TREND Fall 2018 - Page 14

PORTFOLIOS State accountability measures continue to create headaches for teachers as Pre-K and Kindergarten Portfolio scoring experienced technical issues resulting in undeserved 1s for many teachers across the state. As soon as we became aware of the issue, we began working with the TN Department of Education to develop a plan to address teachers’ concer ns with not only the immediate effects of the mis-scores, but the potential for future negative consequences. If you haven’t already heard, there was a major problem with the scoring of Pre-K and Kindergarten portfolios for the 2017-2018 school year. Teachers receive an overall portfolio growth score based on their scores on four separate collections, which look at students’ growth over the course of the school year on specific standards. As soon as the scores were released over the summer, teachers noticed that they received growth scores of 1 for some collections, without any notation as to why from the peer reviewers, despite having submitted the appropriate content and student work examples. In some cases, despite having double-checked their submissions multiple times before submission, there appeared to be files they had uploaded that were missing. For those who received 4s and 5s in every other area, even a single 1 could negatively impact their final score. Not only could this have immediate affects (just look at the teachers in Greene County who are being told they won’t receive score-related bonuses this year due to TNReady failures), but these scores are now on their records. Many worried that potential future employers would see the poor scores and naturally interpret them as a sign of poor growth as a result of ineffective teaching. As the rumor mill began churning, reports of “glitches” and problems with the Educopia platform rolled in. We put in a call to the Department, and their response was a firm denial: “There was no error by our vendor, and there was no computer glitch.” As it turned out, this was technically true. Most of the outlier scores of 1 were the result of what have been dubbed “Scenario 2” submisison errors - most commonly, either student work or the standard selected between point A and point B were mismatched. In those cases, the issue was flagged by a peer reviewer and the teacher received a score of a 1 on that specific collection. However, in most cases there were no comments from the reviewer describing exactly what was mismatched, and many teachers were left unable to determine where the supposed error actually occurred. The first resolution offered by the state was to allow teachers to vacate these scores in their overall TVAAS. This was already a hold harmless year for portfolios, and Public Chapter 751 was passed to protect teachers from employment and compensation decisions based soley on portfolios for the 2018-2019 school year. However, vacating a score would mean substituting their school’s score, which had not yet been released, and many were unsure if that would pan out as an overall better score. Besides, that wasn’t the only concern, of course. All semantics aside, scores needed to be accurate and teachers wanted answers. Unfortunately, some teachers began to be told that they should “file a grievance” as a means of resolution. There are multiple problems with this direction. Such a recommendation would not really resolve the issue, and likely lead to frivolous litigation. There is no state grievance that can be filed and every local system has a different grievance procedure. The only result of a grievance, if approved by the district, would be the aforemnetioned vacating and substitution. Our objective, as with every problem that surfaces for our members, is to identify the issue and find an attainable solution for those problems. In this case, they wanted answers as to why they received the low score - what exactly had they done wrong? If they had really made a mistake, how could they avoid making the same mistake again in the future? What if there wasn’t really a mistake, and the peer reviewer was wrong? How would they know, when there was no