The big news from tonight came before the EAC meeting when BPS sent out an e-mail saying that the deadline for the EAC to choose a plan has been extended through January. BPS has decided that they really need the demand model that MIT is working on in order to evaluate plans. They expect to be able to give the EAC their analysis in mid-January which will give the committee very little time to select a plan to recommend.

As for the meeting, it started with brief presentation from Carleton Jones on how BPS will work with MIT on analysis. They will take best zoned and best non-zoned model and run simulations of the lottery using the MIT demand model. One question that I didn't get a chance to pose to them is how will they know which the best models are before they run the simulations? My understanding is that once the difficult work of creating the demand model is done, running these simulations will be fairly easy. I hope they will run simulations on several models and using several variations, particularly looking at the impact of walk zone preferences (more on that later).

One issue Carleton raised is that they wanted a plan for 2014 in place before parents choose for 2013 so they would be able to choose based on what would happen with school choice grandfathering in the future. Since parents start submitting their school choices in early January, this will not be possible. Carleton suggested that BPS might consider some sibling grandfathering or other measures to make this less of an issue.

The committee then moved to a discussion on voting procedures. There was a fair amount of discussion of whether to allow votes at multiple meetings and how to accommodate members who can't be present for the vote. There was also discussion of what would constitute a majority. Would it be a majority of the entire committee? Of members present? Of members present and voting? While I'm not sure that there was a final decision, it seemed like they generally agreed to the following:

• The vote would take place at a single meeting.
• A quorum consisting of 2/3 of the EAC would need to be present to vote. There are 27 members so 18 would be required.
• A majority of those present and voting would be required for a recommendation to pass.
• Members would vote on all plans under consideration. If a majority was not reached, the plan with the lowest vote total would be dropped and the EAC would vote again on the remaining plans. This would continue until a majority is reached.
• Members who are not present at the vote could submit a ballot ranking the plans to be voted on. This would allow the member to vote in each round by selecting the highest ranked plan still in the running.
• Members not present could also designate another member of the EAC as their proxy. This person could vote on the member's behalf if there were changes made to some of the plans during the meeting that could affect the member's vote.
• Abstentions would not be considered when determining a majority. In other words, if they are voting between two plans and 10 vote for plan A, 8 vote for plan B and 4 abstain, plan A would pass even though it did not have a majority among those present. It would just need a majority of those voting.

While I'm hopeful that the committee will reach consensus or something close to it, I am concerned that these procedures could lead to a plan without broad support being approved if the committee is split at the end. I would like to see the EAC commit to only recommending a plan if a majority of the entire body votes for it. This would require 14 votes regardless of the number of members in attendance. I'm particularly concerned about the idea of not allowing members to explicitly vote against the plans that are under consideration. This could lead to a situation where the EAC recommends a plan that the majority of members present are actually opposed to.

EAC member Miren Uriarte raised concerns along these lines. After the meeting, I suggested to her that the committee allow members to vote for "None of the Above." That would allow members to explicitly vote against all plans under consideration if they found them unacceptable. Unlike abstentions, these votes would be considered when determining whether a majority has been reached. So if 10 vote for plan A, 8 for plan B and 4 for "None of the Above," no plan would be recommended. While this would not be a desirable outcome, it is much better than allowing the possibility of the EAC recommending a plan that most of its members oppose. That would be a disaster politically. Miren felt a "None of the Above" option would address her concerns and said she would raise it to the committee.

Next, the committee spent some time discussing a document that describes the current assignment policy in detail. The committee had seen this document previously, but it had been several months since it had been discussed. The document goes into quite a bit of detail and includes all the exceptions and special considerations that have been added to the system over the years. The committee discussed whether they should make recommendations on removing some of these exceptions. The consensus was that it would depend on what plan was selected.

Helen Dájer brought up the need to consider whether walk zone priority would still make sense under a new plan, particularly if the plan is designed to allow more children to attend schools close to their home. There was not much discussion of this, but there seemed to be general agreement that the impact of walk zone priority should be included in the analysis of the plans. I feel this is very important as walk zone may have a significant negative impact on equitable access to quality schools. Fortunately, I believe that MIT will be able to quantify this affect so that it can be considered.

There was also some discussion of administrative assignments. These assignments happen when a child isn't assigned to any of the schools his or her family listed as choices. This can happen if the family chose only very popular schools, if the child receives a poor lottery number, or if the family applies during one of the later rounds. Hardin Coleman asked what the demographics of kids administratively assigned looked like. I did some quick analysis and it looks like 90.6% of children who were administratively assigned to their current school are black or Hispanic. This compares to 77.0% of the district as a whole. 77.3% qualify for free or reduced price lunch compared to 70.2% of the district as a whole.

The committee then moved on to discussing a draft memo from the chairs. This memo contains information on criteria the committee will use to select a plan, observations and findings, and general recommendations on student assignment and school quality. It also has blank sections for specific recommendations pending selection of a plan. The committee agreed that they need to review this document paragraph by paragraph, but did not have time to do so last night. They scheduled an additional meeting on December 13th (location TBA) to do this. In the meantime, members may submit proposed edits to the document. Also, not discussed at the meeting, but now available on-line is a proposed recommendation for English-language learners.

The committee also decided to dedicate the December 6th meeting to review further analysis that BPS will present at that time. I am not exactly clear on what this will cover, but it seems to be focused on how various features of an assignment plan (walk zone, sibling priority, etc.) could affect the outcome of that plan. They asked Carleton Jones if they needed to dedicate an entire meeting to this discussion and he basically said yes.

The committee ended with a period for public comment. Several people urged the EAC to consider whether walk zone priority should be retained. I talked about my concern with how quality is being measured (see blog post from 11/17). After the meeting, Peter Sloan from BPS told me that they are using a revised measurement that addresses my concerns, but after talking with Peng Shi, I'm not clear whether this is the case. I will look into it further.