ethnicity, ect.) is currently pulled from the STAR reading data. The thinking behind this was ease of use: only two data imports were necessary (math and reading) to articulate the whole school and vertical lookups handle data using the student ID number after that. My method is flawed. Teachers use the “Data Beta” sheet to perform the initial placement and to add subjective scores like behavior and notes. The teacher added information is purely positional to the STAR data. In other words, student X is on row 4, and the teacher added data is also on row 4. The teacher added information will no longer match up to the right students if anything happens to the sort or row placement on the reading sheet. In the next iteration I need to have teachers add their information directly to the STAR reading sheet or a third data sheet that is only student demographics. The teacher added information needs to be anchored to the student ID number. This was lazy thinking on my part. It hasn’t been an issue (except for the next paragraph) because I’ve been careful, but it makes me nervous and is just sloppy.

Cross-Browser Inconsistency
I've discovered that the sheet sorts differently when teachers use a different browser. This was quite a surprise. The all of the data is brought to the “Data Beta” sheet via Queries. I have to test this more and haven’t realized the exact reason, but it is clear that the students sort differently when the sheet is accessed with Firefox or Safari. This is an issue because of the previous paragraph. I've told my teachers to only use Chrome when working with the articulation sheet which avoids any trouble. If I effectively anchor everything using the Student ID this won’t be an issue at all.

The process has been going really well and the sheet has been working. A few things have made the process clunkier than I’d like. Ironically, the biggest issue (not hard linking the teacher added information to Student IDs) was intended to make the setup easier. It was a “penny-wise and pound-foolish” decision. Even so, placement is going better than ever and the issues I've discussed here are fixable. The next iteration will be even better.